Discussion

Genotype‐Associated Metabolomic Variation in P. avium Leaves and Fruits

To evaluate the relative contribution of genotype and environment to metabolite variation in cherry fruits and leaves, we designed a sampling strategy in which fruits and leaves were collected from seven distinct cultivars at a phenological stage corresponding to full fruit maturity. For each cultivar, fruit and leaf samples were obtained from trees growing in orchards located in different geographic areas and sampled across two consecutive growing seasons, thereby increasing the variability of environmental conditions. Sampling followed the strategy described by Commisso et al. (2017), with the additional collection of leaves from the same trees used for fruit sampling. The meteorological data confirmed that the two growing seasons differed, particularly in rainfall regime, number of rainy days, and solar radiation. Within this environmental context, the PCA visualizations indicated that cultivar‐associated clustering was more evident than year‐ or orchard‐associated separation, supporting the presence of a strong genotype‐associated component in fruit metabolomic variation. This observation is consistent with previous reports on sweet cherry fruit metabolomics (Picariello et al. 2016; Commisso et al. 2017; Martini et al. 2017; Di Matteo et al. 2017; Berni et al. 2021). Picariello et al. (2016) compared three sweet cherry cultivars and two sour cherry cultivars, reporting that phenolic composition, particularly anthocyanin profiles, was strongly genotype‐dependent. In our earlier investigation (Commisso et al. 2017), untargeted metabolomic profiling of fruits from 18 sweet cherry cultivars revealed cultivar‐specific associations with distinct classes of phenolic compounds, including hydroxybenzoic and hydroxycinnamic acid derivatives, anthocyanins, flavonoids, proanthocyanidins, and flavan‐3‐ols. Similarly, Martini et al. (2017) found that hydroxycinnamic acids were the most abundant metabolites in fruits of six cultivars, except for “Lapins” and “Durone della Marca”, which accumulated higher levels of anthocyanins and flavan‐3‐ols. Di Matteo et al. (2017) also reported significant variation in total flavonoid content among 26 sweet cherry cultivars grown in southern Italy between 2013 and 2015, while Berni et al. (2021) showed that flavanols and proanthocyanidins were more abundant in two local Tuscan varieties (“Morellona” and “Crognola”) than in the commercial cultivar “Durone” across three growing seasons.

A comparable genotype‐associated component was also detected in leaves. Although leaves are generally regarded as metabolically plastic and responsive to environmental variability, the additional PCA visualizations showed that cultivar‐associated patterns remained detectable across the two growing seasons and orchard locations. To the best of our knowledge, the number of studies exploring the metabolic profile of sweet cherry leaves in relation to genotype is extremely limited. Most previous investigations have focused on the effects of environmental conditions, rootstock, or agricultural practices using one or two cultivars only (Vosnjak et al. 2021; Gerasko et al. 2022; Serapicos et al. 2022; Kubes et al. 2024; Zhang et al. 2024), or have evaluated potential health‐promoting properties of leaf extracts (Jesus et al. 2019), thus preventing a comprehensive assessment of genotype‐metabolome relationships. Studies including more than two cultivars are relatively scarce and have generally aimed at exploiting leaves, considered a by‐product, as natural sources of phenolic compounds. For instance, sweet cherry leaves from “Burlat”, “Kordia”, and “Regina” collected during 2015–2016 (Dziadek et al. 2018) showed cultivar‐ and season‐dependent differences in total polyphenol content. Likewise, leaves from “Kordia”, “Regina”, “Vega”, “Hedelfińska”, “Vanda”, and “Summit” collected in 2016 (Dziadek et al. 2019) displayed genotype‐related variations in vitamin C content, antioxidant capacity, total phenolics, and selected metabolites such as chlorogenic acid, p‐coumaric acid, ferulic acid, and myricetin. Among these metabolites, chlorogenic acid and p‐coumaric acid were more characteristic of the Kordia cultivar, whereas ferulic acid and myricetin were more abundant in Regina than in the other cultivars (Dziadek et al. 2019).

Overall, our findings indicate that genetic background contributes substantially to the organization of phenolic profiles in both fruits and leaves, while year‐ and orchard‐associated variation should be acknowledged as environmental components contributing to the observed metabolomic variability. The comparison between unsupervised and supervised analyses further supports this interpretation. In both organs, cultivar‐associated grouping was evident in PCA, whereas year‐associated variation was mainly resolved by supervised modeling. This indicates that inter‐annual environmental variation contributed to the metabolomic profiles, but did not override the cultivar‐associated structure observed in the exploratory analyses. Orchard‐related variation was also detectable, although it was less clearly structured and less consistently associated with metabolite profiles than cultivar‐ or year‐associated variation.

Therefore, the data do not exclude an environmental contribution to phenylpropanoid variation. Rather, they suggest that, within the environmental range and sampling design considered here, cultivar‐associated metabolic organization remained clearly detectable in both fruits and leaves. This is particularly relevant for leaves, which are generally considered more responsive to environmental conditions than fruits, and indicates that a reproducible genotype‐associated component of phenylpropanoid metabolism can also be observed in vegetative tissues.

Organ‐Specific Accumulation and Structural Diversity of Phenolic Compounds in Prunus avium

Beyond these genotype‐associated patterns, our data also revealed a marked organ‐specific specialization of phenylpropanoid metabolism between fruits and leaves, reflecting possible and distinct physiological roles.

Our results revealed that sweet cherry leaves accumulate substantially higher levels of secondary metabolites than fruits, both in terms of total abundance and structural diversity. Despite the scarcity of data in literature, the few studies investigating the content of secondary metabolites in sweet cherry leaves confirmed that leaves contain more polyphenols than fruits. In detail, leaves of cultivars “Burlat”, “Kordia”, and “Regina”, collected during the 2015 and 2016 growing seasons at the Experimental Station of the Department of Pomology and Apiculture, University of Agriculture in Kraków (Poland), contained markedly higher concentrations of total polyphenols than the corresponding fruits, reaching 5077–12,844 mg per 100 g DW in leaves and 1988–3063 mg per 100 g DW in fruits (Dziadek et al. 2018). In a subsequent study (Dziadek et al. 2019), leaves and fruits collected in 2016 from additional cultivars, such as “Kordia”, “Regina”, “Vega”, “Hedelfińska”, and “Vanda” from Sandomierz, and “Kordia”, “Regina”, and “Summit” from Szczodrkowice, showed the same trend, with leaves containing 6012–15,318 mg per 100 g DW of total polyphenols, while fruits ranged from 1604 to 4045 mg per 100 g DW. In both reports, the difference between leaves and fruits was consistent across cultivars and sites, and, in the former study, across growing seasons, confirming that leaves represent a richer source of phenolic compounds.

Beyond sweet cherry, similar organ‐specific differences have been documented in other fruit‐bearing species. Teleszko and Wojdyło (2015) showed that leaves of Malus domestica , Cydonia oblonga , Vaccinium macrocarpon , and Vaccinium myrtillus contained significantly higher concentrations of polyphenols than fruits. Comparable trends have been reported for Schisandra chinensis (Mocan et al. 2014), reinforcing the view that vegetative tissues, particularly leaves, act as major reservoirs of polyphenolic metabolites.

Within the phenolic fraction, LC–MS profiling revealed that sweet cherry leaves were dominated by hydroxycinnamic acids and flavonols, together accounting for more than half of the total ion signal. In contrast, and in line with previously published data (Martini et al. 2017), anthocyanins, along with hydroxycinnamic acids and flavan‐3‐ols, predominated in mature fruits, representing nearly half of the detected compounds.

Regarding hydroxycinnamic acids, fruits mainly accumulated simpler forms, particularly neochlorogenic acid (5‐O‐caffeoylquinic acid; 7–37 mg per 100 g FW), p‐coumaroylquinic acid (5–19 mg per 100 g FW), and chlorogenic acid (3‐O‐caffeoylquinic acid; 3–4 mg per 100 g FW), in agreement with previous studies (Antognoni et al. 2020; González‐Gómez et al. 2010). Leaves, in contrast, accumulated both simple and more complex hydroxycinnamic acids at much higher concentrations, particularly chlorogenic acid (203–343 mg per 100 g FW), neochlorogenic acid (49–189 mg per 100 g FW), and p‐coumaroylquinic acid (8–82 mg per 100 g FW). Additionally, our LC–MS data revealed the presence of hydroxycinnamic acids conjugated with a compound deriving from the breakdown of L‐ascorbic acid, that is, threonic acid (Ford et al. 2024). These metabolites have already been reported in the moss Physcomitrella patens (Renault et al. 2017) and in leaves of Fagus sylvatica L. (Cadahía et al. 2015) and, at the best of our knowledge, this is the first report of their presence in sweet cherry. More complex derivatives, especially dicaffeoylquinic acid isomers, were also detected in leaves, showing high values. These findings are consistent with previous reports describing the predominance of caffeoylquinic acid derivatives in cherry foliar tissues, although published concentrations are often expressed on a dry‐weight basis or derived from concentrated extracts. For example, Nunes et al. (2021) identified a complex mixture of caffeoylquinic and dicaffeoylquinic isomers as the dominant hydroxycinnamic acids in sweet cherry cv. “Saco” leaves. In detail, the total amount of extracted hydroxycinnamic acids was declared to be 51,345.69 μg g−1 DW in infusions and up to 57,605.22 μg g−1 DW in hydroethanolic extracts, with trans‐5‐caffeoylquinic acid (neochlorogenic acid) as the major compound, followed by chlorogenic acid (3‐O‐caffeoylquinic acid). Likewise, Jesus et al. (2019) reported that hydroxycinnamic acids represent more than 70% and 60% of the total phenolic compounds in the infusion and hydroalcoholic extracts of sweet cherry leaves, respectively.

Hydroxycinnamic acids such as neochlorogenic, chlorogenic, p‐coumaroylquinic, dicaffeoylquinic, and coumaroyl‐caffeoylquinic acids have been suggested to play roles in response towards abiotic stresses, including drought, heavy metal, and temperature stresses, and towards biotic stresses against pathogen attacks by strengthening plant cell walls and acting as antimicrobial agents (Khawula et al. 2023). In particular, hydroxycinnamic acids appear to participate in regulating cell wall plasticity, either by modulating reactive oxygen species (ROS) levels or by serving as precursors of monolignols, thereby influencing the lignification process (Khawula et al. 2023).

Sweet cherry leaves also accumulated high levels of flavonoids, particularly flavonols such as quercetin‐3‐O‐rutinoside and kaempferol‐3‐O‐rutinoside, with concentrations ranging from 70 to 119 mg per 100 g FW and 111 to 161 mg per 100 g FW, respectively. Although not directly comparable due to the use of lyophilized material in other studies, similar trends were reported for sweet cherry cv. “Saco” leaves (Nunes et al. 2021; Jesus et al. 2019). In those works, quercetin‐3‐O‐rutinoside reached 673 mg per 100 g DW (Jesus et al. 2019) and 365 mg per 100 g DW (Nunes et al. 2021) in hydroethanolic leaf extracts, respectively. Kaempferol‐3‐O‐rutinoside reached 313 mg per 100 g DW in hydroethanolic leaf extracts (Jesus et al. 2019) and 130 mg per 100 g DW in extracts after infusion procedure (Nunes et al. 2021). Flavonols, particularly quercetin derivatives, act as effective antioxidants and also modulate key processes related to plant development and stress responses. Their accumulation is enhanced under abiotic stress, and these compounds contribute to the protection of photosynthetic tissues by scavenging singlet oxygen, preserving chloroplast envelope integrity, and inhibiting ROS formation (Agati et al. 2012).

Mature fruits also accumulated other phenolic compounds, particularly anthocyanins and proanthocyanidins. Cyanidin‐3‐O‐glucoside and cyanidin‐3‐O‐rutinoside were the predominant anthocyanins in fruits, with concentrations consistent with previous reports (Usenik et al. 2008; González‐Gómez et al. 2010; Antognoni et al. 2020; Martini et al. 2017). These pigments contribute to fruit coloration and antioxidant protection during ripening, while proanthocyanidins may enhance astringency and participate in defence mechanisms (Mozetič et al. 2004; Clodoveo et al. 2023; Campbell et al. 2024; Zhao et al. 2024).

Overall, the marked enrichment of hydroxycinnamic acids and flavonols in leaves, together with the dominance of anthocyanins and flavan‐3‐ols in fruits, highlights the organ‐specific specialization of phenolic metabolism in sweet cherry, reflecting distinct physiological roles in photoprotection, defence, and fruit quality.

Within this framework, our results suggest that phenylpropanoid metabolism in sweet cherry is organized into tissue‐specific metabolic channels whose structure includes a substantial genotype‐associated component. Importantly, the detection of a reproducible, genotype‐dependent organization also in leaves, a vegetative tissue traditionally regarded as metabolically plastic, indicates that genetic background contributes to phenylpropanoid pathway organization across organs, together with organ identity and environmental context.