Water deficit is a major constraint on pepper (Capsicum annuum) yield, yet the genetic architecture of reproductive-stage drought tolerance remains poorly resolved. We phenotyped a Balkan C. annuum diversity panel (n = 133) and an interspecific backcross inbred line (BIL) population (n = 76) under well-watered (WW) and water-stress (WS) conditions. WS was applied from anthesis of the second truss as a stepwise reduction in irrigation volume relative to WW (30% for 7 days, then 60% thereafter), maintained for 90 days across the reproductive period. We assessed yield components, soluble solids, and stress-tolerance (STI) and stress-susceptibility (SSI) indices. Genome-wide association study (GWAS) identified 104 SNP-trait associations (P < 1×10-5), and QTL mapping detected 38 significant QTLs (1,000 permutations, α = 0.01), with the QTL intervals defined at LOD ≥ 8. Integrating GWAS and QTL mapping under WS revealed overlapping loci on chromosomes 5 and 6, harboring two consensus intergenic SNPs associated with yield components and soluble solids. Haplotype analysis linked chromosome 5 alleles to higher fruit number and soluble solids. At chromosome 6, the G allele at SNP 6_28348737 was enriched in tolerant lines for fruit number. These regions harbor candidate genes for reproductive development and stress response, including GREEN RIPE-LIKE1 (GRL1), CYP77A19, Endoglucanase-like, and FLOWERING PROMOTING FACTOR 1 (FPF1), possibly through cis-regulatory variation. Together, these results advance understanding of the genetic basis of pepper yield under drought and identify candidate breeding markers.

This study decodes how cultivar (cv.), tissue type, and thermal processing affect sweet potato metabolome and antioxidant capacity by integrating UHPLC-HRMS/MS, GNPS molecular networking, UV-Vis spectroscopy, chemometrics, and in vitro assays. A total of 374 metabolites were putatively annotated, with several previously unreported phenolic acids, coumarins, lignans, alkaloids, and apocarotenoids. Multivariate data analyses showed peel-pulp differentiation as the dominant factor, followed by flesh-color-driven cv. segregation, with orange pulps enriched in carotenoids, versus peels' richness in phenolic acids, anthocyanins, and lipids. Peels displayed superior DPPH and FRAP activity, whereas pulps exhibited stronger β-carotene bleaching inhibition. Thermal processing revealed dry-heat methods retained phenolics, carotenoids, and antioxidant activity better than boiling or microwaving in Gendawy cv. pulp. UV-Vis profiling distinguished cvs., tissues, and correlated with phenolics and carotenoids. Pearson correlations highlighted apocarotenoids and hydroxycinnamates as major contributors to antioxidant effects, guiding sweet potato cvs. selection for food and nutraceutical applications.

Common bean (Phaseolus vulgaris L.) is one of the most important grain legumes for direct human consumption. Currently, 60% of its production is estimated to be at risk due to drought. However, the genetic basis of common bean's drought resistance is poorly understood. To this end, we assessed the genetic architecture of drought-responsive changes in a whole genome-sequenced population of 218 common bean accessions. Using multi-omics-based trait evaluation, including lipidomics, photosynthetic and agronomic traits, followed by multi-omics genome-wide association studies (moGWAS), yielded in the detection of a myriad of moQTL for photosynthesis and yield, as well as the levels of various lipids. QTL associated with glycolipids, which are integral to photosynthesis, since they constitute the major membrane components of chloroplasts, were identified. In addition, we molecularly validated several lipid-related candidate genes via P. vulgaris hairy root transformation as well as transient expression in tobacco. In particular, a lipoxygenase and an allene oxide synthase were identified as explaining the variation in triacylglycerol by oxylipin production. These data provide a blueprint for multi-omics-assisted improvement of crop water stress resilience. ### Competing Interest Statement The authors have declared no competing interest.

Optimizing nutrient use efficiency while maintaining fruit quality is a central challenge for sustainable tomato breeding. This study leverages 53 chromosome segment substitution lines (CSSLs) derived from a cross between cultivated tomato and the wild relative Solanum habrochaites (LA1777) to explore the genetic and metabolic basis of adaptation to varying nutrient availability. The CSSLs were evaluated under high-input (HI) and low-input (LI) regimes and characterized for their agro-qualitative performances and primary metabolome profiles. Gas chromatography–mass spectrometry identified 63 compounds revealing a significant metabolic variability driven by input conditions. The plastic response of metabolites highlighted a shift in the accumulation of fruit's carbon- and nitrogen-related compounds: HI conditions enhanced amino acids levels, whereas the LI regime prioritized carbon-rich compounds, resulting in increased sugar content. Using a 51K SNP array, the population was mapped at high resolution across 2614 BINs covering 68% of the genome. Metabolic QTL (mQTL) analysis using stepwise regression and 1000 permutations identified 81 high-confidence loci, including pleiotropic regions on chromosomes 1, 3, and 4 that were associated with amino and organic acid metabolism, corresponding to conserved regions across wild tomato species. Analysis of candidate genes revealed that specific genomic regions responsive to LI govern abiotic stress resistance, while stable hotspots across nutrient regimes highlight the presence of highly canalized genetic hubs. The integration of mapping data with previously identified mQTL and transcriptomic studies validates the biological relevance of these identified loci in coordinating nutrient stress responses and fruit metabolic adaptation. Several lines demonstrated superior metabolite accumulation under stress and varying degrees of plasticity, making them ideal candidates for breeding. These findings provide a high-resolution framework for marker-assisted selection to develop climate-resilient, high-quality tomato cultivars optimized for sustainable agriculture.

Daturodendron absconditum gen. & sp. nov., a monotypic genus of Datureae (Solanaceae), is described based on taxonomic, phylotranscriptomic, and metabolomic evidence. Phylogenetic analyses place Daturodendron as sister to all other members of Datureae, which until now included Datura, Brugmansia, and Trompettia. The tribe has long been a focus of evolutionary studies due to contrasting traits (herbaceous vs. arborescent habit, erect vs. pendent flowers, fleshy indehiscent vs. dry dehiscent fruits) and its characteristic production of tropane alkaloids (TAs), specialized defense compounds with major ethnomedical relevance. Endemic to the Andean regions of Colombia and Peru, Daturodendron is distinguished by its arborescent habit, erect flowers, circumscissile calyx, leathery corollas, and suborbicular to reniform seeds. In D. absconditum, we detected two key TAs, scopolamine and hyoscyamine, consistent with other TA-producing members of Datureae. Ancestral state reconstruction indicates that TA production was likely ancestral in the tribe. Together, these results support the recognition of Daturodendron as a distinct genus within Datureae.

Abstract Tomato wild relatives are valuable genetic resources for trait discovery and understanding the genetic basis of fruit metabolism and quality. Yet, only a fraction of naturally occurring variation has been exploited. Here, we performed metabolite profiling of two large Backcross Inbred Line populations derived from crosses between the wild species S. pennellii accession LA5240 (Lost) and cultivated genotypes LEA (determinate) and TOP (indeterminate), including ∼1400 and ∼500 lines, respectively. High-resolution mapping identified enormous metabolic quantitative trait loci (mQTL), including a new locus on chromosome 12 associated with fruit sucrose accumulation that harbours INVERTASE INHIBITOR 3 ( SlINVINH3 ) protein. Comparative analysis indicated that SlINVINH3 is highly expressed in wild S. pennellii 0716 fruit, whereas a six-amino acid deletion is present in its coding sequence compared with S.pennellii LA5240 and S. lycopersicum . We further demonstrated that in SlINVINH3 -overexpressing tomato plants, only the S. pennellii LA5240 allele led to increased sucrose, accompanied by reduced fructose and glucose levels. Furthermore, the large population size enabled us to assess the epistatic interactions, with approximately 40% of interactions being more-than-additive and 60% less-than-additive. Our results demonstrate the power of permanent exotic populations to reveal hidden metabolic diversity and provide an approach for improving fruit quality through targeted breeding and metabolic engineering.

Phaseolus vulgaris (common bean) is one of the most economically important members of the Fabaceae family and a crop of high nutritional value. Seed and leaf morphological traits are key determinants of yield, shaped by genetic and environmental factors. In this study, we characterized a diversity panel of 434 P. vulgaris accessions for morphological and colorimetric traits. Using genome-wide association studies based on both single-nucleotide polymorphisms and structural variants, we identified 73 high-confidence quantitative trait locus (QTL). In total, we present 114 candidate genes within 25 quantitative trait loci linked to seed and leaf morphology and color, including an OVATE Family Protein 5 (PvOFP5), which was functionally validated as a key regulator of seed size in common bean. We employed a heterologous approach by overexpressing the inferior (Mesoamerican) and superior (Andean) alleles of PvOFP5 in Arabidopsis thaliana wild-type and knockout lines, and confirmed the key role of PvOFP5 in determining seed area. This work provides a comprehensive atlas of genetic associations for bean morphology and color, and highlights PvOFP5 as a promising target for marker-assisted breeding aimed at optimizing seed size.

Drought-induced senescence is a major cause of maize yield loss. While biostimulant priming improves stress tolerance, its molecular basis is unclear. Here we demonstrate that priming maize with the plant-derived biostimulant AgriPrime Stimulus (APS) delays drought-induced leaf senescence at reproductive-stage, resulting in improved cob weight and yield. Integrated physiological, transcriptomic, metabolomic, and phytohormone analyses revealed that APS priming preserves source leaf functionality by maintaining key metabolic processes. APS-primed drought-stressed leaves showed enrichment of photosynthesis-related genes and elevated levels of tricarboxylic acid cycle intermediates, indicating maintained carbon metabolism. APS priming also strengthened cell wall through the induction of genes involved in cellulose, hemicellulose, pectin, cutin, and wax biosynthesis, with increased structural metabolites such as xylose, mannose, and galactonic acid. Delayed senescence was further supported by enhanced redox homeostasis, with upregulation of antioxidant-related genes including superoxide dismutase (SOD3), peroxidases (PRXs), glutathione S-transferases (GSTs), and ascorbate-associated genes (BX13), together with increased levels of protective metabolites such as proline, trehalose, and myo-inositol. In parallel, APS priming suppressed proteolysis and senescence-associated genes (NYC1, NYE1, SAG39, NAC042). Integration of phytohormone and transcriptomic data further revealed maintained growth-promoting hormones alongside reduced abscisic acid and ethylene biosynthesis. Consistent with this reduced catabolic state, APS-primed leaves accumulated amino acids linked to growth, while unprimed drought-stressed leaves accumulated amino acids related to protein degradation. Collectively, these findings show that APS priming preserves source-sink relationships during drought by maintaining leaf longevity, and strengthening sink support, which improves cob weight under water deficit.

Quinoa (Chenopodium quinoa) is a nutrient-rich pseudocereal with diverse specialized metabolites, yet the genetic basis of this metabolic diversity is poorly understood. Here we integrate whole-genome sequencing and multi-tissue metabolic profiling of 603 quinoa accessions. We detected 4688 metabolic features and identified over 1000 metabolites in seeds, leaves, and roots. Using multi-tissue genome-wide association, we mapped the genetic architecture of quinoa metabolome by identifying 584 quantitative trait loci (QTL) and prioritized 219 candidate genes across 58 major QTL governing saponin, betalain, and flavonoid biosynthesis. Moreover, we constructed a drought-responsive multi-omics regulatory network and uncovered additional key genes involved in quinoa stress signaling and metabolic pathways. Finally, we cloned and functional validated the roles of cytochrome P450 76AD1 (CYP76AD1) in betalamate accumulation, UDP-glycosyltransferase (UGT91C1) in flavonoid glycosylation, and CYP72A154 and soyasapogenol B glucuronide galactosyltransferase in saponin biosynthesis. This multi-omic framework provides a high-resolution map of the quinoa metabolome and a foundation for breeding nutrient-rich and stress-resilient quinoa cultivars.

Extensive selection for yield and disease resistance during tomato crop improvement has led to flavor loss in modern commercial tomatoes in comparison with heirloom varieties. To investigate the chemical and genetic architecture of tomato flavor through both domestication and improvement, we analyzed flavor-related chemicals across 558 globally collected accessions comprising wild relatives, semidomesticated and domesticated populations, including uncharacterized Latin American accessions. Key flavor volatiles exhibit major differences across accessions. A genome-wide association study was used to detect associations between genetic loci and flavor-related chemical contents, including sugars, acids, and volatiles. Multiple genetic loci linked to known genes encoding flavor metabolism enzymes, as well as many new loci, were identified. Among the newly identified loci, a gene encoding a previously uncharacterized lipase ( Sl-LIP100 ) was experimentally proven to have an important role in synthesis of lipid-derived flavor volatiles. This enzyme is responsible for synthesis of several important five- and six-carbon flavor volatiles. In sum, this study provides chemical and genetic insights into the evolution of tomato flavor during domestication and subsequent improvement, identifying the complexity of genetic control of fruit flavor chemicals as well as a number of new alleles that can be used to improve the contents of flavor-linked chemicals.

Plant lineages produce distinct alkaloids, indicating the presence of evolutionary mechanisms that prevent the simultaneous metabolism of multiple alkaloid types within the same species. In the Solanaceae family, species produce either tropane alkaloids (TAs) or steroidal glycoalkaloids (SGAs). To investigate the genetic causes underlying this mutually exclusive distribution, we integrated genomic, transcriptomic, and metabolomic data across tens of species representing most Solanaceae clades, focusing on three potential mechanisms: (1) loss of biosynthetic genes, (2) genomic clustering, and (3) regulatory changes. Ancestral trait reconstruction suggests that both pathways were likely present in early Solanaceae lineages but were differentially lost across clades, resulting in transitions between SGA and TA biosynthesis. Notably, SGA and TA genes exhibit contrasting patterns of genomic distribution: SGA genes are organized in large, dynamic biosynthetic clusters, while TA genes are dispersed across the genome. More importantly, SGA genes are ancient and phylogenetically widespread, but are not expressed in clades that produce TAs, implying that a regulatory switch was responsible for the simultaneous silencing of SGA genes in these clades. Finally, we observed the loss and genetic erosion of late-acting SGA and TA genes in clades where they are not expressed. Overall, our study suggests that genomic clustering, regulatory changes, and gene duplication/loss underlie the mutually exclusive distribution of alkaloids in the Solanaceae.

C4 plants have traditionally been classified into NADP-malic enzyme (NADP-ME), NAD-malic enzyme (NAD-ME), and phosphoenolpyruvate carboxykinase (PEPCK) subtypes based on the predominant C4 acid decarboxylating enzyme. To investigate the relative contributions of malate and aspartate to C4 pathway fluxes in each subtype, we performed 13CO2 pulse-chase labelling experiments on four C4 grass species: Zea mays and Setaria viridis (NADP-ME), Panicum miliaceum (NAD-ME), and Megathyrsus maximus (PEPCK). Only a proportion (8-50%) of the total malate pool in the leaves is photosynthetically active, whereas essentially all of the aspartate pool is photosynthetically active. Estimates of metabolic fluxes indicate that approximately two-thirds of the C4 pathway flux is via malate in Z. mays and the remaining third via aspartate, while in S. viridis 50% of the flux is via malate and 50% via aspartate. In P. miliaceum and M. maximus, 91% and 85% of the flux is via aspartate and the remaining 9% and 15% via malate, respectively. The results demonstrate the feasibility of using non-radioactive 13CO2 in pulse-chase labelling experiments to study C4 photosynthesis and to detect C4 pathway fluxes in C3 plants engineered to perform C4 photosynthesis.

Maize primary metabolism drives complex agronomic traits, yet its genetic regulation remains difficult to resolve. Here we integrated genomic, transcriptomic and metabolomic data from 1,404 maize progenies derived from 24 diverse founders to dissect the genetic architecture of primary metabolism. We constructed a high-confidence regulatory network that resolved causal genes underlying metabolic quantitative trait loci and successfully identified targets for improving maize nutritional quality. This systems-level framework further prioritized ZmAVT1A-1, encoding a putative amino acid transporter, as a key regulator of amino acid accumulation. Natural variation and transgenic analyses showed that ZmAVT1A-1 modulates nitrogen partitioning between vegetative tissues and kernels, revealing pleiotropic effects on agronomic traits. These findings illustrated the intricate trade-offs inherent in metabolic regulation. Together, our study provides a comprehensive multiomics resource for decoding metabolic networks and underscores the necessity of a systems approach to navigate the pleiotropic nature of crop improvement targets.

Plant phosphoenolpyruvate carboxylases (PEPCs) are ubiquitously expressed as cytosolic Class-1 PEPC homotetramers composed of 107 kDa plant-type PEPC (PTPC) subunits that are highly sensitive to allosteric inhibition by malate. Class-2 PEPC heterooctameric complexes that are desensitized to malate inhibition also exist in certain sink tissues due to the interaction of a Class-1 PEPC with unrelated 118 kDa bacterial-type PEPC (BTPC) polypeptides. Class-2 PEPCs dynamically associate with the mitochondrial outer envelope and have been hypothesized to support sustained anaplerotic flux and respiratory CO₂ refixation in malate-rich sink tissues, including immature tomato fruit. The current study generated CRISPR-Cas9-edited tomato lines with targeted disruption of the BTPC gene and investigated the impact on fruit development, metabolism, and transcriptional regulation. Immunoblotting and co-immunoprecipitation confirmed the absence of BTPC polypeptides and Class-2 PEPC complexes in the edited lines. Fruits from the edited plants were 25% smaller and 40% lighter and required up to 10 additional days to complete ripening compared to the WT. Metabolomic analysis across ripening stages revealed substantial reductions in malate and citrate, with elevated sugars and amino acids, indicating reprogrammed carbon flux. RNA-seq data showed downregulation of genes for cell wall remodeling, sugar transport, and ethylene-responsive transcription factors. These results provide direct evidence that BTPC is essential for organic acid balance, sugar metabolism, and ripening regulation in tomato. Its absence perturbs metabolic homeostasis and developmental progression, positioning BTPC as a strategic target for enhancing fruit quality traits through genetic engineering.

Seasonal temperature increases in autumn and winter can significantly disrupt frost tolerance in plants. This inducible trait is acquired during cold acclimation, reduced or lost during de-acclimation, and potentially restored during re-acclimation. The functionality of cellular membranes, closely associated with lipid composition, plays a crucial adaptive role in enabling plants to withstand low-temperature stress. However, studies focusing on effects of de-acclimation and re-acclimation on frost tolerance, particularly in forage grasses, remain limited. Here, we assessed frost tolerance in high (HFT) and low frost tolerant (LFT) genotypes of Festuca arundinacea following cold acclimation, de-acclimation, and re-acclimation, based on the assessment of their cellular membranes integrity as reflected by the parameter describing the temperature causing 50% level of electrolyte leakage from cells. Analyses of membrane integrity and lipid composition during cold acclimation revealed that the HFT genotype maintained greater membrane stability through extensive remodeling of glycerolipids, including increased lipid unsaturation, particularly in phosphatidylethanolamine (PE), preserved a balanced monogalactosyldiacylglycerol/digalactosyldiacylglycerol (MGDG/DGDG) ratio, and enhanced triacylglycerol (TAG) accumulation, compared to the LFT genotype. De-acclimation reversed rearrangements in structural lipids and reduced frost tolerance, whereas re-acclimation induced strong genotype-dependent responses, especially in phosphoglycerolipids profiles. Analysis of mRNA accumulation profiles indicated key enzymes potentially involved in lipid remodeling. Major transcriptional changes were observed in phytyl ester synthase 1 (PES1), digalactosyldiacylglycerol synthase (DGD1), and lipoxygenase (LOX) family transcripts, indicating their importance during cold acclimation, de-acclimation, and re-acclimation. Collectively, these findings demonstrate, for the first time, that genotype-dependent lipid remodeling underlies the capacity of F. arundinacea to maintain and recover frost tolerance. We further suggest that stress-induced TAG accumulation may function as a protective metabolic buffer, sequestering excess free fatty acids, limiting lipid oxidation, and supporting membrane stability under low-temperature stress.

Fruit weight is a key determinant of yield in high-value vegetable crops such as tomato. Despite extensive research, the molecular mechanisms underlying this complex trait remain largely elusive, with only a few genes cloned to date based on quantitative trait loci (QTL). Here, we analyzed 2 populations and reanalyzed 3 previously published populations and identified 945 QTL associated with agro-morphological traits, including both previously reported and unidentified loci. We focused on SlGRF10 (GROWTH-REGULATING FACTOR 10) underlying a fruit weight QTL, fw1.2. Loss of SlGRF10 reduced fruit weight by decreasing cell size, without affecting cell number. Analysis of natural variation in SlGRF10 in over 1,000 tomato accessions revealed that increased single-nucleotide polymorphism diversity in SlGRF10 is associated with lower fruit weight. This suggests that putative impaired activity contributes to reduced fruit weight, while breeding-induced reduction of genetic variation may have promoted increased fruit weight. Transcriptome profiling of SlGRF10 knockout lines 7 and 20 days postanthesis identified several differentially expressed genes involved in cell cycle progression. Our findings not only confirm the role of SlGRF10 in regulating tomato fruit weight but also highlight a set of candidate genes associated with key morpho-physiological traits. With fw11.3, named as CELL SIZE REGULATOR, being the only QTL related to cell size determination in tomato fruits so far, SlGRF10 offers a valuable target for precision breeding and enhances our understanding of fruit weight in tomato and related fruit-bearing species.

This study provides the first fully integrated comparison of metabolome, including lipids, primary and specialized secondary metabolites in the wild thallus of Fucus vesiculosus (FV), its derived callus (FVC), and two related brown macroalgae, F. spiralis (FS) and Saccharina latissima (SL), revealing how species identity and dedifferentiation reshape brown algal metabolism. Untargeted ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS/MS) with Global Natural Products Social Molecular Networking (GNPS) molecular networking putatively annotated 106 metabolites across 14 chemical classes, showing species-specific enrichment of peptides, betaine lipids, oxylipins, sulfated phenolics, purine derivatives, amino acids, and chlorophyll catabolites. Venn analysis indicated FS and FV had the richest metabolomes (about 71-72% of metabolites), SL (59%), and FVC markedly reduced (37%), dominated by primary and stress-related metabolites. Gas chromatography-mass spectrometry (GC-MS) profiling revealed nitrogen-rich primary metabolites (88-95%), primarily pyroglutamic acid, alanine, glutamate, and proline. FVC showed complete depletion of tricarboxylic acid cycle intermediates and a shift toward nitrogen assimilation and osmoprotective metabolism. Liquid chromatography-mass spectrometry (LC-MS) lipidomics detected 65 lipid species with distinct species- and tissue-specific patterns; triacylglycerols and galactolipids prevailed in the wild thalli, while callus tissue accumulated sphingolipids, especially ceramides. Multivariate analyses, including principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), distinguished the four systems and identified key metabolic markers, with FVC characterized by suppressed secondary metabolites and elevated nucleotides, amino acids, and MGTA(16:1). Volcano plots revealed 49 metabolites differing between FV and FVC, mostly upregulated in wild thalli. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment highlighted purine metabolism, aromatic amino acid biosynthesis, aminoacyl-tRNA biosynthesis as key pathways underlying species- and culture-dependent metabolic differentiation. This multi-platform metabolomics study offers a comprehensive view of metabolic reprogramming in brown macroalgae, informing chemotaxonomy and marine biotechnology.

Crop wild relatives are used to improve cultivated plants and precise tracking of genetic introgression requires high-quality genome assemblies. Here we present de novo genome assemblies of two wild tomato species - the broadly stress-resistant Solanum pennellii (LA0716) and the salt-resistant Solanum cheesmaniae (LA1039). The improved S. pennellii genome adds 146 Mbp to the twelve chromosomes compared with the original reference. The alignment of the new assemblies with multiple gold-standard assemblies identified shared and species-specific structural variants. Analysis of repeat content demonstrates independent explosions of Tekay retrotransposons in S. pennellii and S. peruvianum. Genome sequencing of 709 recombinant plants derived from male and female backcrosses of three different hybrids reveals higher crossover rate in female meiosis. Conserved female-enhanced recombination regions were discovered and coldspots were attributed to megabase-scale inversions and insertion-deletion polymorphisms. Our S. pennellii and S. cheesmaniae genome assemblies reveal how repeat content diverged in nature and during breeding, and uncovers how both reproductive gender and structural variants dictate recombination landscapes in tomato hybrids.

Raspberry fruits are a healthy food rich in fibre, vitamins, minerals and antioxidants. Raspberries, however, are vulnerable to climate change effects. In particular, heat stress combined with solar radiation leads to white drupelets disorder: appearance of white spots on the fruits, making raspberry fruits unusable. We report that treatment of raspberry plants with a seaweed-based biostimulant, derived from the brown algae Ascophyllum nodosum, prevents the heat-induced sunburn, significantly reducing the number of damaged fruits and the number of white spots caused by the heat stress. Fruits from the biostimulant-primed raspberries were larger and the yield was higher. Comprehensive RNA-seq transcriptional analysis of biostimulant treated and untreated raspberries revealed that the biostimulants treatment induced genes related to auxin signalling and cell wall related proteins including expansins and extensins, supporting the notion that the biostimulant positively regulates growth. Genes encoding proline metabolic process, response to lipids and lipid biosynthetic processes, regulation of pH, and magnesium ion transport were also induced. At the same time, the A. nodosum-based biostimulant suppressed expression of genes encoding heat shock proteins and other stress related proteins. Downregulation of these genes provided molecular confirmation of the lower stress status of fruits collected from the biostimulant-treated plants. Metabolome analysis performed by GC/UHPLC-MS revealed that the fruits collected from biostimulant treated plants had up to three-fold higher levels amino acids, higher levels of the dipeptides HisAsp and ThrAla, and higher content of anthocyanins, two flavonoid glycosides, and Vitamin C. Overall, this demonstrated that the A. nodosum-based biostimulant prevented heat-induced sunburn and lead to production of fruits with higher concentrations of essential nutrients and health promoting metabolites, suggesting additional health benefits for the raspberry consumers.

Small peptides, including dipeptides, remain largely overlooked despite their nutritional and functional potential. Here, we profiled an extensive whole-grain rice (Oryza sativa L.) collection of more than 500 accessions and identified 132 proteinogenic dipeptides enriched in essential amino acids, notably Trp (11.74%), Phe (11.36%), and Leu (9.85%). Germination as a natural bio-enrichment process, enhanced relative dipeptide abundance by up to 10.78-fold, exceeding those observed in wild relatives and non-rice benchmark samples (cereals/pseudocereals, nuts, fruits, and vegetables). Machine learning using a random forest model predicted superior lines exhibiting relatively higher dipeptide peak intensities with 98.84% accuracy. In silico analyses revealed diverse bioactivities, with over half of the bioactive dipeptides targeting angiotensin-converting enzyme (ACE), a prohypertensive central to blood pressure regulation. C-terminal tryptophan-containing dipeptides (Ile-Trp, LysTrp, Tyr-Trp) exhibit the strongest antihypertensive potential. Molecular docking and dynamics simulations confirmed stable interactions with ACE catalytic residues through hydrogen bonding, hydrophobic contacts, and it-stacking. Drug-likeness and ADMET profiling further indicated favorable pharmacokinetic and safety profiles. These findings highlight rice dipeptides as promising nutritional and therapeutic resources for functional food development and drug discovery.