Phenotypic plasticity often serves as an initial adaptive strategy for populations to cope with environmental stress, providing the time needed for slower genetic adjustments to occur across generations (Yeh and Price, 2004; Ghalambor et al., 2007; Ho and Zhang, 2018). These plastic responses can be classified based on their fitness consequences. Adaptive plasticity occurs when the environmentally induced phenotypic change enhances the organism’s fitness in that specific environment (Sultan, 2000; Ghalambor et al., 2007), whereas maladaptive plasticity describes responses that reduce fitness, which can occur when environmental cues are unreliable (Ghalambor et al., 2007).

Several stresses trigger plastic responses, modulating different aspects of the plant physiology. In response to light availability, plants adjust leaf morphology, specific leaf area (SLA), chlorophyll content, and stem elongation to optimize light capture and prevent photodamage (Givnish, 1988; Sultan and Bazzaz, 1993a; Smith, 2000). Under drought, a wide array of responses is triggered, including stomatal regulation to control water loss, osmotic adjustments to maintain cell turgor, and changes in root morphology to enhance water acquisition (Sultan and Bazzaz, 1993b; Chaves et al., 2003; Nicotra and Davidson, 2010). Plants also exhibit remarkable plasticity in root growth and architecture under heterogeneous nutrient availability, concentrating root proliferation in nutrient-rich patches to enable more efficient foraging (Drew, 1975; Hutchings and de Kroon, 1994; López-Bucio et al., 2003). Temperature fluctuations trigger the synthesis of heat shock proteins and cold hardening processes (Thomashow, 1999; Sung et al., 2001), as well as phenological shifts that align reproduction with favorable conditions (Rathcke and Lacey, 1985; Franks et al., 2007).

Finally, biotic pressures such as herbivory or pathogen attack elicit a suite of plastic responses, including chemical and physical defenses that deter attacks and enhance resilience (Agrawal, 1999; Karban and Baldwin, 2007; War et al., 2012).

However, phenotypic plasticity is not without limitations. Its evolution is constrained by physiological costs and trade-offs, as maintaining the underlying mechanisms demands energy and resources. Consequently, some traits exhibit limited plasticity due developmental or structural constraints (DeWitt et al., 1998; Auld et al., 2010; Murren et al., 2015).

Investigating phenotypic plasticity requires carefully designed experiments that partition phenotypic variance into its genetic and environmental components, finding three main different alternatives.

These experimental approaches, often complemented by quantitative genetic analyses, facilitate the statistical partitioning of observed phenotypic variance (P) into its components: genetic variance (G), environmental variance (E), and genotype-by-environment interaction variance (G×E). This partitioning accentuates the fact that plasticity arises from environmental effects on phenotype, while divergent reaction norms reveal unique behaviors of individuals, opening the way for natural selection to act on plastic responses.

G×E describes how different genotypes respond differently to the same environmental gradient, being graphically evidenced by divergent slopes or intercepts of reaction norms (Via and Lande, 1985; Hill and Mackay, 2004; Manuck, 2010; Nicotra et al., 2010). As G×E reveals genetic variation in the shape and magnitude of reaction norms, plasticity is therefore heritable and can evolve through natural selection maximizing fitness across a population’s environmental range (Via and Lande, 1985; Gomulkiewicz and Kirkpatrick, 1992; Scheiner, 1993). When reaction norms for fitness intersect, G×E indicates local adaptation: genotypes perform best in their native habitats but worse elsewhere, reflecting genetically encoded differences in optimal phenotypes rather than plastic adjustments (Hereford, 2009). Moreover, by favoring different genotypes under varying spatial or temporal conditions, G×E promotes the maintenance of genetic diversity within populations, preventing any single genotype from universally dominating (Ellner and Hairston, 1994; Gillespie and Turelli, 1989). This diversity of reaction norms also modulates population resilience: genotypes with broad, adaptive plasticity, or a mosaic of reaction shapes, are better equipped to cope with novel or shifting environments, such as those imposed by climate change (Sultan and Bazzaz, 1993a, 1993b; Reed et al., 2011; Lázaro-Nogal et al., 2015). Understanding G×E can then be applied in plant breeding, developing conservation strategies focused on preserving populations with high adaptive potential (Basford and Cooper, 1998; De Leon et al., 2016; Xu et al., 2017; Cooper and Messina, 2021; Yadesa, 2022).

Summarizing, far from being a mere statistical abstraction, G×E demonstrates that phenotype emerges through the interplay of genetic and environmental influences, and that the sensitivity of a genotype to its environment is itself genetically encoded. Recognizing G×E for adaptive traits reveals the genetic architecture of environmentally responsive phenotypes and opens the way for molecular investigations into the specific genes, regulatory networks, and metabolic pathways underlying plasticity.

F. vesca is a low‐growing, herbaceous perennial plant of the Rosaceae family, easily recognized by its trifoliate serrated leaves. Its flowering period is comprised during temperate seasons, occurring mainly in spring, but some ecotypes like F. vesca f. semperflorens flower continuously (Slovin et al., 2009). Following fertilization, flowers expand into the red accessory fruit (aggregate of achenes) renowned for its aroma and flavor (Yadav et al., 2019; Liu et al., 2020b).

Relevant to its employment as a research model, F. vesca presents a dual reproductive strategy, which can be (i) sexual reproduction via seeds, and (ii) clonal propagation through stolons, commonly known as runners (Staudt, 1999; Schulze et al., 2012). This clonal capacity facilitates rapid colonization of suitable microhabitats, significantly contributing to the species’ local persistence and spatial dynamics, often leading to the formation of extensive patches of interconnected individuals (Roiloa and Retuerto, 2006; Schulze et al., 2012).

F. vesca is native to temperate North America, Europe, and northern Asia, occupying habitats from shaded woodlands to subalpine zones, on soils ranging from calcareous to acidic, and even disturbed sites (Liston et al., 2014; Hilmarsson et al., 2017; Fernie, 2023). Its circumboreal distribution spans from 37°N to 70°N latitude, encompassing broad environmental gradients (Njuguna et al., 2013; Faehn et al., 2023) (Figure 3).

About flowering, there is a well-established negative correlation between latitude and flowering earliness, such that populations from higher latitudes tend to flower earlier than those from more equatorial regions. This pattern reflects adaptation to progressively shorter growing seasons and harsher climates (Heide and Sønsteby, 2007; Still et al., 2023).

F. vesca combines a small diploid genome (2n=2x=14; ~221 Mb) fully sequenced for comparative genomic studies (Shulaev et al., 2011; Zhou et al., 2023), with self-compatibility for generating inbred lines and efficient outcrossing for QTL mapping and recombinant inbred lines (RILs) development (Sargent et al., 2006; Slovin et al., 2009). Its small size makes controlled environment and common garden experiments easier to be conducted (Hancock and Bringhurst, 1978; Sammarco et al., 2022), and efficient clonal propagation capability provides replicates for robust G×E analyses (Sultan, 2000). Moreover, F. vesca amenability to be transformed by Agrobacterium mediated protocols, make it useful for functional genomics research (Razavi et al., 2012; Pantazis et al., 2013; Schaart, 2014).

Beyond basic research, F. vesca has strong significant applied relevance. It is closely related to the octoploid strawberry Fragaria × ananassa, serving as a genetic reservoir, helping breeding programs of the cultivated strawberry, facilitating the development of marker-assisted selection (MAS) and genomic selection (GS) approaches (Denoyes et al., 2017; Liu et al., 2023). Thus, studies of metabolic adaptation in F. vesca not only improve our understanding of plant eco-evolutionary processes but also offer real benefits for sustainability, resilience, and quality of strawberries and other related wild plants. A detailed translational pipeline from F. vesca gene discovery to F. × ananassa breeding is provided in Section 5.3.2.

Although the full extent of plasticity in F. vesca is still under study, its broad ecological distribution suggests a high degree of phenotypic flexibility across multiple biological levels (Hancock and Bringhurst, 1978; De Kort et al., 2020; Still et al., 2023). Evidence indicates that plasticity operates through distinct but interconnected pathways: immediate metabolic adjustments, longer-term epigenetic modifications, and intergenerational plasticity.

Concerning metabolic adjustments, studies on cold adaptation show that a 10-day cold treatment in F. vesca changed its core metabolism, increasing soluble sugars, aspartic acid, and amines like putrescine, while proline, unlikely in many other species, did not increase. Instead, galactinol and raffinose levels significantly rose in leaves and roots, suggesting species specific metabolic plasticity (Rohloff et al., 2012). Comparative studies between cold-tolerant (F. daltoniana) and cold-sensitive (F. vesca) Fragaria species further demonstrated that differential accumulation of abscisic acid (ABA) and glucose signaling through the ABF2 transcription factor gene underpin divergent cold stress responses. F. daltoniana exhibits robust ABA accumulation coupled with significant upregulation of ABF2, which enhances glucose signal transduction and results in high levels of glucose and fructose accumulation that effectively mitigate cellular damage under low temperatures. In contrast, F. vesca, despite ABA activation, shows downregulation of ABF2 expression under cold stress, leading to reduced glucose signal transduction and a more limited metabolic response. This differential regulation of the ABA-ABF2 pathway, combined with higher ABA accumulation in F. daltoniana, explains the species’ superior cold tolerance capacity (Shen et al., 2022).

About longer timescales adaptations, it has been described that genome-wide methylation changes occur under multiple stress conditions including heat, cold, drought, and salinity, resulting in the loss of methylation in centromeric/pericentromeric regions and differential methylation of promoters linked to stress-responsive transcription factors (López et al., 2022). Methylation profiles in wild European populations correlate with climatic origin and pass down through clonal generations, influencing expression of growth and stress-related genes (Sammarco et al., 2024). Moreover, reciprocal transplant experiments demonstrate that modifying methylation patterns changes plant performance under varying environmental conditions, indicating a direct epigenetic role in local adaptation (Gu et al., 2016; Sammarco et al., 2022). Given the tight integration between metabolism, epigenetics, and quick physiological changes, these traits in F. vesca may exhibit greater plasticity than many morphological traits, which are often limited by structural and developmental factors. Understanding how these specific plasticity mechanisms work could lead to ways to understand how plants can withstand climate change.

Primary metabolism comprises conserved pathways essential for survival, growth, development, and reproduction (Buchanan et al., 2015). Beyond their traditional roles in energy provision, carbon skeleton supply, and building-block functions, primary metabolites are increasingly recognized as crucial signaling molecules and regulatory hubs that actively coordinate environmental responses and developmental programs (Salam et al., 2023).

Soluble sugars exemplify this dual role. Their fundamental function in energy supply is well characterized, but their role as signaling molecules is often underestimated, even though they modulate fundamental pathways involved in photosynthesis, carbon partitioning, nitrogen assimilation, growth, defense, and senescence (Hanson and Smeekens, 2009; Li and Sheen, 2016). Glucose and sucrose are notable examples, interacting extensively with hormonal pathways and stress signaling networks and acting as metabolic sensors of the plant’s energy status (Wind et al., 2010; Lastdrager et al., 2014). In strawberry, glucose, fructose and sucrose accumulate substantially under drought stress, indicating that osmotic adjustment through sugar accumulation serves as a core plastic response to water limitation (Sun et al., 2015).

Amino acids constitute another functionally diverse metabolite group. Regarding stress response, proline is the most prominent example, accumulating to high levels under osmotic stress and acting as a compatible solute, antioxidant, and regulator of stress-responsive gene expression (Szabados and Savouré, 2010; Hayat et al., 2012; Forde and Roberts, 2014; Hildebrandt et al., 2015). Notably, F. vesca exhibits species-specific amino acid responses to cold stress that diverge from those observed in model organisms (see Section 3.5). Similarly, GABA (gamma-aminobutyric acid) acts as a metabolic signal linking carbon and nitrogen metabolism (Shelp et al., 2012; Ramesh et al., 2015). Other amino acids also play stress-related roles; for instance, asparagine and glutamine serve as major nitrogen transport compounds and regulate nitrogen-responsive gene expression (Gaufichon et al., 2010; Krapp et al., 2011).

Organic acids, particularly TCA cycle intermediates such as malate, citrate, and fumarate, also have important regulatory role. They influence pH regulation and ion transport, and act as metabolic signals coordinating nitrogen metabolism, stomatal function, and plant-microbe interactions (Fernie and Martinoia, 2009; Meyer et al., 2010; Finkemeier et al., 2013). Moreover, they serve as alternative respiratory substrates and osmoregulatory compounds under stress conditions (Igamberdiev and Eprintsev, 2016).In strawberry, organic acid metabolism displays substantial plasticity during fruit development and ripening, with citric and malic acid levels regulated by sugar content and developmental stage, as reflected in dynamic changes in acid composition correlated with transcriptional reprogramming of metabolic genes (Wang et al., 2025). Geographic studies of wild F. vesca populations reveal substantial variation in the glucose/fructose ratio and in citric and malic acid content across different climatic regions, with both location and growing year significantly metabolite profiles (Akagić et al., 2020).

These findings indicate that primary metabolism is not merely a passive network supplying resources on demand, but an active, dynamic system that integrates environment and the plant’s internal state, contributing directly to the regulation of gene expression and physiological adjustment.

Metabolic plasticity is the capacity of an organism to dynamically reconfigure its metabolic network in response to both internal developmental factors and a environmental stimuli (Sweetlove et al., 2008; Obata and Fernie, 2012; Seebacher and Beaman, 2022). Rapid adjustments occur via allosteric regulation and post‐translational modifications, while slower shifts involve transcriptional changes. These regulatory layers (transcriptional, post‐transcriptional, post‐translational, allosteric, and compartmentalization) operate together to enable both immediate and sustained metabolic responses (Kaiser and Huber, 2001; Huber and Hardin, 2004; Friso and Van Wijk, 2015).

Primary metabolic plasticity is clearly observed in central carbon metabolism, where flux through glycolysis, the TCA cycle, and the pentose phosphate pathway can be rapidly redirected based on cellular demands and environmental conditions (Plaxton and Podestá, 2006; Araújo et al., 2012). This redirection is mediated by multiple regulatory nodes: (i) post-translational modifications (PTMs) of key glycolytic enzymes, such as those controlling fructose-2,6-bisphosphate levels, which modulate glycolytic flux to enhance NADPH production and antioxidant capacity under stress conditions (Smyth et al., 1984; Rider et al., 2004); (ii) allosteric regulation by AMP/ATP ratios and redox status (NADH/NAD⁺), which directly sense the cell’s energetic state (Geigenberger et al., 2010); (iii) transcriptional control by key metabolic regulators, such as the bZIP family transcription factors, which upregulate genes encoding gluconeogenic enzymes and stress-responsive amino acid biosynthetic pathways during sugar depletion, as well as MYB family members, which suppress growth-promoting anabolic pathways while promoting catabolic and defensive pathways under stress (Roy, 2015; Wang et al., 2022a). Similarly, amino acid metabolism shows remarkable plasticity, with pathways for proline, GABA, and branched-chain amino acids being strongly upregulated under abiotic stress (Fait et al., 2008; Trovato et al., 2008). These regulatory nodes integrate information from multiple environmental and internal signals, enabling rapid yet flexible metabolic reprogramming that maintains both immediate survival and long-term fitness during environmental fluctuations (Sammarco et al., 2023) (Mao et al., 2022).

Metabolomics and fluxomics provide snapshots of these dynamic shifts, revealing how primary metabolic pathways are rerouted under stress (Fiehn, 2002; Schauer and Fernie, 2006; Schwender, 2008). ¹³C-labeling experiments have shed light how carbon flux through primary pathways shifts under drought, cold, and nutrient stress (Gauthier et al., 2010; Watanabe et al., 2013). In F. vesca, comparative omics reveal that red genotypes coordinate a developmental switch from proanthocyanidin to anthocyanin biosynthesis by downregulating early phenylpropanoid genes and activating anthocyanin-specific genes during fruit ripening. In contrast, white genotypes fail to execute this switch, maintaining high expression of proanthocyanidin and flavonoid pathway genes while anthocyanin genes remain silent. This regulation is plastic, with environmental conditions modulating the timing and magnitude of the shift (Härtl et al., 2017). Secondary metabolite composition in F. vesca fruits encompasses nearly 100 volatile organic compounds, with genotype-specific profiles showing marked differences, indicating that biosynthetic allocation toward specific volatile classes is under strong genetic and potentially environmental control (Atazhanova et al., 2025).

Changes in primary metabolites often reveal plastic adjustments in growth, development, stress tolerance, defense, and resource allocation.

Under drought or salinity conditions, several sugars, polyols, and amino acids accumulate to maintain cellular turgor and protect against oxidative damage (Bohnert and Jensen, 1996; Szabados and Savouré, 2010; Krasensky and Jonak, 2012). These metabolites also function as molecular chaperones, stabilizing proteins and membranes under stress conditions (Hare et al., 1998; Chen and Murata, 2011). Under resource limitation, sugar allocation shifts from shoots to roots, altering root:shoot ratios and enabling enhanced nutrient acquisition (Rolland et al., 2006; Smith and Stitt, 2007; Eveland and Jackson, 2012).

pH variations also set stressful conditions for plants, not only for their physiological consequences, but the difficulty of obtaining determined nutrients whose availability is tightly influenced by the acidity of the soil. Strategies to face this involves different primary metabolites like malate, whose accumulation acts as a molecular buffer minimizing acidity changes; or citrate and other organic acids, which can chelate toxic metal ions and facilitate nutrient acquisition (Lopez-Millan et al., 2000; Ryan et al., 2001). Under phosphorus deficiency, organic acid exudation from roots helps to solubilize phosphate from soil minerals, allowing its uptake (Vance et al., 2003; Lambers et al., 2006).

Finally, storage compound allocation represents another main plasticity mechanism for plants. Starch accumulation in leaves and roots, along with fructan storage in some species, helps to cope with carbon availability fluctuation, remobilizing them when photosynthetic assimilation is limited (Stitt and Krapp, 1999; Stitt and Zeeman, 2012). The flux between starch synthesis and sucrose export reflects real-time integration of source-sink relationships and environmental conditions.

In summary, changes in the primary metabolome are not just passive consequences of environmental change but are often active, regulated responses that form the mechanistic basis of phenotypic plasticity. Understanding how environmental factors modulate metabolic networks, and how these changes translate into ecologically significant phenotypes, is essential for a comprehensive understanding of plant adaptation strategies and resilience, especially in species like F. vesca, which thrives across a wide spectrum of environmental conditions.

GWAS establishes associations between genetic variants and phenotypic traits by analyzing historical recombination across natural populations (Korte and Farlow, 2013; Visscher et al., 2017). This approach relies on linkage disequilibrium (LD), where a marker near a causal variant serves as its statistical proxy (Slatkin, 2008). Modern approaches combine accurate trait phenotyping with high-density genotyping, while corrections for population structure and relatedness are implemented by using linear mixed models (Yu et al., 2006; Price et al., 2010). Multi-locus methods such as FarmCPU (Liu et al., 2016b) and BLINK (Huang et al., 2019) enhance power for polygenic traits, while integrating GWAS signals with metabolic networks helps prioritize candidate genes and improve biological interpretation (Toubiana et al., 2013).

The application of GWAS to identify genetic loci controlling variation in metabolite levels (mGWAS) has emerged as a powerful strategy for linking genotype to biochemical phenotype (Chan et al., 2010; Illig et al., 2010; Fernie and Tohge, 2017).

As metabolite concentrations are inherently quantitative, they can be measured with high precision employing techniques like mass spectrometry (GC-MS, LC-MS) and nuclear magnetic resonance (NMR) spectroscopy (Marshall and Powers, 2017; Karjalainen et al., 2024), reliable methods that have enabled massive meta-analyses, integrating data from thousands of individuals to increase statistical power (Riedelsheimer et al., 2012b; Karjalainen et al., 2024). In addition, due to the intermediary role of metabolism and the expressed final-term phenotype (view section 4), metabolite levels are often more directly linked to the function of specific genes encoding enzymes, transporters, or regulatory proteins, which allows mGWAS to more easily point to candidate genes (Riedelsheimer et al., 2012b; Wen et al., 2015).

In plants, mGWAS was first applied in Arabidopsis thaliana, identifying loci conferring resistance to pathogens, controlling amino acid levels and revealing genes in biosynthetic pathways and regulatory networks (Aranzana et al., 2005; Chan et al., 2010; Toubiana et al., 2012). Subsequently, it was expanded to other model organisms such as maize, where mGWAS of central carbon metabolites identified multiple loci affecting starch metabolism and revealing genetic networks coordinating carbon partitioning (Riedelsheimer et al., 2012a; Wen et al., 2015). Similarly, studies in rice have mapped loci controlling organic acid levels and amino acid composition, providing insights into nitrogen use efficiency and stress tolerance (Chen et al., 2014; Matsuda et al., 2015).

These early studies established a consistent pattern: primary metabolites exhibit strong genetic control with moderate to high heritability, making them excellent targets for genetic dissection (Riedelsheimer et al., 2012a; Chen et al., 2014; Matsuda et al., 2015; Wen et al., 2015). This heritability is especially relevant given their central role in fundamental life processes. For instance, variation in proline levels identified through mGWAS directly explains drought tolerance differences at the whole-plant level in Arabidopsis (Kesari et al., 2012). Moreover, modern approaches combinate mGWAS results with Mendelian randomization to establish causal relationships between primary metabolite levels and fitness-related traits (Hartwig et al., 2017).

Beyond identifying individual loci, mGWAS is a useful tool for the reconstruction of metabolic network architecture. For example, loci affecting both substrate and product levels can confirm enzymatic functions in amino acid biosynthesis or TCA cycle regulation (Do et al., 2010; Toubiana et al., 2012). Finally, mGWAS can identify regulatory genes controlling entire primary metabolic modules, revealing master regulators of carbon-nitrogen balance or stress-responsive metabolism (Gibon et al., 2006; Nunes-Nesi et al., 2010).

F. vesca is a powerful system for GWAS-based dissection of adaptive and metabolic traits. Its favorable biological characteristics (see Section 3) make it an ideal bridge between ecological/evolutionary research and applied crop improvement (Samad et al., 2017; Davik et al., 2021; Gaston et al., 2021).

However, applying mGWAS to wild F. vesca populations requires careful attention to perennial-specific complications: cryptic population structure arising from historical geographic isolation and local adaptation can generate spurious associations between SNPs and metabolite levels if not rigorously controlled (Gloss et al., 2022). Clonal propagation via stolons means that phenotypic replicates may not represent independent genetic individuals, overestimating effect sizes while breaking statistical assumptions of independent and identically distributed observations (Liu et al., 2016a). Moreover, developmental asynchrony related to photoperiodic dormancy means that metabolite phenotypes are inherently stage-dependent and environment-specific, complicating standardization across genetically diverse populations (Clauw et al., 2025). Nevertheless, wild F. vesca being a perennial specie simultaneously offer unique advantages for mGWAS: clonal genotypes can be phenotyped repeatedly across multiple seasons and environmental conditions, enabling direct estimation of genotype-specific reaction norms (G×E slopes) as quantitative traits themselves, permitting “reaction norm GWAS” to identify genes controlling the magnitude and direction of metabolic plasticity (Liu et al., 2016a; Clauw et al., 2025). Moreover, primary metabolite variation discovered in cultivated strawberry (F. × ananassa), including loci for sugar content, organic acid balance, and amino acid profiles, are readily testable in the diploid F. vesca background, accelerating trait validation and functional follow-up (Vallarino et al., 2019; Alarfaj et al., 2021).