Non-aqueous reversed-phase liquid-chromatography of tocopherols and tocotrienols and their mass spectrometric quantification in pecan nuts

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Food Composition and Analysis journal homepage: www.elsevier.com/locate/jfca

Original research

Non-aqueous reversed-phase liquid-chromatography of tocopherols and tocotrienols and their mass spectrometric quantification in pecan nuts Virginia Pérez-Fernándeza, Mariangela Spagnolia, Anna Roccob, Zeineb Aturkib, Fabio Sciubbaa, ⁎ Flavio Roberto De Salvadorc, Petra Engelc, Roberta Curinia, Alessandra Gentilia, a b c

Dipartimento di Chimica, Sapienza Università di Roma, Piazzale Aldo Moro, 5, 00185, Roma, Italy Istituto di Metodologie Chimiche, Consiglio Nazionale delle Ricerche (C.N.R.), Monterotondo, Roma, Italy CREA-Fruit Tree Research Center, Via Fioranello, 52, 00134, Roma, Italy

A R T I C L E I N F O

A B S T R A C T

Chemical compounds studied in this article: 9 alpha-Tocopherol (PubChem CID: 2116) beta-Tocopherol(PubChem CID: 6857447) gamma-Tocopherol (PubChem CID: 92729) delta-Tocopherol (PubChem CID: 92094) alpha-Tocotrienol (PubChem CID: 5282347) beta-Tocotrienol (PubChem CID: 5282348) gamma-Tocotrienol (PubChem CID: 5282349) delta-Tocotrienol (PubChem CID: 5282350)

An easy and effective analytical method was developed for the simultaneous quantification of four tocopherols (Ts) and four tocotrienols (T3s) in three pecan nut cultivars (Stuart, Sioux, and Pawnee). The analytes were separated on a C30 column kept at 15 °C, under isocratic non-aqueous reversed-phase (NARP) conditions, in only 18 min and detected by atmospheric pressure chemical ionization–tandem mass spectrometry (APCI-MS/MS). The HPLC-APCI-MS/MS method was validated, according to the main FDA guidelines, and then applied for the characterization of the real samples. Analytes were extracted by cold saponification with recoveries greater than 87%. The limits of detection (LOD) and limits of quantification (LOQ) were in the range of 0.3–10 μg/100 mg and 1–30 μg/100 mg, respectively. Compared to other nuts, vitamin E composition of pecan nuts (Carya illinoinensis) has only partially been elucidated. Results have evidenced the prevalence of γ-forms for both Ts and T3 s and clear quantitative differences of the identified vitamers among the studied cultivars. The richest variety in vitamin E was Sioux with a total content of about ∼ 32 mg/100 g wet weight, followed by Stuart (∼16 mg/ 100 g) and Pawnee (∼9 mg/100 g).

Keywords: Vitamin E HPLC APCI-MS Tocopherols Tocotrienols Chromatographic separation Pecan fruits Cultivars Food analysis

1. Introduction Vitamin E consists of a group of eight tocochromanols, all of them of plant origin: four tocopherols (Ts) with a saturated isoprenoid chain, and four tocotrienols (T3s) with an isoprenoid chain containing three trans double bonds. The homologues Ts and T3 s are designated α, β, γ, and δ depending on number and position of the methyl groups on the aromatic ring (see Fig. 1); β and γ tocochromanols have two methyl groups and are positional isomers. Vitamin E is incorporated into cellular membranes in which it plays an important antioxidant function. Both Ts and T3 s act as peroxyl radical scavengers and chain breakers of lipid peroxidation (Esterbauer

et al., 1991). The antioxidant activity increases with the number of methyl groups in the phenolic ring; thus, α forms possess the highest antioxidant activity (α > β > γ > δ) and are very effective in contrasting reactive oxygen species (ROS). However, the unsubstituted C-5 position of γ-T makes it able to trap lipophilic electrophiles such as reactive nitrogen oxide species (RNOS) (Jiang et al., 2001), which are associated with chronic inflammation-related diseases (Cooney et al., 1993). It has also been verified that the antioxidant efficacy of T3 s in membranes is higher than that of Ts, in particular for the α-forms. The determining factors are: i) the more uniform distribution of T3 s in membrane bilayer; ii) the greater recycling activity of the tocotrienoxyl radicals; iii) the closer collocation of T3 s to the membrane surface

Abbreviations: BHT, butylated hydroxytoluene; CV, coefficient of variation; FDA, Food and Drug Administration; FL, Fluorescence; IS, internal standard; HPLC, –APCI, MS/MS, high performance liquid chromatography-atmospheric pressure chemical ionization-tandem mass spectrometry; LOQ, limit of quantification; LOD, limit of detection; MRM, multiple reaction monitoring; MTBE, methyl tert-butyl ether; NARP, non-aqueous reversed phase; NP, normal phase; QC, quality control; R2, coefficient of determination; RNOS, reactive nitrogen oxide species; ROS, reactive oxygen species; RP, reversed phase; T, tocopherol T; T3, tocotrienol ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (A. Gentili). http://dx.doi.org/10.1016/j.jfca.2017.09.002 Received 28 February 2017; Received in revised form 30 August 2017; Accepted 2 September 2017 0889-1575/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Pérez-Fernández, V., Journal of Food Composition and Analysis (2017), http://dx.doi.org/10.1016/j.jfca.2017.09.002

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Fig. 1. Structures of the eight vitamin E homologues.

than 1 h. A complete resolution was achieved on non-alkyl RP columns such as pentafluorophenyl (PFP) columns (Lanina et al., 2007; Viñas et al., 2014; Wong et al., 2014; Górnaś and Siger, 2015; Knecht et al., 2015) and naphthalethyl (π-NAP) columns (Shammugasamy et al., 2013). The good performance offered by both these stationary phases is due to the relative rigidity of the PFP and π-NAP groups which provides superior selectivity for analytes with similar solubility but different size and shape. Although C30 is another kind of stationary phase suitable for separations of geometric and positional isomers, there is only two methods based on it (Saha et al., 2013; Knecht et al., 2015). The method of Knecht et al. (2015) was able to separate all vitamin E homologues in 45 min, keeping the C30 column at 18 °C, employing a flow rate of 0.5 mL/min and using a mobile phase composed by water, methanol and methyl-t-butyl ether (MTBE). A similar mobile phase was used also for the separation on the Core Shell PFP column (Knecht et al., 2015). However, the mobile phase used for both columns is neither completely ideal for HPLC (MTBE tends to form bubbles) nor totally suitable for the analysis of most of real samples. In fact, in the majority of cases, Ts and T3 s are extracted from lipid-rich matrices whose analysis requires a purge step at the end of the chromatographic run to wash the column. The major aim of this work has been the development of a simple and fast non-aqueous reversed-phase (NARP) chromatographic method for the separation and quantification of Ts and T3s. Actually, since not all vitamin E forms have the same function and biological potency, the study of their natural distribution is of special interest and can provide valuable information on the nutritional value of foods. This is especially important for T3s, which are the least studied forms due to the limited availability and high cost of the authentic standards. The needed sensitivity and selectivity to detect the minor components and to characterize complex matrices such as nuts were provided by the APCItandem mass spectrometry (MS/MS). To the best of our knowledge, this is the first application of a HPLC-APCI-MS/MS method to the speciation of vitamin E homologues in different pecan cultivars (Sioux, Stuart and Pawnee) to assess differences in individual and total content of these precious antioxidants.

(Packer et al., 2001). The Dietary Reference Intake of vitamin E for adults is around 15 mg/day (Monsen, 2000). Vegetable oils and lipid-rich plant products are the main natural sources (Syväoja et al., 1986; Piironen et al., 1986); in particular, nuts and seeds cover up to 82% of the daily intakes of vitamin E (Bauernfeind, 1980). Traditionally, the most worldwide consumed nuts are hazelnuts, almonds and walnuts. Pecan is another kind of nut, which is native from North America and is distributed over a large geographic and climatic area, from the United States to Mexico (Thompson and Grauke, 1991; Hall, 2000). Its introduction in North Africa and Europe (Iberian Peninsula, and in Italy at the Botanical Garden of Palermo) was reported in the early 20th century. Regarding the vitamin E composition, pecans (Carya illinoinensis) have been the least studied of nuts, especially in relation to the main cultivars available on the American and European food market (Lee et al., 1998; Kornsteiner et al., 2006; Villarreal-Lozoya et al., 2007). In general, liquid chromatographic techniques are the golden standard for the speciation of tocols. However, so far, the elucidation of their natural distribution in nuts as well as other matrices have mainly been hindered by the difficulties in resolving the β- and γ isomers, which have often been determined globally. The literature has reported not many methods related to the separation of all eight homologous of vitamin E. Some of them were based on Normal Phase (NP) chromatography (Cunha et al., 2006; Pinheiro-Sant’Ana et al., 2011; Kua et al., 2016). This is a chromatographic mode particularly selective towards geometric and positional isomers, which are usually well-resolved on siliceous phases based on their different steric fitting with the adsorption sites. However, NP has two main limitations: i) it is subject to phenomena of tailing; ii) the mobile phase is not ideal for MS detection. In fact, the most used mobile phases contain hexane and low percentages of alcohols or ethers which do not favour either the electrospray or Atmospheric Pressure Chemical Ionization (APCI). On the other hand, selectivity and sensitivity of MS detection is crucial when complex matrices such as food and biological samples have to be studied. Compared to NP, Reversed Phase (RP) offers a series of advantages that include the use of more appropriate mobile phases for MS detection, greater robustness of chromatographic columns, reproducibility of the retention times, faster conditioning and better peak shape. For these reasons, several RP-HPLC methods were developed on C18 columns (Thompson et al., 1980; Stöggl et al., 2005; Barba et al., 2011). Unfortunately, the standard microparticulate C18 phases fail to separate the β- and γ-forms of TS and T3 s because their retention is very similar. However, recently, the partial resolution of the two pairs of isomers was achieved on a conventional C18 column, thermostatically controlled at 7 °C, using a mobile phase composed by water and isopropanol (Irakli et al., 2012). The low flow rate, which was set at 0.3 mL/min to avoid high backpressure values, was responsible for a separation time longer

2. Materials and methods 2.1. Chemicals and materials α-T (all-rac-α-tocopherol; CAS Number 10191-41-0), β-T (all-rac-βtocopherol; CAS Number 148-03-8), γ-T ((+)-γ-tocopherol; CAS Number 54-28-4), δ-T ((+)-δ-tocopherol; CAS Number 119-13-1) and α-T-d6 (ring-5,7-dimethyl-d6; CAS Number 113892-08-3) were purchased from Aldrich-Fluka-Sigma S.r.l. (Milan, Italy). α-T3 (D-α-tocotrienol; CAS Number 58864-81-6), β-T3 (D-β-tocotrienol; CAS Number 490-23-3), γ-T3 (D-γ-tocotrienol; CAS Number 14101-61-2), and δ-T3 2

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

(10: 0.1, wsample/wBHT)), and independently homogenized just before the analysis. The short-term sample stability was investigated for each cultivar by analysing aliquots of a homogenous kernel pool after 8 and 15 days of storage in the dark at −20 °C. The analyte areas were then compared with those of the aliquots analysed just after the pool preparation (time 0). Since the relative standard deviation was lower than 10% in all cases, it was concluded that the samples were stable for at least 15 days. No longer periods were studied. Other authors studied changes in tocopherol content during 16-months' storage of whole kernels at 30 °C and 55% relative humidity; they observed that pecan nuts became rancid after 4 months’ storage (Fourie and Basson, 1989).

(D-δ-tocotrienol; CAS Number 25612-59-3) were obtained from LGC Standards (Middlesex, UK). α-T-d6 was chosen as internal standard (IS) for both Ts and T3s. All chemicals had a purity grade greater than 97%. Methanol was of RS-Plus grade (special grade reagents), 2-propanol and hexane were of RS grade (elevated purity grade); absolute ethanol was of RPE grade (analytical grade). All of these solvents were purchased from Carlo Erba (Milano, Italy) and Aldrich-Fluka-Sigma S.r.l. Potassium hydroxide (KOH) (≥85%) and butylated hydroxytoluene (BHT) (≥99%) were bought by Aldrich-Fluka-Sigma S.r.l. Distilled water was further purified by passing it through a Milli-Q Plus apparatus (Millipore, Bedford, MA USA). 2.2. Standard solutions

2.4. Sample treatment The individual stock solutions of the analytes and IS (1 μg/μL) were prepared dissolving weighed amounts (OHAUS DV215CD Discovery Semi-Micro and Analytical Balance 81 g/210 g capacity, 0.01 mg/ 0.1 mg readability) in methanol. The working solutions were prepared from the individual stock standard solutions, by diluting with methanol to obtain concentrations suitable for the several studies and experiments. The analyte photodegradation was avoided by using amber glassware for all preparations which, when unused, were stored at −18 °C in the dark. According to the results of a stability study of 40 days’ duration, these solutions were prepared monthly.

2.4.1. Overnight cold saponification A 1 g aliquot of homogenized nut was transferred into a 50 mL polypropylene centrifuge tube with a screw-cap (falcon) and spiked with 50 μL of the IS solution (10 ng/μL); thirty minutes were allowed for equilibration. Three mL of absolute ethanol with 0.1% BHT and 1 mL of 50% (w/v) aqueous KOH were poured into the centrifuge tube. After flowing nitrogen into the tube to remove oxygen, the falcon was quickly closed and placed in a water bath kept at 25° C, under magnetic stirring overnight (15 h). When the hydrolysis was completed, the digest was diluted with 3 mL of Milli-Q water and the analytes were extracted with two 4 mL fractions of hexane with 0.1% (w/v) BHT. After the addition of each hexane aliquot, the mixture was stirred for 5 min, vortexed for 5 min and then centrifuged at 8000g at 0 °C for 10 min. The two hexane layers were then combined together and washed with 4 mL portions of Milli-Q water until complete removal of alkalis (after two washings, no colour was observed on phenolphthalein addition). At this point, the extract was poured into a glass tube with a conical bottom (i.d. 2 cm), evaporated up to 500 μL under nitrogen flow at 30 °C in a water bath and diluted to a final volume of 1 mL with 2-propanol containing 0.1% (w/ v) BHT. Eventually, 2 μL of such an extract was injected into the HPLC–MS/MS system.

2.3. Sampling and sample stability In the 90s, the Experimental Institute for Fruit of Rome, now CREAResearch Centre for Fruit, evaluated the possibility of pecan nut cultivations in Italy (Tamponi, 1987), setting an initial collection at the experimental farm of Fiorano (Rome, Italy). Three varieties of pecan nuts, Sioux, Stuart and Pawnee, were supplied by this research centre to Sapienza University for the characterization. All trees have been grown in a plantation of 8 × 6 m planting distance, equipped with a drip irrigation system to satisfy the water requirement of such species. Grass of the orchard floor is managed by mulching and a regular mineral fertilization is applied in Autumn. At the moment, the trees have reached an approximately height of 8 m and provide a yield of about 25 kg of nuts per tree. The cultivars examined in this study have been chosen for their agronomical importance (Grauke and Thompson, 2017). Fig. S1 in the Supplementary material shows the external appearance of each nut variety and the internal fruit. Stuart cultivar origin dates back to 1890. Its nuts, which ripen at the end of October, are oblong elliptic with dark stripes on shell, while kernels are golden to light brown. Their weight is 8–9 g with a meat/ shell ratio of 47–50%. This cultivar is characterized by resistance to several diseases.Sioux was originated by a controlled cross made in 1943. Harvest is at the end of October. Its nuts are oblong round in cross section. The weight is 7- 8 g with a meat/shell ratio of 58–60%. The kernel is cream to golden in colour, with high oil content. This cultivar is characterized by good productivity. Pawnee is the youngest variety resulting from a controlled cross made in 1963. Nuts ripen at the latter half of September, are elliptical and the kernels are cream to golden in colour. Each fruit weighs 8–10 g with a meat/shell ratio of 50–58%. In this work, kernels of eight nuts per cultivar were peeled and grinded to obtain homogenous pools of about 24 g. To this end, each pool was spiked with BHT (in a ratio of 10: 0.1, w/w), homogenized in an Ultra-Turrax homogenizer (Janke & Kunkle, Staufenim Breisgau, Germany) and finally, subdivided in aliquots to be used for the experiments of development, optimization and validation. All subsamples were stored in the dark at −20 °C until analysis. For the characterization of vitamin E homologues, nuts were kept with the shell under dry conditions and in the dark until the sample preparation. Then, five nuts per cultivar were opened, their kernels were weighed, spiked with BHT

2.5. Liquid chromatography- tandem mass spectrometry Liquid chromatography was performed by means of a micro HPLC/ autosampler/vacuum degasser system PE Series 200 (Perkin Elmer, Norwalk, CT). Simultaneous separation of Ts and T3 s was carried out on a ProntoSIL C30 column (4.6 × 250 mm, 3 μm; Bischoff Chromatography, Leonberg, Germany), chilled at 15 ± 0.2 °C (Flexar column oven, Perkin Elmer, Norwalk, CT) and equipped with a guard column of the same type (4.0 mm × 10 mm, 5 μm particle size). NARP mobile phase was methanol: acetonitrile (95:5, v:v) as phase A and 2-propanol: hexane (50:50, v:v) as phase B. The mobile phase flow (1 mL/min) was entirely introduced into the APCI source. The analytes were eluted the first 18 min of the run under isocratic conditions at 100% of phase A; then, in 1 min, the mobile phase composition was changed and maintained at 100% of phase B for other 16 min. The phase B was also used as washing solution of the autosampler injection system. Fig. 2 shows the LC-MRM chromatogram of a working standard solution containing all vitamin E vitamers at a concentration of 10 ng/μL. Under the definitive conditions, it was possible to separate β and γ forms of both Ts and T3s, even if not at the baseline. Analytes were detected and quantified by a 4000 QqQLIT (AB SCIEX, Foster City, CA, USA) mass spectrometer, operated as triple quadrupole. The APCI probe was placed in the Turbo V source; the needle current and probe temperature were set at 3 μA and 450 °C, respectively. Highpurity nitrogen was used as a curtain gas (5 L/min) and collision gas (4 mTorr), while air was used as nebulizer gas (2 L/min) and make up gas (30 psi). 3

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Fig. 2. LC-MRM profiles (quantifier transition) of the eight vitamin E homologues (2 μL injected of a working standard solution at 1 ng/μL).

2.6. Method validation

Mass axis calibration of each quadrupole mass-analyser, Q1 and Q3, was carried out by the infusion of a polypropylene glycol solution at 10 μL/min. Unit mass resolution was established and kept in each massresolving quadrupole by maintaining a full width at half maximum (FWHM) of approximately 0.6 ± 0.1 m/z. The quantitative analysis of the target analytes was performed in multiple reaction monitoring (MRM) mode. The parameters of the APCI source and analysers were optimized for every vitamer by working in flow injection analysis (10 ng injected; 1 mL/min of flow rate). Table 1 lists the quantification channels and the LC–MS/MS parameters (retention time, two MRM transitions for each analyte and their relative abundance) used for the identification of the analytes in the different varieties of pecan nut. The software used for acquiring and elaborating LC–MS data was Analyst 1.5.1 (AB Sciex, Foster City, CA, USA).

The method validation was performed by using homogeneous aliquots taken from the pools of the three pecan varieties, prepared according what explained in Section 2.3. Calibration curves were constructed in matrix to perform the LC-MRM quantitative analysis of the Ts and T3 s in each of the three cultivars and to evaluate sensitivity and linear dynamic range. Method validation also included the study of selectivity, recoveries, precision, limits of detection (LODs), and limits of quantification (LOQs). For each homologue, the most intense MRM transition (quantifier transition, Q) was used for quantification purposes and the second most intense one (qualifier transition, q) for identification/confirmation purpose. Linear regression, means and standard deviations were calculated using Microsoft Excel 2010. 2.6.1. Identification and selectivity The analyte identification was based on the criteria taken from 4

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

signal-to-noise ratio of 3:1). For the LOQ calculation, the following conditions had to be satisfied simultaneously: i) a signal-to-noise ratio ≥ 10:1 for the Q transition; ii) the detection of both MRM transitions with their characteristic relative abundance (ion ratio); this means that the q transition had to be detected with a signal-to-noise ratio ≥ 3:1.

Table 1 LC–MS/MS identification parameters of vitamin E homologues. Vitamin E homologues

Retention timea (min)

MRM transitionsb (m/z)

Ion ratioc (%)

δ-tocotrienol (δ-T3)

6.8

397.4/177.3 397.4/137.1

72

γ-tocotrienol (γ-T3)

8.0

411.5/191.4 411.5/151.2

39

β-tocotrienol (β-T3)

8.5

411.5/191.4 411,5/151,2

28

α-tocotrienol (α-T3)

9,8

425.3/205.2 425.3/165.2

22

δ-tocopherol (δ-T)

10.6

402,4/137,4 402.4/177.3

58

γ-tocopherol (γ-T)

12.8

416.3/191.3 416.3/151.1

30

β-tocopherol (β-T)

13.5

416.3/191.3 416.3/151.1

54

α-tocopherol (α-T)

15.8

430.2/205.4 430.2/165.1

13

α-tocopherol-d6 (α-T-d6; IS)

16.3

436.4/171.2

–

2.6.4. Recovery and precision For the evaluation of recovery and precision, five samples were spiked pre-extraction with known amounts of the analytes and IS. The spiking level of most analytes was chosen so to increase the natural content of each compound by a factor of 2–3, while that of the undetected analytes was established at about 2–3 times their LOQ. After a 30 min equilibration period at room temperature, the samples were submitted to the extraction process. The recoveries were calculated by comparing the results of the five samples with those of a control sample (QC). The QC sample was prepared for each pecan variety by adding the same nominal amounts of analytes and IS at the end of the extraction procedure, just before the final evaporation. Precision was expressed as coefficient of variation (CV). The five replicates analysed on one day were used to estimate the within-run precision, while the between-run precision was valued by repeating the same procedure on two additional days.

a

The retention times are reported as arithmetic average of ten replicates. The first line reports the least intense MRM transition (qualifier,q) and the second line the most intense one (quantifier, Q). c The ion ratio (relative abundance) between the two MRM transitions is calculated as percentage intensity ratio of Iq/IQ; the results are reported as arithmetic average of ten replicates. b

3. Results and discussion 3.1. Optimization of LC–MS conditions Although RP columns are known to have the advantages of better stability and longer durability than NP columns, these last are more efficient in separating the β and γ congeners of Ts and T3s. Moreover, they allow using organic solvents able to solubilize high amounts of lipids co-extracted from lipid-rich foods. A C30 column operating under NARP conditions assembles the advantages of classical RP columns and NP columns. Among the several methods available in the literature, there are only two that, relying on C30 column, are able to separate all vitamin E homologues. Nevertheless, either the run time is long (Knecht et al., 2015) or the aqueous composition of the mobile phase is not suitable to wash lipids from the column (Saha et al., 2013; Knecht et al., 2015). In this work, our intent was to develop a simple and fast NARPHPLC method compatible with MS detection and with the analysis of rich-fat foods. During the first steps of optimization, methanol was used as phase A and a mixture composed of 2-propanol: hexane (50:50, v/v) as phase B. In absence of water, alcohols were efficient in supporting the APCI ionization, while hexane was indispensable in removing the co-extracted lipids from the chromatographic column. Analytes were separated on the ProntoSIL C30 column, kept at 18 °C, by using a gradient elution (t0, B = 0%; t1, B = 0%; t15, B = 75%; t15.1, B = 99.5%; t30, B = 99.5%). Notwithstanding the subambient temperature value, the two β- and γ- pairs were not resolved. Since the C30 selectivity is strongly affected by the temperature, the latter was lowered at 15 °C. Fig. 3 compares the separations of β- and γ- T3 s and Ts at both temperature values. A further lowering of temperature (10 °C) did not produce any further improvement in resolution, while an increase over 16 °C caused loss of selectivity. Changes in the mobile phase composition and elution mode were also tested. These trials were performed in isocratic mode by using 100% methanol and mixtures methanol: acetonitrile in the following proportions: (99:1, v/v); (98:2, v/v); (95:5, v/v); (90:10, v/v). Among all tested NARP conditions (in isocratic as well as in gradient mode), the best resolution was obtained when the composition methanol: acetonitrile (95:5, v/v) was used. The separation was carried out according what described in section 2.5. Analytes eluted within 16 min; the resolution was not in the baseline but to an extent enough to allow the individual quantification of the two pairs of positional isomers (Fig. 2).

Commission Decision 2002/657/EC (Commission Decision 2002/657/ EC, 2002). One of these criteria establishes that if a triple quadrupole does not record a full mass spectrum, two MRM transitions are enough to confirm the identity of an analyte in matrix. By applying such criteria, the occurrence of a target tocol in real samples was verified matching the retention time (tr ± 2.5%) and the relative abundance of the two MRM transitions (ion ratio) with the corresponding average values obtained for the reference standard in solvent (Table 1). The selectivity study was carried on standards, IS and real samples. Since the method is intended to determine more than one analyte, each authentic standard and IS was individually injected into HPLC–MS/MS system at increasing concentrations to exclude the occurrence of interferences, contaminations and/or degradations. 2.6.2. Calibration curves The quantitative analysis of the vitamin E homologues was performed by means of matrix matched external calibration method by using the Q transition. For each pecan variety, seven 1 g aliquots (C0eC6) were taken from the corresponding pool and spiked with 50 μL of a solution containing the IS at a concentration of 10 ng/μL; the aliquots C1eC6 were spiked with the analytes pre-extraction at increasing concentration levels, as shown in Table S1 of the Supplementary material. The extraction procedure was performed according to what described in the Section 2.4. Before constructing the calibration curve of a tocol, the peak area detected in the C0 aliquot and related to the endogenous amount was subtracted from the peak areas of the calibrators C1eC6. Then, the relative peak area (Aanalyte/AIS) was plotted against the fortification level (Fig.S2). 2.6.3. LODs and LOQs For each analyte, LOD and LOQ were calculated using the Q transition, after estimating the endogenous concentration of each analyte. LOD was intended as the amount of substance which, after extraction from a real sample, was able to produce a chromatographic peak three times higher than the noise of the chromatographic baseline (i.e. a 5

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Fig. 3. Comparison between the chromatographic separations of the pairs β-T/γ-T and β-T3/γ-T3 obtained in two different temperature conditions (2 ng injected).

amounts could be observed. Hexane was selected as water-immiscible solvent to extract tocols because it gives two substantial advantages over other solvents: it does not extract soaps (potassium salts of fatty acids) nor form stable emulsions; moreover, its extracts are nearly neutral and few washings with water are enough to remove alkalis definitively. The hexane volume necessary for the analyte recovery was established by using four 4 mL fractions that, after evaporation, were individually analysed. Two fractions were enough to ensure a quantitative transference of the analytes. The optimal temperature for evaporation of the final extract was decided by analysing some “evaporated responses”, i.e. 8 mL hexane solutions spiked with the analytes (500 ng) and submitted to evaporation under nitrogen flow in a water bath at the following temperatures: 30 °C, 35 °C, 40 °C and 45 °C (final volume was 1 mL). The best condition was individuated at 30 °C. Other optimized parameters were volume and diluting solvent of the final extract. Ideally, a lower volume of the final extract results in a higher concentration of analytes and, consequently, in a better sensitivity. Nevertheless, because the unsaponifiable lipid fraction co-extracted by hexane was responsible for a slow evaporation rate of the solvent, the extract evaporation was stopped when a 500 μL volume was reached. The nonpolar nature of hexane made the selection of a more polar diluting solvent crucial to avoid a chromatographic band broadening. Several solvents were tested, including methanol, acetonitrile, 2-propanol and their mixtures. Nevertheless, the need of solubilising residual lipids limited the choice. Clear and transparent extracts were only obtained by diluting the extract to 1 mL with 2propanol: hexane (75:25, v: v), 2-propanol: methanol (75:25, v: v), and 2-propanol. More polar compositions resulted in opaque extracts. Dilution with 2-propanol provided the best signal-to-noise chromatographic ratio when 2-μL of the extract was injected onto column. In fact, since the final composition of the final extract (2-propanol:hexane

Both APCI and ESI are suitable for ionization of Ts and T3 s in both polarity modes. Ionization and fragmentation of Ts have extensively been studied and described in previous papers (Lanina et al., 2007; Gentili and Caretti, 2011; Bartosińska et al., 2016), while few information are available for T3 s (Yu et al., 2007). APCI is more sensitive than ESI of about 10 times. In turn, the APCI detection of Ts in negative mode is similar or more sensitive than the positive one; nevertheless, most analytes generate only one fragment ion characterized by high intensity (Lanina et al., 2007). This means that the selection of two positive MRM transitions fails in the confirmation of the analytes at low concentrations. So, basically for these reasons, we chose to work in positive ion mode. 3.2. Optimization of extraction conditions The extraction of Ts and T3 s from nut kernels implicates several difficulties related to i) the instability of the analytes (UV radiations; slow oxidation by atmospheric oxygen accelerated by light, heat, alkalinity) ii) the high fat content (among 66–78% for pecan varieties (Venkatachalam et al., 2007)) in which vitamin E vitamers are trapped. For these reasons, the hydrolysis of saponifiable lipid fraction is a needed step, but the extraction conditions (KOH concentration and/or temperature) must be opportunely balanced to avoid analyte degradation. In this work, thermal and light stresses were avoided by applying overnight cold saponification at 25 °C in the dark. Concentration and volume of the aqueous KOH solution (0.5 mL, 1 mL and 2 mL of KOH 10%; 0.5 mL and 1 mL of KOH 50%) were established by evaluating the increase of the chromatographic areas of the analytes after hydrolysis. One mL of the KOH 10% solution was the better compromise between moles of hydrolysed triglycerides and analyte preservation. Actually, the T3 s are more susceptible to degradation than the Ts due to their unsaturated side chains and, only in these conditions, their endogenous 6

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

wet weight because of lack of information in their article (final volume of the extract). However, by using APCI-MRM detection, we achieved LODs and LOQs adequate to detect the endogenous amounts of Ts and T3 s in the real samples, maintaining a high identification power also at low concentration levels. Table 3 also groups the absolute recoveries and precision. Withinrun and between-run precision were within ± 20%, in perfect agreement with the FDA guidelines (Guidance for Industry: Bioanalytical Method Validation, 2001) that recommend a threshold of ± 20% for the lowest spike level. The exhaustive validation of a method aimed at analysing micronutrients in food and biological samples should provide the use of certified reference materials (CRMs). Unfortunately, there are not commercially available CRMs containing a complete set of eight natural tocols. There is only the Standard Reference Material 3278 for the determination of tocopherols in edible oils. Nevertheless, oil is not a matrix as complex as nuts and a simpler sample treatment (liquid–liquid extraction) is enough for the full extraction of vitamin E homologues. For this reason, the CRM suitable for oil is not representative for nuts. Thus, in this work, an indirect validation was the only possible way to bypass the unavailability of an appropriate CRM and to evaluate the analytical performance of the developed method.

Table 2 Linear regression parameters. Vitamin E homologues

Linear equationa

R2

α-tocopherol (α-T) β-tocopherol (β-T) γ-tocopherol (γ-T) δ-tocopherol (δ-T) α-tocotrienol (α-T3) β-tocotrienol (β-T3) γ-tocotrienol (γ-T3) δ-tocotrienol (δ-T3)

y = 0.0185×-0.7927 y = 0.0043× + 2.0168 y = 0.0061×-16.127 y = 0.0045×-0.1505 y = 0.0103×-0.3676 y = 0.0341×-0.5825 y = 0.0221×-0.2252 y = 0.0094×-0.321

0.98 0.98 0.92 0.95 0.97 0.99 0.98 0.93

a Regression parameters are referred to the calibration curve constructed by using the Q transition.

(50:50, v/v)) was a little stronger than the starting mobile phase composition (methanol:acetonitrile, 95:5 (v/v)), the low injection volume permitted to contain the band broadening.

3.3. Method validation results The HPLC–MS/MS method for the analysis of vitamin E in different types of pecan nut was validated according to the main FDA guidelines (Guidance for Industry: Bioanalytical Method Validation, 2001). All the validation results are summarized in Table 2 and Table 3. Pecan kernels belonging to the several varieties were selected randomly and analysed to test the method selectivity. The application of the criteria explained in the section 2.6.1 allowed confirming the occurrence of Ts and T3 s in real samples and, at the same time, excluding the presence of potentially interfering substances from matrix. In order to construct the matrix-matched calibration curves and to estimate the linear dynamic range, the spike levels for Ts and T3 s were chosen on the basis of their δ endogenous content, estimated previously in a rough way (see Table S1). An excellent sensitivity (slopes of calibration curves between 0.0043 and 0.0341) and R2 values greater than 0.90 were obtained for all analytes (see Table 2). Since the calibration curves obtained for the three pecan varieties were similar, their slopes and intercepts were averaged and the resulting curve was employed for quantification. The absence of carry-over was assessed by injecting ethanol after the calibrator spiked at the highest point of the calibration curve. Table 3 shows LODs and LOQs, calculated as mean of five replicates. Previous LC methods were based on UV (Fourie and Basson, 1989; Yao et al., 1992; Wyatt et al., 1998; Demir and Cetin, 1999; Kornsteiner et al., 2006; Ryan et al., 2006; Villarreal-Lozoya et al., 2007) and FL detection (Lee et al., 1998). Nevertheless, the majority of them have not provided LODs and LOQs for a comparison; only Lee et al., 1998, for their LC-FL method, reported LODs (0.08 and 0.2 ng/20 μL) and LOQs (0.13–0.39 ng/20 μL) in units that cannot be converted in mg/100 g of

3.4. Characterization of the vitamin e vitamers in three varieties of pecan nuts Nuts in general are fruits rich of polyunsaturated fatty acids, proteins, fibres, carbohydrates and vitamin E (Rainey and Nyquist, 1997; Griel and Kris-Etherton, 2006; Azadmard-Damirchi et al., 2011). Pecan is a kind of nut whose lipophilic antioxidant fraction has not been studied in detail yet, especially in relation to the main commercial varieties. There are over 500 different cultivars of pecans, characterized by slightly differences in size, meat/shell ratio, flavour, texture, colour, shape, etc., which get their names from their discoverers or Native American tribes (Moore, Stuart, Pawnee, Mohawk, Shawnee, etc.) (Grauke and Thompson, 2017). Findings of recent studies suggest a daily intake of this fruit because of their positive effects on the human health. In fact, it has been shown that the high monounsaturated fatty acid content of pecan kernels decreases low-density lipoprotein levels, while the high vitamin E content improves the antioxidant defence of the human organism (Morgan and Clayshulte, 2000; Rajaram et al., 2001; Haddad et al., 2006). Despite these evidences, the number of papers on the vitamin E composition of pecan fruits is quite limited. There is only one paper dealing with the determination of both Ts and T3 s in pecan nuts (Lee et al., 1998). This study investigated the vitamin E composition of a single pecan fruit of an unknown cultivar and did not detect any T3. The differentiation of several pecan varieties (Nacono, Kanza, Desirable, Pawnee, Kiowa, and Shawnee) was tackled by Villarreal-Lozoya et al. (2007), who analysed fatty acids, phenolic compounds and γ-T only. Finally, the T composition was studied by

Table 3 Validation parameters of the LC–MS/MS method developed in this work. Vitamin E homologues

LOD (mg/100 g)

LOQ (mg/100 g)

Recovery (%)

Within-run precision (%)

Between-run precision (%)

α-tocopherol (α-T) β-tocopherol (β-T) γ-tocopherol (γ-T) δ-tocopherol (δ-T) α-tocotrienol (α-T3) β-tocotrienol (β-T3) γ-tocotrienol (γ-T3) δ-tocotrienol (δ-T3)

0.003 0.01 0.01 0.0009 0.005 0.0003 0.0003 0.01

0.01 0.03 0.03 0.003 0.02 0.001 0.001 0.03

92 98 100 107 110 105 105 87

11 9 9 15 20 14 13 19

12 10 10 15 18 16 15 20

7

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Kornsteiner et al. (2006) in several nuts, including pecan. All of these HPLC methods were based on fluorescence (FL) detection (Lee et al., 1998) and UV detection at 295 nm (Kornsteiner et al., 2006; VillarrealLozoya et al., 2007), whose selectivity can be not enough for an unequivocal identification of Ts and T3 s in so lipid-rich matrices as nuts. The analytical approach developed in this work was applied to characterize the vitamin E content in three pecan varieties: Stuart, Sioux and Pawnee. The quantitative results are summarized in Table 4, while Fig. 4 shows the LC-MRM chromatogram obtained for the Sioux cultivar as an example. Clear differences of vitamin E content were observed between the three studied varieties. Sioux was the richest cultivar with a total content of about ∼ 32 mg/100 g wet weight, followed by Stuart (∼16 mg/100 g wet weight) and Pawnee (∼9 mg/100 g wet weight), respectively. The distribution of the vitamers was also different, but γ-T was found as prevalent form in all varieties. Although most foods of plant origin has α-T as major homologue E, seeds and nuts typically contain higher levels of its biosynthetic precursor, γ-T (Lee et al., 1998;

Table 4 Composition of the vitamin E homologues in the three investigated varieties of pecan nuts (five replicates per variety). Analyte

Stuart

Sioux

Pawnee

(mg/100 g wet weight) α-tocopherol (α-T) β-tocopherol (β-T) γ-tocopherol (γ-T) δ-tocopherol (δ-T) α-tocotrienol (α-T3) β-tocotrienol (β-T3) γ-tocotrienol (γ-T3) δ-tocotrienol (δ-T3) a b c

1.31 ± 0.02 ± 13.9 ± 0.85 ± n.d. n.d. LOQb n.d

0.04 0.08 0.3 0.05

0.52 ± 0.06 0.27 ± 0.09 31.1 ± 0.3 n.d n.d. 0.009 ± 0.004 0.025 ± 0.05 n.d

0.40 ± 0.06 0.19 ± 0.06 8.68 ± 0.2 0.15 ± 0.05 n.d < LOQa 0.010 ± 0.07 LODc

β-T3 LOQ: 0.001 mg/100 g. γ-T3 LOQ: 0.001 mg/100 g. δ-T3 LOD: 0.01 mg/100 g.

Fig. 4. LC-MRM profiles (quantifier transition) of the vitamin E homologues identified in the Sioux variety (2 μL injected of the final extract).

8

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Acknowledgements

Wyatt et al., 1998; Wagner et al., 2004; Kornsteiner et al., 2006; Ryan et al., 2006); exceptions are almonds (Lee et al., 1998; Wyatt et al., 1998; Wagner et al., 2004; Kornsteiner et al., 2006) and hazelnuts (Kornsteiner et al., 2006), while for other kind of nuts, such as Brazil nuts, pine nuts and peanuts, the amounts of both forms are comparable (Ryan et al., 2006; Kornsteiner et al., 2006). Pistachios are a very good source of γ-T with an average level around 30 mg/100 g (Ryan et al., 2006; Lee et al., 1998; Kornsteiner et al., 2006). For pecan nuts, the amount of this form strongly depends on the variety (Table 4): Sioux is the richest one, while the Pawnee’s content is about three times lower. The same conclusion was also drawn by Villarreal-Lozoya et al. (2007), who studied the γ-T content in six pecan cultivars (Desirable, Kanza, Kiowa, Nacono, Pawnee, and Shawnee). The authors determined α-T and β-T as well, but they did not reported the punctual concentrations, only specifying that these forms accounted for less than 5% of total T concentration. Our data for the variety Pawnee are in good agreement; moreover, the comparison with the other five cultivars studied by Villarreal-Lozoya et al. (2007) identified the Sioux variety as the most abundant one in γ-T content. Most studies did not taken into account δT or they did not find it in the analysed pecan samples, probably for a sensitivity problem (Lee et al., 1998). We detected δ-T in the Stuart and Pawnee cultivars (Table 4). Such amounts were in agreement with the concentration range defined by Kornsteiner et al. (2006) who analysed an unspecified pecan variety. Eventually, this study has evidenced the occurrence of some T3s, which were either not considered by previous studies or not detected at all (Lee et al., 1998). We identified γ-T3 in all three cultivars, β-T3 was detected in Sioux and Pawnee, while α-T3 was never detected. However, compared to tocopherol content, T3 s contribution to total vitamin E amount is minimum.

Our sincere acknowledgment to our dear colleague Anna Rocco, who prematurely passed away. She cooperated to this last work with her expertise in food chemistry. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jfca.2017.09.002. References Azadmard-Damirchi, S., Emami, S., Hesari, J., Peighambardoust, S.H., Nemati, M., 2011. Nuts composition and their health benefits. World Academy Sci. Eng. Technol. 5, 544–548. Barba, F.J., Esteve, M.J., Frigola, A., 2011. Determination of vitamins E α-, γ- and δtocopherol) and D (cholecalciferol and ergocalciferol) by liquid chromatography in milk, fruit juice and vegetable beverage. Eur. Food Res. Technol. 232, 829–836. Bartosińska, E., Buszewska-Forajta, M., Siluk, D., 2016. GC–MS and LC-MS approaches for determination of tocopherols and tocotrienols in biological and food matrices. J. Pharm. Biomed. Anal. 127, 156–169. Bauernfeind, J.C., 1980. Tocopherols in foods. In: Machlin, L.J. (Ed.), Vitamin E: A Comprehensive Treatise. Marcel Dekker, Inc, New York, NY, USA, pp. 99. Cooney, R.V., Franke, A.A., Harwood, P.J., Hatch-Pigott, V., Custer, L.J., Mordan, L.J., 1993. Gamma-tocopherol detoxification of nitrogen dioxide: superiority to alphatocopherol. Proc. Natl. Acad. Sci. U. S. A. 90, 1771–1775. Cunha, S.C., Amaral, J.S., Fernandes, J.O., Oliveira, M.B.P., 2006. Quantification of tocopherols and tocotrienols in Portuguese olive oils using HPLC with three different detection systems. J. Agric. Food Chem. 54, 3351–3356. Demir, C., Cetin, M., 1999. Determination of tocopherols, fatty acids and oxidative stability of pecan: walnut and sunflower oils. Deutsche Lebensmittel-Rundschau 95, 278–282. Esterbauer, H., Dieber-Rotheneder, M., Striegl, G., Waeg, G., 1991. Role of vitamin E in preventing the oxidation of low-density lipoprotein. Am. J. Clin. Nutr. 53, 314S–321S. Fourie, P.C., Basson, D.S., 1989. Changes in the tocopherol content of almond: pecan and macadamia kernels during storage. J. Am. Oil Chemists’ Society 66, 1113–1115. Górnaś, P., Siger, A., 2015. Simplified sample preparation and rapid detection by RPHPLC/FLD of tocopherols and tocotrienols in margarines: preliminary screening of plant fats—potential quality markers. Eur. J. Lipid Sci. Technol. 117, 1589–1597. Gentili, A., Caretti, F., 2011. Evaluation of a method based on liquid chromatography–diode array detector–tandem mass spectrometry for a rapid and comprehensive characterization of the fat-soluble vitamin and carotenoid profile of selected plant foods. J. Chromatogr. A 1218, 684–697. Grauke, L.J., Thompson, T.E., 2017. Pecan Breeding & Genetics, Agricultural Research Service, U.S. Dept. of Agriculture. (Retrieved February 4, 2017 from:http:// cgru.usda.gov/carya/pecans/cvintro.htm). Griel, A.E., Kris-Etherton, P.M., 2006. Tree nuts and the lipid profile: a review of clinical studies. Br. J. Nutr. 96 (Suppl. 2), S68–S78. Haddad, E., Jambazian, P., Karunia, M., Tanzman, J., Sabaté, J., 2006. A pecan-enriched diet increases γ-tocopherol/cholesterol and decreases thiobarbituric acid reactive substances in plasma of adults. Nutr. Res. 26, 397–402. Hall, G.D., 2000. Pecan food potential in prehistoric North America. Econ. Bot. 54, 103–112. Irakli, M.N., Samanidou, V.F., Papadoyannis, I.N., 2012. Optimization and validation of the reversed-phase high-performance liquid chromatography with fluorescence detection method for the separation of tocopherol and tocotrienol isomers in cereals, employing a novel sorbent material. J. Agric. Food Chem. 60, 2076–2082. Jiang, Q., Christen, S., Shigenaga, M.K., Ames, B.N., 2001. γ-Tocopherol, the major form of vitamin E in the US diet, deserves more attention. Am. J. Clin. Nutr. 74, 714–722. Knecht, K., Sandfuchs, K., Kulling, S.E., Bunzel, D., 2015. Tocopherol and tocotrienol analysis in raw and cooked vegetables: a validated method with emphasis on sample preparation. Food Chem. 169, 20–27. Kornsteiner, M., Wagner, K.H., Elmadfa, I., 2006. Tocopherols and total phenolics in 10 different nut types. Food Chem. 98, 381–387. Kua, Y.L., Gan, S., Morris, A., Ng, H.K., 2016. A validated, rapid: simple and economical high-performance liquid-chromatography method to quantify palm tocopherol and tocotrienols. J. Food Compos. Anal. 53, 22–29. Lanina, S.A., Toledo, P., Sampels, S., Kamal-Eldin, A., Jastrebova, J.A., 2007. Comparison of reversed-phase liquid chromatography–mass spectrometry with electrospray and atmospheric pressure chemical ionization for analysis of dietary tocopherols. J. Chromatogr. A 1157, 159–170. Lee, J., Landen, W.O., Phillipos Jr., R.D., Eitenmiller, R.R., 1998. Application of Direct Solvent Extraction to the LC Quantification of Vitamin E in peanuts, peanut butter and selected nuts. Peanut Sci. 25, 123–128. Monsen, E.R., 2000. Dietary reference intakes for the antioxidant nutrients: vitamin C vitamin E, selenium, and carotenoids. J. Am. Diet. Assoc. 100, 637–640. Morgan, W.A., Clayshulte, B.J., 2000. Pecans lower low density lipoprotein cholesterol in people with normal lipid levels. J. Am. Dietetic Assoc. 100, 312–318. Packer, L., Weber, S.U., Rimbach, G., 2001. Molecular aspects of α-tocotrienol antioxidant action and cell signalling. J. Nutr. 131, 369S–373S.

4. Conclusions Recent studies have suggested to increase the vitamin E intake significantly and to update nutrient databases on Ts and T3 s levels in foods, not limiting them only to α-T (Wagner et al., 2004; Traber, 2014). For such reasons, efforts have been dedicated to study the distribution of the vitamin E homologues in foods characterized by high bioavailability. Pecan nuts are excellent sources of vitamin E. Nevertheless, data on their detailed composition as well as comparative studies on commercially pecan cultivars are lacking. This paper describes the development of a LC/MS-based method for the speciation of vitamin E homologues in three pecan cultivars of commercial relevance: Stuart, Sioux and Pawnee. A key advantage of this method is the good chromatographic resolution obtained by using a C30 column kept at 15 °C. By working in isocratic mode with a mobile phase composed by a methanol: acetonitrile (95:5, v/v) solution, the analytes were eluted within the first 18 min. However, a final step with a 2-propanol: hexane (50:50, v: v) solution was indispensable to wash the co-extracted lipids from the column. After validation, the method was applied to elucidate the natural distribution of the vitamin E homologues in the real samples. Besides confirming γ-T as major component, the high method sensitivity was also able to detect the minor homologues: so data on δ-T and T3 s were provided for the first time. The γ form was the most abundant one also for T3s. Although Stuart is the most common variety that can be found on the market, Sioux has been resulted the richest one.

Funding The present work has been carried out under the project “Research for the improvements of the fruit trees protected cultivation in Southern Italy” − FRU·MED.; subproject VA. FRU. SE. ME., funded by the Ministry of Agricultural, Food and Forestry Policies (Italy). 9

Journal of Food Composition and Analysis xxx (xxxx) xxx–xxx

V. Pérez-Fernández et al.

Thompson, J.N., Hatina, G., Maxwell, W.B., 1980. High-performance liquid chromatographic determination of vitamin A in margarine, milk, partially skimmed milk, and skimmed milk. J. Assoc. Off. Anal. Chem. 63, 894–898. Traber, M.G., 2014. Vitamin E inadequacy in humans: causes and consequences. Advan. Nutri. Int. Rev. J. 5, 503–514. Venkatachalam, M., Kshirsagar, H.H., Seeram, N.P., Heber, D., Thompson, T.E., Roux, K.H., Sathe, S.K., 2007. Biochemical composition and immunological comparison of select pecan [Carya illinoinensis (Wangenh.) K. Koch] cultivars. J. Agric. Food Chem. 55, 9899–9907. Viñas, P., Bravo-Bravo, M., López-García, I., Pastor-Belda, M., Hernández-Córdoba, M., 2014. Pressurized liquid extraction and dispersive liquid? liquid microextraction for determination of tocopherols and tocotrienols in plant foods by liquid chromatography with fluorescence and atmospheric pressure chemical ionization-mass spectrometry detection. Talanta 119, 98–104. Villarreal-Lozoya, J.E., Lombardini, L., Cisneros-Zevallos, L., 2007. Phytochemical constituents and antioxidant capacity of different pecan [Carya illinoinensis (Wangenh.) K Koch] cultivars. Food Chem. 102, 1241–1249. Wagner, K.H., Kamal-Eldin, A., Elmadfa, I., 2004. Gamma-tocopherol?an underestimated vitamin? Ann. Nutr. Metab. 48, 169–188. Wong, Y.F., Makahleh, A., Saad, B., Ibrahim, M.N.M., Rahim, A.A., Brosse, N.1, 2014. UPLC method for the determination of vitamin E homologues and derivatives in vegetable oils, margarines and supplement capsules using pentafluorophenyl column. Talanta 130, 299–306. Wyatt, C.J., Carballido, S.P., Mendez, R.O., 1998. α-and γ-tocopherol content of selected foods in the Mexican diet: effect of cooking losses. J. Agric. Food Chem. 46, 4657–4661. Yao, F., Dull, G., Eitenmiller, R., 1992. Tocopherol quantification by HPLC in pecans and relationship to kernel quality during storage. J. Food Sci. 57, 1194–1197. Yu, S., Nehus, Z.T., Badger, T.M., Fang, N., 2007. Quantification of vitamin E and γoryzanol components in rice germ and bran. J. Agric. Food Chem. 55, 7308–7313.

Piironen, V., Syväoja, E.-L., Varo, P., Salminen, K., Koivistoinen, P., 1986. Tocopherols and tocotrienols in cereal products from Finland. Cereal Chem. 63, 78–81. Pinheiro-Sant’Ana, H.M., Guinazi, M., da Silva Oliveira, D., Della Lucia, C.M., de Lazzari Reis, B., Brandão, S.C.C., 2011. Method for simultaneous analysis of eight vitamin E isomers in various foods by high performance liquid chromatography and fluorescence detection. J. Chromatogr. A 1218, 8496–8502. Rainey, C., Nyquist, L., 1997. Nuts-Nutrition and health benefits of daily use. Nutr. Today 32, 157–163. Rajaram, S., Burke, K., Connell, B., Myint, T., Sabaté, J., 2001. A monounsaturated fatty acid-rich pecan-enriched diet favorably alters the serum lipid profile of healthy men and women. J. Nutr. 131, 2275–2279. Ryan, E., Galvin, K., O'Connor, T.P., Maguire, A.R., O'Brien, N.M., 2006. Fatty acid profile tocopherol, squalene and phytosterol content of brazil, pecan, pine, pistachio and cashew nuts. Int. J. Food Sci. Nutr. 57, 219–228. Saha, S., Walia, S., Kundu, A., Pathak, N., 2013. Effect of mobile phase on resolution of the isomers and homologues of tocopherols on a triacontyl stationary phase. Anal. Bioanal. Chem. 405, 9285–9295. Shammugasamy, B., Ramakrishnan, Y., Ghazali, H.M., Muhammad, K., 2013. Combination of saponification and dispersive liquid–liquid microextraction for the determination of tocopherols and tocotrienols in cereals by reversed-phase highperformance liquid chromatography. J. Chromatogr. A 1300, 31–37. Stöggl, W., Huck, C., Wongyai, S., Scherz, H., Bonn, G., 2005. Simultaneous determination of carotenoids, tocopherols, and gamma-oryzanol in crude rice bran oil by liquid chromatography coupled to diode array and mass spectrometric detection employing silica C30 stationary phases. J. Sep. Sci. 28, 1712–1718. Syväoja, E.L., Piironen, V., Varo, P., Koivistoinen, P., Salminen, K., 1986. Tocopherols and tocotrienols in Finnish foods: oils and fats. J. Am. Oil Chemistsä Society 63, 328–329. Tamponi, G., 1987. Prime Osservazioni Su Cultivar Di Pecan Introdotte Dagli U.S.A. 43. L'Informatore Agrario, pp. 53–59. Thompson, T.E., Grauke, L.J., 1991. Pecans and other hickories (Carya). Genet. Resour.Temp. Fruit Nut Crops 290, 839–906.

10