Eur Food Res Technol DOI 10.1007/s00217-017-2880-8
ORIGINAL PAPER
Determination of melatonin levels in different cherry cultivars by high-performance liquid chromatography coupled to electrochemical detection T. Rosado1,2 · I. Henriques1 · Eugenia Gallardo1,2 · A. P. Duarte1,2
Received: 12 January 2017 / Accepted: 18 March 2017 © Springer-Verlag Berlin Heidelberg 2017
Abstract The determination of melatonin has been reported in several fruits, including cherries. This study had the goal to evaluate melatonin levels present in five different cherry cultivars from the region of Fundão in Portugal, a well-known location for this fruit cultivation. Three of the five cultivars are evaluated for the first time regarding melatonin levels (Saco, Summit and Sunburst). To determine the presence of this hormone, a reliable analytical method was developed and fully validated, and an SPE procedure was applied and optimized using Oasis® HLB cartridges to concentrate melatonin and clean-up the samples before analysis by high-performance liquid chromatography coupled to electrochemical detection. The extraction efficiency from fresh matrices varied from 74 to 91%, while the absolute recoveries (SPE clean-up) of melatonin ranged from 61 to 75%. The procedure was considered linear for concentrations ranging from 0.025 to 4 µg/mL with a mean R2 value of 0.9974 and with calibrators’ accuracy (mean relative error) within a ±15% interval for all concentrations. The concentrations of melatonin found in the different cherry cultivars ranged from 11 to 28 ng/g of fresh fruit, and Burlat cultivar was the one presenting the highest level of this hormone.
T. Rosado and I. Henriques contributed equally to this paper. * Eugenia Gallardo
[email protected] 1
Centro de Investigação em Ciências da Saúde (CICS‑UBI), Universidade da Beira Interior, Av. Infante D. Henrique, 6200‑506 Covilhã, Portugal
2
Laboratório de Fármaco‑Toxicologia‑UBIMedical, Universidade da Beira Interior, Covilhã, Portugal
Keywords Melatonin · Cherries · Solid-phase extraction · HPLC-ECD
Introduction Melatonin (N-acetyl-5-methoxy-tryptamine), which is a neuro hormone produced by the pineal gland but also found in bacteria [1], protists [2], and plants [3, 4], being described as responsible for numerous aspects of the biological and physiologic regulation of body functions [5, 6]. Among the large number of health benefits described, the most known are the regulation of circadian rhythm and alleviation of sleep disorders [5, 7–10], depression treatment [11] being also reported as a potent free radical scavenger and a broad-spectrum antioxidant [6, 12–14]. The diminishing of neuro-degenerative diseases, such as Parkinson’s and Alzheimer’s [15], and anticancer properties [16] have as well been associated to this biogenic indoleamine. Regarding the plants, melatonin, as in vertebrates, acts as a free radical scavenger and possibly in photoperiodism [17, 18]. Melatonin possesses both hydrophilic and lipophilic characteristics [19, 20] and when consumed in plant products is absorbed, easily penetrates biological membranes, enters all subcellular compartments, and consequently leads to its physiological effects [4, 21]. Due to its wide biological properties, the interest of melatonin study in different types of plants and fruits increased substantially [22], although limited by difficulties due to the low concentrations of this compound in the matrices, the limited number of reliable analytical methods, and the complexity of the biological matrices [23]. The concentration can vary among species and in different parts of the plants, ranging from picograms to
13
Vol.:(0123456789)
micrograms per gram of plant tissue [24]. Some plants, especially in flowers and seeds, can reveal melatonin concentrations much higher than those normally found in vertebrate tissues (except for the pineal gland) [17]. The determination of this hormone has been reported in several fruits, such as apples, grapes, bananas, pineapples, tomatoes, and cherries [21, 24–26]. Cherries, in particular, are a nutritionally dense food rich in anthocyanins, quercetin, hydroxycinnamates, potassium, fiber, vitamin, carotenoids, and melatonin [27], and have been involved in a human study relating its consumption to a significant rise in the melatonin metabolite (6-sulfatoxymelatonin in urine) [28]. In Portugal, the region of Fundão is well known for cherry orchards, which are a staple of the local economy. There are more than 100 cherry cultivars, among which Burlat, Saco, Summit, Sunburst, and Sweetheart are quite common cultivars, hence of great interest to determine the presence of melatonin in the latter. In fact, Saco is an autochthonous cultivar from the Fundão region and the oldest and most traditional variety. This fruit already have been object of study to quantify melatonin by González-Gómez et al. [21] and Kirakosyan et al. [29], both applying high-performance liquid chromatography coupled with mass spectrometry (HPLC-MS), and Burkhardt et al. [18] with the use of high-performance liquid chromatography coupled with electrochemical detection (HPLC-ECD). Feng et al. [26] review the presence of melatonin in different fruit sources and the chromatographic and non-chromatographic techniques developed for its determination, reporting HPLC as the most widely used. The present work describes a fast, sensitive, and fully validated method with the previous optimization of the extraction procedure to determine melatonin in five cherry cultivars of the region of Fundão using the HPLC-ECD technique. Three cherry cultivars (Saco, Summit, and Sunburst) are studied for the first time regarding melatonin presence, and its values are compared with other present in the literature.
Materials and methods Reagents and materials Melatonin analytical standard (≥98% purity grade) was obtained from Sigma–Aldrich (Sintra, Portugal). Solvents and other reagents used during protocol presented analytical grade. Ethyl acetate, ethylenediaminetetraacetic acid (EDTA), and hydrochloric and phosphoric acid were purchased from VWR, Carnaxide, Portugal. Acetonitrile, hexane, methanol, dichloromethane, ammonium hydroxide, and formic acid were purchased from Enzymatic,
13
Eur Food Res Technol
Santo Antão do Tojal, Portugal. Deionised (DI) water was obtained from a Milli-Q System (Millipore, Billerica, MA, USA). Sodium dihydrogen phosphate (Panreac, Barcelona, Espanha); isopropanol (Laborspirit, Santo Antão do Tojal, Portugal); and sodium octyl sulfate (OSA) (Sigma–Aldrich; Sintra, Portugal) were also used in the present work. To pre-concentrate melatonin present in cherry cultivars, Oasis® HLB 3 mL (Waters, Lisboa, Portugal) SPE cartridges were used, although S trata®-X 33 μ 60 mg/3 mL (Tecnocroma, Caldas da Rainha, Portugal) SPE cartridges have been also used during the extraction procedure optimization phase. In addition, 0.2 μm, Whatman filters (VWR, Carnaxide, Portugal) were used to filter the extracts prior to injection. Stock and working solutions A stock melatonin solution was prepared at 1 mg/mL in methanol, after which several working solutions were made by proper dilution. The working solutions were prepared at 1, 10, 100, and 200 μg/mL according to the linearity range in the analytical method, and these solutions were stored in the absence of light at at 4 °C. Cherry cultivars specimens In the present work, the five cherry cultivars studied from the region of Fundão were Burlat, Saco, Summit, Sunburst, and Sweetheart. These specimens were harvested during the months of May, June, and July of 2015, and kindly offered by the Fundão co-op, Cerfundão Lda. These samples were properly stored at −20 °C until analysis. Extraction of melatonin from cherry tissues Each cherry cultivar was separated, after which the core was removed, and the fruit tissues sliced into very small pieces and stored at −80 °C during 24 h. Subsequently, a re-frozen step was performed in liquid nitrogen (−196 °C) during 48 h to optimize the to-be-applied lyophilization step. After this first preparation of the cherries, 2.00 g of each lyophilized cultivar were measured and 10 mL of methanol were added. These solutions were sonicated at 30 °C during 30 min and then centrifuged at 9652 rpm during 30 min with temperature set to 4 °C. The supernatant was collected and evaporated under vacuum obtaining a dry extract. Pre‑concentration of melatonin The pre-concentration of melatonin was carried out by SPE using Oasis® HLB 3 mL cartridges. This extraction procedure was previously optimized (see “Results and
Eur Food Res Technol
discussion” section), and the final conditions were as follows: 2 mL of methanol followed by 2 mL of deionized water were used for conditioning the cartridge; the sample load was carried out by passing through the cherry cultivar extract (2 mL) previously diluted with 8 mL of deionized water; 2 mL of 5% methanol was used to wash the column, after which it was dried under full vacuum during 10 min followed by elution of the analytes with 2 mL of ethyl acetate. The eluted was, then, evaporated to dryness at 35 °C under a gentle stream of nitrogen and the dry extract was dissolved in 500 µL of mobile phase. A filtering step was added using 0.2 μm Whatman filters and 20 µL were injected into the HPLC system. Chromatographic conditions An high-performance liquid chromatography system (HPLC) 1260 with quaternary pump from Agilent technologies (Soquimica, Lisboa, Portugal) was set to perform the chromatographic analysis coupled to electrochemical detection (ECD) carried with a Coulochem III ESA system (Dias de Sousa S.A., Lisboa Portugal) and with a coulometric cell (5011A) from ESA (Dias de Sousa S.A., Lisboa Portugal). The chosen oxidation potentials to detect the analyte were +600 mV (E1) and +300 mV (E2) applying 1 μA of sensitivity. Melatonin was separated with a Zorbax 300SB-C18 (5 μm, 4.6 × 250 mm i.d.) analytical column from Agilent Technologies (Soquímica, Lisboa, Portugal). The HPLC-ECD worked on isocratic mode with a mobile phase composed by two solutions. Solution 1 corresponded to 85% of the total mobile phase with 75 mM NaH2PO4, 1.7 mM OSA, and 25 μM EDTA in deionized water, pH 3.5, while solution 2 (15%) was acetonitrile. Mobile phase rate was 1 mL/min and sampler and column temperatures were set to 4 and 30 °C, respectively, each sample with a runtime of 25 min. Analytical method validation The developed analytical method was fully validated according to the guiding principles of the Food and Drug Administration (FDA) [30] and International Conference on Harmonization (ICH) [31]. The guiding principles included the 5 day validation protocol and the study of parameters such as selectivity, linearity and limits, intraand inter-day precision and accuracy, carry over effect, and extraction efficiency. To study the potential interferences from cherry matrix at the retention time of melatonin, the selectivity of the analytical method was evaluated by the analysis of several compounds also present in the same matrix. The
molecules analyzed were citric acid, β-carotene, catechin, epicatechin, quercetin, and sugars such as glucose, sucrose and sorbitol. Work solutions of β-carotene, catechin, epicatechin, and quercetin were prepared in methanol to a final concentration of 100 μg/mL. Regarding citric acid and sugars, the work solutions were prepared in deionized water also to a final concentration of 100 μg/ mL. Each work solution was then diluted with mobile phase (1:4) prior to injection and individual analysis by HPLC-ECD. The method would be considered selective if melatonin could not be identified in the mentioned solutions. Linearity of the method was established in the range of 0.025–4 μg/mL (six calibrators evenly distributed,). Since melatonin is present in all cherry cultivars, hence, the impossibility of using cherry matrix to make the calibration curves by spiking the analyte, the linearity was evaluated in 5% methanol:water (1:3). By plotting the melatonin peak area against its concentration, we obtained the calibration curves which were evaluated considering the acceptance criteria that included a determination coefficient (R2) value of, at least, 0.99 and the calibrators’ accuracy within ±15% (except at the lower limit of quantification (LLOQ), where ±20% was considered acceptable). Together with each calibration curve, a blank sample and three quality control (QC) samples at the concentrations of 0.075, 0.75, and 3 μg/mL (n = 3) were also analyzed. The lowest concentration measured with adequate precision and accuracy, i.e., with a coefficient of variation (CV, %) of less than 20% and a relative error (RE, %) within ±20% of the nominal concentration was defined as LLOQ. The limit of detection (LOD) was not systematically evaluated, but it was assumed as the same as the LLOQ (n = 5). Intra-day precision was evaluated by analyzing in the same day 6 replicates at the concentration levels of 0.025, 1, 2, and 4 μg/mL. Inter-day precision was evaluated at nine concentration levels within the 5-day period. The accuracy of the method was characterized in terms of the mean RE between the measured and the spiked concentrations; the accepted limit was 15% for all concentrations, except at the LLOQ, where 20% was accepted. Carryover effect was evaluated by injection of mobile phase immediately after analysis of the highest level calibrator (4 μg/mL). The absolute recovery was evaluated by preparing two sets of samples (n = 3) at the same concentration levels used for QC. Set 1 was spiked with melatonin after extraction of a blank sample (representing 100% recovery), and set 2 spiked with melatonin before extraction. The comparison of the peak areas set 2 with those of the corresponding peaks in set 1 allowed the evaluation of the recoveries.
13
Table 1 Extraction procedures for Strata®-X 33 and Oasis® HLB
Eur Food Res Technol SPE step
Strata®-X
Conditioning
2 mL methanol 1 mL methanol 2 mL deionized water 1 mL deionized water 2 mL 5% methanol spiked with melatonin (2.5 μg/mL) 2 mL 5% methanol 1 mL 5% methanol 2 mL methanol 1 mL 2% formic acid in methanol
Load Wash Elution
Fig. 1 Graphical representation of the absolute peak areas obtained for the different SPE cartridges (n = 3)
Results and discussion Solid‑phase extraction optimization Different types of sorbents are currently available to carry out SPE, but the optimal performance of any SPE extraction will depend on the proper type of sorbent that will control important parameters such as selectivity, affinity, and capacity [32]. Among the most common sorbents described for SPE extraction of melatonin in fruits, C18 silica seems to be the favourite [25, 29, 33, 34]. In the present work, polymeric sorbents that present a lower particle size, hence a larger surface area, were studied [35]. Table 2 Different wash and elution procedures applied on Oasis® HLB cartridges
The two SPE cartridges chosen to study were Strata®-X 33 and O asis® HLB, both 60 mg/3mL and commonly applied to acidic, basic, and neutral compounds [36]. Following the manufacturer’s instructions for each cartridge, the extraction procedures applied (Table 1) resulted in no significant differences (F 1.4 = 0.41, p
