Antibacterial and antibiofilm activity of a triterpenoid-rich Eucalyptus globulus leaf extract against gram-positive bacteria Authors: Chiara Spaggiari, Costanza Spadini, Laura Righetti, Nicolò Mezzasalma, Alessandro Cupola, Marco Pieroni, Giannamaria Annunziato, Clotilde Silvia Cabassi, Gabriele Costantino Published: 2026 ABSTRACT AbstractEucalyptus globulus leaf extract (EGL) is a phytocomplex associated with antimicrobial and anti-inflammatory effects, but a comprehensive characterization of its chemical composition and its relationship with biological activity remains limited. In this study, we combined untargeted metabolomic profiling with functional antibacterial and antibiofilm assays to characterize EGL and to investigate the relationship between its chemical composition and biological activity. Untargeted UHPLC–TWIMS–QTOF analysis revealed a complex profile dominated by polyphenols and pentacyclic triterpenoids. Targeted quantification identified asiatic acid and ursolic acid as major triterpenoid components, present at concentrations of 47.29 ± 0.13 µg/mL and 16.12 ± 0.32 µg/mL, respectively. The antibacterial and antibiofilm activities of EGL were evaluated against clinically relevant Gram-positive pathogens, including Streptococcus pyogenes, Staphylococcus aureus, methicillin-resistant S. aureus and Streptococcus mutans. The two major triterpenoids identified in the extract, asiatic acid and ursolic acid, were tested separately as reference compounds to explore their individual antibacterial and antibiofilm activity in comparison with the whole extract. EGL inhibited bacterial growth with MIC values ranging from 62.5 µg/mL to 437.5 µg/mL, while the triterpenoids showed stronger activity, with MIC values of 2 µg/mL for ursolic acid and 4 µg/mL for asiatic acid against Streptococcus pyogenes. In antibiofilm assays, asiatic acid and ursolic acid showed MBIC values between 2 and 32 µg/mL, whereas the whole extract displayed partial inhibitory activity. Pre-exposure of bacterial cultures to EGL or its major triterpenoids enhanced antibiofilm efficacy, reducing MBIC values by up to fourfold. Overall, these results indicate that EGL is a chemically characterized multicomponent system with measurable antibacterial and antibiofilm activity against Gram-positive pathogens. The study highlights the importance of integrating chemical profiling with functional assays to better understand the activity of complex phytocomplexes and supports further investigation of phytochemical-based strategies targeting early stages of biofilm formation.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-50090-4.Keywords: Eucalyptus globulus leaves extract, Antibiofilm activity, Gram-positive pathogens, Ursolic acid, Asiatic acid, UHPLC-HRMS metabolomicsSubject terms: Biochemistry, Biotechnology, Drug discovery, Microbiology, Plant sciences Introduction Eucalyptus globulus, commonly known as the blue gum tree, is a widely cultivated species of the Myrtaceae family and one of the most extensively studied members of the Eucalyptus genus. Originally native to Australia, it is now grown worldwide across subtropical and Mediterranean regions due to its rapid growth and industrial relevance, particularly in the paper industry. The leaves, often regarded as a byproduct of harvesting, are increasingly recognized as a valuable biomass enriched in bioactive phytochemicals with potential pharmaceutical, cosmetic, and food applications1. Traditionally, E. globulus leaves have been used in various ethnomedicinal contexts, especially in African and Mediterranean regions, for the treatment of wounds, respiratory conditions, and microbial infections2. Recent phytochemical and pharmacological studies support these uses, showing that leaf extracts contain a complex mixture of secondary metabolites, including terpenoids, flavonoids, tannins, and phenolic compounds. These constituents are associated with a broad range of biological activities, particularly antimicrobial, anti-inflammatory, and antioxidant effects3–5. A growing number of in vitro and in vivo studies have reported the efficacy of E. globulus leaf extract (EGL) against a broad spectrum of pathogenic microorganisms, suggesting its potential as a complementary or adjunctive strategy to conventional antibiotics. Among its bioactive constituents, the triterpenoids ursolic acid (UA) and asiatic acid (AA) have received increasing attention due to their well-known anti-inflammatory and antioxidant properties, as well as their emerging antimicrobial activity against clinically relevant bacterial species6,7. Despite this growing interest, most studies have focused on essential oils or isolated phytochemicals, whereas studies that combine comprehensive metabolomic characterization with antibacterial activity of EGL extracts remain limited8. Biofilm-associated infections represent a major clinical challenge, as bacterial communities embedded within extracellular polymeric matrices show increased tolerance to both antibiotics and host immune responses9. It is estimated that up to 80% of chronic and recurrent bacterial infections involve biofilm formation, making the prevention and early disruption of biofilms a key therapeutic priority10. The rise of antimicrobial resistance (AMR) further complicates this scenario, reducing the effectiveness of conventional antibiotics and underscoring the need for new strategies that target both planktonic bacteria and biofilm-associated phenotypes11. Understanding how chemically complex plant extracts influence bacterial behaviour is therefore an important step toward developing new phytochemical-based antimicrobial approaches. Gram-positive bacteria were selected as experimental models because several clinically relevant species in this group, including Staphylococcus and Streptococcus, are well known for forming persistent biofilms and causing chronic or recurrent infections12. In addition, these pathogens are frequently associated with AMR, making them relevant targets for evaluating alternative antibacterial strategies13. In this context, the present study combines metabolomic profiling with functional microbiological assays to investigate the antibacterial and antibiofilm activity of a chemically characterized EGL. Ultra-high-performance liquid chromatography coupled with high-resolution mass spectrometry (UHPLC-HESI-HRMS) was used to define the metabolic profile of the extract and to quantify two representative triterpenoids, ursolic acid (UA) and asiatic acid (AA). The activities of EGL and the two reference triterpenoids were evaluated against four clinically relevant Gram-positive pathogens known for their capacity to form persistent biofilms: Streptococcus pyogenes, Staphylococcus aureus, methicillin-resistant S. aureus (MRSA), and Streptococcus mutans. While AA and UA were tested as representative constituents to provide a comparative framework, the primary aim was to evaluate the behaviour of the whole phytocomplex. Increasing evidence indicates that bacterial metabolic pathways contribute to stress adaptation and biofilm formation14. In particular, pentacyclic triterpenoids such as ursolic acid (UA) and asiatic acid (AA) have been linked to effects on early biofilm development, through mechanisms that do not necessarily involve direct growth suppression15. Given that these compounds are present in EGL, we investigated whether the extract could exhibit comparable effects on early stages of biofilm formation. Overall, this integrated approach provides a framework for interpreting the biological activity of EGL in relation to its chemical complexity, highlighting its ability to modulate early biofilm formation in Gram-positive pathogens and supporting the exploration of phytochemical-based strategies targeting the initial stages of biofilm establishment. Results and discussion Untargeted metabolomic results The untargeted metabolomic analysis revealed a complex, highly resolved chemical profile of EGL. The extract was analyzed using UHPLC-TWIMS-QTOF in both positive and negative electrospray ionization modes to maximize metabolome coverage (Fig. 1). The EGL extract was prepared as described in the "Methods" section ("Sample preparation" section). Base peak chromatograms obtained by UHPLC–TWIMS–QTOF analysis of the quality control (QC) pooled sample of Eucalyptus globulus leaf extract. Data were acquired in both positive (green) and negative (purple) electrospray ionization modes. The QC sample, generated by pooling all extract replicates, was used to assess analytical stability and to provide an overview of the global metabolite profile across the chromatographic run. Initial peak detection revealed 5873 and 9556 features in negative and positive ionization modes, respectively. After quality filtering to remove poorly reproducible signals (CV% < 30%), features that showed robust detection across quality control samples and met stringent statistical criteria (ANOVA FDR-adjusted p-value < 0.01) were retained for downstream annotation. This process yielded 2487 features in negative-ion mode and 8355 in positive-ion mode. All retained features were subsequently subjected to putative identification using accurate mass, isotopic pattern, fragmentation spectra and, when available, collision cross section (CCS) information. Metabolite annotation and identification Putative annotation of significant features revealed a chemically diverse metabolite profile, based on matching accurate masses, isotope patterns, adduct types, MS/MS spectra and, when available, CCS values against public and in-house spectral libraries. A complete list of annotated metabolites for both ionization modes is provided in Supplementary Tables S1 (ESI-) and S2 (ESI+). Across the two datasets, a broad diversity of chemical classes was observed, confirming the high phytochemical complexity of EGL. Negative-ion mode predominantly revealed phenolic acids, flavonoids, glycosylated derivatives, hydrolysable tannins, terpenoids and coumarin-related structures. Positive-ion modes provided complementary coverage, including flavonoids, coumarins, anthraquinones, phenolic glycosides, terpenoids, fatty acid derivatives and several triterpenoid subclasses. A summary of chemical classes and representative structures is provided in Table 1. Table 1Summary of annotated metabolite classes in Eucalyptus globulus Leaf extract.Chemical classRepresentative compoundsDetected in ESI−Detected in ESI+Phenolic acidsGallic acid, Chlorogenic acidYesYesFlavonoidsKaempferol, Rutin, Myricitrin, HyperosideYesYesHydrolysable tanninsTellimagrandin II, PedunculaginYesNoTriterpenoidsUrsolic acid, Asiatic acid, MacrocarpalsYesYesCoumarinsMethylesculetin, ScopolineYesYesTerpenoidsEuglobal-type terpenoids, Ganoderone AYesYesAnthraquinonesAloe-emodin derivativesNoYesFatty acid derivativesLinoleoyl derivativesNoYesOligopeptidesShort linear peptidesNoYes Summary of annotated metabolite classes in Eucalyptus globulus Leaf extract. Annotation confidence was supported by low mass errors (< 5ppm), consistent fragmentation patterns (score 60–98), and retention-time alignment with known phytochemical classes. According to the Metabolomics Standard Initiative (MSI), most compounds were assigned to Level 2 or Level 3. In contrast, Level 1 confidence was achieved only for metabolites confirmed with authentic standards (asiatic and ursolic acid). Overall, the combined ESI-/ESI+ analysis highlighted a metabolomic fingerprint characterized by polyphenols, flavonoids, tannins, coumarins and pentacyclic triterpenoids, chemical families broadly associated with antimicrobial and antibiofilm properties, providing a coherent chemical context for the observed biological activity of EGL. Targeted quantification of ursolic acid and asiatic acid Targeted quantification of the two representative triterpenoids, asiatic acid (AA) and ursolic acid (UA), revealed measurable concentrations of these compounds in EGL using UHPLC-HESI-HRMS. Calibration curves generated from authentic analytical standards exhibited excellent linearity across the tested ranges (R² > 0.999 for both compounds)16. The analysis revealed that AA was the most abundant triterpenoid, at 47.29 ± 0.13 µg/mL. In comparison, UA was present at 16.12 ± 0.32 µg/mL, indicating that AA was approximately threefold higher than UA in the extract. In addition to their quantification in the extract, AA and UA were evaluated independently in the antibacterial and antibiofilm assays using commercially available analytical standards rather than compounds purified from the extract. This approach allowed comparison between the activity of the whole extract and that of two representative triterpenoid constituents. The presence of AA and UA at measurable concentrations suggests that these triterpenoids contribute to the biological activity of EGL, although the overall effect likely arises from the combined action of multiple phytochemical constituents. Antimicrobial activity The antibacterial activity of EGL and its major triterpenoids, UA and AA, was assessed against the following reference strains: Streptococcus mutans ATCC 35,668, Staphylococcus aureus ATCC 25,923, Methicillin-resistant Staphylococcus aureus (MRSA) ATCC 43,300, and Streptococcus pyogenes ATCC 19,615. These strains were selected because they represent clinically relevant Gram-positive pathogens frequently associated with antimicrobial resistance and biofilm-related infections17–19. Both MIC values and percentage growth inhibition for each tested concentration were reported, as these parameters offer complementary information. While MIC indicates the lowest concentration needed to completely inhibit bacterial growth, percentage growth inhibition describes the gradual reduction in bacterial growth across the tested concentration range. As shown in Figs. 2 and 3, EGL, UA, and AA exhibited a steady dose-dependent decrease in inhibitory activity, with complete inhibition only at the highest tested concentrations. These findings are consistent with the MIC values reported in Table 2. Specifically, EGL showed the strongest antibacterial activity against S. mutans (MIC = 62.5 ± 0 µg/mL) and S. pyogenes (MIC = 250 ± 0 µg/mL). Conversely, higher MIC values were observed against S. aureus (312.5 ± 115.7 µg/mL) and MRSA (437.5 ± 115.7 µg/mL), indicating reduced susceptibility in these strains. Inhibition percentages (%) of planktonic bacterial growth for each tested concentration of Eucalyptus globulus leaf extract (EGL-L; 1000 –1.95 µg/mL) against the tested reference bacterial strains. *SA= Staphylococcus aureus, MRSA= methicillin-resistant Staphylococcus aureus, SP= Streptococcus pyogenes and SM= Streptococcus mutans. Inhibition percentages (%) of planktonic bacterial growth for each tested concentration of Ursolic acid (AU; 256 –0.5 µg/mL) and Asiatic acid (AA; 256 –0.5 µg/mL) against the tested reference bacterial strains. *SA= Staphylococcus aureus, MRSA= Methicillin-resistant Staphylococcus aureus, SP= Streptococcus pyogenes and SM= Streptococcus mutans. Table 2Minimal Inhibition Concentration (MIC) expressed in µg/ml of ursolic acid (UA), asiatic acid (AA), and Eucalyptus globulus leaf extract (EGL) against tested reference bacterial strains.CompoundMIC (µg/ml ) S. mutans S. aureus MRSA S. pyogenes UA18 ± 9.3228 ± 7.4116 ± 02 ± 0AA18 ± 5.668 ± 08 ± 04 ± 0.71EGL62.5 ± 0312.5 ± 115.73437,5 ± 115.73250 ± 0 Minimal Inhibition Concentration (MIC) expressed in µg/ml of ursolic acid (UA), asiatic acid (AA), and Eucalyptus globulus leaf extract (EGL) against tested reference bacterial strains. Among the two pentacyclic triterpenes, UA exhibited the strongest antimicrobial activity against S. pyogenes, with a MIC value of 2 ± 0 µg/mL. In addition, it was also effective against S. aureus, MRSA, and S. mutans, with MIC values ranging from 16 to 28 µg/mL. Similarly, AA showed its strongest activity against S. pyogenes (MIC = 4 ± 0.71 µg/mL). In comparison, inhibitory effects were also observed against S. aureus and MRSA (MIC = 8 ± 0 µg/mL) and against S. mutans (MIC = 18 ± 5.66 µg/mL). Overall, both UA and AA displayed stronger antibacterial activity against Streptococcus species, particularly S. pyogenes. These findings are consistent with the activity profile observed for EGL and suggest that its antibacterial effects may be partially attributable to its triterpenoid constituents. Biofilm inhibition The antibiofilm activity of EGL, UA, and AA was assessed through MBIC assays against the same Gram-positive pathogens20,21. The data reported in Tables 3 and 4 show that the tested compounds exhibited varying levels of antibiofilm activity across bacterial species. Table 3Minimum biofilm inhibitory concentrations (MBIC) (µg/mL) of asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) against tested reference bacterial strains.Compound S. mutans S. aureus MRSA S. pyogenes UA4 ± 0 µg/mL––16 ± 0 µg/mLAA8 ± 0 µg/mL32 ± 0 µg/mL32 ± 0 µg/mL8 ± 0 µg/mLEGL extract–––– Minimum biofilm inhibitory concentrations (MBIC) (µg/mL) of asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) against tested reference bacterial strains. Table 4MBIC90, MBIC80, and MBIC50 (µg/mL) of UA and EGL extract against the tested reference bacterial strains.CompoundBacteriaMBIC90 µg/mLMBIC80µg/mLMBIC50µg/mLUA S. aureus 2566432MRSA1286432EGL extract S. mutans ––31.25 S. aureus 5007.813.9MRSA2502503.9 S. pyogenes 10001257.81 MBIC90, MBIC80, and MBIC50 (µg/mL) of UA and EGL extract against the tested reference bacterial strains. Overall, EGL exhibited lower antibiofilm activity than its main triterpenoid constituents, as complete MBIC values were not achieved for all tested strains, and only partial biofilm inhibition was observed. The highest antibiofilm activity of EGL was observed against Staphylococcus aureus (MBIC90=500 µg/mL, MBIC80=7.81 µg/mL and MBIC50=3.9 µg/mL) and MRSA (MBIC90=250 µg/mL and MBIC50=3.9 µg/mL). Lower antibiofilm activity was observed against S. pyogenes, requiring higher concentrations to reach MBIC90 (1000 µg/mL), MBIC80 (125 µg/mL), and MBIC50 (7.81 µg/mL). Against S. mutans, only MBIC50 was reached at 31.25 µg/mL, indicating a markedly lower antibiofilm efficacy against this strain. AA and UA demonstrated notable inhibitory biofilm activity. Specifically, AA showed a broader biofilm-inhibitory profile, with MBIC values of 8 ± 0 µg/mL for S. mutans and S. pyogenes, and 32 ± 0 µg/mL for S. aureus and MRSA. UA exhibited a more selective inhibitory profile among the bacterial strains tested, achieving measurable MBIC values only for members of the Streptococcaceae family (4.0 ± 0 µg/mL for S. mutans and 14 ± 3.7 µg/mL for S. pyogenes). However, complete MBIC values were not observed for S. aureus or MRSA, where only partial inhibition of biofilm formation was recorded (MBIC₉₀ = 256 µg/mL for S. aureus and 128 µg/mL for MRSA). Overall, both UA and AA exhibited stronger antibiofilm activity than the whole extract. Their similar activity profiles may reflect shared structural features, although strain-specific factors likely contribute to the observed variability among the tested bacteria. The lower activity observed for EGL may be related to the dilution of active constituents within the phytocomplex and to the contribution of multiple components with different modes of action. This behavior is consistent with the presence of these triterpenoids in EGL, although their contribution alone does not fully account for the activity of the extract. MBIC with pre-incubation MBIC assays under pre-incubation conditions were conducted to evaluate the effect of early exposure to UA, AA, and EGL on biofilm formation in Gram-positive bacteria. In this experiment, bacterial cultures were exposed to the tested compounds 24 h before biofilm induction, allowing evaluation of their effects during the early stages of biofilm formation22–24. In addition to the reference compounds AA and UA, the whole EGL extract was evaluated to compare the activity of the phytocomplex with that of two representative triterpenoid constituents. Previous studies have reported that ursolic acid can reduce biofilm formation through mechanisms that do not necessarily involve direct inhibition of bacterial growth. Moreover, modulation of metabolic pathways, including those related to sulfur metabolism and redox homeostasis, has been suggested to influence biofilm development. Although these mechanisms have been mainly described in Gram-negative models, conserved metabolic processes are also relevant in Gram-positive bacteria and may contribute to early-stage biofilm formation25–27. Table 5; Fig. 4 report the MBIC values obtained under pre-incubation conditions. Overall, AA exhibited the strongest antibiofilm activity, inhibiting biofilm formation in all tested bacterial strains. Specifically, AA showed pre-incubation MBIC values of 2 ± 0 µg/mL for S. mutans and 8 ± 0 µg/mL for both S. aureus and S. pyogenes. For MRSA, a pre-incubation MBIC value of 16 µg/mL was obtained. UA displayed MBIC activity under pre-incubation conditions only against S. aureus and S. pyogenes, with values of 16 µg/mL and 8 µg/mL, respectively. In contrast, EGL exhibited MBIC activity only against S. aureus (250 µg/mL), indicating substantially lower antibiofilm potency than the purified triterpenoids. Comparison of MBIC values obtained with and without pre-incubation (Table 6) revealed an overall increase in antibiofilm activity following early exposure to the tested compounds. In particular, UA pre-treatment enhanced biofilm inhibition against S. pyogenes, reducing the MBIC from 16 to 8 µg/mL, and enabled UA to reach an MBIC for S. aureus (16 µg/mL). In contrast, under standard conditions, only an MBIC₉₀ value had previously been obtained (256 µg/mL). A similar trend was observed for AA, which exhibited the most pronounced enhancement under pre-incubation conditions. Pre-treatment with AA resulted in a several-fold reduction of MBIC values across all tested bacterial strains (ranging from 2- to 4-fold, depending on the organism), confirming its strong ability to disrupt early biofilm establishment. These observations are consistent with the possibility that prolonged exposure to AA and UA may interfere with early regulatory processes involved in biofilm development. Although transcriptional regulation in Gram-positive bacteria differs from that described in Gram-negative models, conserved metabolic pathways involved in redox balance and biofilm physiology could represent common targets affected during the initial stages of biofilm formation28. Table 5Results obtained after 24-hour pre-incubation of gram-positive bacteria with asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) and tested reference bacterial strains.Compound S. mutans S. aureus MRSA S. pyogenes UA4 ± 0 µg/ml16 ± 0 µg/ml> 16 ± 0 µg/ml8 ± 0 µg/mlAA2 ± 0 µg/ml8 ± 0 µg/ml16 µg/ml8 ± 0 µg/mlEGL> 500 µg/ml250 ± 0 µg/ml> 250 ± 0 µg/ml> 250 µg/ml Results obtained after 24-hour pre-incubation of gram-positive bacteria with asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) and tested reference bacterial strains. The minimum biofilm inhibitory concentration (MBIC-PRE) of asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) was tested against reference bacterial strains. Table 6Comparison of MBIC and MBIC in pre-incubation of asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) on reference bacterial strains.Compound S. mutans S. aureus MRSA S. pyogenes MBIC (µg/mL)MBIC-Preincubation (µg/mL)MBIC (µg/mL)MBIC-Preincubation (µg/mL)MBIC (µg/mL)MBIC-Preincubation (µg/mL)MBIC (µg/mL)MBIC-Preincubation (µg/mL)UA4 ± 04 ± 0> 256 (MBIC90 = 256 µg/mL)16 ± 0> 256 ± 0 (MBIC90 = 128 µg/mL)> 16 ± 0 (93.37% inhibition at 16 µg/mL)16 ± 08 ± 0AA8 ± 02 ± 032 ± 08 ± 032 ± 0168 ± 08 ± 0EGL> 1000 (MBIC50 = 31.25 µg/mL)> 500 (96.57% inhibition at 62.5 µg/mL)> 1000 (MBIC90 = 500 µg/mL)250 ± 0> 1000 (MBIC90 at 250 µg/mL)> 250 ± 0 (98.64% inhibition at 250 µg/mL)> 1000 (MBIC90 at 1000 µg/mL)> 250 ± 0 (90.8% inhibition at 62.5 µg/mL) Comparison of MBIC and MBIC in pre-incubation of asiatic acid (AA), ursolic acid (UA), and Eucalyptus globulus leaf extract (EGL) on reference bacterial strains. Lastly, EGL pre-incubation resulted in modest improvements, since pre-incubation MBIC was reached only against S. aureus. This partial enhancement may reflect the contribution of multiple constituents in the extract, including AA and UA, whose individual activities are attenuated within the phytocomplex. Taken together, these results indicate that early exposure to pentacyclic triterpenoids, as well as to the whole extract, enhances antibiofilm activity under pre-incubation conditions, likely by affecting early-stage physiological processes required for biofilm establishment. The more pronounced effects observed for the isolated compounds are consistent with their greater direct activity, whereas the extract appears to exert a more moderate yet potentially broader modulatory effect, reflecting the combined contributions of multiple constituents rather than that of a single dominant compound. The specific molecular mechanisms underlying these observations remain to be elucidated. Correlation between metabolomic composition and biological activity The integration of untargeted metabolomic profiling with functional microbiological data offers a valuable framework for understanding the biological behaviour of EGL. Although the metabolomic analysis was mainly qualitative, the distribution and relative abundance of the primary metabolite classes provide a clear chemical context for the antibacterial and antibiofilm effects observed in the biological assays. Similar integrative approaches combining phytochemical characterization with biological activity assessment have been reported for other medicinal plants. For instance, metabolomic and bioactivity-guided studies of Cissus incisa extracts identified bioactive metabolites, including sphingolipids and other constituents with antibacterial and cytotoxic properties, demonstrating how detailed chemical profiling can support the interpretation of biological effects in plant-derived extracts29,30. The targeted quantification confirmed the presence of two representative pentacyclic triterpenoids, AA and UA, in the extract at measurable concentrations. These compounds are known to possess antibacterial and antibiofilm properties, and their activity is consistent with the effects observed for the whole extract, supporting their contribution to the antibacterial and antibiofilm behavior of EGL. However, their individual activity does not fully account for the overall biological response. Plant extracts frequently exhibit biological activity through additive or synergistic interactions among multiple constituents. Accordingly, although AA and UA are likely to contribute to the antibacterial effects of EGL, their detected concentrations do not imply that these compounds alone account for the overall biological activity. Rather, the observed effects are more plausibly explained by the combined contribution of multiple phytochemical components. Untargeted analysis also revealed a substantial representation of phenolic acids, flavonoids, and other polyphenolic scaffolds. While these metabolites are not individually potent antibacterials at the concentrations typically present in crude extracts, their combined presence may contribute to the gradual, concentration-dependent antibiofilm effects observed for EGL. Notably, several of the detected metabolite classes are commonly associated with redox-active chemical scaffolds31. The enrichment of such compounds in the metabolomic profile provides a plausible chemical context for the enhanced antibiofilm effects observed under pre-incubation conditions, in which prolonged exposure to EGL increased inhibition of biofilm formation despite limited bactericidal activity. While specific molecular targets cannot be identified based on the present data, the convergence between metabolomic composition and time-dependent biological responses supports the hypothesis that EGL interferes with early physiological processes involved in biofilm establishment. The alignment between metabolomic signatures and functional assay outcomes indicates that the antibacterial and antibiofilm properties of EGL arise from the combined contribution of multiple metabolite classes. This multicomponent behaviour is consistent with the observed hierarchy of biological effects, which are stronger for isolated triterpenoids and more moderate yet reproducible for the whole extract. Together, these findings support the interpretation of EGL as a chemically characterized, multicomponent system whose biological activity reflects the combined contribution of multiple constituents rather than the action of a single dominant compound. Mammalian cell viability The effects of EGL, UA, and AA on the viability of MDBK (Madin–Darby Bovine Kidney) cells were evaluated. The results for the extract and the pentacyclic triterpenes are shown in Figs. 5 and 6. The EGL extract did not significantly affect cell viability across the tested concentration range in this cell model. Cell viability remained high even at the highest rate (1000 µg/mL), reaching 151.38%. Similarly, viability values of 114.06% and 93.1% were recorded at 500 and 250 µg/mL, respectively. At lower concentrations (125–7.8 µg/mL), cell viability consistently exceeded 90%, ranging from 95.38% to 113.97%. Cytotoxicity activity of Eucalyptus globulus leaf extract (EGL-L) on MDBK cell lines. Cytotoxicity activity of asiatic acid (AA) and ursolic acid (UA)on MDBK cell lines. These findings may reflect an apparent increase in metabolic activity, although potential interference of extract components with the MTT assay cannot be excluded. However, it is well documented that naturally colored compounds can interfere with colourimetric viability assays such as MTT, potentially leading to overestimated viability values32. This concern is less relevant for UA and AA, as these compounds do not exhibit significant intrinsic absorbance at the assay wavelength. Specifically, UA displayed marked cytotoxicity at concentrations ≥ 16 µg/mL, with cell survival dropping to ≤ 40%. In contrast, concentrations ≤ 8 µg/mL were associated with cell viability ≥ 80%, exceeding the commonly accepted toxicity threshold33. Notably, at concentrations of UA at or below 8 µg/mL, where UA exhibited antibiofilm activity, cell viability values were 89.94% (8 µg/mL), 118.46% (4 µg/mL), and 99.86% (2 µg/mL), indicating good tolerability at biologically relevant concentrations. Importantly, these concentrations correspond to the range in which UA displayed antibiofilm activity, indicating a favourable overlap between antibacterial efficacy and acceptable mammalian cell viability. AA exhibited lower cytotoxicity than UA, as all concentrations ≤ 32 µg/mL maintained a cell survival rate ≥ 70%. Overall, these results highlight a key difference between the isolated triterpenoids and the complete phytocomplex. Specifically, both acids exhibited substantial cytotoxicity at high concentrations, whereas the whole extract demonstrated good apparent biocompatibility across the tested range. However, given the potential limitations of the MTT assay discussed above, further studies will be required to confirm these observations and better assess the safety profile of the extract and its active constituents. Methods Chemicals Asiatic acid (AA) and ursolic acid (UA) were used as purified reference compounds and purchased from Sigma-Aldrich (Merck, Schnelldorf, Germany; purity ≥ 98%, HPLC grade). LC–MS grade methanol and acetonitrile were purchased from Scharlab Italia S.r.l. (Milan, Italy). Ultrapure water was obtained using a Milli-Q system (Millipore, Bedford, MA, USA). MS-grade ammonium acetate, acetic acid, and formic acid were purchased from Fisher Chemical (Thermo Fisher Scientific Inc., San Jose, CA, USA). Plant material collection and identification Leaves of Eucalyptus globulus (Labill., 1800) were collected in March 2023 from eight cultivated trees located in the Botanical Garden of the University INES-Ruhengeri, Musanze, Rwanda (GPS coordinates: −1.5014° S, 29.6108° E), with the permission of the pro-tempore Vice Chancellor, Dr. Fr. Fabien Hagenimana. The collection of plant material complied with relevant institutional, national, and international regulations. The species is not listed as endangered in the International Union for Conservation of Nature (IUCN) Red List and is not included in the appendices of the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES). Leaf collection does not harm the plants or compromise their survival. As the material was collected from cultivated trees within the University Botanical Garden, no additional government permits were required. Botanical identification was carried out by Dr. Aimable Nsanzurwimo, botanist at the University INES-Ruhengeri (Musanze, Rwanda). Before use, leaves were washed with tap water and air-dried at room temperature for three weeks. The dried material was manually ground using a mortar and pestle and subsequently passed through an 800 μm sieve to obtain a more uniform particle size fraction. Sample preparation Dried leaves (20 g) were independently extracted in triplicate by hydroalcoholic maceration using 70% ethanol (plant material: solvent ratio 1:10, w/v) in hermetically sealed flasks. The extraction was performed for 72 h at 20–25 °C under continuous magnetic stirring (100 rpm). After extraction, the three hydroalcoholic solutions were filtered through Whatman No. 1 filter paper using a Buchner funnel. The organic solvent was removed under reduced pressure using a rotary evaporator (Buchi), and the remaining aqueous phase was frozen with liquid nitrogen and subsequently lyophilized (− 56 °C, 1.0 mbar) for 36 h (Alpha 1–2 LDplus freeze dryer, Martin Christ, Germany). The dried extracts obtained from the three independent extractions were pooled, aliquoted and stored until use. For UHPLC–TWIMS–QTOF analysis, samples were reconstituted in 70% methanol at a concentration of 1 mg/mL. UHPLC-TWIMS-QTOF An ACQUITY I-Class UPLC separation system coupled with a VION IMS QTOF mass spectrometer (Waters, Wilmslow, UK) equipped with an electrospray ionization (ESI) interface was employed for untargeted metabolomics. Extract samples, prepared as described in "Sample preparation" section, were injected (1 µL) and chromatographically separated using a reversed-phase C18 BEH ACQUITY column (2.1 × 100 mm, 1.7 μm; Waters, Milford, MA, USA). A gradient profile was applied using water containing 5 mM ammonium formate (eluent A) and acetonitrile (eluent B), both acidified with 0.1% formic acid. Initial conditions were set at 5% B. After a 1.5 min isocratic step, a linear gradient was applied to 100% B over 13.5 min. The system was held at 100% B for 5 min for column washing, followed by re-equilibration to initial conditions over 5 min, resulting in a total run time of 25 min. The column temperature was maintained at 45 °C, and a flow rate of 0.35 mL/min was set. Mass spectrometry data were acquired in both positive and negative electrospray ionization modes over an m/z range of 50 − 1000 (see Fig. 1). Source parameters were as follows: capillary voltage 1.5 kV (positive) and 2.0 kV (negative), source temperature 120 °C, desolvation temperature 650 °C, and desolvation gas flow 950 L/h. The TOF analyser was operated in sensitivity mode, and data were acquired using HDMSE, a data-independent acquisition (DIA) approach coupled with ion mobility separation. Ion mobility parameters were set as follows: nitrogen flow rate 90 mL/min (3.2 mbar), wave velocity 650 m/s, and wave height 40 V. The instrument was calibrated using the Major Mix IMS calibration kit (Waters, Wilmslow, UK) to allow for CCS values to be determined in nitrogen. The calibration covered the CCS range from 130 to 306 Å2. The TOF was also calibrated before data acquisition and covered the mass range from 151 Da to 1013 Da. TOF and CCS calibrations were performed for both positive and negative ion modes. Data acquisition was conducted using UNIFI 1.8 (Waters, Wilmslow, UK). Quality control (QC) samples were prepared by pooling aliquots of all extracts and were periodically injected throughout the analytical sequence to monitor system stability and data reproducibility. Data processing Data processing and compound identification were conducted using Progenesis QI Informatics (Nonlinear Dynamics, Newcastle, UK). Each UHPLC-MS run was imported as an ion-intensity map, including m/z (50–1000) and retention time, and the data were then aligned along the retention time axis (0–20 min). An aggregate run comprising all samples was generated and used for peak picking, ensuring consistent ion detection across all runs. Isotope and adduct deconvolution were applied to reduce data redundancy. Quality filtering was performed by retaining features with a coefficient of variation (CV) ≤ 30% across quality control (QC) samples. QC samples were prepared by pooling aliquots of all extracts and injected periodically throughout the analytical sequence to monitor system stability. Statistical analysis was carried out using analysis of variance (ANOVA), and features meeting stringent criteria (FDR-adjusted p-value < 0.01) were retained for further analysis. In parallel, multivariate supervised models, including partial least-squares discriminant analysis (PLS-DA), were built and validated using SIMCA software (v. 16.0.2, Sartorius Stedim Data Analytics, Sweden) to identify discriminative metabolic patterns. Cross-validation of the PLS-DA model was performed using a leave-one-third-out approach, and permutation testing was applied to exclude overfitting by inspecting model parameters (goodness-of-fit R²Y and goodness-of-prediction Q²Y). Variable Importance in Projection (VIP) scores were further used to identify compounds with the highest discriminative potential (VIP values > 1.2). Features meeting both criteria (ANOVA p < 0.01 and VIP > 1.2) were retained for annotation. Metabolite annotation was performed using publicly available databases, including the Lipid Metabolites and Pathways Strategy (LIPID MAPS) and the Human Metabolome Database (HMDB), combined with fragmentation patterns, retention times, and CCS. Based on the Metabolomics Standards Initiative32, metabolites were annotated as level II (putatively identified compounds). Quantification of selected compounds HPLC-MS analysis was performed using a 2695 Alliance separation system (Waters, Milford, MA, USA) equipped with a QuattroTM API triple quadrupole mass spectrometer with an electrospray source (Micromass/Water, Manchester, UK). Chromatographic separation was achieved using an XSelect® HSS T3 column (250 × 2.1 mm, 5 μm). The mobile phase consisted of water (eluent A) and acetonitrile (eluent B), both containing 0.1% formic acid. The gradient started at 100% A, held for 5 min, then increased linearly to 100% B at 8 min, was maintained for 5 min, and then returned to initial conditions. Column re-equilibration was achieved over 6 min, resulting in a total run time of 20 min. The column temperature was maintained at 30 °C, and the flow rate was set at 0.20 mL/min. Mass spectrometry was performed in negative ESI mode under the following conditions: capillary voltage 2.5 kV, cone voltage 25 V, extractor voltage 2 V, source temperature 120 °C, and desolvation temperature 350 °C. Nitrogen was used as the desolvation gas, and argon was used as the collision gas. Selective ion monitoring (SIM) mode was employed for quantification. Ursolic acid was monitored at m/z 455.4 [M–H]⁻ and asiatic acid at m/z 487.7 [M–H]⁻. External calibration curves were constructed using standard solutions of AA and UA and used for quantitative analysis of these compounds in the EGL extract. Calibration curves showed good linearity within the tested concentration range (R² > 0.99). Tested bacteria and inoculum preparation The following reference strains were used in this study: Streptococcus mutans ATCC 35,668, Staphylococcus aureus ATCC 25,923, methicillin-resistant Staphylococcus aureus (MRSA) ATCC 43,300, and Streptococcus pyogenes ATCC 19,615. Bacterial inocula were prepared according to Clinical and Laboratory Standards Institute (CLSI) guidelines34. All microbiological assays were performed within 30 min of inoculum standardization. Briefly, five colonies from fresh solid cultures of each strain were transferred into sterile tubes containing Mueller Hinton broth (MHB) and incubated at 37 °C for 18–24 h under appropriate atmospheric conditions (aerobic for staphylococci and 5% CO₂ for streptococci). After incubation, cultures were centrifuged at 2000 rpm for 20 min at 4 °C. The resulting bacterial pellets were resuspended in 10 mM phosphate buffer (PB, pH 7.0). This washing step was introduced to remove residual growth medium and ensure reproducible starting conditions across experiments. The bacterial suspension was adjusted spectrophotometrically to an optical density (OD600) of 0.08–0.13 using a 1 cm path-length cuvette, corresponding to approximately a 0.5 McFarland standard (≈ 1.5 × 10⁸ CFU/mL). This suspension was then diluted 1:100 in sterile MHB to obtain the working inoculum used for antimicrobial assays. Aliquots of 50 µL of the resulting bacterial suspension were inoculated into each well to achieve a final bacterial concentration of 5 × 10⁵ CFU/mL. Minimum inhibitory concentration Minimum inhibitory concentrations (MICs) of the tested compounds were determined using a broth microdilution method, in accordance with previously published procedures6,35 and consistent with CLSI guidelines. Stock solutions of EGL-L, UA, and AA were prepared in dimethyl sulfoxide (DMSO) at concentrations of 100 mg/mL for EGL-L and 25.6 mg/mL for UA and AA. Twofold serial dilutions of each compound were prepared in DMSO in 96-well microtiter plates (Greiner, Milan, Italy): from 100 mg/mL to 0.195 mg/mL of EGL-L and from 25.6 mg/mL to 0.05 mg/mL of UA and AA. Since the final concentration of DMSO in assay wells should not exceed 1% (v/v) to not affect bacterial growth, aliquots of 1 µL of each compound diluted in DMSO were added to 49 µL of MHB per well, followed by 50 µL of bacterial suspension adjusted to 1 × 10⁶ CFU/mL, resulting in a final inoculum of 5 × 10⁵ CFU/mL. Consequently, the final concentration ranges were 1000–1.95 µg/mL for EGL-L and 256–0.5 µg/mL for UA and AA. Growth and sterility controls were included for each strain and compound: growth controls with and without 1% of DMSO were prepared to verify not to affect bacterial growth. Plates were incubated at 37 °C for 24 h under strain-specific conditions. Each assay was performed in three independent experiments with three technical replicates per condition. After incubation, bacterial growth was assessed both visually using a microtiter plate reading mirror and spectrophotometrically by measuring optical density at 620 nm with a Multiskan™ FC microplate reader (Thermo Fisher Scientific). MIC values were defined as the lowest concentration that showed no visible bacterial growth and were reported as the mean ± standard deviation (SD) across independent experiments. Escherichia coli ATCC 25,922 was periodically tested as a quality control strain to ensure procedural accuracy. Minimum biofilm inhibitory concentration (MBIC) assay The Minimum Biofilm Inhibitory Concentrations (MBICs) of the tested compounds were determined following previously described methods23,36. Briefly, 178 µL of tryptic soy broth (TSB) supplemented with 1% (w/v) glucose was dispensed into each well of a flat-bottom 96-well microtiter plate (Greiner, Milan, Italy). Twofold serial dilutions of EGL-L, UA, and AA stock solutions in DMSO were prepared in a separate plate, and 2 µL of each dilution was added to the corresponding wells. Final concentration ranges were 1000 –1.95 µg/mL for EGL-L and 256 –0.5 µg/mL for AA and UA. The final concentration of DMSO in assay wells did not exceed 1% (v/v) and control wells with and without 1% of DMSO were prepared to verify not to affect biofilm formation. Bacterial pellets were prepared as described above, resuspended in phosphate buffer, and adjusted to a concentration of 1 × 107 CFU/mL by spectrophotometry. Subsequently, 20 µL of the bacterial suspension was added to each well. Growth controls without DMSO consisted of 180 µL of TSB supplemented with 1% glucose plus 20 µL of bacterial suspension (1 × 10⁷ CFU/mL), while controls with 1% of DMSO consisted in 178 µL of TSB with 1% of glucose, 2 µL of DMSO, and 20 µL of bacterial suspension. Sterility controls contained 200 µL of TSB supplemented with 1% glucose only. Plates were incubated at 37 °C for 24 h under strain-specific conditions. After incubation, planktonic cells were removed, and wells were gently washed with sterile buffer. Biofilm formation was quantified using crystal violet staining, followed by solubilization and measurement of optical density at 620 nm using a microplate reader. The percentage of biofilm inhibition was calculated according to the following equation: where ODGC is the mean optical density of the growth control, and ODx is the mean optical density of the tested concentration. MBIC50, MBIC80, and MBIC90 were defined as the lowest concentrations of each compound that inhibited biofilm formation by 50%, 80%, and 90%, respectively, relative to the growth control. All experiments were performed in triplicate and repeated independently three times. MBIC with pre-incubation To evaluate the effects of early exposure to EGL, UA, and AA on biofilm formation, a pre-incubation protocol was implemented before the standard MBIC assay. Briefly, 4–5 colonies of each strain were inoculated into Mueller–Hinton broth (MHB) and incubated at 37 °C for 24 h. Cultures were centrifuged (2000 rpm, 20 min, 4 °C), and bacterial pellets were resuspended in phosphate buffer and adjusted spectrophotometrically to approximately 1 × 10⁸ CFU/mL (OD600 = 0.08–0.13). An aliquot of 50 µL of the 10⁸ CFU/mL bacterial suspension was added to 9.95 mL of TSB supplemented with 1% glucose containing UA or AA at 64, 32, 16, or 8 µg/mL, or EGL at 2000, 1000, 500, or 250 µg/mL, resulting in a final bacterial concentration of 5 × 10⁵ CFU/mL. Growth controls were prepared under identical conditions without compounds. Cultures were incubated at 37 °C on an orbital shaker (200 rpm) until they reached a concentration of 1 × 10⁸ CFU/mL (OD600 = 0.08–0.13). At this stage, 80 µL of each culture was transferred into 7.92 mL of fresh TSB supplemented with 1% glucose containing the corresponding compound concentrations, yielding a working inoculum of approximately 1 × 10⁶ CFU/mL. Aliquots of 200 µL were then dispensed into flat-bottom 96-well microtiter plates and incubated at 37 °C for 24 h under strain-specific conditions. Biofilm formation was evaluated as described above ("Minimum biofilm inhibitory concentration (MBIC) assay" section), following Stepanovic et al.36. After incubation, biofilm formation was quantified using a resazurin-based viability assay followed by crystal violet staining, as described in "Minimum biofilm inhibitory concentration (MBIC) assay" section. All experiments were performed in triplicate and repeated independently three times. Mammalian cell viability assay The Madin–Darby Bovine Kidney (MDBK) cell line was obtained from the American Type Culture Collection (ATCC, Manassas, VA, USA; CRL-6071) and maintained according to the supplier’s recommendations. Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% (v/v) fetal bovine serum (FBS), 1% penicillin–streptomycin and incubated at 37 °C in a humidified atmosphere containing 5% CO₂. For viability assays, MDBK cells were seeded in 96-well plates at a density of approximately 1 × 10⁴ cells per well and allowed to adhere for 24 h. Cells were then treated with increasing concentrations of EGL, UA, and AA for 24 h. The seeding density was optimized according to the specific experimental conditions, including cell type, growth rate, and assay requirements37,38. Cells were then treated with increasing concentrations of EGL, UA, and AA for 24 h. The final concentration of DMSO in culture medium did not exceed 1% (v/v). Cell viability was assessed using the MTT assay, as reported by Donofrio et al.39, with minor modifications. Briefly, after treatment, 10 µL of 5 mg/mL MTT solution was added to each well, and the plate was incubated for 3–4 h at 37 °C. The yellow tetrazolium MTT salt is reduced in metabolically active cells to form insoluble purple formazan crystals, which are solubilized by adding 100 µL of a solubilization solution (10% SDS in 0.01 M HCl) to each well and incubating overnight. Cell viability was expressed as a percentage relative to untreated control cells. All experiments were performed in triplicate and repeated at least three times independently. Statistical analysis and data visualization Statistical analyses were performed using IBM SPSS v.23.0 (SPSS Italia, Bologna, Italy) and TIBCO Statistica software (TIBCO Software Inc., Palo Alto, CA, USA). Data are presented as mean ± standard deviation (SD), unless otherwise specified. Differences between groups were evaluated using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. Statistical significance was set at p < 0.05. Inferential statistical analysis was applied to metabolomic datasets. Microbiological data (MIC and MBIC values) are presented as mean ± SD obtained from three independent experiments performed in triplicate and were not subjected to inferential statistical analysis, as these assays are based on threshold-defined microbiological endpoints. No statistical comparison was performed for MIC and MBIC values, as these represent threshold-based microbiological endpoints. Data visualization of microbiological results was performed using R (version 4.4.2) within the RStudio environment. Heatmaps were generated using the ComplexHeatmap package, with data manipulation and formatting carried out using the tidyverse package. Color scaling was customized using the circlize package to reflect inhibition and MIC values. Conclusions This study provides an integrated characterization of EGL by combining untargeted metabolomics, targeted quantification of triterpenoids, and functional antibacterial and antibiofilm assays. The metabolomic analysis revealed a chemically rich phytocomplex characterized by polyphenols and pentacyclic triterpenoids, with asiatic acid and ursolic acid identified as major constituents at quantifiable, reproducible levels. Functionally, the isolated triterpenoids exhibited pronounced antibacterial and antibiofilm activity against Gram-positive pathogens, whereas the whole extract displayed moderate but consistent inhibitory effects. Notably, enhanced antibiofilm activity was observed under pre-exposure conditions, suggesting that EGL and its major constituents can interfere with early stages of biofilm establishment rather than acting solely through direct bactericidal effects. Notably, selected constituents, such as asiatic acid and ursolic acid, exhibited antibacterial and antibiofilm activities consistent with those observed for the whole extract, supporting their contribution, though they do not fully account for the overall biological response. Although the present work provides an integrated analytical and functional characterization of EGL, several limitations should be acknowledged. First, the biological evaluation was performed using in vitro microbiological assays and reference bacterial strains; therefore, the observed antibacterial and antibiofilm effects require further validation in more complex biological models. Second, while metabolomic profiling allowed the identification of major metabolite classes and the quantification of key triterpenoids, the specific molecular mechanisms underlying the observed antibiofilm activity remain to be elucidated. In particular, further transcriptomic or proteomic investigations will be necessary to determine whether the compounds interfere with defined metabolic or regulatory pathways involved in biofilm establishment. Overall, these findings highlight EGL as a chemically complex system whose biological activity reflects its compositional diversity and its ability to modulate early stages of biofilm development, presenting EGL as a chemically complex system capable of influencing early biofilm formation, rather than acting through a single dominant component. Electronic Supplementary Material Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements This work has been carried out in the frame of the ALIFAR project of the University of Parma, funded by the Italian Ministry of University through the program “Dipartimenti di Eccellenza 2023–2027”. Author contributions Chiara Spaggiari and Costanza Spadini contributed equally to this work.Chiara Spaggiari, Costanza Spadini, Laura Righetti, Nicolò Mezzasalma and Alessandro Cupola performed the experimental work and data acquisition. Costanza Spadini, Nicolò Mezzasalma, Alessandro Cupola and Clotilde Silvia Cabassi carried out the microbiological assays. Giannamaria Annunziato, Chiara Spaggiari, Costanza Spadini, Laura Righetti and Marco Pieroni contributed to data analysis and interpretation. Giannamaria Annunziato, Costanza Spadini and Nicolò Mezzasalma wrote the manuscript. Giannamaria Annunziato coordinated the research activities. Gabriele Costantino conceived the study, coordinated the project and acquired the funding.All authors critically revised the manuscript and approved the final version. Data availability The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request, Giannamaria Annunziato (Email: giannamaria.annunziato@unipr.it). Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Chiara Spaggiari and Costanza Spadini are co-first authors. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement The datasets generated and/or analyzed during the current study are available from the corresponding author upon reasonable request, Giannamaria Annunziato (Email: giannamaria.annunziato@unipr.it).