Metabolomic Profiling and Antioxidant Properties of Chilean Eucryphia cordifolia Cav.: Insights from Leaves, Flowers, and Monofloral Honey

Abstract

This study aimed to characterize the metabolomic profile of monofloral honey from Eucryphia cordifolia (ulmo) and evaluate the potential transfer of bioactive compounds from the plant parts, including the leaves and flowers, to the honey. Using UHPLC/Q-TOF-MS analysis, various flavonoids and phenolic acids were identified and quantified in extracts from the leaves, flowers, and honey from E. cordifolia. Given their rich polyphenolic composition, E. cordifolia leaves were included in this study to assess their potential contribution to the antioxidant properties and chemical markers of ulmo honey. Additionally, the polyphenolic compounds in honey samples were quantified. Chromatographic analysis via UHPLC-MS/MS revealed that ulmo honey contains phenolic acids such as gallic, syringic, ferulic, chlorogenic, caffeic, and coumaric acid, as well as flavonoids including pinocembrin, quercetin, luteolin, kaempferol, epicatechin, apigenin, and isorhamnetin. The results indicate that pinocembrin and gallic acid are the main chemical markers of ulmo honey, while isorhamnetin could complement its characterization as a complementary marker. UHPLC/Q-TOF-MS analysis was also utilized to compare the compounds present in the honey with those found in the plant parts (leaves and flowers), respectively. A total of 10 shared compounds were identified, 9 of which were preliminarily identified, while 1 remains unknown. Notably, dihydroquercetin 3-O-rhamnoside, quercetin 3-O-rhamnoside, cyanidin 3-(p-coumaroyl)-glucoside, and eupatorin were detected in ulmo honey for the first time. Along with gallic acid, pinocembrin, and isorhamnetin, these compounds could contribute to a characteristic fingerprint for identifying the botanical origin of the honey. Overall, these findings provide valuable insights into the chemical composition of ulmo honey and its potential application as a functional product with antioxidant properties.

1. Introduction

Honey is a natural food product with a sweet taste that is produced by the honeybee (Apis mellifera L.) and known for its high nutritional and medicinal value. In the Egyptian and Greek cultures, it was used to treat various ailments, like stomach ulcers and skin wounds [1]. Honey is a natural matrix composed of approximately 80–85% total carbohydrates (about 40% fructose, 30% glucose, and 10–15% maltose, sucrose, and other sugars), with the exact composition largely depending on the plant species from which the nectar or exudate was collected, as well as the environmental conditions and climate [2].

Chile ranks 29th among global honey producers with a production of 11,600 tons of honey annually (0.68% of global production), of which 10,000 tons are exported [3]. Chilean honey is highly appreciated due to its nematicidal and antifungal properties and its known chemical characteristics and biological attributes. A particular species, ulmo (Eucryphia cordifolia Cav.), possesses flowers that are highly appealing to bees (Figure 1). In recent times, ulmo honey has garnered recognition for its antioxidant and antimicrobial characteristics. The monofloral honey of Eucryphia cordifolia has been widely used in traditional medicine in the Mapuche-Chilean culture, especially for the treatment of infections. The ulmo tree is distributed in southern Chile from the Ñuble Region to the Los Lagos Region, being a typical tree of the Valdivian forest. Its flowers produce a large amount of nectar, making it an excellent honey tree and providing a product with a pleasant aroma and characteristic golden color (Figure 1) [4]. The ulmo (E. cordifolia) blooms between January and March, depending on the latitude and altitude. During this period, bees collect nectar from its flowers to produce honey. The ulmo honey harvest typically takes place between March and April.

Recently, a study conducted by our group reported that the leaves and monofloral honey of E. cordifolia exhibit antioxidant and antibacterial activities [5,6]. Another study suggests that ulmo honey may possess healing properties, along with potential anticancer, antioxidant, and antibacterial effects [4]. The presence of phenolic compounds, flavonoids, tannins, triterpenes, diterpenes, steroids, and volatile oils in the leaves and stems of E. cordifolia has been reported, evidencing its bioactive potential [7,8,9,10,11]. Therefore, their inclusion in this study allows for a more comprehensive evaluation of the plant’s potential impact on the composition and bioactivity of the resulting honey. However, information on the metabolomic identification of its flowers and the honey derived from this specimen is still limited. Considering that ulmo honey originates from floral nectar, it is essential to evaluate its metabolomic profile and determine whether its bioactive compounds present antioxidant properties. To date, the antioxidant properties of E. cordifolia flowers have not been studied, which represents a gap in knowledge about the chemistry of this species and its potential influence on honey composition.

Various techniques have been developed to verify the floral origin of honey, aiming to ensure its authenticity and distinguish it from other types of honey. Among these methodologies, melissopalynology has been widely used to identify the characteristic pollen profile of the species in question. Additionally, infrared spectroscopy and chemometric analysis have been used to predict the phenolic and flavonoid content in ulmo honey [12]. Furthermore, sensory analysis and SPME-GC-MS have enabled the identification of distinctive compounds in monofloral ulmo honey [13]. In recent years, metabolomic profiling using mass spectrometry has emerged as an advanced tool for detailed chemical characterization.

In this study, metabolomic analysis using high-resolution mass spectrometry (LC-MS/MS) with a QTOF analyzer was employed to identify and characterize flavonoids, phenolic acids, and anthocyanins in E. cordifolia. This technology provides a comprehensive profile of the metabolites present in ulmo honey and allows comparison with leaf and flower extracts of the species. Therefore, the main objective of this research was to evaluate the relationship between the chemical compounds present in the plant and their possible transfer to honey, aiming to identify potential chemical markers that contribute to its botanical authentication.

2. Materials and Methods

2.1. Chemicals and Reagents

Folin–Ciocalteu’s phenol reagent, Ethyl acetate, n-hexane, and methanol were purchased from Merck (Darmstadt, Germany). 1,1-diphenyl-2-picrylhydrazyl (DPPH), potassium persulfate (K₂S₂O₈), 2,2′-azino-bis(3-ethylbenzothiazoline-6-sulphonate) (ABTS), 1,3,5-triphenyltetrazolium chloride (TPTZ), Trolox (6-hydroxy2,5,7,8-tetramethylchroman-2-carboxylic acid), FeSO₄·7H₂O, gallic acid, cinnamic acid, syringic acid, ferulic acid, chlorogenic acid, caffeic acid, p-coumaric acid, benzoic acid, pinocembrin, rutin, quercetin, luteolin, kaempferol, epicatechin, catechin, apigenin, myricetin, isorhamnetin, taxifolin, chrysin, galanganin, genistein, and hesperetin were all purchased from Sigma Chemical Co. (St. Louis, MO, USA).

2.2. Plant Material and Extraction

The plant material (leaves and flowers) of E. cordifolia was collected in January 2020 in Puerto Varas (Chile). Ethyl acetate and aqueous fractions from the methanolic extract of E. cordifolia leaves were previously obtained [10]. The flowers were dried at room temperature in the shade. The dry and powdered material was subsequently extracted separately with n-hexane, ethyl acetate, and methanol by maceration followed by filtration. The extracts were prepared at room temperature by stirring for 48 h. The extracts were concentrated to dryness in a rotary evaporator (Rotavapor^® R-100, BÜCHI Labortechnik AG, Flawil, Switzerland) under reduced pressure at 45 °C. Subsequently, they were stored at 4 °C for further analysis. Also, n-hexane was used to defat the flower samples, and the resulting extract was not included in the analysis.

2.3. Honey Collection

Honey samples from ulmo honey (UH055, UH056, and UH057) were obtained from local beekeepers in the Mediterranean climate zone of southern Chile (Los Lagos: Región X, Los Ríos: Región XIV). For comparison, three samples of monofloral honey (Tiaca, Guindo Santo, Alfalfa Chilota; 3 samples of each honey) were collected from southern Chile. Only honeys containing ≥ 45% ulmo pollen, as determined by melissopalynological analysis, were selected. The samples were stored at −20 °C until the study was conducted. The water content, pH value, diastase activity and sugar content of the ulmo honey samples were determined according to the methods proposed by the Association of Official Analytical Chemists (AOAC) [14].

2.4. Honey Extract Preparation

Ulmo honey samples (1 g) were dissolved in 10 mL of distilled water and allowed to stand for approximately 1 h. The sample was added to an Oasis HLB cartridge, which was previously conditioned with 5 mL of methanol and 5 mL of distilled water, respectively. The washing step was finished by adding 5 mL of distilled water, extracting mainly interfering sugars, and allowing the sample to dry for approximately 15 min under vacuum. The phenolic compounds were eluted with 5 mL of methanol. The methanol extract was filtered through a 0.45 μm filter and stored at −20 °C for further analysis, as described in Section 2.5 and Section 2.6 [15].

2.5. Quantitative UHPLC-MS/MS Chromatographic Analysis

All samples were analyzed with an ABSciex triple Quad 4500 mass spectrometer equipped with an electrospray (TurboV) interface coupled to an Eksigent Ekspert Ultra LC100 automatic sampling system with an Ekspert Ultra LC100-XL (AB/Sciex Concord, Concord, ON, Canada). A LiChrospher 100 RP-18 (125 mm × 4 mm id, 5 µm) column was utilized, and 0.1% formic acid in water (A) and methanol (B) with the following gradient compositions were used as a mobile phase: 0–1 min, 15% B; 1–17 min, 15–100% B; 17–21 min 100–100% B; 21–22 min, 100–15% B; 22–25 min, 15–15% B. Ten microliters of each sample were injected at a flow rate of 0.5 mL/min at 30 °C. Calibration curves were recorded for each compound using available standards in the 10–250 μg/L range.

2.6. UHPLC/Q-TOF-MS/MS

Chromatographic profiling was performed on an Agilent 1260 Infinity UHPLC system (Agilent, Waldbronn, Germany) coupled to a Triple TOF 5600 QTOF-MS detector (AB SCIEX, Toronto, ON, Canada) with an electrospray ionization (ESI) source. Data acquisition and processing were performed using the Analyst TF 1.5, PeakView™ 1.2, and LibraryView™ (AB Sciex, Toronto, ON, Canada) software, respectively. For analysis, 2 mg per mL of the plant parts was dissolved in distilled water/methanol (1:1 v/v) and then 5 µL of the filtered solution (PTFE filter, Merck KGaA, Darmstadt, Germany) was injected.

Liquid chromatography was performed using an Acquity UPLC C18 1.7 μm, 2.1 mm × 50 mm column (Waters Corporation, Milford, MA, USA). A gradient of 0.1% formic acid (solvent A) and methanol with 0.1% formic acid (solvent B) was used as follows: between 0 and 13 min, 90% (A) and 10% (B), then between 13 and 15 min, 100% (B), and, finally, between 15 and 22 min, 90% (A) and 10% (B). The flow rate was adjusted to 0.4 mL/min and the separation was finished in 35 min.

MS data were recorded from 80 to 1200 m/z in negative mode and directed information dependent acquisition (IDA) analysis was performed in survey scan type (TOF-MS) using dependent scan (production) with −50 V collision energy voltage. The temperature was 400 °C with 25 psi curtain gas (CU), while 50 psi was applied for both ion source gas 1 (GC1) and ion source 2 gas (GS2). The ion spray voltage was −4500 V, the disaggregation potential was (DP) 90 V, and the collision energy voltage was set to (CE) −50 V (Supplementary Materials, Figures S1–S5).

2.7. Antioxidant Activity Assays

2.7.1. 2,2-Diphenyl-1-picrylhydrazyl Radical (DPPH)

The DPPH• radical was determined by the decolorization method [16] and adapted to 96-well plates. Briefly, 50 µL of sample was mixed with 150 µL of a DPPH solution (0.15 mM) in methanol and kept in the dark for 30 min. Subsequently, the absorbance was measured at 517 nm in a microplate reader (Cytation™ 5, BioTek, Santa Clara, CA, USA). The results are expressed in TEAC, the antioxidant activity equivalent to Trolox (μmol Trolox/g extract). The synthetic antioxidant reference, Trolox, at a concentration of 25–200 µM in methanol solution, was tested under the same conditions.

2.7.2. 2,2′-Azino-bis(3-ethylbenzothiazoline-6-sulphonate) (ABTS)

The ABTS assay was carried out by means of the iron reduction method [17], and adapted to 96-well plates. A stock solution of ABTS was prepared by mixing ABTS and potassium persulfate (K₂S₂O₈) in distilled water, considering a final concentration of 7 mM for ABTS and a final concentration of 3.6 mM for K₂S₂O₈. The reaction was then incubated at room temperature for 24 h in the dark. Subsequently, the ABTS stock solution was diluted to a final concentration of 156 µM to obtain a final absorbance of 0.70 ± 0.02 at 732 nm. Then, 50 µL of sample solution was mixed with 150 µL of reagent solution and the absorbance was measured at 732 nm after 30 min in the dark using a microplate reader (Cytation™ 5, BioTek, Santa Clara, CA, USA). The calibration curve was built with Trolox, and the results were expressed as micromoles of Trolox equivalents per gram of dry extract (µmol TE/g).

2.7.3. Ferric Ion Reducing Antioxidant Power Assay (FRAP)

The FRAP assay was performed according to the reported procedure [18]. Briefly, the FRAP reagent was prepared in situ by mixing 300 mM acetate buffer (pH 3.6), a solution of 10 mM 2,4,6-tripyridyl-s-triazine (TPTZ) in 40 mM hydrochloric acid and 20 mM ferric chloride in a 10:1:1 ratio (v/v/v). Then, 20 µL of each extract was mixed with 180 µL of FRAP reagent in 96-well plates. The mixture was left for 30 min in the dark and the absorbance at 595 nm was measured in a microplate reader (Cytation™ 5, BioTek, Santa Clara, CA, USA). Ferrous sulfate heptahydrate (FeSO₄·7H₂O) was used as standard to perform the calibration curve and results were expressed as millimoles of Fe per gram of dry sample (mmol Fe/g).

2.8. Total Phenolic Content (Folin-Ciocalteau)

The phenolic content was analyzed as previously reported [18]. A total of 20 µL of each sample was mixed with 100 µL of Folin–Ciocalteau reagent (1:10 v/v) and 80 µL of a Na₂CO₃ solution (7.5%), then incubated for 60 min at room temperature, after which the absorbance of the resulting blue solution was measured at 760 nm in a microplate reader (Cytation™ 5, BioTek, Santa Clara, CA, USA). The obtained absorbance values were replaced in the equation of the standard curve of gallic acid. The results of the total phenol content are expressed in mg of gallic acid equivalent per gram of dry extract (mg GAE/g).

2.9. Statistical Analysis

All assays were conducted in triplicate, with the results being expressed as the median ± standard deviation (SD) using Microsoft Excel (Microsoft 365, Microsoft Corporation, Redmond, WA, USA). Statistical differences between groups were considered significant at p < 0.05 and were analyzed using one-way ANOVA followed by Tukey’s post hoc test in the commercial software Minitab 19.

3. Results

3.1. Identification of Compounds in E. cordifolia Extracts by LC QTOF-MS/MS

In this study, the fingerprint was generated by LC-QTOF-MS/MS, which allowed for the determination of several families of compounds in the extracts of the leaves, flowers, and honey of E. cordifolia. The metabolomic study of the three ulmo honey samples enabled the identification and characterization of their chemical profiles, highlighting the compounds present in each. For the construction of the compound table, only metabolites shared among the three samples were considered, ensuring a more consistent and representative depiction of the characteristic chemical profile of E. cordifolia honey. The tentative identification of the metabolites was carried out by the analysis of full scan mass spectra, base peak chromatograms, and fragmentation patterns in comparison to the PubChem database, FooDB, and data from the literature.

3.1.1. Identification of Metabolites in the Ethyl Acetate Fraction of Leaves

Eighteen compounds were detected, and 14 compounds were tentatively identified in the ethyl acetate fraction (EtOAcF) from the methanolic extract of E. cordifolia leaves. Of these compounds, two are phenolic acids (1, 15), eight are flavonoids (5, 6, 7, 8, 9, 10, 11, and 18), three are anthocyanins (12, 13 and 14), and one is a polyphenolic compound (17); four compounds remained unidentified, respectively (2, 3, 4, and 16) (

Table 1. Identification of metabolites in EtOAc fraction of E. cordifolia leaves.

Peak	Retention Time (min)	Tentative Identification	Molecular Formula	Theoretical Mass (m/z)	Measured Mass [M−H]− or [M−H]+ (m/z)	Accuracy (ppm)	MSn Ions	Reference
1	2.97	Galloyl glucose	C13H15O10−	331.0671	331.0669	−0.4	168.8574	[19]
2	10.06	Unknown	C16H19O9−	355.1035	355.1037	0.7
3	10.17	Unknown	C16H17O8−	337.0929	337.0926	−0.8
4	11.05	Unknown	C9H7O3−	163.0401	163.0401	0.3
5	11.39	Dihydroquercetin 3-O-rhamnoside	C21H21O11−	449.1089	449.1087	−0.5	284.8040, 256.8297, 151.8702	FooDB FDB007557
6	11.45	Quercetin 3-O-glucoside	C21H19O12−	463.0882	463.0884	0.4	300.7885, 270.7986, 254.8152	[20]
7	11.6	Hesperidin	C28H33O15−	609.1825	609.1835	1.7	300.8283, 285.8131	[21]
8	11.77	Quercetin 3-O-pentoside	C20H17O11−	433.0776	433.0774	−0.4	299.7795	[22]
9	12.04	Quercetin 3-O-rhamnoside	C21H19O11−	447.0933	447.0929	−0.9	300.7888, 270.7983, 254.8150, 150.8639	[23]
10	12.25	Isorhamnetin	C16H11O7−	315.0510	315.0510	−0.1	270.8001, 242.8017, 226.8453	FooDB FDB002422
11	12,67	Kaempferol 3-O-rhamnoside	C21H19O10−	431.0984	431.0985	0.4	284.8040	FooDB FDB01691
12	12.96	Delphinidin 3-O-(6-p-coumaroyl-glucoside)	C30H27O14+	611.1328	611.1300	−4.7	302.7946, 272.7952, 255.8279, 151.8627	[24]
13	13.09	Pelargonidin-3-O-beta-D-p-coumaroylglucoside	C30H26O12+	578.1352	578.1355	0.6	431.7574, 413.7599, 271.8191	FooDB FDB031089
14	13.36	Cyanidin 3-(p-coumaroyl)-glucoside	C30H27O13+	595.1301	595.1306	0.9	286.8026, 254.8140	Pubchem 102514898
15	15.28	Vanillic acid	C8H7O4−	167.0350	167.0350	0.4	136.8979, 108.9227	FooDB FDB0008416
16	15.58	Unknown	C39H59O5−	607.4368	607.4365	−0.5	264.9243, 96.8608
17	16.17	Carvacrol	C10H13O−	149.0972	149.0971	−0.5	116.9226, 105.0340, 93.0704	FooDB FDB013868
18	16.53	Eupatorin	C18H15O7−	343.0823	343.0821	−0.7	214.8393	[25]

, Supplementary Materials, Figure S1).

Anthocyanin

Peak 12, with a pseudomolecular ion at M+H⁺ at m/z: 611.1300, was identified as delphinidin 3-O-(6-p-coumaroyl-glucoside) (C₃₀H₂₇O₁₄⁺). Peak 13 exhibited a molecular ion at M+H⁺ with m/z: 578.1355 and showed a major fragment with m/z: 271.8191 that may correspond to pelargonidin; this compound could be tentatively identified as Pelargonidin-3-O-beta-D-p-coumaroylglucoside (C₃₀H₂₆O₁₂⁺). Peak 14 showed a molecular ion at M+H⁺ with m/z 595.1306 and produced a major fragment with m/z 286.8026 (cyanidin); this compound was identified as cyanidin 3-(p-coumaroyl)-glucoside (C₃₀H₂₆O₁₃⁺).

Flavonoids

Peak 5, with a M−H⁻ ion at m/z: 449.1087, was identified as dihydroquercetin 3-O-rhamnoside (C₂₁H₂₁O₁₁⁻) and peaks 6–11 were identified as quercetin 3-O-glucoside (C₂₁H₁₉O₁₂⁻), hesperidin (C₂₈H₃₃O₁₅⁻), quercetin 3-O-pentoside (C₂₀H₁₇O₁₁⁻), quercetin 3-O-rhamnoside (C₂₁H₁₉O₁₁⁻), isorhamnetin (C₁₆H₁₁O₇⁻), and kaempferol 3-O-rhamnoside (C₂₁H₁₉O₁₀⁻), respectively. Finally, peak 18 was identified as eupatorin (C₁₈H₁₅O₇⁻).

Phenolic Acids

Peak 1 was identified as galloyl glucose (C₁₃H₁₅O₁₀⁻), and peak 15 as vanillic acid (C₈H₇O₄⁻).

Other Polyphenols

Peak 17, with a pseudomolecular ion at m/z: 149.0971, was identified as carvacrol (C₁₀H₁₃O⁻).

Unknown

Peaks 2, 3, 4, and 16 remain unidentified to date.

3.1.2. Identification of Metabolites in the Aqueous Fraction of Leaves

Twelve compounds were detected in this fraction and 10 compounds were tentatively identified in the aqueous fraction (AqF) from the methanolic extract of E. cordifolia leaves. Of these, two are procyanidins (1, 2), five are flavonoids (3, 4, 5, 6, 7), one is a phenolic acid (8), one is an anthocyanin (9), one is a polyphenol (12), and two compounds remain unidentified (10 and 11) (

Table 2. Identification of metabolites in the aqueous fraction of E. cordifolia leaves.

Peak	Retention Time (min)	Tentative Identification	Molecular Formula	Theoretical Mass (m/z)	Measured Mass [M−H]− (m/z)	Accuracy (ppm)	MSn Ions	Reference
1	7.51	Procyanidin dimer B1	C30H25O12−	577.1352	577.1366	2.5	406.7550, 288.8324, 124.9046	[26]
2	7.64	Procyanidin trimer EEC	C45H37O18−	865.1985	865.2006	2.4	-	[27]
3	11.51	Dihydroquercetin 3-O-rhamnoside	C21H21O11−	449.1089	449.1085	−0.9	284.8044, 270.7905, 151.8745	FooDB FDB007557
4	11.57	Quercetin 3-O-glucoside (Isoquercetin)	C21H19O12−	463.0882	463.0874	−1.8	300.7866, 270.7982, 254.8142, 150.8548	[20]
5	11.72	Hesperidin	C28H33O15−	609.1825	609.1830	0.8	300.8245, 285.8117, 256.8657	[21]
6	11.86	Quercetin 3-O-pentoside	C20H17O11−	433.0776	433.0776	−0.1	-	[22]
7	12.00	Quercetin 3-O-rhamnoside	C21H19O11−	447.0933	447.0925	−1.8	300.7885, 270.7970, 254.8149, 150.8641	[23]
8	14.93	Vanillic acid	C8H7O4−	167.0350	167.0356	4.0	136.8934, 108.9212	PubChem 8468
9	15.10	Cyanidin 3-(p-coumaroyl)-glucoside	C30H27O13+	595.1301	595.1297	−0.7	286.8026, 254.8140	Pubchem 102514898
10	15.31	Unknown	C9H11O2−	151.0765	151.0764	−0.2	-
11	15.65	Unknown	C39H59O5−	607.4368	607.4360	−1.2	264.9257, 96.8624
12	16.25	Carvacrol	C10H13O−	149.0972	149.0971	−0.6	116.9226, 133.1017 105.0340	PubChem 10364

, Supplementary Materials, Figure S2).

Anthocyanin

Peak 9 showed a molecular ion at M+H⁺ with m/z 595.1297 and was identified as cyanidin 3-(p-coumaroyl)-glucoside (C₃₀H₂₆O₁₃⁺).

Flavonoids

Peak 3, with a M−H⁻ ion at m/z: 449.1085, was identified as dihydroquercetin 3-O-rhamnoside (C₂₁H₂₁O₁₁⁻), and peaks 4–7 were identified as quercetin 3-O-glucoside (C₂₁H₁₉O₁₂⁻), hesperidin (C₂₈H₃₃O₁₅⁻), quercetin 3-O-pentoside (C₂₀H₁₇O₁₁⁻), and quercetin 3-O-rhamnoside (C₂₁H₁₉O₁₁⁻), respectively.

Phenolic Acids

Peak 8 was identified as vanillic acid (C₈H₇O₄⁻).

Procyanidins

Peak 1, with a pseudomolecular ion at m/z 577.1366, was identified as procyanidin dimer B1 (C₃₀H₂₅O₁₂⁻). Peak 2 showed a deprotonated molecule at m/z 865.2006 and was identified as an isomer of a procyanidin trimer EEC with a MW of 866.

Other Polyphenols

Peak 12, with a pseudomolecular ion at m/z: 149.0971, was identified as carvacrol (C₁₀H₁₃O⁻).

Unknown

Peaks 10 and 11 remain unidentified.

3.1.3. Identification of Metabolites in the Ethyl Acetate Extract of Flowers

Twenty-seven compounds were detected, and 21 compounds were tentatively identified in the ethyl acetate extract (EtOAcE) of E. cordifolia flowers. Of these, 2 compounds are anthocyanins (21 and 23), 10 are flavonoids (5, 10, 13, 14, 15, 16, 17, 18, 19, 27), 5 are phenolic acids (1, 2, 7, 8 and 24), 2 are procyanidins (6, 12), and 2 are polyphenol (4 and 26). Finally, six unidentified compounds (3, 9, 11, 20, 22, and 25) were observed (

Table 3. Identification of metabolites in the EtOAc extract of flowers.

Peak	Retention Time (min)	Tentative Identification	Molecular Formula	Theoretical Mass (m/z)	Measured Mass [M−H]− (m/z)	Accuracy (ppm)	MSn Ions	Reference
1	2.14	3-Feruloylquinic acid	C17H19O9−	367.1035	367.1049	3.9	172.8436, 112.9139, 94.9186	FooDB FDB000249
2	2.93	Galloyl glucose	C13H15O10−	331.0671	331.0669	−0.5	168.8608	[19]
3	4.18	Unknown	C6H5O3−	125.0244	125.0244	−0.3	-
4	7.25	Resorcinol	C6H5O2−	109.0295	109.0292	−2.4	-	PubChem 5054
5	8.35	(-)-Catechin	C15H13O6−	289.0718	289.0717	−0.1	149.8930, 122.9261	[27]
6	9.54	Procyanidin trimer EEC	C45H37O18−	865.1985	865.1995	1.1	-	[27]
7	10.08	Ferulic acid 4-O-glucoside	C16H19O9−	355.1035	355.1034	−0.1	160.8772, 132.9035	FooDB FDB000256
8	10.19	5-p-Coumaroylquinic acid	C16H17O8−	337.0929	337.0937	2.4	-	FooDB FDB000236
9	10.73	Unknown	C22H17O10−	441.0827	441.0822	−1.1	-
10	10.89	Isorhamnetin 3-O-glucoside	C22H21O12−	477.1039	477.1039	0.1	313.7841, 271.8078	PubChem 5318645
11	11.18	Unknown	C30H23O12−	575.1195	575.1198	0.5	406.7520
12	11.4	Bis-8,8-Catechinyl methane	C31H27O12−	591.1508	591.1508	0.0	288.8312, 244.8721, 136.8944	PubChem 46182787
13	11.41	Dihydroquercetin 3-O-rhamnoside	C21H21O11−	449.1089	449.1087	−0.5	284.8020, 216.8605, 150.8997	FooDB FDB007557
14	11.47	Quercetin 3-O-glucoside	C21H19O12−	463.0882	463.0881	−0.1	300.7875, 270.7981, 254.8146	[20]
15	11.79	Quercetin 3-O-pentoside	C20H17O11−	433.0776	433.0775	−0.4	299.7797, 278.7976, 254.8143	[22]
16	12.06	Quercetin 3-O-rhamnoside	C21H19O11−	447.0933	447.0929	−0.9	300.7869, 270.7972, 254.8137, 150.8626	[23]
17	12.27	Ishormanetin	C16H11O7−	315.0510	315.0507	−1.2	270.7969, 242.8214, 226.8384	FooDB FDB002422
18	12.28	Kaempferol 3-O-arabinoside	C20H17O10−	417.0827	417.0824	−0.7	283.7958, 254.8146, 226.8385	PubChem 5481882
19	12.7	Kaempferol 3-O-rhamnoside	C21H19O10−	431.0984	431.0986	0.4	284.8033, 254.8159, 226.8400	FooDB FDB01691
20	12.84	Unknown	C23H21O12−	489.1039	489.1050	2.3	-
21	13.11	Pelargonidin-3-O-beta-D-p-coumaroylglucoside	C30H26O12+	578.1352	578.1353	0.3	431.7581, 413.7604, 271.8189
22	13.12	Unknown	C26H31O14−	567.1719	567.1718	−0.3	458.7602, 149.8549
23	13.36	Petunidin 3-O-(6-p-coumaroyl-glucoside)	C31H29O14−	625.1485	625.1455	−4.7	316.7941, 299.7759	FooDB FDB000041
24	15.22	Vanillic acid	C8H7O4−	167.0350	167.0350	0.4	136.8979, 108.9227	FooDB FDB0008416
25	15.59	Unknown	C39H59O5−	607.4368	607.4365	−0.5	264.9243
26	16.18	Carvacrol	C10H13O−	149.0972	149.0971	−0.5	93.0704	FooDB FDB013868
27	16.51	Eupatorin	C18H15O7−	343.0823	343.0821	−0.7	214.8393	[25]

, Supplementary Materials, Figure S3).

Anthocyanin

Peak 21 exhibited a molecular ion [M + H]⁺ with m/z: 578.1355 and produced a major fragment at m/z: 271.8189 that may correspond to pelargonidin; this compound could be tentatively identified as Pelargonidin-3-O-beta-D-p-coumaroylglucoside (C₃₀H₂₆O₁₂⁺). Peak 23, with a pseudomolecular ion at m/z: 624.1455, was identified as Petunidin 3-O-(6-p-coumaroyl-glucoside) (C₃₁H₂₉O₁₄⁺).

Flavonoids

Peak 5, with a [M−H]⁻ ion at m/z: 289.0717, was identified as (-)-catechin (C₁₅H₁₃O₆⁻); peak 10, showing a pseudomolecular ion at m/z: 447.1039, was identified as the isorhamnetin 3-O-glucoside (C₂₂H₂₁O₁₂⁻). Peaks 13–19 were identified as flavonol, flavononol, and flavone derivatives, respectively. Peak 13 showed a pseudomolecular ion at m/z: 449.1087 and was identified as the dihydroquercetin 3-O-rhamnoside (C₂₁H₂₁O₁₁⁻). Peak 14 (m/z: 463.0881, [M−H]⁻) was consistent with the molecular formula (C₂₁H₁₉O₁₂⁻), corresponding to quercetin 3-O-glucoside. Peak 15, with a parent ion at m/z: 433.0775, was identified as quercetin 3-O-pentoside (C₂₀H₁₇O₁₁⁻). Peak 16 produced a molecular ion at m/z 447.0929 and MSn ions at m/z 300.7869 (M-rhamnose), and was identified as quercetin 3-O-rhamnoside (C₂₁H₁₉O₁₁⁻). Peak 17, with a [M−H]⁻ ion at m/z 315.0507, was tentatively identified as isorhamnetin (C₁₆H₁₁O₇⁻). Peaks 18 and 19 were identified as kaempferol derivatives. Peak 18 produced a molecular ion at m/z 417.0824 and a kaempferol MS2 fragment at m/z 283.7958, which pointed to the presence of kaempferol 3-O-arabinoside (C₂₀H₁₇O₁₀⁻). Peak 19, with a pseudomolecular ion at m/z 431.0986, which in turn produced a kaempferol ion at m/z 284.8083 (M-hexose), was proposed to be an O-glycosylated kaempferol derivative (kaempferol 3-O-rhamnoside). Finally, peak 27 was identified as eupatorin (C₁₈H₁₅O₇⁻).

Phenolic Acids

Peak 1 was identified as 3-feruloylquinic acid (C₁₇H₁₉O₉⁻), peak 2 as galloyl glucose (C₁₃H₁₅O₁₀⁻), peak 7 as ferulic acid 4-O-glucoside (C₁₆H₁₉O₉⁻), peak 8 as 5-p-coumaroylquinic acid (C₁₆H₁₇O₈⁻), and peak 24, with pseudomolecular ions at m/z: 167.0350, was identified as vanillic acid (C₈H₇O₄⁻).

Procyanidins

Peak 6 showed a deprotonated molecule at m/z 865.1995 and was identified as an isomer of a procyanidin trimer EEC with a MW of 866. Peak 12, with a [M−H]⁻ ion at m/z: 591.1508, was identified as Bis-8,8-Catechinylmethane (C₃₁H₂₇O₁₂⁻).

Other Polyphenols

Peak 4 was identified as resorcinol (C₆H₅O₂⁻) and peak 26, with a pseudomolecular ion at m/z: 149.0971, was identified as carvacrol (C₁₀H₁₃O⁻).

Unknown

Peaks 3, 9, 11, 20, 22, and 25 remain unidentified.

3.1.4. Identification of Metabolites in the Methanolic Extract of Flowers

Twenty-six compounds were detected, and 23 compounds were tentatively identified in the methanolic extract (MeOHE) of E. cordifolia flowers. Of these, 3 are anthocyanins’ compounds (18, 20, and 22), 13 are flavonoids (3, 9, 10, 11, 12, 13, 14, 15, 16, 21, 23, 24, 26), 4 are phenolic acids (1, 4, 6, 7), 2 are polyphenols (2, 19), and one is a procyanidin (5). Finally, three compounds remained unidentified (8, 17, and 25) (

Table 4. Identification of metabolites in the MeOH extract of flowers.

Peak	Retention Time (min)	Tentative Identification	Molecular Formula	Theoretical Mass (m/z)	Measured Mass [M−H]− (m/z)	Accuracy (ppm)	MSn Ions	Reference
1	4.25	Gallic acid	C7H5O5−	169.0142	169.0143	0.3	125.0244, 94.9150	FooDB FDB000662
2	7.39	Resorcinol	C6H5O2−	109.0295	109.0296	0.6	107.9144, 90.9250	PubChem 5054
3	8.44	(-)-Catechin	C15H13O6−	289.0718	289.0717	−0.1	149.8930, 122.9261	[27]
4	10.18	Ferulic acid 4-O-glucoside	C16H19O9−	355.1035	355.1034	−0.1	160.8772, 132.9035	FooDB FDB000256
5	10.19	Bis-8,8-Catechinylmethane	C31H27O12−	591.1508	591.1508	0.0	288.8312, 244.8721, 136.8944
6	10.28	5-p-Coumaroylquinic acid	C16H17O8−	337.0929	337.0927	−0.7	190.8914, 92.9427	FooDB FDB000236
7	11.18	p-Coumaric acid	C9H7O3−	163.0401	163.0401	0.5	118.9340, 92.9390	FooDB FDB002593
8	11.29	Unknown	C30H23O12−	575.1195	575.1200	0.9	-
9	11.52	Dihydroquercetin 3-O-rhamnoside	C21H21O11−	449.1089	449.1087	−0.5	284.8020, 216.8605, 150.8997	FooDB FDB007557
10	11.58	Dihydroquercetin 3-O-glucoside	C21H21O12−	465.1039	465.1039	0.1	301.7841, 150.8616	FooDB FDB021739
11	11.59	Quercetin 3-O-glucoside	C21H19O12−	463.0882	463.0881	−0.1	300.7875, 270.7981, 254.8146	[20]
12	11.9	Quercetin 3-O-pentoside	C20H17O11−	433.0776	433.0775	−0.4	299.7797, 278.7976, 254.8143	[22]
13	12.18	Quercetin 3-O-rhamnoside	C21H19O11−	447.0933	447.0929	−0.9	300.7869, 270.7972, 254.8137, 150.8626	[23]
14	12.41	Isorhamnetin	C16H11O7−	315.0510	315.0509	−0.3		FooDB FDB000604
15	12.59	Isorhamnetin 3-O-glucoside	C22H21O12−	477.1039	477.1036	−0.5		[28]
16	12.8	Kaempferol 3-O-rhamnoside	C21H19O10−	431.0984	431.0986	0.4	284.8033, 254.8159, 226.8400	FooDB FDB01691
17	13.02	Unknown	C23H21O7−	409.1293	409.1290	−0.6	216.8965, 160.9086, 1498904
18	13.08	Delphinidin 3-O-(6-p-coumaroyl-glucoside)	C30H26O14−	610.1328	610.1299	−4.7	301.7882, 150.8636	[29]
19	13.12	Homovanillyl alcohol	C9H11O3−	167.0714	167.0714	0.0	121.9223, 107.9141	FooDB FDB018391
20	13.21	Pelargonidin-3-O-beta-D-p-coumaroylglucoside	C30H26O12+	578.1352	578.1359	1.2	431.7609, 270.8202, 144.8942
21	13.42	Naringenin	C15H11O5−	271.0612	271.0614	0.7	118.9360, 106.9058, 92.9381	[30]
22	13.47	Petunidin 3-O-(6-p-coumaroyl-glucoside)	C31H29O14−	624.1485	624.1455	−4.7	477.7326, 315.7944, 299.7743	FooDB FDB000041
23	13.97	Apigenin	C15H9O5−	269.0455	269.0457	0.7	148.8828, 120.9104, 116.9188, 106.9067	[31]
24	14.83	Pinocembrin	C15H11O4−	255.0663	255.0661	−0.7	184.8977, 144.9281	[32]
25	15.06	Unknown	C20H17O6−	353.1031	353.1034	0.9	248.8433, 220.8707, 96.8616
26	16.62	Eupatorin	C18H15O7−	343.0823	343.0821	−0.7	214.8563	[25]

, Supplementary Materials, Figure S4).

Anthocyanin

Peak 18, with a pseudomolecular ion at m/z: 610.1299, was identified as delphinidin 3-O-(6-p-coumaroyl-glucoside) (C₃₀H₂₆O₁₄⁻). Peak 20 exhibited a molecular ion at M+H⁺ with m/z: 578.1359 and produced a major fragment with m/z: 270.8202 that may correspond to pelargonidin; this compound could be tentatively identified as Pelargonidin-3-O-beta-D-p-coumaroylglucoside (C₃₀H₂₆O₁₂⁺). Peak 22, showing a pseudomolecular ion at m/z: 625.1455, was identified as petunidin 3-O-(6-p-coumaroyl-glucoside) (C₃₁H₂₉O₁₄⁺).

Flavonoids

Peak 3, with a [M−H]⁻ ion at m/z: 289.0717, was identified as (-)-catechin (C₁₅H₁₃O₆⁻). Peaks 9 and 10 were identified as dihydroquercetin derivatives. Peak 9 produced a molecular ion at m/z 449.1087, which pointed to the presence of dihydroquercetin 3-O-rhamnoside (C₂₁H₂₁O₁₁⁻). Peak 10, with a pseudomolecular ion at m/z 465.1039, which in turn produced a dihydroquercetin ion at m/z 301.7841 (M-hexose), was proposed to be an O-glycosylated dihydroquercetin derivative (Dihydroquercetin 3-O-glucoside). Peaks 11–13 were identified as flavonol derivatives. Peak 11, with a parent ion at m/z: 463.0881, was identified as quercetin 3-O-glucoside (C₂₁H₁₉O₁₂⁻); peak 12 showed a pseudomolecular ion at m/z: 433.0775 and was tentatively identified as quercetin 3-O-pentoside (C₂₀H₁₇O₁₁⁻); and peak 13 produced a molecular ion at m/z 447.0929 and MSn ions at m/z 300.7869 (M-rhamnose), and was identified as quercetin 3-O-rhamnoside (C₂₁H₁₉O₁₁⁻). Peak 14, with a [M−H]⁻ ion at m/z 315.0509, was identified as isorhamnetin (C₁₆H₁₁O₇⁻). Peaks 15 and 16 were identified as kaempferol derivatives. Peak 15 produced a molecular ion at m/z 477.1036 and was identified as isorhamnetin 3-O-glucoside (C₂₂H₂₁O₁₂⁻). Peak 16, with a pseudomolecular ion at m/z 431.0986, which in turn produced a kaempferol ion at m/z 284.8083 (M-hexose), was proposed to be an O-glycosylated kaempferol derivative (kaempferol 3-O-rhamnoside, C₂₁H₁₉O₁₀⁻). Finally, peaks 21 and 24 were identified as the flavanone aglycones naringenin and pinocembrin, while peaks 23 and 26 were identified as the flavone aglycones apigenin and eupatorin, respectively. MS-MS analysis of all of these compounds showed characteristic daughter ions (Table 4).

Phenolic Acids

Peak 4 was identified as ferulic acid 4-O-glucoside (C₁₆H₁₉O₉⁻), peak 6 as 5-p-coumaroylquinic acid (C₁₆H₁₇O₈⁻), and peak 7 as p-coumaric acid (C₉H₇O₃-). Additionally, peak 1 was identified as gallic acid (C₇H₅O₅⁻).

Procyanidins

Peak 5, with a [M−H]⁻ ion at m/z: 591.1508, was identified as Bis-8,8-Catechinylmethane (C₃₁H₂₇O₁₂⁻).

Other Polyphenols

Peak 2 was identified as resorcinol (C₆H₅O₂⁻) and peak 19 as homovanillyl alcohol (C₉H₁₁O₃⁻).

Unknown

Peaks 8, 17, and 25 remain unidentified.

3.1.5. Identification of Metabolites in the Honey Extract

The three samples that were evaluated contained between 75% and 77% E. cordifolia pollen grains. According to the Codex Alimentarius [33], the ulmo honey in this study has a high quality, with a pH of 4.5, a water content of 19.3%, a reducing sugar content of 81.3%, and a relatively high diastase activity of 23.1 units according to Schade. Twenty-four compounds were detected, and 22 compounds were tentatively identified in the honey extract. Of these, 4 are polyphenolic compounds (1, 2, 8, 9), 15 are flavonoids (3, 4, 5, 7, 10, 12, 13, 14, 15, 17, 18, 19, 20, 22, 24), 2 are coumarins (6, 16), and 1 is an anthocyanin (21). Finally, two compounds remained unidentified (11 and 23) (

Table 5. Identification of metabolites in the honey extract.

Peak	Retention Time (min)	Tentative Identification	Molecular Formula	Theoretical Mass (m/z)	Measured Mass [M−H]− (m/z)	Accuracy (ppm)	MSn Ions	Reference
1	10.13	p-Anisaldehyde	C8H7O2−	135.0452	135.0452	0.6	88.9467	PubChem 31244
2	10.31	Syringaldehyde	C9H9O4−	181.0506	181.0508	1.0	120.9122, 91.9305	FooDB FDB000822
3	10.69	Myricetin 3-O-rutinoside	C27H29O17−	625.1410	625.1413	0.4	299.7784, 270.7989	FooDB FDB016619
4	11.29	Quercetin 3-O-rutinoside	C27H29O16−	609.1461	609.1455	−1.1	301.0351, 283.7972, 254.8131	[28]
5	11.73	Dihydroquercetin 3-O-rhamnoside	C21H21O11−	449.1089	449.1072	−3.9	284.8033, 150.8959	FooDB FDB007557
6	12.24	Mellein	C10H9O3−	177.0557	177.0558	0.5	116.9188, 104.9287	FooDB FDB000337
7	12.38	Quercetin 3-O-rhamnoside	C21H19O11−	447.0933	447.0920	−2.9	300.7859, 270.7964, 254.8124, 150.8624	[23]
8	13.36	Eugenol	C10H11O2−	163.0765	163.0767	1.8	147.0452	FooDB FDB012171
9	13.37	Homovanillyl alcohol	C9H11O3−	167.0714	167.0715	0.9	122.9267, 108.9180	FooDB FDB018391
10	13.43	Quercetin	C15H9O7−	301.0354	301.0351	−1.0	150.8646, 120.5446	[34]
11	13.6	Unknown	C16H13O5−	285.0768	285.0770	0.5	251.8283
12	13.61	Formononetin	C16H11O4−	267.0663	267.0664	0.3	250.8207, 194.8696, 166.8975	FooDB FDB012219
13	13.78	Isorhamnetin	C16H11O7−	315.0510	315.0502	−2.6	270.7952, 254.8131, 242.8218	FooDB FDB002422
14	13.89	Naringenin	C15H11O5−	271.0612	271.0614	0.6	118.936, 106.9058, 92.9381	FooDB FDB000678
15	14.44	Kaempferide	C16H11O6−	299.0561	299.0559	−0.7	254.8137, 226.8389	FooDB FDB016499
16	14.96	Umbelliferone	C9H5O3−	161.0244	161.0245	0.7	143.0502, 133.0303	[35]
17	15.01	Cirsimaritin	C17H13O6−	313.0718	313.0717	−0.1	252.8350, 207.8788, 196.8851	[35]
18	15.06	3,7-Dimethyl quercetin	C17H13O7−	329.0667	329.0667	0.2	298.7730, 270.7949	[35]
19	15.07	Pinocembrin	C15H11O4−	255.0663	255.0665	0.9	212.8693, 184.8948	FooDB FDB002758
20	15.29	Daidzein	C15H9O4−	253.0506	253.0509	0.9	252.8379, 207.8706, 179.8956, 166.8974	FooDB FDB
21	15.37	Cyanidin 3-(p-coumaroyl)-glucoside	C30H27O13+	595.1301	595.1297	−3.3	286.8026, 254.8140	Pubchem 102514898
22	15.44	Apigenin	C15H9O5−	269.0455	269.0456	0.1	268.8264, 251.8329, 226.8345, 170.8903	[31]
23	15.93	Unknown	C39H59O5−	607.4368	607.4354	−2.4	264.8231, 96.8628
24	16.8	Eupatorin	C18H15O7−	343.0823	343.0813	−3.1	214.8398	[25]

, Supplementary Materials, Figure S5).

Anthocyanin

Peak 21 showed a molecular ion at M+H⁺ with m/z 595.1297 and produced a major fragment with m/z 286.8026 (cyanidin) and was identified as cyanidin 3-(p-coumaroyl)-glucoside (C₃₀H₂₆O₁₃⁺).

Coumarin

Peak 6, with a pseudomolecular ion at m/z: 177.0558, was identified as mellein (C₁₀H₉O₃⁻), and peak 16, with a pseudo-molecular ion at m/z: 161.0245, was identified as umbelliferone (C₉H₅O₃⁻).

Flavonoids

Peak 3, with a M−H⁻ ion at m/z: 625.1413, was identified as myricetin 3-O-rutinoside (C₂₇H₂₉O₁₇⁻), and peaks 4–5 were identified as quercetin 3-O-rutinoside (C₂₇H₂₉O₁₆⁻) and dihydroquercetin 3-O-rhamnoside (C₂₁H₂₁O₁₁⁻). Peak 7, with a [M−H]⁻ ion at m/z 447.0920 and MS2 ions at m/z 300.7859 (M-rhamnose), was tentatively identified as quercetin 3-O-rhamnoside (C₂₁H₁₉O₁₁⁻), while peak 10 was assigned to quercetin (C₁₆H₁₁O₄⁻). Peaks 12–15 were identified as formononetin (C₁₆H₁₁O₄⁻), isorhamnetin (C₁₆H₁₁O₇⁻), naringenin (C₁₅H₁₁O₅⁻), and kaempferide (C₁₆H₁₁O₆⁻), respectively. Peaks 17–18 were assigned to flavone cirsimaritin (C₁₇H₁₃O₆⁻) and 3,7-dimethylquercetin (C₁₇H₁₃O₇⁻), while peak 19 was assigned to flavanone pinocembrin (C₁₅H₁₁O₄⁻) and peak 20 to daidzein (C₁₅H₉O₄⁻). Finally, peak 22 was assigned to flavone apigenin (C₁₅H₉O₅⁻) and peak 24 was identified as eupatorin (C₁₈H₁₅O₇⁻), respectively.

Other Polyphenols

Peak 1 was identified as p-anisaldehyde (C₈H₇O₂⁻) and peak 2 as syringaldehyde (C₉H₉O₄⁻). Peaks 8–9 were assigned to eugenol (C₁₀H₁₁O₂⁻) and homovanillyl alcohol (C₉H₁₁O₃⁻), respectively.

Unknown

Peaks 11 and 23 remain unidentified.

3.2. Quantification of Phenolic Compounds

Phenolic acids and flavonoids are key contributors to the color, aroma, flavor, and health benefits of honey. As shown in

Table 6. Quantitation of compounds in E. cordifolia ethyl acetate, aqueous fraction, and methanolic extract of honey (mg/kg).

Compound	EtOAc F. Leaves	Aqueous F. Leaves	EtOAc E. Flower	MeOH E. Flower	UH055	UH056	UH057
Gallic acid	36.39 ± 0.41	173.02 ± 0.23	33.05 ± 0.22	1378.69 ± 0.38	15.78 ± 0.19	32.80 ± 0.20	3.72 ± 0.16
Cinnamic acid	nd	nd	nd	nd	d	d	d
Syringic acid	nd	nd	nd	nd	0.01	0.02	0.02
Ferulic acid	0.01	nd	0.01	nd	0.02	0.03	0.02
Chlorogenic acid	nd	0.25 ± 0.02	0.01 ± 0.01	nd	nd	nd	0.01 ± 0.01
Caffeic acid	0.03 ± 0.01	nd	0.03 ± 0.01	0.08 ± 0.01	0.04 ± 0.01	0.17 ± 0.01	0.12 ± 0.01
p-coumaric acid	0.02 ± 0.01	nd	0.07 ± 0.01	0.28 ± 0.01	0.05 ± 0.01	0.13 ± 0.01	0.06 ± 0.01
Benzoic acid	nd	nd	d	d	d	d	d
Pinocembrin	1.02 ± 0.08	nd	10.28 ± 0.09	123.24 ± 0.15	83.60 ± 0.16	1196.53 ± 0.24	231.44 ± 0.09
Rutin	nd	0.05 ± 0.01	0.02 ± 0.01	nd	0.02 ± 0.01	0.02 ± 0.01	0.01 ± 0.01
Quercetin	1.20 ± 0.02	nd	0.38 ± 0.01	3.57 ± 0.02	0.06 ± 0.01	0.06 ± 0.01	nd
Luteolin	0.73 ± 0.02	0.19 ± 0.02	0.13 ± 0.02	0.45 ± 0.02	0.02 ± 0.01	0.02 ± 0.01	0.02 ± 0.01
Kaempferol	0.05 ± 0.01	nd	0.04 ± 0.01	0.21 ± 0.01	0.05 ± 0.01	0.07 ± 0.01	0.03 ± 0.01
Epicatechin	19.72 ± 0.08	17.22 ± 0.06	0.19 ± 0.02	3.69 ± 0.02	0.25 ± 0.01	nd	0.04 ± 0.01
Catechin	13.06 ± 0.07	5.40 ± 0.04	0.12 ± 0.02	1.19 ± 0.01	0.10 ± 0.01	0.04 ± 0.01	0.03 ± 0.01
Apigenin	0.04 ± 0.01	nd	0.02 ± 0.01	0.07 ± 0.01	0.02 ± 0.01	0.03 ± 0.01	0.02 ± 0.01
Myricetin	0.03 ± 0.01	nd	0.02 ± 0.01	0.04 ± 0.01	nd	nd	nd
Isorhamnetin	42.55 ± 0.06	nd	2.44 ± 0.03	7.58 ± 0.04	1.39 ± 0.02	2.89 ± 0.02	1.16± 0.02
Taxifolin	d	d	d	d	d	d	d
Chrysin	d	nd	d	d	d	d	d
Galanganin	nd	nd	nd	d	d	d	d
Genistein	d	d	d	d	d	d	d
Hesperetin	d	d	d	d	d	d	d

d: detected; nd: not detected.

, 22 phenolic standards were employed to quantify their presence in fractions and extracts obtained from the leaves, flowers, and honey of ulmo.

The results in Table 6 show significant differences (p < 0.05) in the concentrations of phenolic and flavonoid compounds among the E. cordifolia extracts and ulmo honey samples. Statistical analysis using one-way ANOVA followed by Tukey’s test revealed that the methanolic extract of flowers had the highest concentration of gallic acid (1378.69 ± 0.38 mg/kg), which was significantly higher than all other fractions/extracts and honey samples that were analyzed. In contrast, in ulmo honey, the concentration of this compound ranged from 3.72 ± 0.16 mg/kg to 32.80 ± 0.20 mg/kg, suggesting a possible transfer from the plant to the honey in lower proportions. Pinocembrin was one of the most abundant flavonoids in the ulmo honey, with sample UH056 showing the highest concentration (1196.53 ± 0.24 mg/kg), followed by sample UH057 (231.44 ± 0.09 mg/kg). These differences were statistically significant (p < 0.05) and suggest variability in the chemical composition of the honey, which is possibly influenced by environmental and floral factors. In the plant extracts, pinocembrin was found at a higher concentration in the methanolic extract of the flowers (123.24 ± 0.15 mg/kg), supporting its potential botanical origin in E. cordifolia. Other compounds, such as isorhamnetin and quercetin, also showed significant differences among the samples. Isorhamnetin was found at higher concentrations in the leaf fractions (42.55 ± 0.06 mg/kg) compared to the honey, where it ranged from 1.16 ± 0.02 to 2.89 ± 0.02 mg/kg. Meanwhile, quercetin was more abundant in the methanolic extract of the flowers (3.57 ± 0.02 mg/kg) and was detected only in low concentrations in the honey samples (≤0.06 mg/kg). Overall, the variability in the concentration of these compounds suggests that, while some metabolites are transferred from E. cordifolia to honey, the proportion will be significantly lower, possibly due to degradation processes or chemical modifications during honey production. Taken together, these results indicate that pinocembrin and gallic acid can be considered the main chemical markers of ulmo honey, while isorhamnetin may complement their characterization. Notably, the leaves of E. cordifolia were found to be a rich source of polyphenolic compounds, underscoring their potential influence on the chemical profile and antioxidant properties of ulmo honey. This highlights the importance of considering not only the floral nectar but also other plant parts when evaluating the bioactive composition of honey. However, further studies are needed to assess their stability and specificity in larger samples of E. cordifolia honey.

3.3. Antioxidant Activity and Total Phenolic Content

The fractions and extracts of leaves and flowers were evaluated in vitro to determine their antioxidant activity using the DPPH, ABTS, and FRAP methods and are expressed as µmol Trolox/g of dry extract (

Table 7. Content of phenols and antioxidant activity of the different extracts of E. cordifolia.

Fractions/Extract	Total Phenolic Content (mg GAE/g DE)	DPPH		ABTS		FRAP (mmol Fe/g)
		(µmol TE/g DE)	CI50 (µg/mL)	(µmol TE/g DE)	CI50 (µg/mL)
EtOAcF. leaves	302.19 ± 0.71 a	174.91 ± 0.59 c,d	45.05 ± 1.58 c	146.74 ± 0.16 b	41.25 ± 0.46 c	2.74 ± 0.02 a
Aqueous F. leaves	184.82 ± 11.02 b	174.25 ± 1.23 d	40.16 ± 1.66 d	147.16 ± 0.08 b	39.36 ± 1.65 d	2.66 ± 0.04 b
Ethyl acetate E. flowers	85.57 ± 2.21 c	240.79 ± 6.75 b	281.83 ± 5.28 a	182.73 ± 3.23 a	127.48 ± 5.36 a	0.77 ± 0.04 d
Methanol E. flowers	185.79 ± 10.19 b	279.46 ± 8.17 a	109.74 ± 1.74 b	185.25 ± 4.31 a	46.40 ± 3.60 b	1.99 ± 0.30 c
UH055	40.0 ± 2.0 e	94.0 ± 3.0 f	-	75 ± 7.0 d	-	0.49 ± 0.02 f
UH056	58.0 ± 5.0 d	120.0 ± 3.0 e	-	114 ± 15 c	-	0.23 ± 0.01 g
UH057	35.0 ± 2.0 e	187.0 ± 3.0 c	-	122 ± 13 c	-	0.59 ± 0.02 e

Different letters in the same column indicate significant differences: p < 0.05, n = 3. -: not evaluated.

). In addition, the total phenolics content was determined by the Folin–Ciocalteau method. Our results showed that the EtOAc and aqueous leaf fractions were more potent antioxidants than the EtOAc and MeOH extracts of E. cordifolia flowers. On the other hand, the ulmo 056 honey (UH056) showed the highest TPC value (58 mg GAE/100 g honey), while the ulmo 057 (UH057) honey presented the lowest TPC value (35 mg GAE/100 g honey). The values of antioxidant capacity, evaluated by the DPPH, ABTS, and FRAP methods, were between 94 (UH055) and 187 μmol TE/g ulmo honey (UH057) for DPPH, between 75 (UH055) and 122 μmol TE/g UH (UH057) for ABTS, and between 0.23 (UH056) and 0.59 0 mmol Fe/g UH (UH057) for FRAP. The results show statistically significant differences (p < 0.05) among the leaf fractions, flower extracts, and honey samples in terms of their antioxidant capacity and total phenolic content. Analysis of variance (ANOVA), followed by Tukey’s test, confirmed that the flower extracts, particularly those obtained with methanol, exhibited the highest antioxidant activity in the DPPH and ABTS assays, with significantly higher values than the leaf fractions and honey samples. Similarly, the total phenolic content of the flower extracts was significantly higher compared to that of the honey (p < 0.05), suggesting a greater concentration of bioactive compounds in the plant. Among the honey samples that were analyzed, sample UH057 showed the highest antioxidant capacity, significantly differing from the other honey samples (UH55 and UH056). These differences suggest that the chemical composition of ulmo honey may be influenced by environmental and floral factors, highlighting the need for further studies to correlate the presence of specific metabolites with its botanical origin and biological functionality.

4. Discussion

Honey has been consumed by humans since ancient times, mainly due to its high nutritional and medicinal values. In addition, honey is a mixture of hundreds of compounds that are found in very low concentrations, e.g., polyphenols, which give the different varieties of honey different colors, smells, flavors, and bioactivities. According to their geographical origin and floral source, the trace compound composition of honey differs considerably [35]. In this sense, it is necessary to identify chemical markers that can serve to distinguish the individual characteristics of honey.

A metabolomic study is a qualitative and quantitative analytical approach to detecting metabolites in different biological samples. The use of targeted metabolomics for quantitative analysis provides comprehensive information on biological samples. While previous studies have reported the presence of both volatile and non-volatile/semi-volatile compounds in ulmo honey, no studies have examined the chemical composition of its floral source. This study represents the first metabolomic profiling of E. cordifolia, providing extensive and detailed identification of the compounds present in its flowers, leaves, and honey.

Using the established methods, the honey extract and the fractions and extracts of leaves and flowers of E. cordifolia were analyzed by UHPLC-Q-TOF-MS. Analysis by LC-MS resulted in 24 tentative compound identifications in the honey, 22 in the leaf fractions, and 37 in the flower extracts of E. cordifolia. The Venn diagram in Figure 2 compares the metabolic profiles of these samples, revealing 10 compounds shared between the plant parts and the honey, among which 5 were common to both the leaves and flowers. These shared compounds were further scrutinized to identify potential botanical origin markers. Among the six candidate compounds that were present in the leaves, five were successfully identified through exact mass and high-resolution MS2 fragmentation patterns, compared with chromatographic data and [M−H]⁻ data reported in the literature, as dihydroquercetin 3-O-rhamnoside, quercetin 3-O-rhamnoside, isorhamnetin, cyanidin 3-(p-coumaroyl)-glucoside, and eupatorin. On the other hand, nine were found to be common in flowers and honey, of which five were also shared with the leaves, as mentioned above. The remaining four compounds were identified as homovanillyl alcohol, naringenin, apigenin, and pinocembrin. Due to the shared presence of these compounds in both the plant parts and ulmo honey, and their absence in other honeys produced in the same region, the compounds dihydroquercetin 3-O-rhamnoside (astilbin), quercetin 3-O-rhamnoside (quercitrin), isorhamnetin, and eupatorin are potential candidates as chemical markers for ulmo honey (E. cordifolia).

The five flavonoids quercitrin, astilbin, isorharmnetin, cyanidin 3-(p-coumaroyl)-glucoside, and eupatorin are present in the ulmo honey analyzed herein and have not been reported to be simultaneously present in any type of Chilean honey to date.

On the other hand, the identification of quercitrin and astilbine in honey agrees with their reported isolation from the leaves of E. cordifolia [10]. As an additional contribution to the research, a set of phenolic compounds was quantified in the honey and flowers of E. cordifolia, along with an analysis of its leaves. This complementary analysis provided a broader perspective on the chemical composition of the species, exploring the potential transfer of metabolites from the plant to its honey. Although the primary objective of this study was to obtain a metabolomic profile of the honey of E. cordifolia and its relationship with the compounds present in the plant, the evaluation of phenolic compounds in the leaves offered valuable insights into the chemical variability of the species and its potential influence on the honey produced from its nectar. The identification of isorhamnetin by UPLC-MS/MS validated the findings obtained through UHPLC/Q-TOF-MS, providing additional confirmation that the presence of these compounds in honey originates from the E. cordifolia species.

The results in Table 6 highlight the variability in the composition of phenolic compounds of E. cordifolia extracts and the different honeys analyzed. Compounds such as gallic acid and pinocembrin were found at significantly higher concentrations in the ulmo honeys, particularly in sample UH056, which contained 1196.53 mg/kg of pinocembrin, suggesting the potential of this compound as a distinctive chemical marker for this variety. In contrast, compounds such as quercetin and ferulic acid were present in lower but still detectable concentrations, indicating their contribution to the unique chemical profile of each sample. This variability reflects the influence of the botanical and geographical origins of honey on its chemical composition and underscores the value of metabolomic analysis for their characterization. Previously, gallic acid, caffeic acid, coumaric acid, pinocembrin, chrysin, quercetin, abscisic acid, luteolin, and apigenin have been detected in E. cordifolia honey [6]. It is important to highlight that the polyphenols identified in ulmo honey have biological activity. For example, gallic acid has been reported to have antimicrobial, anticancer, and antiviral activities. Pinocembrin is a characteristic flavonoid of propolis and has been previously detected in honey from E. cordifolia [6] and European honey samples [36]. In addition, pinocembrin has shown anti-inflammatory, anti-asthmatic, and antitumor activities [37]. Similarly, isorhamnetin exerts beneficial effects in the areas of cardiovascular and cerebrovascular protection, antitumor activity, anti-inflammatory action, antioxidant properties, organ protection, obesity prevention, etc. Table 7 demonstrates that both E. cordifolia fractions and honeys analyzed herein exhibit antioxidant activity, which is closely correlated with their total phenol content [38]. The ethyl acetate and methanolic extracts of E. cordifolia flowers stood out for their high antioxidant activity, particularly in the DPPH and ABTS assays, with values ranging from 240.79 to 279.46 µmol TE/g DE in the DPPH and from 182.73 to 185.25 µmol TE/g DE in the ABTS assay, respectively. Meanwhile, the ulmo honeys exhibited moderate antioxidant activity, with sample UH057 showing superior performance in the DPPH and ABTS assays, which is potentially attributable to its higher content of bioactive compounds such as pinocembrin. These findings highlight the significance of phenolic compounds in determining the bioactivity of honeys and extracts, emphasizing their functional potential and their role in differentiating bee products.

5. Conclusions

This study contributes to the understanding of the metabolomics of the Eucryphia cordifolia plant, an endemic species of Chile. Chilean ulmo is an important source of monofloral honey with antioxidant properties. The three samples analyzed in this study feature various molecules from different plant parts and honey, including 22 compounds from the leaf fractions, 37 from the flower extracts, and 24 in the respective honeys of E. cordifolia. By examining and comparing the chemical profiles of ulmo honey and flowers using untargeted metabolomics, we report for the first time that dihydroquercetin 3-O-rhamnoside (astilbine), quercetin 3-O-rhamnoside (quercitrin), isorharmnetin, cyanidin 3-(p-coumaroyl)-glucoside, and eupatorin could serve as chemical markers. Notably, the leaves of E. cordifolia displayed a diverse and bioactive metabolomic profile, contributing significantly to the species’ overall chemical composition. Their rich contents of flavonoids, phenolic acids, and anthocyanins suggest that they may play an indirect but relevant role in the antioxidant potential of ulmo honey. These compounds could potentially act as chemical markers of floral origin for the honey of this species; however, a larger number of honey samples should be analyzed to determine a correlation between the compounds existing in the plant parts and honey of E. cordifolia.