Integrative multi-scale analysis reveals a candidate melatonin receptor associated with early salinity response networks in barley Authors: Roohollah Shamloo-Dashtpagerdi, Angelica Lindlöf, Ahmad Tahmasebi Published: 2026 ABSTRACT Melatonin serves as a crucial signaling regulator in plants, playing a vital role in their growth, development, and stress responses. Although receptor-mediated signaling has been well characterized for several plant hormones, the integration of melatonin perception into early stress response networks in temperate cereals remains unclear. Here, we employed an integrative, multi-scale approach combining computational modeling with molecular, physiological, and biochemical analyses to investigate the involvement of a candidate melatonin receptor, HvPMTR1, and its associated heterotrimeric G-protein α-subunit, HvGα1, in barley (Hordeum vulgare L.) during early salinity responses. Structural modeling using molecular docking and ligand-tunneling simulations, and membrane topology predictions, supported HvPMTR1 as a seven-transmembrane protein with features compatible with melatonin interaction. Exogenous melatonin application triggered coordinated upregulation of HvPMTR1 and HvGα1, accompanied by controlled reactive oxygen species (ROS) dynamics, enhanced antioxidant capacity, and modulation of stomatal behavior. Under early salinity stress, this signaling axis was more strongly activated in a salinity-tolerant genotype compared with a sensitive genotype and was associated with improved water status and sustained photosynthetic performance. Multivariate analyses further positioned HvPMTR1 and HvGα1 expression dynamics, alongside endogenous melatonin levels, within a signaling framework associated with stress resilience. Collectively, these findings suggest that putative melatonin perception is integrated into early salinity-responsive signaling networks in barley, highlighting a potential regulatory module coordinating molecular signaling with physiological adaptation in a temperate cereal crop.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-55742-z. Introduction Melatonin (N-acetyl-5-methoxytryptamine) is an ancient and highly conserved indoleamine that emerged early in the evolution of life and is now found across nearly all biological kingdoms1. First identified in animals as a regulator of circadian rhythms and neuroendocrine processes, melatonin was later discovered in higher plants, where it is commonly referred to as phytomelatonin2. Over the past three decades, accumulating evidence has expanded its functional repertoire far beyond that of a classical antioxidant; phytomelatonin is now recognized as a key regulator of plant growth, development, and environmental adaptation3,4. Its biosynthetic pathway is dynamically responsive to a wide range of stresses, supporting the view that melatonin acts as an active signaling molecule rather than a passive metabolic byproduct. In this capacity, phytomelatonin modulates antioxidant enzyme systems, regulates stomatal behavior, influences the expression of ion transporters, and interacts with major phytohormonal networks, including ABA, auxin, jasmonic acid, and ethylene, forming an intricate regulatory node within the plant stress response machinery5–8. Despite the extensive documentation of melatonin’s physiological roles, the mechanism by which plants perceive and transduce melatonin signals remained unresolved until relatively recently. A major conceptual breakthrough emerged with the discovery of PMTR1 (Phytomelatonin Receptor 1; also known as CAND2) in Arabidopsis thaliana9. AtPMTR1 encodes a GPCR-like, plasma-membrane-localized protein that interacts directly with the heterotrimeric G-protein α-subunit GPA1. Upon melatonin binding, the AtPMTR1-GPA1 module initiates rapid downstream signaling events, including Ca2⁺ influx, controlled ROS production, and MAP kinase activation, ultimately regulating stomatal closure and enhancing abiotic stress tolerance9. These findings suggested a receptor-mediated component of melatonin signaling in plants, positioning melatonin as a bona fide hormone-like signaling molecule. Subsequent studies have extended this receptor-based signaling paradigm beyond Arabidopsis to tropical monocots. In rice (Oryza sativa), a functional PMTR1 ortholog (OsPMTR1) was shown to physically interact with RGA1, the rice Gα subunit, activating a signaling cascade that strengthens ROS-scavenging capacity, stabilizes Na⁺/K⁺ homeostasis, and improves salinity tolerance10. While PMTR1-mediated signaling has been characterized in the dicot model A. thaliana and the tropical monocot O. sativa, evidence for such a signaling module in temperate cereals, particularly in barley, remains limited. Those findings cannot be directly extrapolated to temperate cereals like barley due to fundamental differences in their evolutionary trajectories and adaptive strategies11,12, Leaving open the question of whether temporal cereals share a common receptor-based mechanism for melatonin perception under salinity stress. Barley is one of the most widely cultivated cereal crops and remains a key genetic and physiological model for temperate grasses13,14. Physiologically, barley exhibits distinct osmotic adjustment patterns and ion homeostasis mechanisms compared to rice, which is typically more sensitive to salinity15,16. Furthermore, the signaling architecture of temperate cereals often involves divergent G-protein coupled dynamics and distinct temporal regulation of early stress perception17,18. Given that barley has evolved in temperate environments with fluctuating osmotic pressures, investigating its specific receptor-mediated signaling is essential to understand the unique upstream early-warning systems that dictate its superior resilience. In addition, accumulating evidence indicates that melatonin plays a significant role in barley’s response to salinity. Exogenous melatonin application has been shown to enhance antioxidant activity, stabilize photosynthetic performance, and improve ionic balance under salinity stress19–21. In parallel, endogenous melatonin biosynthesis is dynamically regulated during salinity exposure, with the recently described SNAC1-ASMT1 module emerging as a determinant of melatonin production and salinity-tolerance capacity in barley22. However, these studies primarily focus on downstream responses and do not address how melatonin signals are perceived upstream of these responses. This gap limits our mechanistic understanding of melatonin perception in monocots and hinders the development of receptor-informed strategies to enhance stress tolerance. To address these knowledge gaps, we adopted an integrative multi-scale approach to explore how melatonin perception may be connected to early salinity signaling in barley. Specifically, we combined advanced in silico structural modeling, including AlphaFold3-based predictions and ligand-tunneling simulations, with transcriptional, biochemical, and physiological analyses to examine the candidate receptor HvPMTR1 and its relationship with melatonin and the HvGα1. We further investigated how this signaling axis is associated with coordinated regulation of ROS, antioxidant responses, stomatal behavior, and photosynthetic performance under salinity stress. By integrating computational predictions with experimental observations across biological scales and genotypes, our study places the melatonin-HvPMTR1-HvGα1 module as the primary statistical discriminator within an early salinity-responsive melatonin signaling framework. This systems-level perspective provides a foundation for receptor-informed strategies to enhance stress tolerance in temperate cereal crops. Materials and methods Identification of barley melatonin receptor candidate(s) The CDS of A. thaliana PMTR1 (AtPMTR1, TAIR: AT3G05010) was used as a query to identify homologous sequence(s) in the barley genome using BLASTn in the BarleyOmics database (https://www.barleymdb.cn/home/) against the Morex_v3_CDS dataset (E-value ≤ 1 × 10−5). The deduced amino acid sequence of the putatively identified HvPMTR1 was obtained using CLC Genomics Workbench 22. The basic physicochemical properties of the protein, including molecular weight (Mw), isoelectric point (pI), instability index, aliphatic index, and grand average of hydropathicity (GRAVY), were calculated using the ProtParam tool (https://web.expasy.org/protparam/) on the ExPASy server. Functional domains and conserved motifs were identified using Pfam (http://pfam.xfam.org/), and post-translational modification sites and secondary-structure elements, including α-helices, β-strands, and coils, were predicted with CLC Genomics Workbench 22. The subcellular localization was predicted using WoLF PSORT (https://wolfpsort.hgc.jp/). Protein topology and transmembrane helices were annotated using the PROTTER tool (http://wlab.ethz.ch/protter/start/). Phylogenetic analysis To elucidate evolutionary relationships, homologous protein sequences from several selected plant species (including monocots (Oryza sativa, Zea mays, Sorghum bicolor, and Brachypodium distachyon), dicots (A. thaliana, A. lyratha, Brassica oleracea, and B. rapa, and an outgroup (Juglans regia)) were retrieved from NCBI using BLASTp. Multiple sequence alignment was performed in CLC Genomics Workbench 22, and a phylogenetic tree was constructed using the Neighbor-Joining (NJ) method with 1000 bootstrap replicates in the same software. Molecular docking investigation The 2D structure of melatonin was retrieved from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) and used as the ligand. The HvPMTR1 protein sequence was used to generate a predicted 3D structure through AlphaFold3 (https://alphafoldserver.com/)23, and saved in PDB format. The Orientations of Proteins in Membranes (OPM) database and the Positioning of Proteins in Membranes (PPM3) web server (https://opm.phar.umich.edu/ppm_server3_cgopm)24 were used to determine the spatial orientation of HvPMTR1 within the plasma membrane. Molecular docking was carried out using AutoDock Vina25 through the SwissDock server (https://www.swissdock.ch/) with the following parameters: Center_x = 1, Center_y = -2.0, Center_z = 12. Size_x = 20, Size_y = 20, Size_z = 20. Sampling exhaustivity = 64. The docking affinity (kcal/mol) of melatonin to the putative HvPMTR1 receptor was evaluated to predict binding stability. The protein–ligand complex was visualized and analyzed using PyMOL v3.1.1 (Schrödinger, LLC). Interaction distances were measured between the melatonin functional groups and the side chains of key residues. Analysis of ligand access channels To identify potential transport pathways for melatonin from the extracellular environment to the binding pocket of HvPMTR1, a geometric tunnel analysis was performed using the MOLEonline web server (https://mole.upol.cz/)26. The refined 3D model of the receptor, complexed with melatonin, was used as the starting structure. The search for tunnels was initiated by defining a starting point at the center of gravity of the key residue Phe133, which was identified as critical in the binding site. The tunnel calculation parameters were set with a minimum bottleneck radius of 1.2 Å and a probe radius of 3 Å to ensure the detection of functionally relevant cavities. The identified pathways were characterized based on their geometric and physicochemical properties, including length, bottleneck radius, hydropathy, and polarity. The most energetically and geometrically favorable channel was selected for further analysis. Visualization of the identified tunnel and its spatial orientation relative to the docked ligand was performed using PyMOL v3.1.1 (Schrödinger, LLC). The final molecular surfaces were rendered with a focus on the spatial overlap between the tunnel trajectory and the ligand-binding cavity to confirm the accessibility of the predicted docking site. In silico analysis of the HvPMTR1 promoter sequence The 1500 base pair (bp) promoter region upstream of the transcription start site (TSS) of the HvPMTR1 gene was retrieved from the BarleyOmics database (https://www.barleymdb.cn/home/)27. Putative cis-regulatory elements (CREs) were identified using the PlantCARE database (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/)28. The sequence was submitted in FASTA format using default parameters, and the results were manually curated to categorize the predicted CREs by known biological function and number of occurrences. RNA-Seq data retrieval for HvPMTR1 expression profiling The corresponding gene accession number for HvPMTR1 (HORVU.MOREX.r3.5HG0511510/HORVU5Hr1G096100) was used to query the Morex Gene Atlas database (https://ics.hutton.ac.uk/morexgeneatlas/index.html) to retrieve barley gene expression data across different conditions and treatments (Table 1). Expression data, provided as Transcripts Per Million (TPM), were downloaded and analyzed statistically to identify significant expression differences under the selected conditions.Table 1List of experiments used to retrieve expression data of the HvPMTR1 gene from the Morex Gene Atlas database.TreatmenttissueTime point/LevelNo. of replicates per treatmentABARoot1, 3, 6, 24 h3NaClShoot0, 150, 300 mM3NaClRoot0, 150, 300 mM3ZnShootControl and treated3ZnRootControl and treated3CuShootControl and treated3CuRootControl and treated3CdShootControl and treated3CdRootControl and treated3ColdLeafControl and treated4 Table 1List of experiments used to retrieve expression data of the HvPMTR1 gene from the Morex Gene Atlas database.TreatmenttissueTime point/LevelNo. of replicates per treatmentABARoot1, 3, 6, 24 h3NaClShoot0, 150, 300 mM3NaClRoot0, 150, 300 mM3ZnShootControl and treated3ZnRootControl and treated3CuShootControl and treated3CuRootControl and treated3CdShootControl and treated3CdRootControl and treated3ColdLeafControl and treated4 List of experiments used to retrieve expression data of the HvPMTR1 gene from the Morex Gene Atlas database. Experiment I: melatonin treatment Plant material, growth conditions, and experimental design Barley genotypes used in this study were the salt-tolerant cultivar Nosrat and the salt-sensitive cultivar Shirin22,29. Seeds were obtained from the Darab County Agricultural Research Station, affiliated with the Fars Province Agricultural Research Organization in Iran and were surface sterilized prior to cultivation. were surface-sterilized in 1% (v/v) sodium hypochlorite for 10 min, rinsed thoroughly with distilled water, and pre-germinated on moist filter paper for 48 h. Seedlings were transplanted into pots (17 cm height × 10 cm diameter) containing 1 kg of autoclaved sandy-clay loam soil (particle size distribution: 52% coarse sand, 38% clay, 8% silt; pH 6.8 ± 0.2) supplemented with slow-release fertilizer (NPK 14:14:14, manufacturer’s rate). Each pot contained one plant. Plants were grown in a controlled greenhouse under the following conditions: 16 h light/8 h dark photoperiod, day/night temperatures of 22 ± 2 °C/18 ± 2 °C, and relative humidity of 55–65%. Watering and nutrient supply were applied uniformly to maintain non-limiting soil moisture until treatments were initiated. Pots were randomized within the greenhouse, and the arrangement was rearranged weekly to minimize position effects. Melatonin solution and treatment application A working solution of melatonin at 100 µM was prepared by dissolving melatonin in a small volume of ethanol and then diluting it with distilled water. The final ethanol concentration was maintained below 0.1% (v/v). Plants at the five-leaf stage were treated by foliar spray until runoff. To avoid cross-contamination, different sprayers were used for control and melatonin, and treated plants were spatially separated for 30 min after spraying to allow leaf surfaces to dry. The application was performed at the start of the photoperiod (within the same 30-min window for all plants) to reduce diurnal variation. Sampling times were defined relative to the time of spray: 0 h (control) (immediately before treatment), 1 h, 3 h, and 6 h after application. The experiment included four independent biological replicates (pots) per genotype × treatment combination. At each sampling time, the youngest fully expanded leaf was harvested between 09:00 and 10:30 AM to minimize diurnal effects on melatonin and gas-exchange variables. For each pot (biological replicate), a 2–3 cm disc (for biochemical assays) or ~ 100 mg fresh tissue (for RNA extraction) was excised, immediately flash-frozen in liquid nitrogen, and stored at − 80 °C until analysis. For gas exchange, a separate intact leaf was measured on each plant immediately before tissue harvest, to avoid perturbing leaf physiology during biochemical assays. Gene expression analysis (qRT-PCR) Total RNA was extracted from ~ 100 mg frozen leaf tissue using the Column RNA Isolation Kit, DENAzist Asia, Iran, according to the manufacturer’s instructions. RNA concentration and purity were measured with a NanoDrop spectrophotometer (260/280 and 260/230 ratios), and integrity was checked by agarose gel electrophoresis. Residual genomic DNA was enzymatically degraded using recombinant DNase I (Thermo Scientific, USA) before reverse transcription. First-strand cDNA synthesis was performed from 2 µg DNase-treated RNA template via Easy cDNA Synthesis Kit, ParsGen, Iran, following the manufacturer’s protocol. Gene-specific primers for HvPMTR1, HvGα1, and reference gene (Elongation Factor 1α (HvEF1α); GenBank: NC_058519) were designed using AlleleID 7.7 (Premier Biosoft). Primer sequences and amplicon sizes are reported in the supplementary table S1. qPCR reactions were performed on a Bioer LineGene 9600 thermal cycler (Bioer Technology, China) using SYBR Green I chemistry (2X Real-Time PCR Master Mix, ParsGen, Iran). All reactions were run in two technical replicates. Relative expression was calculated with the 2−ΔΔCt method30. ΔΔCt was calculated relative to the mean ΔCt of the susceptible genotype as calibrator, and data are reported as fold change relative to the calibrator. Physiochemical assays Frozen leaf tissue was ground to fine powder under liquid nitrogen and homogenized in ice-cold extraction buffer appropriate for enzymatic assays (50 mM phosphate buffer, pH 7.0, containing 1 mM EDTA and 1% polyvinylpyrrolidone). Homogenates were centrifuged at 12,000 × g for 15 min at 4 °C, and the supernatant was collected for enzyme assays. Catalase (CAT) activity CAT (E.C. 1.11.1.6) activity was measured spectrophotometrically by monitoring the breakdown of H₂O₂ at 240 nm as described by Aebi31. The reaction mixtures consisted of 50 mM potassium phosphate buffer (pH 7.0), 15 mM H₂O₂, and 20 µL of crude enzyme extract. The decline in absorbance was monitored, and the activity was reported as µmol H₂O₂ min−1 mg−1 protein or U mg−1 protein, with one unit defined as µmol H₂O₂ decomposed per min. Hydrogen peroxide (H₂O₂) content The concentration of H₂O₂ in leaf samples was assessed using the potassium iodide (KI) method32. Fresh foliar tissue (100 mg) was finely ground in liquid nitrogen and mixed with 1 mL of ice-cold 0.1% (w/v) trichloroacetic acid (TCA). The resulting homogenate was centrifuged (12,000 × g for 15 min at 4 °C) to separate cellular debris. A 0.5 mL sample of the supernatant was mixed with 0.5 mL of 10 mM potassium phosphate buffer (pH 7.0) and 1 mL of 1 M KI to start the oxidative coupling reaction. After incubating in the dark for 10 min, the chromogenic development was measured spectrophotometrically at 390 nm. The results were expressed as µmol H₂O₂ g−1 FW based on a standard curve. Stomatal conductance measurement Stomatal conductance (gs, mmol H₂O m−2 s−1) was taken on the same portion of an intact, fully expanded leaf using an LI-6400XT Portable Photosynthesis System (LI-COR Biosciences, USA). The measurement settings were: leaf chamber CO₂ = 400 µmol mol−1, chamber temperature = 25 °C, PPFD = 1000 µmol m−2 s−1 (or actual greenhouse light if lower), flow rate 500 µmol s−1, and measurement allowed to stabilize for 1–2 min before recording. Data handling and statistical analysis All statistical analyses were carried out in STATISTICA v.12 (TIBCO Software Inc., Palo Alto, CA, USA). Data are presented as mean ± standard error (SE) of four biological replicates. Data were checked for normality (Shapiro–Wilk test) and homogeneity of variance (Levene’s test). Regarding the melatonin treatment experiment, data were analyzed with ANOVA. Post-hoc mean comparisons were performed with Tukey’s HSD test at α = 0.05. For qPCR, expression values (log₂ fold change) were analyzed similarly. Principal component analysis (PCA) was performed to explore relationships among molecular, biochemical, and physiological variables. Experiment II: salinity treatment The two contrasting barley genotypes, Nosrat (salt-tolerant) and Shirin (salt-sensitive), were grown under similar greenhouse conditions and at the same developmental stage as described for the melatonin experiment. Salinity stress was imposed by applying a 150 mM NaCl solution directly to the soil at the 5-leaf stage. Pots were irrigated to field capacity with a saline solution, and samples were collected at 12, 24, and 48 h after treatment. Four biological replicates (four independent pots) were used per genotype × treatment. Gene expression analysis (HvPMTR1 and HvGα1) Total RNA extraction, DNase I treatment, cDNA synthesis, and qPCR procedures were identical to those described for the melatonin experiment, using the same primer pairs and reference gene. Relative expression values were calculated using the 2−ΔΔCt method with genotype-matched controls (0 h) as calibrators. Evaluation of endogenous melatonin To quantify endogenous melatonin levels, leaf samples from both barley genotypes were extracted with 1.0 mL of chloroform:methanol (30:1 v/v) at low temperature. The resulting extract was centrifuged at 10,000 × g for 15 min at 4 °C to separate the phases. The supernatant was carefully transferred into a new tube and evaporated under a nitrogen stream to remove the solvents. The dried residue was reconstituted in 1.0 mL of a 1:1 (v/v) methanol mixture and filtered through a 0.22 μm nylon membrane. A final dissolution in a methanol:water mixture (1:1 v/v) was performed, followed by another filtration step through a 0.22 μm nylon filter. Endogenous melatonin levels in leaf extracts were measured using an ELISA Kit (EK-DSM; Buhlmann Laboratories AG, Schonenbuch, Switzerland) according to the manufacturer’s protocols. Oxidative status analyses Assays of CAT enzyme activity and H₂O₂ content were performed using the same protocols, reagents, and spectrophotometric conditions described in the melatonin experiment. In addition, as a marker for lipid peroxidation, malondialdehyde (MDA) levels were quantified using the thiobarbituric acid (TBA) assay33. Physiological evaluations The relative density of chlorophyll in the third fully developed leaf was assessed non-destructively using a SPAD-502Plus chlorophyll meter (Delta T, UK). For each leaf, triplicate SPAD readings were recorded along the midrib axis (basal, medial, and apical regions) and averaged to reduce positional variability. The photosynthetic performance was measured on mature leaves exposed to sunlight (third phyllotactic position) utilizing an LI-6400XT Portable Photosynthesis System (LI-COR Biosciences, USA) under controlled conditions, which included a reference CO₂ concentration of 400 µmol mol−1, a photosynthetic photon flux density (PPFD) of 1000 µmol m−2 s−1, a leaf chamber temperature of 25 °C, relative humidity of 60 ± 5%, and a flow rate of 500 µmol s−1. Leaves were allowed to acclimate in the cuvette for 1–2 min until steady-state gas exchange was established (Aₙ variation < 2% over 60 s). The parameters measured included the net CO₂ assimilation rate (Aₙ; µmol CO₂ m−2 s−1), stomatal conductance (gₛ; mol H₂O m−2 s−1), and the transpiration rate (E; mmol H₂O m−2 s−1). All measurements were taken between 10:00 and 14:00 h (during the solar noon period) to reduce diurnal variability. Ion homeostasis After drying, the samples were ground into a fine powder and weighed again. The powdered samples were ashed at 120 °C for 12 h. The levels of Na+ and K+, and the K+/Na+ ratio, were measured using a flame photometer34. Statistical analysis All statistical analyses were conducted in STATISTICA v.12 (TIBCO Software Inc., Palo Alto, CA, USA). The data were analyzed using two-way ANOVA with genotype and time as fixed factors. When ANOVA indicated significant effects (p < 0.05), Tukey’s HSD test was applied for pairwise comparisons. Normality and homogeneity of variances were verified using the Shapiro–Wilk and Levene’s tests, respectively. PCA was used to visualize sample clustering across time points. In addition, Partial Least Squares Discriminant Analysis (PLS-DA) and Chi-squared Automatic Interaction Detection (CHAID) were applied to assess the relative contribution of each variable to genotype separation. To apply PLS-DA, the following parameters were used: Stop conditions: Automatic (to automatically determine the number of components to keep); Algorithm: Precise (to use the slower but more precise algorithm, which recenters or standardizes the jackknife training sets); Cross validation: Jackknife (LOO). In addition, Variable importance in projection (VIP) scores were extracted from PLS-DA to identify the most informative traits (95% confidence interval). To perform CHAID, the following parameters were used: Method: CHAID, Measure: Pearson’s Chi-square, Tree parameters: Automatic (to choose the best parameters which provide a minimal cross-validation error to build the final model), Maximum tree depth: 3, Merge threshold: 5%, Cross Validation: Number of folds: 2. In addition, variable importance was assessed from the tree based on contribution to splits variables. Results Identification and sequence characterization of a putative melatonin receptor in barley Using the A. thaliana PMTR1 coding sequence as a query, a homologous sequence was successfully identified in the barley genome through BLASTn analysis. The identified sequence and its predicted protein, named HvPMTR1, showed high similarity to AtPMTR1 and its orthologs in other monocot species. The predicted protein contained 439 amino acids with a calculated molecular weight of approximately 48.2 kDa and a theoretical isoelectric point (pI) of 9.16, suggesting a slightly basic nature. Sequence analysis revealed a hydrophobic N-terminal segment consistent with a signal peptide / signal-anchor. Notably, signal peptide prediction (SignalP) did not identify a clear cleavage site, indicating that the N-terminal hydrophobic stretch is likely acting as a signal-anchor rather than a cleavable signal peptide. This topology is compatible with many membrane receptors, in which the N-terminus remains associated with the membrane and contributes to extracellular domain architecture or receptor targeting. The GRAVY value (0.625) and the high aliphatic index indicated a hydrophobic, thermostable protein, consistent with a membrane-localized receptor. Domain and motif analyses revealed seven predicted transmembrane helices, a characteristic feature of G protein-coupled receptor (GPCR)-like proteins (Fig. 1A). Multiple post-translational modification sites, including potential phosphorylation and glycosylation motifs, were identified and may play roles in receptor activation or signal transduction. The subcellular localization predicted by WoLF PSORT strongly indicated plasma membrane targeting (score = 9), further supporting the potential role of HvPMTR1 as a receptor protein.Fig. 1(A) Membrane topology of HvPMTR1. Schematic representation of the seven predicted transmembrane helices and the orientation of the N- and C-termini. The N-terminal region displays a signal-anchor motif without a cleavage site (red). Cysteines involved in disulfide bridges are shown in orange. Conserved YYSEM[KR]DAGFF motif is shown as yellow. Topology was generated using the PROTTER web server. (B) A neighbor-joining phylogenetic tree was constructed from aligned amino acid sequences of PMTR1 genes from H. vulgare and various other plant species. The values at the nodes represent bootstrap support (%) from 1000 replicates. (A) Membrane topology of HvPMTR1. Schematic representation of the seven predicted transmembrane helices and the orientation of the N- and C-termini. The N-terminal region displays a signal-anchor motif without a cleavage site (red). Cysteines involved in disulfide bridges are shown in orange. Conserved YYSEM[KR]DAGFF motif is shown as yellow. Topology was generated using the PROTTER web server. (B) A neighbor-joining phylogenetic tree was constructed from aligned amino acid sequences of PMTR1 genes from H. vulgare and various other plant species. The values at the nodes represent bootstrap support (%) from 1000 replicates. Phylogenetic analysis reveals evolutionary conservation among monocots To gain insights into the evolutionary relationships of HvPMTR1, a phylogenetic tree was constructed from homologous protein sequences of selected monocot and dicot species (Supplementary 2). The analysis revealed two major clades corresponding to monocots and dicots, with J. regia serving as an outgroup. Within the monocot clade, H. vulgare grouped closely with B. distachyon and O. sativa, while Z. mays and S. bicolor formed a separate subcluster (Fig. 1B). This phylogenetic clustering pattern is consistent with the evolutionary divergence of the Poaceae family. The clear separation between monocots and dicots, and the strong bootstrap support (≥ 90%) for key branches, confirmed the reliability of the clustering. These findings suggest that HvPMTR1 shares a common ancestral origin with other known or putative melatonin receptors in grasses and maintains structural and functional conservation. Structural modeling and molecular docking of melatonin with HvPMTR1 The 3D model of HvPMTR1 was predicted by AlphaFold. Orientation analysis using the OPM and PPM3 servers positioned the protein stably within the plasma membrane, with an energy minimum (Emin) of approximately –67 kcal/mol, a membrane thickness of ~ 31 Å, and a tilt angle of ~ 15°. These parameters are within the expected range for functional GPCR-type receptors embedded in plant plasma membranes. Because signal peptide analysis indicated an N-terminal signal-anchor rather than a cleavable signal peptide, the docking experiments were performed on the full-length, membrane-oriented model (i.e., including the N-terminal segment). This approach preserves the native topology of the extracellular domain and avoids artefacts that might arise from the arbitrary removal of an anchor sequence. The molecular docking simulation of melatonin into the predicted binding pocket of HvPMTR1 yielded a binding affinity of − 6.145 kcal/mol, indicating a stable and thermodynamically favorable complex. Detailed interaction analysis (Fig. 2) reveals that the binding is primarily driven by hydrophobic forces and Van der Waals interactions rather than classical hydrogen bonding. The indole ring of melatonin is positioned in a deep hydrophobic cavity, where it undergoes significant π–π stacking interactions with the aromatic side chain of Phe133. Additionally, Tyr143 provides a hydrophobic shield at the distal end of the pocket, further stabilizing the ligand's orientation. While no formal hydrogen bonds were detected within a 3.3 Å cutoff, close polar contacts were observed between the acetyl group of melatonin and the side chain of Ser75. This suggests that the ligand's orientation is highly constrained by the pocket's shape (geometric fit). These findings suggest that HvPMTR1 binds melatonin via a "lock-and-key" mechanism, dominated by non-polar surface complementarity.Fig. 2Molecular docking analysis and binding mode of melatonin within the putative HvPMTR1 receptor. (Left) Overview of the receptor-ligand complex embedded in the membrane (indicated by yellow circles representing the phospholipid headgroups). The protein is shown in a blue surface representation, with the binding pocket highlighted in light blue. (Right) Magnified view of the binding site showing the detailed interaction network between melatonin (magenta sticks) and key residues, represented by dashed lines. (Bottom) Summary table of the critical amino acids forming the binding pocket and contributing to the stability of the complex. Molecular docking analysis and binding mode of melatonin within the putative HvPMTR1 receptor. (Left) Overview of the receptor-ligand complex embedded in the membrane (indicated by yellow circles representing the phospholipid headgroups). The protein is shown in a blue surface representation, with the binding pocket highlighted in light blue. (Right) Magnified view of the binding site showing the detailed interaction network between melatonin (magenta sticks) and key residues, represented by dashed lines. (Bottom) Summary table of the critical amino acids forming the binding pocket and contributing to the stability of the complex. Tunnel analysis To evaluate the ligand’s entry path, a geometric tunnel analysis was performed starting from the Phe133 residue, which was an important amino acid in the predicted pocket of HvPMTR1. A high-priority access channel was identified with a length of 10.3 Å (Fig. 3). The tunnel features a bottleneck radius of 2.2 Å, which is sterically compatible with the molecular dimensions of melatonin. Physicochemical profiling of the channel (supplementary table 3) showed a balanced hydropathy index (0.18) and moderate polarity (8.71), suggesting that the tunnel provides a favorable environment for the translocation of melatonin from the extracellular space to the internal binding site. These findings suggest that the predicted docking pose lies within a highly accessible, functionally relevant cavity of the HvPMTR1 receptor.Fig. 3Molecular architecture of the melatonin access channel in HvPMTR1. The receptor is shown in gray surface representation. The predicted binding pocket is highlighted in yellow and contains the docked melatonin molecule (shown in magenta sticks/spheres). The identified transport pathway, calculated from Phe133 to the protein surface, is visualized as a cyan transparent tunnel. The close spatial alignment between the tunnel terminus and the binding pocket illustrates the ligand’s entry route from the extracellular environment to the deep hydrophobic core of the receptor. Molecular architecture of the melatonin access channel in HvPMTR1. The receptor is shown in gray surface representation. The predicted binding pocket is highlighted in yellow and contains the docked melatonin molecule (shown in magenta sticks/spheres). The identified transport pathway, calculated from Phe133 to the protein surface, is visualized as a cyan transparent tunnel. The close spatial alignment between the tunnel terminus and the binding pocket illustrates the ligand’s entry route from the extracellular environment to the deep hydrophobic core of the receptor. Cis-regulatory element analysis of the HvPMTR1 promoter In silico analysis of the promoter region of the HvPMTR1 gene was performed using the PlantCARE database to identify potential cis-regulatory elements (CREs). The analysis revealed a rich array of putative elements associated with various biological processes and stress responses. These include core promoter elements (TATA-box and multiple CAAT-boxes), elements implicated in light responsiveness (e.g., G-box, Sp1, TCT-motif), hormone-responsive elements (such as ABRE for abscisic acid, TGACG/CGTCA-motifs for methyl jasmonate, and TGA-element for auxin), and several stress- and defense-related elements (including MYB, MYC, TC-rich repeats, WUN-motif, and W box). The presence of these elements suggests that environmental cues, phytohormones, and various abiotic/biotic stresses likely regulate HvPMTR1 expression. A summary of the key predicted cis-regulatory elements is provided in Table 2.Table 2Summary of key cis-regulatory elements identified in the HvPMTR1 promoter region.CategoryElement namePredicted function/biological roleCore promoter elementsTATA-boxCore promoter element around -30 relative to TSSCAAT-boxCommon cis-acting element in promoter and enhancer regionsLight responsivenessG-boxInvolved in light responsivenessTCT-motifPart of a light-responsive elementGATA-motifPart of a light-responsive elementSp1Light-responsive elementCCGTCC-boxRegulatory element (potentially light-related)Hormone responsivenessABA: ABREInvolved in abscisic acid (ABA) responsivenessMeJA: TGACG/CGTCAInvolved in methyl jasmonate (MeJA) responsivenessAuxin: TGA-elementAuxin-responsive elementAuxin/SA: as-1Activation sequence-1Stress and defense responsivenessDrought/Stress: MYBMYB transcription factor binding siteDrought/ABA: MYCMYC transcription factor binding siteDefense/Stress: TC-rich repeatsInvolved in defense and stress responsivenessWounding: WUN-motifWound-responsive elementGeneral Stress: STREStress-responsive element (osmotic, heat, etc.)Defense: W boxBinding site for WRKY transcription factorsOthercircadianCis-acting element involved in circadian control Table 2Summary of key cis-regulatory elements identified in the HvPMTR1 promoter region.CategoryElement namePredicted function/biological roleCore promoter elementsTATA-boxCore promoter element around -30 relative to TSSCAAT-boxCommon cis-acting element in promoter and enhancer regionsLight responsivenessG-boxInvolved in light responsivenessTCT-motifPart of a light-responsive elementGATA-motifPart of a light-responsive elementSp1Light-responsive elementCCGTCC-boxRegulatory element (potentially light-related)Hormone responsivenessABA: ABREInvolved in abscisic acid (ABA) responsivenessMeJA: TGACG/CGTCAInvolved in methyl jasmonate (MeJA) responsivenessAuxin: TGA-elementAuxin-responsive elementAuxin/SA: as-1Activation sequence-1Stress and defense responsivenessDrought/Stress: MYBMYB transcription factor binding siteDrought/ABA: MYCMYC transcription factor binding siteDefense/Stress: TC-rich repeatsInvolved in defense and stress responsivenessWounding: WUN-motifWound-responsive elementGeneral Stress: STREStress-responsive element (osmotic, heat, etc.)Defense: W boxBinding site for WRKY transcription factorsOthercircadianCis-acting element involved in circadian control Summary of key cis-regulatory elements identified in the HvPMTR1 promoter region. Expression profiles of HvPMTR1 Data from 10 separate RNA-Seq experiments (Supplementary 4) showed that, overall, HvPMTR1 is responsive to diverse stimuli. ABA treatment in roots induces a biphasic temporal response of HvPMTR1 expression (Fig. 4). In the early phase (1–6 h), statistically significant downregulation was observed at 1 to 6 h compared to control conditions. In the late phase (24 h), a significant upregulation (p-value ≤ 0.01) was observed, indicating a delayed induction mechanism. The gene exhibits a concentration-dependent, tissue-specific response to NaCl (Fig. 4). In shoots, 300 mM NaCl induces significant upregulation (p-value ≤ 0.05), whereas 150 mM NaCl shows no significant effect. In roots, 150 mM NaCl significantly downregulated HvPMTR1 (p-value ≤ 0.01), whereas a slight, non-statistically significant increase was observed at 300 mM NaCl. In shoot, HvPMTR1 was strongly upregulated by exposure to heavy metals, including Zn, Cu, and Cd (Fig. 4). In root, no significant changes were observed in response to Zn treatment. However, significant upregulation is observed in both Cu and Cd treatments (Fig. 4). Cold treatment in leaves does not significantly affect HvPMTR1 expression (Fig. 4), suggesting that HvPMTR1 is not responsive to low-temperature stress in this tissue.Fig. 4Gene expression levels of HvPMTR1 across various tissues/conditions. Values are presented as the average of Transcripts Per Million (TPM) ± SE. Green bars: ABA treatments, red bars: NaCl treatment, yellow bars: Heavy metals (Zn, Cu, and Cd) treatments, blue bars: Cold treatments. ns: non-significant, *: significant (p-value ≤ 0.05), **: significant (p-value ≤ 0.01). Gene expression levels of HvPMTR1 across various tissues/conditions. Values are presented as the average of Transcripts Per Million (TPM) ± SE. Green bars: ABA treatments, red bars: NaCl treatment, yellow bars: Heavy metals (Zn, Cu, and Cd) treatments, blue bars: Cold treatments. ns: non-significant, *: significant (p-value ≤ 0.05), **: significant (p-value ≤ 0.01). By integrating the promoter analysis findings (Section "Cis-regulatory element analysis of the HvPMTR1 promoter") with the gene expression results presented in this section, it can be inferred that this gene may respond to diverse stimuli, particularly abiotic stress signals. Experiment I: melatonin treatment Expression profiles of HvPMTR1 and HvGα1 genes Exogenous melatonin application induced distinct transcriptional responses in the two barley genotypes (Fig. 5). In the tolerant genotype (Nosrat), HvPMTR1 expression significantly increased within 1 h after treatment, reaching a maximum 2.7-fold relative to control, followed by a statistically significant decline at 3 h and 6 h. In contrast, its expression in Shirin significantly increased more modestly, becoming detectable only at 3 h (1.6-fold), with no significant changes at other time points. A similar but stronger trend was observed for HvGα1. Its expression in Nosrat increased significantly at 1 h (3.1-fold) and then gradually returned to basal levels, whereas only a minor, non-significant increase at 3 h was detected in Shirin. These results indicate a rapid and strong activation of the melatonin signaling cascade components, namely HvPMTR1–Gα1, in the tolerant genotype.Fig. 5Expression profiles of HvPMTR1 and HvGα1 genes in leaf tissue of two contrasting H. vulgare genotypes in response to melatonin treatment. Values are presented as means ± SE (n = 4) for each time point (Control, 1 h, 3 h, and 6 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Expression profiles of HvPMTR1 and HvGα1 genes in leaf tissue of two contrasting H. vulgare genotypes in response to melatonin treatment. Values are presented as means ± SE (n = 4) for each time point (Control, 1 h, 3 h, and 6 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Physiochemical responses to melatonin treatments: oxidative status and stomatal conductance Melatonin treatment enhanced the antioxidant response in both genotypes, but with a more pronounced effect in Nosrat. In Nosrat, CAT activity increased significantly at all time points. It peaked at 1 h after treatment (up to 146 U mg−1 protein), then returned to near-control values by 6 h. In contrast, in Shirin, the increase in CAT activity was weaker and delayed to 3 h (Fig. 6A). Consistent with this, hydrogen peroxide (H₂O₂) levels increased at the early time point (1 h) in both genotypes but remained higher in Shirin, suggesting less efficient ROS scavenging. The lower accumulation of H₂O₂ in Nosrat after 3–6 h reflects a more effective melatonin-mediated antioxidant activation (Fig. 6B). The amount of gs significantly decreased in Nosrat at 1 h, but there was no detectable change in Shirin at this time point (Fig. 6C). Moreover, gs values in Nosrat remained on lower levels than in Shirin throughout the observation period, consistent with melatonin-mediated partial stomatal closure, which may aid water conservation.Fig. 6Changes in physiochemical characteristics of two contrasting H. vulgare genotypes in response to melatonin treatment. (A) Catalase (CAT) activity, (B) H2O2 content, and (D) Stomatal conductance (gs). Values are presented as means ± SE (n = 4) for each time point (Control, 1 h, 3 h, and 6 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Changes in physiochemical characteristics of two contrasting H. vulgare genotypes in response to melatonin treatment. (A) Catalase (CAT) activity, (B) H2O2 content, and (D) Stomatal conductance (gs). Values are presented as means ± SE (n = 4) for each time point (Control, 1 h, 3 h, and 6 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Collectively, the results indicate that melatonin triggers a more rapid and stronger transcriptional and biochemical response in the tolerant genotype Nosrat, characterized by higher HvPMTR1 and HvGα1 expression, and enhanced antioxidant activity. In contrast, the sensitive genotype Shirin exhibited a weaker and delayed activation pattern. These findings support the hypothesis that HvPMTR1 is an active component of the melatonin signaling pathway in barley. Principal component analysis (PCA) To visualize the multivariate relationships among biochemical and molecular traits in response to melatonin treatment, PCA was performed using expression of HvPMTR1, expression of HvGα1, CAT, H₂O₂, and gs as active variables. The first two principal components (D1 and D2) explained 97.61% of the total variance (60.87% and 36.74%, respectively), indicating a robust model (Fig. 7). In the biplot, samples of the tolerant genotype (Nosrat) were clearly separated from those of the sensitive genotype (Shirin) along the second principal component (D2). The variables HvPMTR1, HvGα1, CAT, and H₂O₂ were positively correlated with each other and were high in Nosrat, indicating that these variables drive the separation of the genotypes. In contrast, gs showed a negative loading on D2, was negatively correlated with HvPMTR1, HvGα1, CAT, and H₂O₂, and aligned more with Shirin's response, consistent with the generally higher stomatal conductance observed in the sensitive genotype. D1 reflected the overall magnitude of the melatonin response, with positive association for HvPMTR1, HvGα1, CAT activity, and H₂O₂, and a negative association for gs. This pattern indicates that Nosrat plants exhibit coordinated upregulation of the putative melatonin receptor signaling (HvPMTR1 and HvGα1) and antioxidant capacity (CAT activity and controlled H₂O₂ levels). In contrast, Shirin plants exhibit higher gs, but lower activation of the melatonin-associated defense network.Fig. 7PCA analysis of some physiochemical characteristics of two contrasting H. vulgare genotypes in response to melatonin treatment. The positions of measured traits and genotypes are shown. The blue region corresponds to the salinity-tolerant genotype, and the red area corresponds to the salinity-susceptible genotype. PCA analysis of some physiochemical characteristics of two contrasting H. vulgare genotypes in response to melatonin treatment. The positions of measured traits and genotypes are shown. The blue region corresponds to the salinity-tolerant genotype, and the red area corresponds to the salinity-susceptible genotype. Experiment II: salinity stress Expression patterns of HvPMTR1 and HvGα1 under salinity Exposure to 150 mM NaCl caused significant transcriptional changes in both HvPMTR1 and HvGα1, with apparent genotypic differences. In the tolerant genotype (Nosrat), HvPMTR1 expression increased markedly after 12 h of salt exposure, peaked at 24 h (3.3-fold), and then declined by 48 h (Fig. 8). A similar temporal pattern was observed for HvGα1, which peaked (fivefold) at 24 h and had declined at 48 h. In contrast, the sensitive genotype (Shirin) exhibited only modest increases in HvPMTR1 and HvGα1 expression at 48 h, with less than 1.5-fold and 1.9-fold, respectively.Fig. 8Expression profiles of HvPMTR1 and HvGα1 genes in leaf tissue of two contrasting H. vulgare genotypes in response to salinity. Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Expression profiles of HvPMTR1 and HvGα1 genes in leaf tissue of two contrasting H. vulgare genotypes in response to salinity. Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). The coordinated upregulation of HvPMTR1 and HvGα1 in Nosrat suggests activation of melatonin receptor-mediated signaling and G-protein coupling as an early adaptive response to salt stress. Endogenous melatonin content and oxidative status Endogenous melatonin levels in leaves of the tolerant genotype were consistently higher than those of the sensitive genotype throughout the experiment, peaking at 24 h after salinity stress. In contrast, endogenous melatonin levels in the sensitive genotype showed no significant change until 24 h after salt treatment; however, increased significantly at 48 h (Fig. 9A).Fig. 9Changes in physiochemical characteristics of two contrasting H. vulgare genotypes in response to salinity. (A) Endogenous melatonin content, (B) Catalase activity, (C) H2O2 content, and (D) Malondialdehyde (MDA) content. Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Changes in physiochemical characteristics of two contrasting H. vulgare genotypes in response to salinity. (A) Endogenous melatonin content, (B) Catalase activity, (C) H2O2 content, and (D) Malondialdehyde (MDA) content. Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Different letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Salinity stress stimulated antioxidant enzyme activity, particularly in Nosrat. CAT activity increased progressively up to 48 h (172 U mg−1 protein). This increase was accompanied by a moderate accumulation of H₂O₂ and MDA, both of which were significantly lower than in Shirin. The sensitive genotype showed a smaller CAT response and higher H₂O₂ and MDA levels, indicating oxidative stress and less efficient detoxification (Fig. 9B–D). Physiological traits, photosynthetic responses, and Ionic balance Physiological traits were markedly influenced by salinity exposure (Table 3). Relative water content (RWC) declined gradually in both genotypes but remained consistently higher in Nosrat (90.2% at 24 h, respectively 81.7% at 48 h) than in Shirin (80.2%, respectively 69.3% for the same time points). SPAD values, net CO₂ assimilation (Aₙ), transpiration (E), and stomatal conductance (gs) also decreased under salinity treatment; only the magnitude of decline was smaller in Nosrat.Table 3Photosynthetic parameters, relative water content (RWC), and ion homeostasis in Nosrat and Shirin genotypes in response to salinity.SPAD valueAN (µmol CO2 m-2 s-1)E (mmol H2O m-2 s-1)gs (mmol H2O m-2 s-1)RWCK+/Na+NosratShirinNosratShirinNosratShirinNosratShirinNosratShirinNosratShirinNon-stress47.24a47.26a21.58a20.94a7.72a7.16a101.62a102.02a96.11a95.51a3.71a3.68a12 h46.63a44.08b20.37b19.11c7.04a6.21b95.37b95.03b93.24b92.36b3.52b3.37b24 h41.35c39.72d17.18d16.08e6.07c5.56d81.34d88.52c90.25c80.24d2.83c2.55d48 h34.81e31.07f.12.63f.10.39f.4.18e2.02f.61.48f.75.15e81.72d69.27e1.61e0.84f.Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Table 3Photosynthetic parameters, relative water content (RWC), and ion homeostasis in Nosrat and Shirin genotypes in response to salinity.SPAD valueAN (µmol CO2 m-2 s-1)E (mmol H2O m-2 s-1)gs (mmol H2O m-2 s-1)RWCK+/Na+NosratShirinNosratShirinNosratShirinNosratShirinNosratShirinNosratShirinNon-stress47.24a47.26a21.58a20.94a7.72a7.16a101.62a102.02a96.11a95.51a3.71a3.68a12 h46.63a44.08b20.37b19.11c7.04a6.21b95.37b95.03b93.24b92.36b3.52b3.37b24 h41.35c39.72d17.18d16.08e6.07c5.56d81.34d88.52c90.25c80.24d2.83c2.55d48 h34.81e31.07f.12.63f.10.39f.4.18e2.02f.61.48f.75.15e81.72d69.27e1.61e0.84f.Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). Photosynthetic parameters, relative water content (RWC), and ion homeostasis in Nosrat and Shirin genotypes in response to salinity. Values are presented as means ± SE (n = 4) for each time point (Non-stress (0 h), 12 h, 24 h, and 48 h). Letters indicate statistically significant differences among treatments based on Tukey’s HSD test (P ≤ 0.05). For example, Aₙ in Nosrat remained above 12 µmol CO₂ m−2 s−1 after 48 h, while in Shirin it dropped to nearly half that value. These results indicate that Nosrat retained higher photosynthetic efficiency and water status under salinity, consistent with its biochemical protection mechanisms. Salinity stress led to substantial Na⁺ accumulation in both genotypes (Table 3). Still, Nosrat maintained a significantly higher K⁺/Na⁺ ratio throughout the treatment (2.8-fold at 24 h and 1.6-fold at 48 h), whereas Shirin dropped below 1.0-fold by 48 h. This ionic homeostasis is a well-known indicator of salinity tolerance, reinforcing Nosrat's physiological superiority under salinity. Multivariate discrimination (PLS-DA) A PLS-DA model was applied to evaluate the contribution of molecular and physiological traits to genotypic discrimination. The first two latent variables (LV1 and LV2) accounted for more than 85% of the total variance, clearly separating the tolerant (Nosrat) and sensitive (Shirin) genotypes. Variable importance in projection (VIP) scores indicated that HvPMTR1 expression (VIP = 1.7), melatonin content (VIP = 1.46), HvGα1 expression (VIP = 1.31), and RWC (VIP = 1.09) were the strongest discriminators among genotypes. CAT activity and transpiration rate also contributed moderately (VIP ≈ 1.0), whereas gs, SPAD value, and K⁺/Na⁺ had a lower influence (VIP < 0.7) (Fig. 10A). These findings point toward involvement of the putative melatonin–HvPMTR1–HvGα1 signaling module in regulating downstream antioxidant and osmotic adjustments associated with salinity tolerance.Fig. 10Multivariate discrimination and classification tree analysis of evaluated molecular and physiological traits of two contrasting H. vulgare genotypes in response to salinity. (A) Variable Importance in Projection (VIP) scores derived from the PLS-DA model. Variables with VIP > 1.0 (indicated by the red dashed line) are considered the most significant contributors to genotypic discrimination, highlighting the melatonin signaling axis (HvPMTR1, Melatonin, and HvGα1) and Relative Water Content (RWC). (B) Classification tree based on the Chi-square Automatic Interaction Detector (CHAID) algorithm. The tree identifies HvPMTR1 expression as the primary splitter (Node 0) and RWC as the secondary splitter, thereby providing a hierarchical model to differentiate tolerant and sensitive genotypes with high purity and accuracy. Multivariate discrimination and classification tree analysis of evaluated molecular and physiological traits of two contrasting H. vulgare genotypes in response to salinity. (A) Variable Importance in Projection (VIP) scores derived from the PLS-DA model. Variables with VIP > 1.0 (indicated by the red dashed line) are considered the most significant contributors to genotypic discrimination, highlighting the melatonin signaling axis (HvPMTR1, Melatonin, and HvGα1) and Relative Water Content (RWC). (B) Classification tree based on the Chi-square Automatic Interaction Detector (CHAID) algorithm. The tree identifies HvPMTR1 expression as the primary splitter (Node 0) and RWC as the secondary splitter, thereby providing a hierarchical model to differentiate tolerant and sensitive genotypes with high purity and accuracy. Classification tree (CHAID analysis) To identify the most discriminating traits differentiating the genotypes, a CHAID classification tree was constructed from all measured variables. The first split occurred on HvPMTR1 expression (threshold = 1.87). Samples with HvPMTR1 ≤ 1.87 were predominantly classified as Shirin (75%), whereas those with higher HvPMTR1 expression were classified as Nosrat. A secondary split on RWC further improved classification accuracy, with purity levels exceeding 80%. Thus, HvPMTR1 expression and leaf water status were primary statistical discriminators that classify genotypes based on their salinity tolerance (Fig. 10B). Discussion Salinity tolerance in plants is a complex, multilevel trait that arises from the integration of early signal perception, intracellular transduction, and coordinated physiological adjustment35,36. Although melatonin has been repeatedly implicated as a key regulator of these processes, particularly through its antioxidant and stomatal effects, the molecular basis of melatonin perception in crops remains largely unresolved. Recent discoveries in Arabidopsis and rice suggest that melatonin signaling may be initiated through receptor-mediated mechanisms involving G-protein activation9,10; however, whether such a pathway operates in temperate cereals remains to be established. In this context, the present study provides integrative evidence, combining computational prediction with molecular, biochemical, and physiological analyses, to characterize a candidate melatonin-associated receptor in barley and associate its transcriptional dynamics with early salinity responses. Rather than proposing definitive receptor functionality, our findings delineate a coherent signaling framework in which the putative HvPMTR1, G-protein signaling, endogenous melatonin dynamics, and downstream stress-responsive traits are coordinately engaged, particularly in a salinity-tolerant genetic background. Structural and computational evidence supporting HvPMTR1 as a melatonin receptor The identification of HvPMTR1 as a candidate melatonin-associated receptor in barley is strongly supported by multiple independent layers of computational evidence converging on a GPCR-like signaling role. HvPMTR1 encodes a seven-transmembrane protein with predicted plasma membrane localization, a defining structural feature shared with the Arabidopsis melatonin receptor AtPMTR1/CAND2 and its rice ortholog OsPMTR19,10. The presence of a hydrophobic N-terminal signal-anchor, rather than a cleavable signal peptide, is consistent with membrane-embedded receptor architecture and supports a topology in which the N-terminus contributes to extracellular ligand perception. This feature has also been reported for other plant GPCR-like proteins and suggests conservation of membrane insertion mechanisms across monocot and dicot melatonin receptors37,38. Phylogenetic analysis further reinforces this structural inference by placing HvPMTR1 firmly within the monocot PMTR1 clade, clustering closely with Brachypodium distachyon and O. sativa orthologs. The clear separation between monocot and dicot PMTR1 proteins, supported by high bootstrap values, indicates that HvPMTR1 likely evolved from a common ancestral receptor and retained core functional features despite lineage-specific divergence. Such evolutionary conservation argues against HvPMTR1 representing a nonspecific membrane protein and instead supports its classification as a bona fide member of the PMTR1 receptor family. Beyond sequence similarity, structural modeling and membrane-orientation analyses provide functional insight into the receptor’s potential to bind melatonin. The AlphaFold-predicted three-dimensional structure of HvPMTR1 adopts a stable conformation within the lipid bilayer, with membrane thickness, tilt angle, and energetic parameters comparable to experimentally validated GPCR-type receptors. Importantly, molecular docking simulations revealed a thermodynamically favorable interaction between HvPMTR1 and melatonin, characterized by a binding affinity consistent with low-micromolar ligand–receptor interactions reported for plant signaling molecules39,40. The binding mode is dominated by hydrophobic and π–π stacking interactions, particularly involving Phe133 and Tyr143, which stabilize the indole ring of melatonin within a deep hydrophobic cavity. This interaction pattern suggests that HvPMTR1 is predicted to accommodate melatonin through a conserved structural logic41. While structural studies on AtPMTR1 and OsPMTR1 have primarily utilized molecular docking to define binding affinities, recent advances in computational structural biology have highlighted the importance of compatible pathways that facilitate ligand entry into the hydrophobic core of transmembrane receptors42,43. For instance, in other plant GPCR-like proteins, the presence of specific gating tunnels has been shown to regulate the accessibility of small signaling molecules to their respective binding pockets44. However, detailed profiling of such access channels for PMTR1 orthologs in temperate cereals remains unexplored. Accordingly, the functional plausibility of melatonin binding site is further strengthened by tunnel analysis, which identified a sterically and physicochemically compatible access channel connecting the extracellular environment to the predicted ligand-binding pocket. The dimensions, polarity, and hydrophobicity of this tunnel are well-suited to melatonin translocation. Unlike previous reports on AtPMTR1 and OsPMTR1, which primarily focused on static docking, our tunneling simulation reveals the dynamic pathway through which melatonin reaches the binding pocket. The specific physicochemical profile of this channel in barley suggests a refined evolutionary adaptation of the receptor to efficiently perceive ligands under the osmotic constraints of temperate cereal environments. Together, these results suggest that the putative HvPMTR1 possesses both the structural features required for melatonin recognition and a feasible pathway for ligand entry, a critical yet often overlooked aspect of in silico receptor validation. While computational approaches alone cannot definitively establish receptor functionality, the strength of the present evidence lies in its internal consistency and alignment with experimentally validated melatonin receptors in other plant species9,10,45. When considered alongside the rapid transcriptional activation of HvPMTR1 following melatonin treatment and salinity stress, the structural data provide a compelling rationale for positioning HvPMTR1 as a functional component of receptor-mediated melatonin signaling in barley. Thus, HvPMTR1 emerges as a potent candidate receptor that bridges evolutionary conservation, structural feasibility, and stress-responsive regulation, laying the groundwork for future functional validation through genetic and biochemical approaches. Coordinated early signaling links putative HvPMTR1 regulation to G-protein responses, ROS homeostasis, and stomatal dynamics Short-term melatonin application revealed rapid, genotype-dependent induction of receptor-associated signaling components in barley, providing further molecular and physiological evidence for the involvement of the candidate HvPMTR1 in early melatonin perception. In the salinity-tolerant genotype, HvPMTR1 transcripts accumulated within 1 h of melatonin treatment, followed by induction of the heterotrimeric G-protein α-subunit gene HvGα1. The rapid (1 h) coordinated induction of HvPMTR1 and HvGα1 in the tolerant genotype highlights a highly primed signaling axis that has not been characterized with such temporal resolution in other cereals. This ready-to-act state of the melatonin perception machinery in barley differentiates our findings from studies that focus on late-stage stress responses, providing a new perspective on how early perception dictates long-term survival. The tight temporal coupling between receptor and G-protein activation is a hallmark of receptor-mediated signaling and mirrors the PMTR1–GPA1 module described in Arabidopsis and rice, where melatonin binding initiates downstream signal transduction within minutes to hours9,10. In contrast, the delayed and attenuated transcriptional responses observed in the sensitive genotype suggest a reduced capacity to efficiently perceive or amplify the melatonin signal, rather than an absence of melatonin responsiveness per se. The transcriptional activation of HvPMTR1 and HvGα1 was accompanied by coordinated changes in redox status, consistent with controlled ROS signaling rather than generalized oxidative stress. In the tolerant genotype, melatonin induced a transient increase in antioxidant capacity, as reflected by enhanced CAT activity, thereby effectively constraining H₂O₂ accumulation at later time points. This pattern aligns with the emerging view that melatonin-triggered ROS production functions as a signaling intermediate that must be tightly regulated to avoid cellular damage46,47. The comparatively higher H₂O₂ levels and weaker antioxidant induction in the sensitive genotype indicate a decoupling between signal initiation and ROS detoxification, potentially compromising downstream adaptive responses. Stomatal conductance dynamics further support the operation of a conserved melatonin signaling module in barley. The modest but rapid reduction in stomatal conductance observed in the tolerant genotype following melatonin application is consistent with PMTR1-dependent regulation of guard cell behavior reported in Arabidopsis9. In that system, melatonin perception via PMTR1 activates G-protein signaling and ROS-mediated Ca2⁺ influx, culminating in partial stomatal closure. Although direct measurements of Ca2⁺ fluxes were not performed in the present study, the coordinated regulation of HvPMTR1, HvGα1, ROS homeostasis, and stomatal conductance strongly suggests that a similar signaling logic operates in barley. The absence of a comparable stomatal response in the sensitive genotype further highlights the potential importance of efficient signal transduction over ligand presence alone. Multivariate analysis reinforced this conceptual interpretation by demonstrating that HvPMTR1 expression, HvGα1 expression, antioxidant activity, and ROS levels form a tightly associated trait module that discriminates tolerant from sensitive genotypes during early melatonin responses. The strong separation of genotypes along principal components dominated by these variables indicates that melatonin-induced signaling operates as an integrated network rather than as isolated molecular or physiological events. Collectively, these findings provide robust integrative evidence that HvPMTR1 participates in an early melatonin signaling cascade in barley, associating its expression dynamics with potential G-protein pathways, redox control, and stomatal adjustment; processes that are central to the plant’s capacity to anticipate and mitigate osmotic and ionic stress36,48,49. Receptor-associated melatonin signaling integrates receptor activation with physiological resilience Under salinity stress, the coordinated regulation of endogenous melatonin levels and HvPMTR1-associated signaling components reveals a tightly integrated mechanism underlying genotypic differences in salt tolerance. The salinity-tolerant genotype exhibited earlier and greater accumulation of endogenous melatonin, concomitant with pronounced transcriptional activation of HvPMTR1 and HvGα1 during the first 24 h of salt exposure. This temporal alignment suggests that increased melatonin biosynthesis is rapidly translated into receptor-mediated signal transduction, rather than acting solely through passive antioxidant buffering. In contrast, the delayed elevation of melatonin in the sensitive genotype was not matched by a comparable induction of HvPMTR1 or HvGα1, indicating that late melatonin accumulation alone is insufficient to trigger an effective adaptive response. The functional consequences of this differential signaling capacity were evident in the regulation of redox homeostasis and membrane integrity. In the tolerant genotype, sustained CAT activation constrained H₂O₂ accumulation and limited lipid peroxidation, as reflected by lower MDA levels throughout the salinity treatment. This pattern is consistent with a model in which melatonin perception through HvPMTR1 initiates controlled ROS signaling that primes antioxidant defenses while preventing oxidative damage. By contrast, the sensitive genotype showed weaker antioxidant induction and excessive ROS accumulation, resulting in increased membrane damage and compromised cellular function. Physiological performance under salinity further supports the potential role of HvPMTR1-mediated melatonin signaling in stress tolerance. The tolerant genotype maintained higher RWC, photosynthetic capacity, and stomatal regulation compared with the sensitive genotype, despite similar external salinity exposure. These traits are closely linked to early stomatal adjustment and efficient osmotic regulation, processes that are known to be influenced by melatonin signaling in other plant systems50–52. Notably, the ability of the tolerant genotype to partially restrict stomatal conductance while sustaining CO₂ assimilation suggests an optimized balance between water conservation and carbon gain, a hallmark of adaptive stress responses53. Ion homeostasis provided additional evidence linking receptor-associated signaling to whole-plant salinity tolerance. The higher K⁺/Na⁺ ratio observed in the tolerant genotype throughout salinity exposure indicates superior control of ionic balance, a critical determinant of cellular viability under salinity stress29,54,55. While melatonin has been reported to modulate ion transporter expression and activity56, the present results suggest that such effects are likely embedded within a broader receptor-mediated signaling network that coordinates redox status, water relations, and ion transport rather than acting through isolated pathways. The integrative nature of this signaling framework was most clearly captured by multivariate analyses. Both PLS-DA and CHAID classification consistently identified HvPMTR1 expression, endogenous melatonin content, and HvGα1 expression as the most informative variables discriminating tolerant and sensitive genotypes, with relative water content emerging as a key downstream physiological outcome. The application of the PLS-DA and CHAID algorithms provides a robust hierarchical classification model that illuminates the predictive structure of early salinity responses. By identifying HvPMTR1 expression as the primary statistical discriminator for salinity tolerance in this hierarchical framework, we provide empirical evidence that the coordinated efficiency of receptor-associated perception, rather than the mere abundance of melatonin or antioxidant capacity, is an important regulatory component linked to the salinity-tolerance trait. This hierarchical positioning of HvPMTR1 at the apex of the statistical classification model offers a comprehensive conceptual advance over existing linear response models in monocots. Together, these findings demonstrate that salinity tolerance in barley is partly associated with the early induction and coordinated responses of endogenous melatonin production alongside HvPMTR1-dependent signaling traits. Rather than reflecting differences in melatonin abundance alone, the contrasting responses of the two genotypes highlight the importance of candidate receptor-mediated perception pathways and downstream signal integration in determining the effectiveness of melatonin-driven stress adaptation. It is important to emphasize that this study specifically focuses on the early signaling window to capture the initial perception and rapid adaptive responses of barley. Our experimental design, targeting the 0–6 h window for melatonin treatment and the 12–48 h window for salinity stress, was specifically chosen to isolate upstream regulatory events from late-stage stress damage. The rapid induction of the HvPMTR1-HvGα1 axis within this timeframe suggests its involvement in an early-warning system that primes the plant's antioxidant and physiological machinery before extensive cellular compromise occurs. Accordingly, based on integrating structural predictions, short-term signaling responses, and salinity-induced physiological outcomes, we propose a receptor-mediated conceptual model of melatonin signaling in barley that links early perception to whole-plant stress adaptation. In this proposed framework, endogenous or exogenously supplied melatonin is predicted to be perceived at the plasma membrane by the candidate receptor HvPMTR1, which subsequently interfaces with HvGα1. This coordinated module is suggested to initiate a tightly regulated ROS signaling phase that activates antioxidant defenses, modulates stomatal behavior, and ultimately improves water status, sustains photosynthesis, and maintains ionic homeostasis under salinity stress. The efficiency of this signaling cascade, rather than melatonin abundance alone, emerges as a prominent regulatory component distinguishing salt-tolerant and salt-sensitive genotypes. Notably, despite similarities with OsPMTR1 in rice, the barley HvPMTR1 module exhibits distinct transcriptional signatures and a specific structural access channel that may reflect the unique physiological requirements of temperate grasses. Our findings suggest that the melatonin-G-protein interface in barley may have adapted to facilitate a more rapid, early-warning signal, which is crucial for surviving the sudden osmotic shifts typical of its cultivation environments8. Such an architecture allows plants to rapidly translate environmental cues into adaptive responses while maintaining tight control over potentially damaging intermediates such as ROS. Despite the strength of the integrative evidence presented here, several limitations warrant consideration. The putative receptor-like role of HvPMTR1 is inferred from structural compatibility, evolutionary conservation, and coordinated downstream responses. Nevertheless, direct biochemical validation of melatonin-binding kinetics and functional genetic manipulation of HvPMTR1 remain necessary to establish causality definitively. Future studies employing loss- and gain-of-function approaches, ligand-binding assays, and real-time monitoring of Ca2⁺ and ROS dynamics in guard cells will be essential to fully resolve the downstream signaling events following HvPMTR1 activation. In addition, exploring allelic variation of HvPMTR1 across diverse barley germplasm may reveal naturally occurring variants associated with enhanced stress tolerance. Furthermore, while we focused on the early signaling window (0–48 h) to capture the primary perception events, future long-term studies are needed to determine how this early perception dictates grain yield and final biomass under sustained field-level salinity stress. In conclusion, this study provides evidence for a melatonin-associated receptor-mediated signaling pathway in barley and positions HvPMTR1 as a key component linking melatonin perception to early salinity signaling. By bridging computational structural biology with molecular and physiological analyses across contrasting genotypes, our findings establish a mechanistic framework for understanding melatonin-driven stress adaptation in monocot crops and open new avenues for exploiting melatonin signaling pathways to enhance crop resilience. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1. Supplementary Material 1. Supplementary Material 2. Supplementary Material 2. Supplementary Material 3. Supplementary Material 3. Supplementary Material 4. Supplementary Material 4.