HPLC-HRMS untargeted metabolomic characterization of Honeys produced in the Venice Lagoon

Honey is a food that honeybees produce starting from the nectar of blossoms. The main components of honey are sugars, particularly glucose (38%) and fructose (31%), and water (17%). The minor components of honey, as amino acids, organic acids and polyphenols, despite being present in low amount, are very important to define the quality and the organoleptic properties of honey [1]. Moreover, since the nectar composition directly influences honey features, each honey presents different physicochemical, organoleptic, and nutraceutical properties depending by its floral origin [2]. In this context, the development of analytical methods for the assessment of honey quality, in relation to its nutraceutical properties, is an emerging field, due to the increasing awareness of consumers. The variability of floral sources, depending by seasons and geographical and climatic conditions, determine a high differentiation in the composition of honey, which need to be investigated to valorize the peculiarities of each product. The metabolomic approach, being able to qualitatively analyze a high number of known and unknown compounds, demonstrated to be advantageous in the characterization of complex matrices, and especially in the nutraceutical field [3]. In this study, HPLC-HRMS technology has been applied for the untargeted characterization of 4 kinds of honeys of different floral origins (salt marsh, spring multifloral, autumn multifloral and acacia). These honeys are produced in the S. Erasmo island by “Miele del doge”, in the northern lagoon of Venice. Sample treatment procedure has been developed in order to: -remove sugars, which prevent the determination of minor honey components; -assure the extraction of flavonoids, considered as target bioactive compounds of honey, evaluating the recovery of four compounds (apigenin, kaempferol, quercetin, naringenin). The analysis was conducted by a metabolomic approach and permitted to identify marker compounds for each honey typology and to find a relation between honey composition and botanical origin of honey.The sample treatment procedure was optimized by selecting the most appropriate stationary phase for solid phase extraction and removal of matrix effect. Tests were conducted on commercial glucose-fructose syrup to which known amount of four flavonoids were added. Three kinds of phases were tested: Styrene-divinylbenzene (Strata X, 6 mL, 33 µm Phenomenex, California USA), N-vinylpyrrolidone-divinylbenzene (Oasis HLB, 6 mL, 30 µm) and Sep-Pak C18 (3 mL, 55-105 µm) both of Waters Corporation, Massachusetts USA. Moreover, different sample amounts (300 and 400 mg) and different elution volumes (3 and 5 mL) were tested to select the method with the highest recovery for the target flavonoids. The final preanalytical protocol included the solubilization of 400 mg of honey in 1.5 mL of ultrapure water acidified with 0.1 of formic acid (AF). The solution was homogenized for 30 minutes in ultrasonic bath and then centrifugated; the supernatant was extracted by C18-SPE, washed with acidified water and eluted with 5mL of MeOH. The samples were then concentrated in thermostated bath until the final volume of 500 µL. The instrumental analysis has been conducted using an UltiMate 3000 (Dionex) coupled to an ESI-LTQ Orbitrap XL (Thermo Fisher Scientific, Waltham, MA, USA), as described by Scalabrin et al., 2015. The column employed was a C18 SB-Aq Narrow Bore RR 2.1 x 150 mm, 3.5 μm (Agilent Technologies, Wilmington, USA). The column was eluted by AF 0.01% and ACN acidified with AF 0.01%. The ion source was operated in both negative and positive polarity at 300°C. The MS analysis were performed in full scan mode, in a mass range of 90-1500 Da, at a high mass accuracy (<5 ppm). The resolving power was 60000. To trace a complete profile of fragmentation of the analyzed ions, data dependent analysis was carried out. The identification of compounds was based on the most probable molecular formula and by comparison of fragmentation spectra with literature and available online libraries. The identification level assigned was in accordance with the protocol of Sumner et al., 2007 [5]. The method was selected based on the best performance obtained by glucose-fructose syrup, spiked with flavonoid standards. StrataX SPE were excluded because the Felhing assay highlighted the presence of sugars, therefore they were not effective in removing sample matrix. C18-SPE proved to be efficient in the extraction of kaempferol, which was not retained by HLB-SPE. The results highlighted that the best method was represented by C18 extraction, combined with 400 mg of samples weighted and the elution volume of 5 mL. The selected method permitted the analysis of flavonoids in honey samples. Particularly, naringenin, quercetin and rutin showed the highest intensities in “salt marsh” honey while apigenin demonstrated to be particularly high in acacia honey. In addition, the method permitted the analysis of other 80 compounds of nutraceutical interest, related to the botanical origin of honey, especially: flavonoids, organic acids, phenolic acids, terpenes, iridoids. Among all the compounds, some biomarkers for each honey typology were identified. Acacia honey presented high levels of abscisic acid and of oleandrose, a sugar related to Nerium oleander; this plant, start flowering together with Robinia pseudoacacia, in May, continuing until September. The spring multifloral honey was characterized by dihydroconiferin attributable to Taraxacum officinale, and isosuspensolide, an iridoid probably related to the flowering of Viburnum. The autumn multifloral showed high levels of the iridoid genipin gentibioside, found in the plants of Gardenia jasminoides. The salt marsh honey showed the presence of compounds related to the particular vegetation of this area(Fig.2): 7,2'-Dihydroxy-6-methoxyisoflavone, which is found in Salicornia europaea, and the flavone chrysin, which is present in many plants and could be related to Artemisia campestris. Chrysin is known also to characterize the acacia honey, indeed its presence in this kind of honey was higher than in multiflorals. Luteolin glucosides are present in all the typologies of honey in similar levels and are probably related to the plants of Cynara scolymus, which are typical cultivations of this island.1. Schievano E, Morelato E, Facchin C, Mammi S (2013) J Agric Food Chem 61:1747–1755. 2. Di Marco G, Manfredini A, Leonardi D, Canuti L, Impei S, Gismondi A, Canini A (2017) Plant Biosyst 151:450–463. 3. Montoro P, D’urso G, Kowalczyk A, Tuberoso CIG (2021) Molecules 26:1–14. 4. Scalabrin E, Radaelli M, Rizzato G, Bogani P, Buiatti M, Gambaro A, Capodaglio G (2015) Anal Bioanal Chem 407:6357–68. 5. Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin C a., Fan TW-M, Fiehn O, Goodacre R, Griffin JL, Hankemeier T, Hardy N, Harnly J, Higashi R, Kopka J, Lane AN, Lindon JC, Marriott P, Nicholls AW, Reily MD, Thaden JJ, Viant MR (2007) Metabolomics 3:211–221