Abstract

   Drought is a major abiotic stress that limits oat growth. This study
   investigated the phenotypic, physiological, and proteomic differences
   between drought-resistant (Grain King [G]) and drought-susceptible
   (XiYue [X]) oat varieties under drought stress (soil water content of
   15% ± 5% of field water-holding capacity) and normal conditions (soil
   water content of 75% ± 5% of field water-holding capacity). Phenotypic
   analysis showed that plant height, aboveground biomass, and underground
   biomass decreased under drought stress in both varieties, with variety
   X exhibiting a greater reduction. Physiological analysis revealed
   increased malondialdehyde content, soluble sugar (SS) content, and
   superoxide dismutase (SOD) and peroxidase (POD) activities in both
   varieties under drought stress, though variety X showed smaller
   increases. Proteomic analysis identified 151 differentially accumulated
   proteins (DAPs) in variety G and 792 in variety X. Further analyses
   showed that the DAPs in variety G, which were highly correlated with
   POD and SOD activities and SS content, were primarily involved in
   energy metabolism, protein translation, RNA processing, amino acid
   metabolism, and protein folding. Conversely, in variety X, the DAPs
   were mainly associated with RNA processing, protein stabilization,
   plant photosynthesis, intracellular signal transduction, and protein
   folding. Further analysis suggested that variety G significantly
   upregulated proteases related to photosynthesis, catalysts involved in
   citrulline synthesis, temperature-induced lipid transport proteins,
   fibrillin proteins linked to stress tolerance signal transduction and
   response, and shearing factors involved in mRNA shearing—proteins that
   were not significantly upregulated in variety X. These proteins may
   play essential roles in protecting oats from drought stress. Overall,
   this research elucidates the drought resistance mechanisms of different
   oat varieties at the protein level.

Introduction

   Drought is a prominent threat to agriculture worldwide [[32]1]. When
   crops experience prolonged water deficits, they evolve complex
   mechanisms to sense water availability in their environment and
   activate stress responses at the morphological, physiological,
   biochemical, and molecular levels, reprogramming their metabolism and
   growth to establish new homeostasis [[33]2–[34]4]. Morphologically,
   most plants respond to drought stress through adaptive changes in root,
   stem, and leaf morphology; leaf orientation; stomatal closure; root
   system deepening; reduced branch growth; and leaf senescence and
   dormancy [[35]3–[36]6]. Physiologically and biochemically, drought
   stress induces the production and accumulation of organic solutes,
   including proline, betaine, sugar, and sugar alcohols, which prevent
   membrane disintegration and enzyme inactivation [[37]7–[38]9]. At the
   molecular level, drought stress activates multiple drought-related
   functional genes (e.g., GhNAC4, ZmGA20ox3, and TaWRKY31) and induces
   the synthesis and expression of stress effectors and regulatory
   proteins involved in transcription, reactive oxygen species scavenging,
   antioxidant regulation, and signal transduction pathway activation
   [[39]10–[40]14]. These mechanisms enable plants to enhance their
   ability to avoid damage (stress avoidance mechanisms) or maintain
   metabolic functions under drought conditions (stress tolerance
   mechanisms) [[41]15].

   Oat (Avena sativa L.) is one of the six most grown cereal crops
   worldwide, with over 10 million hectares under cultivation; it is
   primarily used as food, feed, and forage [[42]15,[43]16]. Oat contains
   highly valuable compounds for industrial applications, including
   glucan, oil, and protein [[44]17,[45]18]. In China, policies such as
   “grain to feed” and the rapid development of animal husbandry have
   significantly increased the demand for high-quality forage oats,
   leading to expanded planting areas. However, drought has severely
   impacted oat productivity, and it has been shown that oats are
   predominantly grown in drought-prone or low-nutritional areas that are
   unsuitable for major food crop production in most countries [[46]19].
   Therefore, it is crucial to conduct research on drought-related genes
   and proteins in oats to analyze their drought resistance mechanisms at
   the molecular level, better guide the selection and breeding of
   drought-resistant varieties, and improve their production.

   Proteomics involves the study of functions, recognition, and regulation
   of proteins in cells, subcellular compartments, and tissues. Numerous
   proteomic studies have elucidated the effects of drought stress on
   plant growth and development, as well as the protein expression
   strategies employed by various plants in response to drought. For
   instance, drought stress decreases rubisco-binding protein levels and
   inhibits photosynthesis in alfalfa leaves [[47]20]. In maize, drought
   upregulates protective and stress-related proteins, primarily chaperone
   proteins and dehydrins, and causes one-to-one changes in antioxidant
   enzyme activities and detoxification protein levels [[48]21].
   Similarly, drought stress reduces photosynthesis, carbohydrate levels,
   protein levels, and defense- and energy metabolism-related protein
   levels in a drought-tolerant safflower variety, with these parameters
   recovering upon rehydration; however, no significant differences are
   observed in a drought-sensitive variety under drought and rehydration
   conditions [[49]22]. In soybean roots, proteins involved in
   osmoregulation, defense signaling, and programmed cell death, expressed
   in response to short-term drought stress, are critical for drought
   adaptation [[50]23]. Additionally, under mild drought stress, osmolyte
   accumulation enhances resistance, whereas under moderate and severe
   drought stress, unsaturated fatty acid oxidation and glucose and
   galactose accumulation increase, accompanied by suppression of
   β-coumarin synthesis and the terpene precursor 2,3-epoxide. Triterpene
   (glycyrrhetinic acid) accumulation is also affected [[51]24]. Thus,
   proteomic studies are essential for understanding the complex
   regulatory mechanisms underlying plant drought resistance. In this
   study, tandem mass tag (TMT) chemical marker quantitative proteomics
   technology was used to characterize and quantify proteins in the
   seedling leaves of various drought-resistant forage oat varieties under
   drought stress and normal water supply conditions. Differential protein
   accumulation profiles were screened through comparative analysis to
   identify key proteins involved in drought stress stimulation and
   regulatory pathways in forage oats. The findings provide a foundation
   for further research on the mechanisms of oat response to drought
   stress.

Materials and methods

Plant materials

   The materials used in this study were commercial varieties of forage
   oats Avena sativa cv. ‘Grain King’ [G] (Baili International Grass,
   Beijing Co., Ltd.) and A. sativa cv. ‘XiYue’ [X] (Ningxia Daxinong Seed
   Co.). A previous study determined the phenotypic traits of seven oat
   varieties under drought stress and normal water supply conditions and
   screened varieties with different drought tolerance levels; among them,
   varieties G and X were identified as drought-resistant and
   drought-susceptible, respectively [[52]25].

Stress treatment

   Two treatments, drought stress and normal water supply (control), were
   set up for each variety, with three replicates planned for each
   treatment. Normal irrigation was implemented from sowing to 15 days
   after seedling emergence, maintaining the soil water content at 75% ±
   5% of the field water-holding capacity. Drought stress was applied 15
   days after seedling emergence, with the soil water content reduced to
   15% ± 5% of the field water-holding capacity. This soil moisture level
   was sustained for 3 days, after which the relevant indicators were
   measured. For the control conditions, the soil water content was
   maintained at 75% ± 5% of the field water-holding capacity until
   measurements. The drought and control treatments for variety G are
   indicated as GD and GW, whereas those for variety X are indicated as XD
   and XW, respectively.

Plant growth conditions

   The drought experiment was conducted in a plant growth room at Qinghai
   University under controlled environmental conditions: 20/25°C
   (night/day), 50%–60% relative humidity, 500 μmol m⁻^2·s⁻^1 of light
   intensity, and a photoperiod of 8 h/16 h (night/day). For each variety,
   240 seeds of uniform size and fullness were selected. The seeds were
   sterilized using 5% Na[2]ClO[3], rinsed three times with distilled
   water, dried, and sown on non-porous pots (24.5 × 35 × 30 cm) filled
   with soil. The soil’s basic physicochemical properties are listed in
   [53]Table 1. Fifteen days after emergence, seedlings were thinned out,
   leaving 20 healthy plants per pot for the experiments.

Table 1. Basic physical and chemical properties of the soils.

   Organic matter (g/kg) Total nitrogen (mg/kg) Total phosphorus (%) Total
   potassium (%) Hydrolytic
   nitrogen
   (mg/kg) Fast-acting potassium (mg/kg) Effective phosphorus (mg/kg)
   Water-soluble salt
   Total amount (g/kg) pH
   89.33 3.42 × 10^3 0.17 3.00 241.08 1147.00 273.74 4.78 7.75
   [54]Open in a new tab

Morphological and physiological index determination methods

   Five seedlings were randomly selected from each pot for phenotypic
   indicators. The roots were cleaned, and the plant height was measured
   using a scale. The plants’ aboveground and underground parts were then
   separated, dried at 80°C ± 2°C until constant weight, and measured.

   The malondialdehyde (MDA) content was determined based on a
   colorimetric technique using thiobarbituric acid, as described by Na
   Zhang [[55]26]. Additionally, the soluble sugar (SS) content was
   determined by referring to the anthrone method described by Jun Wang
   [[56]27]. Furthermore, the total superoxide dismutase (SOD) and
   peroxidase (POD) activities were determined using the SOD (A001–3–2)
   and POD (A084–3-1) test kits (Nanjing Jiancheng Institute of Biological
   Engineering, Nanjing, China) and water-soluble tetrazolium-1 (WST-1)
   dye with colorimetric methods.

Statistics and analysis of phenotypic and physiological data

   The data were initially processed using Microsoft Excel 2016 and
   analyzed using SPSS19.0 statistical software for ANOVA (p < 0.05). The
   data were cartographed using OriginPro 2021(v 9.8.0.200) and R28 (v
   3.4.3) statistical software.

Proteomic analysis

Sample preparation.

   Sample leaves (≥100mg) were first ground to a powder with a small
   amount of liquid nitrogen, added to 200 µL of lysis buffer (150 mM
   tris(hydroxymethyl)aminomethane hydrochloride [Tris–HCl], 4% sodium
   dodecyl sulfate, 100 mM dithiothreitol [DTT], pH 7.8), sonicated,
   boiled for 5 min, and precipitated with a trichloroacetic acid-acetone
   solution. After centrifugation at 16,000 ×g (1 min, 4°C), the tubes
   were washed twice with cold acetone and air dried. Each sample tube was
   complemented with 150 µL of lysis buffer, sonicated, and centrifuged at
   16,000 ×g for 15 min to remove undissolved cell debris. The
   supernatants were collected for quantification using the bicinchoninic
   acid protein kit (Bio-Rad, USA).

Protein digestion.

   Protein digestion was performed according to the method described by
   Wisniewski et al. [[57]28]. DTT, detergents, and other
   low-molecular-weight fractions were removed by centrifugal
   ultrafiltration (Microcon units, 30 kD) using 200 μL of uric acid (UA)
   buffer (150 mM Tris–HCl, 8 M urea, pH 8.0). Next, 100 μL of 0.05 M
   iodoacetamide was added to the UA buffer to block the reduced cysteine
   residues, and samples were incubated in the dark for 20 min. The
   filters were washed three times with 100 μL of UA buffer and twice with
   100 μL of 25 mM NH[4]HCO[3]. Finally, the protein suspension was
   digested using 4 μg trypsin in 40 μL of 25 mM NH[4]HCO[3] overnight at
   37°C, and the filtrate was obtained. The peptide concentration was
   determined using the nanodrop OD[280] method.

TMT labeling of peptides and fractionation.

   Peptides were labeled with the TMT reagent according to the
   manufacturer’s instructions (Thermo Fisher Scientific). Each sample
   (100 μg peptide equivalent) was reacted with one tube of TMT reagent.
   After dissolving the sample in 100 μL of 0.05 M triethylammonium
   bicarbonate solution at pH 8.5, the TMT reagent was dissolved in 41 μL
   of anhydrous acetonitrile. The reaction mixture was incubated for 1 h
   at room temperature, 8 μL of 5% hydroxylamine was added, and the sample
   was incubated for 15 min to quench the reaction. The 12
   Multiplex-labeled samples were pooled together and lyophilized. The
   TMT-labeled peptide mixture was separated on a Waters XBridge BEH130
   column (C18, 2.1 × 150 mm, 3.5 μm) on an Agilent 1290 high-performance
   liquid chromatograph operating at 0.3 mL/min. Buffer A included 10 mM
   ammonium formate, and buffer B comprised 10 mM ammonium formate and 90%
   acetonitrile. Both buffers were adjusted to pH 10 with ammonium
   hydroxide. Thirty fractions were collected for each peptide mixture and
   linked in series to 15 fractions. The fractions were dried and
   subjected to liquid chromatography-mass spectrometry (LC-MS) analysis.

LC-MS analysis.

   For each supernatant, 1 μg of each fraction was analyzed using a
   nanoliter flow rate Easy nLC1200 chromatography system (Thermo
   Scientific). The buffers used in the analysis were solution A (0.1%
   formic acid aqueous solution) and solution B (0.1% formic acid,
   acetonitrile and water mixture [with 95% acetonitrile]). The column was
   first equilibrated using 100% of liquid A. The sample was loaded onto a
   Trap Column (100 µm × 20 mm, C18, 5 µm, Dr. Maisch GmbH) and passed
   through the analytical column (75 µm × 150 mm, C18, 3 µm, Dr. Maisch
   GmbH) for gradient separation at a flow rate of 300 nL/min. The liquid
   phase separation gradient settings were as follows: 0–2 min, liquid B
   linear gradient from 2%–8%; 2–71 min, liquid B linear gradient from
   8%–28%; 71–79 min, liquid B linear gradient from 28%–40%; 79–81 min,
   liquid B linear gradient from 40%–100%; 81–90 min, liquid B at 100%.
   The peptides were analyzed via data-dependent acquisition-mass
   spectrometry using a Q-Exactive HF-X mass spectrometer (Thermo
   Scientific) after separation. The analysis time was 90 min, the
   detection mode was positive-ion, the parent ion scan range was 400–1800
   m/z, and the primary mass spectrometry resolution settings were as
   follows: 60,000 @m/z 200; automatic gain control (AGC) target, 3e6; and
   primary maximum injection time, 50 ms. For the peptide segments’
   secondary mass spectrometry, 20 secondary mass spectra of the highest
   intensity parent ions were triggered after each full scan with the
   following secondary mass resolution: 45,000 @ m/z 200; AGC target, 1e5;
   secondary maximum injection time, 50 ms; MS2 activation type,
   high-energy collisional dissociation; isolation window, 1.2 m/z; and
   normalized collision energy, 32.

Database query and analysis.

   The obtained LC-MS/MS raw files were imported into the search engine in
   Thermo Scientific Proteome Discoverer software (v2.4, Thermo
   Scientific) for database search. Because no oat-specific database is
   available, we used the Uniprot-Pooideae database
   [147368]_924681_20211014.fasta (download date: October 14, 2023;
   available on [58]https://www.uniprot.org/taxonomy/147368) Protein Data
   Bank with the 924681 protein sequence. An initial search was set in a
   precursor to the 6 ppm mass window. The study followed the rules of
   enzymatic cleavage of trypsin/P, allowing a fragment ion mass tolerance
   value of 20 ppm and two maximum missing cleavage sites with the
   following modification sets: fixed modifications, carbamoylmethyl (C),
   TMT16plex (K), TMT16plex (N term); variable modifications, oxidation
   (M), acetyl (Protein N term). Peptides require at least six amino acids
   and one unique peptide per protein. For peptide and protein
   identification, the false discovery rate (FDR) was set to 1%. The mass
   spectrometry data for all proteins can be obtained from the iProX
   database website (IPX0010483000/PXD058831). The TMT reporter ion
   intensity was used for quantification, and the relative quantification
   of the sample proteins was performed using the MaxQuant algorithm
   ([59]http://www.maxquant.org) [[60]29].

Bioinformatics analysis

   Bioinformatics data were analyzed using Perseus44 (v1.6.5.0) software,
   Microsoft Excel 2016, and R28 (v 3.4.3) statistical software. DAPs were
   screened with the cutoff of a ratio fold-change of > 1.20 or < 0.83 and
   p < 0.05. Additionally, proteins with significant variance features
   were identified using hierarchical clustering (Euclidean distance) and
   analysis of variance (ANOVA). For the enrichment analysis of the
   specified protein functions, annotation information was extracted from
   the UniProtKB/Swiss-Prot, Kyoto Encyclopedia of Genes and Genome
   (KEGG), and Gene Ontology (GO) databases. GO and KEGG enrichment
   analyses were conducted using Fisher’s exact test, and FDR correction
   for multiple testing was also performed. GO terms were grouped into
   three categories: biological process (BP), molecular function (MF), and
   cellular component (CC). Correlations between sample features and
   protein expression changes were analyzed using weighted gene
   co-expression network analysis (WGCNA), and the expression data were
   clustered hierarchically based on the protein level.

Results

Effects of drought stress on oat growth

   Different drought-tolerant oat varieties show varied growth responses
   to drought stress ([61]S1 Table). Under drought stress, GD and XD
   displayed different degrees of reduction in plant height. No
   significant difference was observed between GD and GW, while the plant
   height of XD was significantly reduced (by 7.9%) compared to XW (p <
   0.05) ([62]Fig 1A). Additionally, the aboveground biomass of GD was
   9.5% lower than that of GW, while the aboveground biomass of XD was
   14.6% lower than that of XW, with no significant differences between
   treatments within varieties (p > 0.05) ([63]Fig 1B). Furthermore, the
   underground biomass of GD was 8.0% lower than that of GW, whereas the
   underground biomass of XD was 6.9% lower than that of XW, with no
   significant differences between treatments within varieties (p > 0.05)
   ([64]Fig 1C). Variety G exhibited greater drought resistance than
   variety X.

Fig 1. Changes in the growth of two oat varieties under drought stress and
normal water supply treatments.

   [65]Fig 1
   [66]Open in a new tab

   A Plant height. B Aboveground biomass. C Underground biomass. Different
   lowercase letters above the columns indicate significant differences (p
   < 0.05) between drought stress and normal water supply treatments based
   on ANOVA.

   G is forage oats Avena sativa cv,‘Grain King’, X is forage oats Avena
   sativa cv, ‘XiYue’.

Effects of drought stress on the physiological indicators of oats

   Several physiological indicators were measured after drought stress to
   further clarify the drought resistance of the two oat varieties ([67]S2
   Table). Under drought stress, the MDA content of GD was significantly
   higher (48.6%) than that of GW, while the MDA content of XD was
   significantly higher (28.9%) than that of XW (p < 0.05) ([68]Fig 2A).
   The SS content of GD was significantly higher (by 68.5%) than that of
   GW, while it was significantly higher in XD (by 37.2%) than in XW (p <
   0.05) ([69]Fig 2B). The SOD activity of GD was significantly higher (by
   81.3%) than that of GW, and that of XD was significantly higher (by
   14.6%) than that of XW (p < 0.05) ([70]Fig 2C). The POD activity of GD
   was significantly higher (by 101.7%) than that of GW, while it was
   significantly higher in XD (by 82.5%) than in XW (p < 0.05) ([71]Fig
   2D). This analysis suggests that drought stress affects the
   physiological indicators of oats, as revealed by the significant
   increase in osmoregulatory substances and antioxidant enzyme activities
   under drought stress. However, significant differences were observed
   between varieties under different treatments: variety G displayed a
   higher change margin in MDA, SS, SOD, and POD parameters and higher
   drought tolerance than variety X under drought stress conditions.

Fig 2. Changes in the physiological indexes of two oat varieties under
drought stress and normal water supply treatments.

   [72]Fig 2
   [73]Open in a new tab

   A MDA content. B SS content. C SOD activity. D POD activity. Different
   lowercase letters above the columns indicate significant differences (p
   < 0.05) between drought stress and normal water supply treatments based
   on ANOVA. G is forage oats Avena sativa cv,‘Grain King’, X is forage
   oats Avena sativa cv, ‘XiYue’.

Proteomics information identification

   Seedling samples of the two oat varieties under different drought
   stress treatments were used, and the peptide spectral matching, unique
   peptide, protein groups, and quantified protein were determined as
   17,690, 4,946, 1,821, and 1,817, respectively. Sample expression
   analysis and principal component analysis revealed that the 12 samples
   were classified into four main groups, demonstrating that the samples’
   protein expression levels were biologically reproducible ([74]Fig 3).

Fig 3. Normalized sample plots for each treatment for data quality control.

   [75]Fig 3
   [76]Open in a new tab

   A Normalized density line box plot. B Normalized principal component
   plot.

DAP analysis

   The quantitative analysis of proteins from the two drought-resistant
   varieties under different treatments revealed 846 DAPs (p < 0.05;
   cutoff at ratio fold-change > 1.2 or < 1/1.2), among which 724 were
   upregulated and 122 downregulated. GD had 140 upregulated and 11
   downregulated DAPs compared to GW, while XD had 666 upregulated and 126
   downregulated DAPs compared to XW. These results indicate remarkable
   differences in protein expression between the two oat varieties under
   drought stress conditions ([77]Fig 4A, [78]B).

Fig 4. Volcano map of DAPs.

   [79]Fig 4
   [80]Open in a new tab

   A DAPs between GD and GW. B DAPs between XD and XW.

   Red and green indicate significantly upregulated and downregulated
   proteins, respectively.

Identification of WGCNA modules related to drought resistance in oat
varieties

   Nine WGCNA modules were identified using WGCNA of sample
   characteristics and protein expression changes for the 1,817 quantified
   proteins ([81]Fig 5). The results of the module-trait relationship
   indicated that 55 proteins in the “MEpink” module were highly
   correlated with POD (r = 0.82, p = 0.001) and SOD (r = 0.79, p = 0.002)
   activities and SS content (r = 0.79, p = 0.002), whereas they were
   moderately correlated with MDA content (r = 0.65, p = 0.02). The same
   proteins were not significantly correlated with plant height (r =
   −0.37, p = 0.2), aboveground biomass (r = −0.3 p = 0.3), and
   underground biomass (r = −0.2, p = 0.5). This observation may be due to
   a response threshold for drought stress in the phenotypic,
   physiological, and molecular aspects of plant tissues and organs.
   Within a certain threshold, drought stress only activates the stress
   response at the physiological and molecular levels of oats, and the
   effects on morphological and growth traits are not particularly
   evident.

Fig 5. WGCNA of the sample characteristics and proteins of the two oat
varieties under drought stress.

   [82]Fig 5
   [83]Open in a new tab

   A Stratified clustering of the nine WGCNA modules. B Correlation
   analysis of sample characteristics and proteins with changes in
   expression. Significant differences are indicated in parentheses, while
   correlation coefficients are indicated outside parentheses. PH, plant
   height; AGB, aboveground biomass; UGB, underground biomass.

   In the MEpink module, the drought-resistant variety GD had 17 DAPs
   compared to GW, whereas the drought-susceptible variety XD had 5 DAPs
   compared to XW; the two varieties had two common DAPs, which were
   upregulated. These findings indicate that the 15 DAPs (excluding the
   two shared ones) in the MEpink module of the drought-tolerant variety
   may be essential in oat drought tolerance, and their synergistic
   expression can improve drought tolerance.

GO enrichment analysis

   The GO enrichment analysis of DAPs in the leaves of the two oat
   varieties indicated that all DAPs were enriched in biological process
   (BP), cellular component (CC), and molecular function (MF). Specific
   proteins with different abundances are presented in [84]Fig 6. In
   variety G ([85]Fig 6A), BP was mainly enriched in the following GO
   terms: “cellular process,” “metabolic process,” “response to stimulus,”
   “regulation of biological processes,” “biological regulation,” and
   “localization” (and 11 other aspects). Additionally, CC was enriched in
   “anatomical entity” and “protein-containing complex.” Furthermore, MF
   mainly included the terms “catalytic activity,” “binding,” “structural
   molecule activity,” and “antioxidant activity.” In variety X ([86]Fig
   6B), BP was mainly enriched in “cellular process,” “metabolic process,”
   “response to stimulus,” “localization,” “biological regulation,” and
   “regulation of biological processes.” In CC, only two terms were
   enriched: “anatomical entity” and “protein-containing complex.”
   Furthermore, the DAPs in MF primarily had 10 functions, including
   “catalytic activity,” “binding,” “structural molecule activity,”
   “antioxidant activity,” “translation regulator activity,” and
   “transporter activity.” However, the 20 DAPs of the MEpink module were
   not significantly enriched in the GO database. The above results
   indicate that in terms of CC, MF, and BP, DAP enrichment levels were
   markedly different between the two oat varieties under drought stress
   conditions, revealing large differences in the protein levels of the
   two varieties in response to drought stress.

Fig 6. GO enrichment of DAPs, including BP, CC, and MF.

   [87]Fig 6
   [88]Open in a new tab

   A GO enrichment of DAPs between GD and GW. B GO enrichment of DAPs
   between XD and XW.

KEGG pathway enrichment analysis

   The KEGG pathway enrichment analysis revealed that DAPs were enriched
   in 46 and 97 KEGG metabolic pathways in the G and X varieties,
   respectively, under drought stress. In variety G ([89]Fig 7A), DAPs
   were significantly enriched in 27 KEGG pathways under drought stress (p
   < 0.05), including metabolic pathways, photosynthesis, nitrogen
   metabolism, porphyrin and chlorophyll metabolism, oxidative
   phosphorylation, and protein processing in the endoplasmic reticulum.
   In variety X ([90]Fig 7B), DAPs were significantly enriched in 51 KEGG
   pathways under drought stress (p < 0.05), including metabolic pathways,
   carbon metabolism, photosynthesis, biosynthesis of secondary
   metabolites, glyoxylate and dicarboxylate metabolism, and carbon
   fixation by photosynthetic organisms.

Fig 7. KEGG pathway enrichment bubble map of the top 20 DAPs.

   [91]Fig 7
   [92]Open in a new tab

   A KEGG pathway enrichment map of the top 20 DAPs between GD and GW. B
   KEGG pathway enrichment map of the top 20 DAPs between XD and XW. The
   vertical coordinate indicates the enriched pathway, while the
   horizontal coordinate indicates the negative logarithmic transformation
   of the p-value. p < 0.05 indicates that the function is significantly
   enriched, with smaller p-values indicating significantly higher
   functional enrichment. The color of the circle denotes the rich factor:
   Rich factor = (a/b)/(c/d), where a is the number of DAPs annotated to
   the term, b is the total number of DAPs annotated to the term, c is the
   number of background proteins of the term, and d is the total number of
   background proteins of the term. A larger Rich factor indicates a
   higher level of enrichment, and the size of the circle indicates the
   number of differentially abundant proteins enriched in the pathway.

   The KEGG enrichment analysis of the 20 DAPs in the MEpink module
   indicated enrichment in 12 and 3 KEGG metabolic pathways in the G and X
   varieties, respectively, under drought stress. Among them, DAPs of
   variety G were significantly enriched in five KEGG pathways (p < 0.05),
   including the mRNA surveillance pathway, ribosome, phenylalanine
   metabolism, photosynthesis, and phagosome, whereas DAPs of variety X
   were significantly enriched in three KEGG pathways (p < 0.05),
   including photosynthesis, phagosome, and plant-pathogen interactions.
   Thus, the KEGG enrichment pathways of DAPs differed between the two oat
   varieties, implying that they may involve metabolic pathways in
   response to drought stress.

Protein division in the MEpink module

   In the MEpink module, 17 DAPs were identified in variety G under
   drought stress. All were upregulated, except for two DAPs that were
   shared with variety X (a chaperone protein DnaJ [ID: M8A0F5] and an
   uncharacterized protein [ID: A0A446J984]). The remaining DAPs were
   mostly involved in energy metabolism, protein translation, RNA
   processing, and amino acid metabolism: a 9-kDa polypeptide (ID:
   A0A0Q3IET4), carbamoyl-phosphate synthase (ID: [93]Q9AXS0),
   temperature-induced lipid (TIL) transport protein (ID: I1IBR5),
   photosystem I subunit VII (ID: A0A446S4S2), PAP_fibrillin
   domain-containing protein (ID: A0A446MFG9), SpoU_sub_bind
   domain-containing protein (ID: A0A0Q3FXR3), SRP54 domain-containing
   protein (ID: A0A446KTE5), isoleucyl-tRNA synthetase (ID: I1H766), and
   seven uncharacterized proteins (ID: A0A446SZM6, A0A453AN53, A0A510B3S2,
   A0A446QWZ0, M5BPX4, A0A3B6FW66, and A0A453N829). Variety X had three
   DAPs (excluding the two shared with variety G) under drought stress,
   all of which were upregulated: peptide chain release factor (ID:
   M8CEY9) and two uncharacterized proteins (ID: A0A3B5YS83 and
   A0A3B6A287) primarily involved in RNA processing, protein
   stabilization, plant photosynthesis, and intracellular signal
   transduction.

Discussion

Common pathways to cope with drought stress in different oat varieties

   Different oat varieties of the same species exhibit common drought
   tolerance pathways at the protein level. One such pathway involves the
   DnaJ protein, a heat stress protein primarily associated with protein
   folding, translocation, secretion, localization, and the reversion and
   degradation of denatured proteins. It plays a crucial role in
   maintaining intracellular protein homeostasis under adverse conditions
   and stabilizing protein complexes [[94]30]. Studies on DnaJ protein
   expression in pepper placenta have shown that it may participate in the
   plant’s response to abiotic stress during the biosynthesis of
   stimulatory compounds [[95]31]. Wang’s investigation into the function
   of the tomato (Lycopersicon esculentum) chloroplast-targeted DnaJ
   protein (LeCDJ2) and its ectopic expression in transgenic tobacco
   revealed that salicylic acid, drought, and pathogen attacks induced
   LeCDJ2 expression. Furthermore, the ectopic expression of LeCDJ2 in
   transgenic tobacco reduced the production of superoxide anion radicals
   (O[2]−) and hydrogen peroxide (H[2]O[2]) under drought stress. The
   overexpression of the tomato chloroplast-targeted DnaJ protein also
   improved the drought and cyanobacterial resistance of transgenic
   tobacco [[96]32]. In the present study, DnaJ protein expression was
   significantly upregulated in both oat varieties, with higher expression
   observed in the drought-susceptible variety compared to the
   drought-resistant one. This finding suggests that the
   drought-susceptible variety may activate its response mechanisms to
   water deficit earlier than the drought-resistant variety, consistent
   with previous reports [[97]33,[98]34].

Different pathways to cope with drought stress in different oat varieties.

   Photosystem I (PSI) is a multisubunit membrane complex that catalyzes
   electron transfer involving iron redox proteins, which bind to fumarate
   and nitrate reduction and reduced nicotinamide adenine dinucleotide
   phosphate (NADPH). In higher plants, the PSI complex includes 4–80 kDa
   polypeptides in addition to light-harvesting complexes I (LHCI), among
   which the 9-kDa polypeptide subunit is a protease that may play a
   functional role under unfavorable growth conditions [[99]35]. In this
   study, the expression of the 9-kDa polypeptide subunit was
   significantly upregulated in the strongly drought-resistant variety G,
   whereas it remained insignificant in the drought-susceptible variety X.
   This finding suggests that strongly drought-tolerant oat varieties may
   respond to drought stress by enhancing energy production during
   cellular photosynthesis, thereby reestablishing homeostasis and complex
   energy metabolic pathways. This observation aligns with previous
   research showing that the expression of photosynthesis-related proteins
   is upregulated in drought-resistant maize varieties [[100]36].

   Arginine serves as a precursor for the production of compounds that act
   as second messengers, such as nitric oxide (NO) and polyamines
   (including spermidine and putrescine), which play crucial roles in
   regulating stress resistance in plants [[101]37–[102]40]. For instance,
   exogenous chitosan and spermine have been shown to mitigate
   drought-induced oxidative damage and enhance the content of total
   phenols and flavonoids in white clover (Trifolium repens L.) [[103]40].
   The carbamoyl-phosphate synthase subunit is a catalytic enzyme
   essential for converting ornithine to citrulline during arginine
   biosynthesis and plays a role in plant resistance mechanisms [[104]41].
   This observation was further supported by the findings of the present
   study, where the enzyme was significantly upregulated under drought
   stress in the drought-resistant variety. Its activity catalyzed the
   synthesis of ornithine into citrulline, releasing energy to sustain
   plant growth under adverse conditions [[105]40]. Conversely, this
   enzyme was not significantly upregulated in the drought-susceptible
   variety.

   Lipoproteins are critical enzymes of the lutein cycle and play a key
   role in protecting the photosynthetic apparatus from oxidative damage
   caused by excessive light exposure [[106]42]. Two homologous cDNAs from
   wheat (TaTIL) and Arabidopsis (AtTIL) encode lipoproteins containing
   three structurally conserved regions that potentially serve as ligands
   with diverse structures and functions for the synthesis of phytosteroid
   hormones, such as 2,4-epi-oligolactone, which enhance plant tolerance
   to heat and cold stress [[107]43,[108]44]. AtTIL transport proteins are
   also involved in regulating plant tolerance to oxidative stress
   [[109]45], with plant TIL transport proteins counteracting severe heat
   stress-induced lipid peroxidation [[110]46]. Additionally, AtTIL, a
   lipid-like transport protein gene, is implicated in plant salt
   tolerance [[111]47]. However, the role of TIL transport proteins in
   drought tolerance in plants has not been previously reported. In this
   study, TIL transport proteins were significantly upregulated in the
   drought-tolerant variety under drought stress, whereas their relative
   abundance did not change remarkably in the drought-susceptible variety.
   This finding suggests that TIL transport proteins may play a role in
   regulating plant drought tolerance and improving drought resistance.

   Plastoglobuli, lipoprotein particles ranging in diameter from 30 nm to
   5 μm, are present in plastids [[112]48]. A family of fibrillin (FBN)
   proteins has been identified within these lipoproteins. The FBN gene is
   induced by various stresses, including high temperature, low
   temperature, drought, salt, herbicides, and strong light, mediating
   ABA-related resistance signaling and regulating the synthesis of
   jasmonic acid, triacylglycerol, and plastoquinone [[113]49–[114]52]. In
   this study, the expression of the PAP_fibrillin domain-containing
   protein in variety G was significantly upregulated under drought
   stress, consistent with the increased expression of FBN proteins under
   drought stress observed in Arabidopsis, tomato, potato, tobacco, and
   other plants [[115]49,[116]53,[117]54]. This finding suggests that the
   PAP_fibrillin domain-containing protein may play a role in signal
   transduction and response to drought stress in oat chloroplasts.
   Further research is necessary to determine whether this protein has a
   critical role in recognizing and responding to drought stress and to
   elucidate its specific functions.

   Serine/arginine-rich proteins are primarily involved in the assembly
   and splicing of eukaryotic mRNA precursors. They are critical
   components of the mRNA spliceosome, playing a unique regulatory role in
   splicing mechanisms [[118]55]. Plants respond to stress by regulating
   splicing factors; for instance, variable splicing of the OsWRKY45
   allele is involved in the drought stress response [[119]56], and the
   overexpression of ScMYBAS1–2 and ScMYBAS1–3 splice transcripts promotes
   changes in plant growth under water deficiency and drought conditions
   [[120]57]. In this study, the SRP54 domain-containing protein was
   significantly upregulated under drought stress in variety G, but not in
   variety X. This finding suggests that the SRP54 domain-containing
   protein may play a critical role in mediating the oat response to
   drought stress as a splicing factor.

   In the MEpink module, seven uncharacterized proteins were significantly
   upregulated under drought stress in variety G but not in variety X.
   These proteins, with unknown structures and functions, may play a
   crucial role in the regulatory network of drought stress and contribute
   to improving drought resistance in oats. Further studies on these
   uncharacterized proteins are needed to elucidate the molecular
   mechanisms underlying the oat response to drought stress.

Conclusions

   Our study investigated the mechanisms of drought tolerance in
   drought-resistant (Grain King) and drought-susceptible (XiYue) oat
   varieties from phenotypic, physiological, and proteomic perspectives
   ([121]Fig. 8). Phenotypic and physiological data demonstrated that the
   drought-resistant oat variety outperformed the drought-susceptible
   variety under drought stress. Proteomic analysis suggested that the
   high drought resistance of Grain King is likely associated with the
   upregulation of proteins, including the 9-kDa polypeptide, the subunit
   of carbamoyl-phosphate synthase, temperature-induced lipid transport
   protein, PAP fibrillin domain-containing protein, and SRP54
   domain-containing protein. This study elucidates the relationships
   between physiological and proteomic factors contributing to drought
   resistance and advances our understanding of drought tolerance
   mechanisms in different oat varieties.

Fig 8. Different responses of two oat varieties to drought stress.

   [122]Fig 8
   [123]Open in a new tab

Supporting information

   S1 Table

   Raw data of fig 1

   (DOCX)
   [124]pone.0322022.s001.docx^ (19.1KB, docx)
   S2 Table

   Raw data of fig 2.

   (DOCX)
   [125]pone.0322022.s002.docx^ (19.1KB, docx)

Acknowledgments