Food Chemistry 139 (2013) 289–299

Contents lists available at SciVerse ScienceDirect

Food Chemistry

journal homepage: www.elsevier .com/locate / foodchem

Analytical Methods

Antioxidant capacity, polyphenolic content and tandem HPLC–DAD–ESI/MSprofiling of phenolic compounds from the South American berriesLuma apiculata and L. chequén

Mario J. Simirgiotis a,⇑, Jorge Bórquez a, Guillermo Schmeda-Hirschmann b

a Laboratorio de Productos Naturales, Departamento de Química, Facultad de Ciencias Básicas, Universidad de Antofagasta, Casilla 170, Antofagasta, Chileb Laboratorio de Química de Productos Naturales, Instituto de Química de Recursos Naturales, Universidad de Talca, Casilla 747, Talca, Chile

a r t i c l e i n f o

Article history:Received 16 January 2012Received in revised form 26 December 2012Accepted 28 January 2013Available online 13 February 2013

Keywords:Luma apiculataLuma chequénMyrtaceaeArrayánSouth American berriesHPLC–DAD–MSPhenolicsAntioxidants

0308-8146/$ - see front matter � 2013 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.foodchem.2013.01.089

⇑ Corresponding author. Tel.: +56 55 637229; fax: +E-mail address: mario.simirgiotis@uantof.cl (M.J. S

a b s t r a c t

Native Myrtaceae fruits were gathered by South American Amerindians as a food source. At present, thereis still some regional consume of the small berries from trees belonging to genus Luma that occurs insouthern Chile and Argentina. The aerial parts and berries from Luma apiculata and Luma chequen wereinvestigated for phenolic constituents and antioxidant capacity. A high performance electrospray ionisa-tion mass spectrometry method was developed for the rapid identification of phenolics in polar extractsfrom both species. Thirty-one phenolic compounds were detected and 27 were identified or tentativelycharacterised based on photodiode array UV–vis spectra (DAD), ESI–MS–MS spectrometric data and spik-ing experiments with authentic standards. Twelve phenolic compounds were detected in L. apiculatafruits and 12 in the aerial parts while L. chequen yielded 10 compounds in fruits and 16 in aerial parts,respectively. From the compounds occurring in both Luma species, seven were identified as tannins ortheir monomers, 15 were flavonol derivatives and five were anthocyanins. The whole berry and aerialparts extracts presented high antioxidant capacity in the DPPH assay (IC50 of 10.41 ± 0.02 and2.44 ± 0.03 lg/mL for L. apiculata, 12.89 ± 0.05 and 3.22 ± 0.05 for L. chequen, respectively), which canbe related to the diverse range of phenolics detected. The antioxidant capacity together with the highpolyphenolic contents and compounds identified can support at least in part, their use as botanical drugs.From the compounds identified in both species, 3-O-(600-O-galloyl)-hexose derivatives of myricetin, quer-cetin, laricitrin and isorhamnetin are reported for the first time for the genus Luma.

� 2013 Elsevier Ltd. All rights reserved.

1. Introduction 1995; Muñoz, Barrera, & Meza, 1981). Despite the well known uses

The consumption of fruits belonging to the Myrtaceae family isa common and ancient practice in South America. The Amerindianpopulations gathered the fruits and the largest in size with the besttaste were incorporated into South American culinary traditions allover the continent. Several edible Myrtaceae fruits including theChilean berry ‘‘murtilla’’ (Ugni molinae Turczaninov) and themurtilla-like berry Myrteola nummularia (Poiret) Berg. have beenshown to be a good source of polyphenolic antioxidants(Arancibia-Avila et al., 2011; Reynertson, Yang, Jiang, Basile, &Kennelly, 2008).

The Myrtaceae Luma apiculata (DC.) Burret and Luma chequén(Molina) A. Gray are trees with edible black-coloured berriesoccurring in southern Chile and Argentina. The berries from bothare half the size of commercial blueberries with a more intense col-our but similar aspect and consistence, and have been employed toprepare ‘‘chicha’’, a Mapuche fermented beverage (Hoffmann,

ll rights reserved.

56 55 637457.imirgiotis).

and health benefits (Murillo, 1889) of these berries, especially L.apiculata, their polyphenolic composition and antioxidant activityhave not been reported.

The medicinal properties of ‘‘Arrayan’’ leaves (L. apiculata, syn.Eugenia apiculata DC. or Myrceugenella apiculata (DC.) Kausel,Hoffmann, 1995), include aromatic, astringent, balsamic andanti-inflammatory activity (Murillo, 1889). In addition, inhibitoryactivity of the xanthine oxidase enzyme has been reported forthese and other Chilean Myrtaceae with similar medicinal uses,including the treatment of gout, in Chile and Paraguay (Theoduloz,Franco, Ferro, & Schmeda Hirschmann, 1988; Theoduloz, Pacheco,& Schmeda Hirschmann, 1991). Leaves from the related species L.chequén A. Gray (syn: Myrceugenella chequen (Mol.) Kaus have beenused as an astringent (de Mösbach, 1991).

In the last few years, several biological samples such as alcoholicplant and fruit extracts containing complex mixtures of small andmedium size phenolic and other molecules including very polarand thermally labile constituents have been analysed with thedevelopment of reliable LC–MS/MS equipment (Steinmann & Ganz-era, 2011; Wright, 2011). Indeed, the use of liquid chromatography

290 M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299

(HPLC, UPLC) coupled to diverse mass spectrometers such as hybridquadrupole time of flight (Q-TOF) or electrospray ionization-iontrap (Q-ESI) analyzers with complementary properties have beenused in the last years for metabolic profiling and biological analysis(Aliferis & Chrysayi-Tokousbalides, 2011; Kang et al., 2011; Mattoliet al., 2011).

The LC–MS methods proved to be superior to GC–MS since noprior derivatisation of polar samples (bearing hydroxyl and car-boxyl groups) is required (Hao, Zhao, & Yang, 2007). Quality con-trol of herbal drugs and medicinal plants is currently performedwith ESI–MS (Steinmann & Ganzera, 2011). HPLC–ESI–MS wasused to analyse carotenoids (Maoka, 2009), anthocyanins (Barnes,Nguyen, Shen, & Schug, 2009), phenolic acids (Fischer, Carle, &Kammerer, 2011) and alkaloids (He et al., 2011) in edible fruits.Chilean native berries such as calafate (Berberis spp.) (Ruiz et al.,2010) maqui (Aristotelia chilensis) and murta (U. molinae) (Rubilaret al., 2006) were also analysed using this precise technique.

Despite the traditional use of L. apiculata and L. chequen, we werenot able to find studies about phenolics constituents or antioxidantcapacity of edible fruits from either species. The main goals and nov-elty of this work is the profiling of phenolics as well as the measure-ment of antioxidant capacity and polyphenolic content of extractsfrom the berry fruits and leaves of the native Chilean L. apiculataand L. chequen, which is a continuation of our studies on SouthAmerican food plants (Simirgiotis & Schmeda-Hirschmann, 2010a).

2. Materials and methods

2.1. Chemicals and plant material

Folin–Ciocalteu phenol reagent (2 N), Na2CO3, AlCl3, FeCl3,

NaNO2, NaOH, D(+) glucose, D(+) galactose, L(+) rhamnose, D(�) ri-bose, quercetin, sodium acetate, HPLC-grade water, HPLC-gradeacetonitrile, thin layer chromatography (TLC, Kieselgel F254) plates,reagent grade MeOH and formic acid were obtained from Merck(Darmstadt, Germany). Myricetin 3-O-rhamnoside (myricitrin),quercetin 3-O-rhamnoside (quercitrin), cyanidin, myricetin, isorh-amnetin, syringetin, petunidin, pelargonidin, peonidin, malvidinand their 3-O-glucosides (all standards with purity higher than95% by HPLC) were purchased either from ChromaDex (SantaAna, CA, USA) or Extrasynthèse (Genay, France). Gallic acid, TPTZ(2,4,6-tri(2-pyridyl)1,3,5-triazine), Trolox and DPPH (1,1-diphe-nyl-2-picrylhydrazyl radical) were purchased from Sigma–AldrichChemical Co. (USA).

Aerial parts and ripe fruits of L. apiculata (DC.) Burret (local name:Arrayán), and L. chequén (Molina) A. Gray (local name: Chequén),were collected by Luis Bermedo Guzmán and Mario J. Simirgiotisin Re-Re, Región del Bio-Bio, Chile in May 2011. Voucher herbariumspecimens and fruit samples were deposited at the Laboratorio deProductos Naturales, Universidad de Antofagasta, Antofagasta, Chile,with the numbers La-111505-1 and Lc-111505-2, respectively.

2.2. Sample preparation

Fresh Luma fruits and aerial parts (leaves and stems) were sep-arately homogenised in a blender and freeze-dried (Labconco

Table 1Total phenolic content (TPC), total flavonoid content (TFC), total anthocyanin content (TACpercent w/w extraction yield of Luma methanolic extracts on the basis of freeze-dried sta

Species and plant parta TPC (mg/g) TFC (mg/g) TAC (mg/g)

L. apiculata fruits 29.44 ± 0.10 13.31 ± 0.01 21.03 ± 2.14L. chequén fruits 5.15 ± 0.00 1.51 ± 0.00 1.57 ± 0.00L. apiculata aerial parts 179.83 ± 0.38 139.70 ± 1.48 -L. chequen aerial parts 327.09 ± 0.80 126.54 ± 1.15 -

a Measurements are expressed as mean ± SD of three parallel determinations (All valu

Freezone 4.5 L, Kansas, MO, USA). One gram of lyophilised materialwas finally pulverised in a mortar and extracted thrice with 25 mLof 0.1% HCl in MeOH in the dark for 1 h each time. The extractswere combined, filtered and evaporated in vacuo (40 �C). The ex-tracts were suspended in 10 mL ultrapure water and loaded ontoa reverse phase solid phase extraction cartridge (SPE, Varian BondElut C-18, 500 mg/6 mL). The cartridge was rinsed with water(10 mL) and phenolic compounds were eluted with 10 mL MeOHacidified with 0.1% HCl. The solutions were evaporated to drynessunder reduced pressure to give 73.60 mg of L. apiculata fruits,93.60 mg of L. apiculata aerial parts, 67.4 mg of L. chequén fruitsand 40.6 mg of L. chequén aerial parts, respectively (for extractionyields see Table 1). The extracts were then dissolved in 2 mL 0.1%HCl in MeOH, filtered through a 0.45 lm micropore membrane(PTFE, Waters) before use and 10 ll were injected into the HPLCinstrument for analysis.

2.3. HPLC analysis

A Merck-Hitachi (LaChrom, Tokyo, Japan) instrument equippedwith an L-7100 pump, an L-7455 UV diode array detector, a D-7000chromato-integrator and a column compartment was used foranalyses. The sample was separated on a Purospher star-C18 col-umn (250 mm � 5 mm, 4.6 mm i.d., Merck, Germany). The mobilephase consisted of 10% formic acid in water (A) and acetonitrile (B).A gradient program was used for HPLC–DAD and ESI-MS as fol-lows: 90% A in the first 4 min, linear gradient to 75% A over25 min, then linear gradient back to initial conditions for other15 min. The mobile phase flow rate was 1 mL/min. The columntemperature was set at 25 �C; the detector was monitored at520 nm for anthocyanins and 320–280 nm for other compoundswhile UV spectra from 200 to 600 nm were recorded for peakcharacterisation.

2.4. Mass spectrometric conditions

An Esquire 4000 Ion Trap mass spectrometer (Bruker Daltoniks,Germany) was connected to an Agilent 1100 HPLC instrument viaESI interface for HPLC–ESI-MS analysis. Full scan mass spectrawere measured between m/z 150 and 2000 u in positive ion modefor anthocyanins and negative ion mode for other compounds.High purity nitrogen was used as nebuliser gas at 27.5 psi, 350 �Cand at a flow rate of 8 l/min. The mass spectrometric conditionsfor negative ion mode were: electrospray needle, 4000 V; end plateoffset, �500 V; skimmer 1, �56.0 V; skimmer 2, �6.0 V; capillaryexit offset, �84.6 V; and the operating conditions for positive ionmode were: electrospray needle, 4000 V; end plate offset,�500 V; skimmer 1, 56.0 V; skimmer 2, 6.0 V; capillary exit offset,84.6 V; capillary exit, 140.6 V. Collisionally induced dissociation(CID) spectra were obtained with a fragmentation amplitude of1.00 V (MS/MS) using ultrahigh pure helium as the collision gas.

2.5. Alkaline and acid hydrolysis of MeOH extracts

To verify acylation of the flavonol glycoside derivatives 18, 23,24 and 28 from the HPLC fingerprint, the SPE MeOH extracts

), ferric reducing antioxidant power (FRAP), scavenging of the free radical DPPH andrting material.

FRAP (lmol/g) DPPH (IC50, lg/mL) w/w extraction yield (%)

93.4 ± 0.0 10.41 ± 0.02 7.3676.2 ± 0.0 12.89 ± 0.05 9.36

170.5 ± 0.1 2.44 ± 0.03 6.74135.6 ± 0.3 3.22 ± 0.05 4.06

es are significantly different at p < 0.05).

M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299 291

obtained as explained above (2 mL) was hydrolysed with 2 mL of2 mol/L sodium hydroxide as previously reported (Simirgiotis, Cal-igari, & Schmeda-Hirschmann, 2009). The mixture was kept for16 h at room temperature, neutralised with concentrated hydro-chloric acid, filtered (0.45 lm, PTFE Waters) and directly analysed(10 ll) by HPLC–DAD and ESI–MS–MS. Another portion of pro-cessed SPE MeOH extracts (2 mL) was dissolved in 4 mol/L HCl(2 mL) in order to further confirm the identity of flavonol aglyconesby HPLC. After stirring the solution at 90 �C for 60 min, water wasadded (10 mL), the solvent and hydrochloric acid were removedunder reduced pressure and the remaining aqueous solution(10 mL) was again submitted to SPE to wash the remaining acidand sugars. The identification of the sugars in the aqueous solutionwas performed by TLC using different sugars as standards as re-ported by Figueirinha, Paranhos, Pérez-Alonso, Santos-Buelga,and Batista (2008). After SPE, the resulting methanolic solution(2 mL) was filtered (0.45 lm, PTFE Waters) and directly analysed(10 ll) by HPLC–DAD–ESI/MS. In order to measure the recoveryof phenolics for the total extraction procedure, a standard anthocy-anin (pelargonidin, 0.5 mg/mL) a standard flavonoid (quercetin,0.5 mg/mL) and a standard phenolic acid (gallic acid, 0.5 mg/mL)were added to a fresh sample (one gram) of freeze-dried L. apicula-ta fruits (three times) used as matrix, extracted, SPE processed asabove and recovery was calculated using HPLC–DAD.

2.6. Antioxidant assessment

2.6.1. Free radical scavenging activityThe free radical scavenging activity of the extracts was deter-

mined by the DPPH� assay as previously described (Simirgiotis &Schmeda-Hirschmann, 2010a), with some modifications. DPPHradical absorbs at 517 nm, but upon reduction by an antioxidantcompound its absorption decreases. Briefly, 50 lL of processedSPE MeOH extract or pure compound prepared at different concen-trations was added to 2 mL of fresh 0.1 mM solution of DPPH inmethanol and allowed to react at 37 �C in the dark. After 30 minthe absorbance was measured at 517 nm. The DPPH scavengingability as percentage was calculated as: DPPH scavenging abil-ity = (Acontrol � Asample/Acontrol) � 100. Afterwards, a curve of % DPPHbleaching activity versus concentration was plotted and IC50 valueswere calculated. IC50 denotes the concentration of sample requiredto scavenge 50% of DPPH free radicals. The lower the IC50 value themore powerful the antioxidant capacity. If IC50 6 50 lg/mL thesample has high antioxidant capacity, if 50 lg/mL < IC50 6 100 lg/mL the sample has moderate antioxidant capacity and if IC50 > 200 -lg/mL the sample has no relevant antioxidant capacity. Gallic acid(from 1.0 to 125.0 lg/mL, R2 = 0.991) and quercetin (from 1.0 to125.0 lg/mL, R2 = 0.993) were used as standard antioxidant com-pounds, and were determined to have IC50 values of 1.1 lg/ml(6.8 lmol/L) and 7.5 lg/ml (24.8 lmol/L), respectively.

2.6.2. Ferric reducing antioxidant powerThe determination of ferric reducing antioxidant power or ferric

reducing ability (FRAP assay) of the extracts was performed as de-scribed by (Benzie & Strain, 1996) with some modifications. Thestock solutions prepared were 300 mM acetate buffer pH 3.6,10 mM TPTZ (2,4,6-tri(2-pyridyl)-s-triazine) solution in 40 mMHCl, and 20 mM FeCl3�6H2O solution. Plant extracts or standardmethanolic Trolox solutions (150 lL) were incubated at 37 �C with2 mL of the FRAP solution (prepared by mixing 25 mL acetate buf-fer, 5 mL TPTZ solution, and 10 mL FeCl3�6H2O solution) for 30 minin the dark. Absorbance of the blue ferrous tripyridyltriazinecomplex formed was then read at 593 nm. Quantification was per-formed using a standard calibration curve of antioxidant Trolox(from 0.2 to 2.5 lmol/mL, R2: 0.995). Samples were analysed intriplicate and results are expressed in lmol TE/gram dry mass.

2.7. Polyphenol, flavonoid and anthocyanin contents

’The total polyphenolic contents (TPC) of Luma fruits and leaveswere determined by the Folin–Ciocalteau method (Simirgiotis,Caligari, et al., 2009; Simirgiotis, Theoduloz, Caligari, and Schme-da-Hirschmann, 2009) with some modifications. An aliquot of eachprocessed SPE extract (200 ll), was added to the Folin–Ciocalteaureagent (2 mL, 1:10 v/v in purified water) and after 5 min of reac-tion at room temperature (25 �C), 2 mL of a 100 g/L solution of Na2-

CO3 was added. Sixty minutes later the absorbance was measuredat 710 nm. A calibration curve was performed with the standardgallic acid (concentrations ranging from 16 to 500 lg/mL,R2 = 0.999) and the results expressed as mg gallic acid equiva-lents/g dry mass.

Determination of total flavonoid content (TFC) of the methano-lic extracts was performed as reported previously (Simirgiotiset al., 2008) using the AlCl3 colorimetric method. Quantificationwas expressed by reporting the absorbance in the calibration graphof quercetin, which was used as the flavonoid standard (from 0.1 to65.0 lg/mL, R2 = 0.994). Results are expressed as mg quercetinequivalents/g dry mass. The assessment of total anthocyanin con-tent (TAC) was carried out as described by (Lee, Durst, & Wrolstad,2005). Absorbance was measured at 510 and 700 nm in buffers atpH 1.0 and 4.5. Pigment concentration is expressed as mg cyanidin3-glucoside equivalents/g dry mass and calculated using theformula:

TA ðmg=gÞ ¼ A�MW� DF� 103e� 1

where A = (A510nm � A700nm) pH 1.0 � (A510nm � A700nm) pH 4.5;MW (molecular weight) = 449.2 g/mol; DF = dilution factor;1 = cuvette pathlength in cm; e = 26,900 L/mol cm, molar extinc-tion coefficient for cyanidin 3-O-b-D-glucoside. 103: factor to con-vert g to mg. All spectrometric measurements were performedusing a Unico 2800 UV–vis spectrophotometer (Shangai, Unicoinstruments, Co., Ltd.).

2.8. Statistical analysis

The statistical analysis was carried out using the originPro 9.0software packages (Originlab Corporation, Northampton, MA,USA). The determination was repeated at least three times for eachsample solution. Analysis of variance was performed using ANOVA.Significant differences between means were determined by stu-dent’s t-test (p values < 0.05 were regarded as significant).

3. Results and discussion

In the present study, we assessed the polyphenolic profile ofaerial parts and fruits of L. apiculata and L. chequen collected inthe Bio-Bio Region, Chile, and evaluated its antioxidant capacityas well as the total phenolic, total flavonoid and total anthocyanincontent by spectrophotometric methods. The fresh fruits and aerialparts were extracted with methanol and the resulting extractswere processed by solid phase extraction. The weight/weightextraction yields of the extracts were 7.36%, 9.36%, 6.74% and4.06% for L. apiculata fruits, L. chequén fruits, L. apiculata aerial partsand L. chequen aerial parts, respectively. The identity of phenoliccompounds from the extracts was investigated by high-perfor-mance liquid chromatography paired with UV photodiode array(HPLC–DAD) and triple quadrupole ion trap-electrospray ionisa-tion tandem mass spectrometry (HPLC–ESI/MS). Anthocyaninswere monitored in ESI positive mode while other compounds weremeasured in negative mode. While L. chequén is reported to pro-duce several flavanones (Labbe et al., 1992), these compounds

292 M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299

were not identified in our Luma samples. The pattern of methoxy-lated flavonoids glycosides (laricitrin, myricetin, isorhamnetin) de-tected in Luma resembles that reported in Vitis vinifera cv. PetitVerdot grapes (Castillo-Muñoz et al., 2009).

3.1. Total phenolic, anthocyanin and flavonoid contents andantioxidant capacity of Luma extracts

In this study, the phenolic profiles of methanolic extracts fromLuma fruits and aerial parts were compared by HPLC–DAD(Fig. 1). Antioxidant capacity of the extracts was measured and cor-related with the total phenolic, flavonoid and anthocyanin con-tents. The total phenolic, anthocyanin and flavonoid contents aswell as the extraction yield and antioxidant capacity measuredby the bleaching of DPPH radical and ferric reducing antioxidantpower are given in Table 1. The flavonoid content of the L. apiculatafruit extract forms about 45% of its total phenolic content, while forL. chequen is only 29% (Table 1). The aerial parts of Luma had highertotal phenolic content (179.83 ± 0.38 and 327.09 ± 0.80 mg/g gallicacid equivalents for L apiculata and L. chequén, respectively) thanthe fruits (Table 1) with total phenolic content values similar tothat reported for green tea (213 ± 5.9 mg/g gallic acid equivalents)and two times the values reported for mate leaves (112.1 ± 4.1 mg/g gallic acid equivalents) (Piccinelli, De Simone, Passi, & Rastrelli,2004). Anthocyanins identified were present in both Luma fruits,but concentration of these pigments were different, (Total antho-cyanin content: 21.03 ± 2.14 mg/g for L. apiculata and1.57 ± 0.00 mg/g for L. chequen fruits) which is in accordance with

Fig. 1. HPLC–DAD chromatograms of Luma extracts. (a) Chromatograms at 280 nm. (A) L.(b) chromatograms at 520 nm. (E) L. apiculata fruits, (F) L. chequén fruits.

the different antioxidant capacity (10.41 ± 0.02/12.89 ± 0.05 lg/mLin the DPPH assay and 93.4 ± 0.0/76.2 ± 0.0 lmol Trolox/g in theFRAP assay, respectively, Table 1). The TAC value for L. apiculatafruits (21.03 ± 2.14 mg/g) was almost three times of that reportedfor the blueberries Vaccinium uliginosum (9.01 ± 0.06 mg/g offreeze-dried powder) (Li et al., 2011). The small Korean fruits (Lir-iope platyphylla) of a comparable shape, size and colour to Lumafruits (Fig. 2) produced similar HPLC–DAD anthocyanin fingerprint(Lee & Choung, 2011).

The phenolic content correlated with antioxidant capacity (R2:0.778 for TP/DPPH assay) while the antioxidant assays correlatedwith one another (R2: 0.922). The phenolic content of L. apiculatafruits was even higher than those reported for 17 berry cultivarsincluding raspberries (Rubus idaeus), blackberries (Rubus fructico-sus), red currant (Ribes sativum), gooseberry (Ribes glossularia)and cornelian cherry (Cormus mas) (Pantelidis, Vasilakakis, Man-ganaris, & Diamantidis, 2007). The L. apiculata aerial parts showeda FRAP value (170.5 ± 0.1 lmol TE/g) close to those reported for 50red and 18 green pepper collections (FRAP mean values: 185 and157 lmol TE/g dry weight, respectively), while L. chequén FRAP val-ues for aerial parts and fruits (135.6 ± 0.3 and 76.2 ± 0.0 lmol TE/g,respectively) were higher than common foods like beet(86 ± 29 lmol TE/g), cauliflower (51 ± 12 lmol TE/g), spinach(64 ± 13 lmol TE/g), broccoli (41 ± 11 lmol TE/g), tomato(56 ± 8 lmol TE/g) and white cabbage (39 ± 17 lmol TE/g), amongothers (Ou, Huang, Hampsch-Woodill, Flanagan, & Deemer, 2002).Phenolic and flavonoid content of both Luma leaves were higherthan those of the fruits (Table 1). The values were also higher than

apiculata fruits, (B) L. chequén fruits, (C) L. apiculata leaves, (D) L. chequén leaves, and

Fig. 1. (continued)

M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299 293

those reported for seven black and fresh tea leaves from Asia wherethe highest value reported was for the brand Ouvagalia tea with110 ± 10 mg/g GAE and 80 ± 7 mg/g quercetin equivalents (Luxi-mon-Ramma et al., 2005). L. apiculata aerial parts showed a pheno-lic content close to that reported for Rosa chinensis methanolextract (189 mg ± 13 GAE/g dry weight) (Cai, Xing, Sun, Zhan, &Corke, 2005).

3.2. Identification of phenolic constituents

Phenolics occurring in Luma fruits and aerial parts extracts wereseparated by HPLC and UV–vis spectra were obtained using adiode-array detector. HPLC fingerprints were generated (Fig. 1)and phenolic compounds subsequently analysed by ESI–MS–MS.A preliminary analysis of DAD spectrum obtained for the peaksgave a first indication of the family of phenolic compounds (Simir-giotis, Caligari, et al., 2009; Simirgiotis, Theoduloz, et al., 2009).Some compounds were identified by co-elution with standardphenolics. For those compounds not commercially available, fullscan mode followed by ESI–MS–MS experiments in negative mode

was a powerful tool for their characterisation. The 31 compoundsdetected and 27 identified or tentatively identified are listed in Ta-ble 2, along with UV–vis and MS data. Fig. 2 shows structures ofseveral compounds identified while Fig. 3 shows structures and fullMS and MS–MS spectra of some representative compounds. Peaks1–8 and 10–12 were tentatively identified as tannins (hydrolysableor proanthocyanins) or their monomers, peaks 30 and 31 weresimple flavonols while peaks 15, 16, 18, 20–29 were glycosyl flavo-nol derivatives, and among those, peaks 15, 20, 21 and 25 wereflavonols acylated with gallic acid. Peaks 9, 13, 14, 17 and 19 wereanthocyanins, and peaks 2, 8, 10 and 12 remain as unidentifiedphenolics. The identification of peaks is listed below.

3.3. Tannins and flavanol derivatives

Peaks 1–8 and 10–12 were tentatively identified as hydrolysa-ble tannins, epimeric procyanidins or flavanol derivatives. Peak 1was identified as the hydrolysable tannin hexahydroxydiphenoyl-glucose (HHDP-glucose, (Fig. 2) with a MW 482, a [M�H]� ion at

Fig. 2. Structures of compounds identified in Chilean Luma berries. (a) Tannins, peaks 1, 5, 6 and 7 and corresponding fragmentation pattern. (b) Anthocyanins: peaks 9, 13,14, 17 and 19. (c) Flavonol/glycoside derivatives: peaks 15, 16, 18, 20–31.

294 M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299

m/z 481 yielding diagnostic fragments at m/z 301, 283, 257 and229) as reported (Salminena, Ossipova, & Pihlajaa, 2002; Simirgio-tis & Schmeda-Hirschmann, 2010a). Peak 3 was also a hydrolysabletannin with a molecular ion at m/z 783, an MS2 ion at m/z 481, pro-ducing a daughter MS3 ion at m/z 301 (with MS4 ions at m/z 283,257 and 229 assigned to one HHDP unit or ellagic acid) identifiedas a bis-HHDP-glucose/hexose as previously reported (Fischeret al., 2011; Simirgiotis & Schmeda-Hirschmann, 2010b). Peak 4was assigned as the ellagitannin castalagin or its isomer vescalaginboth with a [M�H]� at m/z 933 (Figs. 2 and 3a) and MS ions at m/z631 (Fig 2, loss of HHDP unit), 481 (loss of gallic acid moiety) and301 (loss of galloyl-glucosyl moiety from the parent MS2 ion at m/z631 (Simirgiotis & Schmeda-Hirschmann, 2010b). Peak 5 was iden-tified as an ellagic acid hexoside (pedunculagin I) showing an[M�H]� ion at m/z 633 and fragment ions at m/z 615, 481 and301 (Fischer et al., 2011). Furthermore, peak 6 was identified as an-other bis-HHDP-hexose derivative with a mass difference of 32 U([M�H]� ion at m/z 815, C34H23O24) and the same UV data and

MS–MS fragments as for compounds 3 and 5. This compoundwas characterised as the ellagitannin furosinin (Takuo, 2005). Peak7 showed a [M�H]� ion at m/z 577 (Fig. 3a) and MSn ions at m/z425 (RDA rearrangement from one heterocicle of the dimer), m/z407 (loss of water from fragment at m/z 425) and m/z 289 (epicat-echin, diagnostic fragments at m/z 245, 205 and 179 (Stoggl, Huck,& Bonn, 2004) and was identified as procyanidin B1 by comparisonwith literature (Sun, Liang, Bin, Li, & Duan, 2007) and spikingexperiments with authentic compound. MS/MS analysis of peak11 with a molecular ion at m/z 457 showed MS2 ions at m/z 331,169 (gallic acid moiety), and 305 (deprotonated epigallocatechin).This compound was identified as epigallocatechin gallate (Mark-owicz Bastos et al., 2007) by comparison with an authentic sample.

3.4. Anthocyanins

Five known anthocyanins were identified in the fruits, (peaks 9,13, 14, 17 and 19, Fig. 3b) with molecular ions in positive mode at

Fig. 2. (continued)

M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299 295

m/z 465, 449, 479, 463 and 493 and showing characteristic MS2

ions at m/z 303 (MS3 ion at m/z 257), 287 (MS3 ions at m/z 213,147), 317 (MS3 ion at m/z 302), 301 (MS3 ion at m/z 286) and331 (MS3 ion at m/z 299, 179) respectively, corresponding todelphinidin 3-O-glucoside (kmax: 275-341sh-512), cyanidin-3-O-glucoside, (kmax: 278-503), petunidin-3-O-glucoside (kmax: 275-343sh-512), peonidin-3-O-glucoside, (kmax: 268-357sh-503), andmalvidin-3-O-glucoside (kmax: 275-343sh-512), respectively. Theidentity was corroborated by co-elution with standard anthocya-nins and literature data. After extraction and SPE method therecovery of an external standard compound (pelargonidin) was97 ± 7% by HPLC.

3.5. Flavonol derivatives

Peaks 15, 16, 18, 20–31 were identified as flavonol derivativessince the shape of the UV spectra were similar to those reported(Mabry, Markham, & Thomas, 1970; Simirgiotis, Caligari, et al.,2009; Simirgiotis, Theoduloz, et al., 2009; Sun et al., 2007). Thelinkages of gallic acid moieties for the 3-O-acylated flavonols werelocated in position 600 of the sugar hydroxyl by characteristic MSfragmentation (Ferreres et al., 2008). MS–MS analysis of all of thosecompounds showed characteristic ions at m/z 179 and 151 (Fig. 3c)which were confirmed to be produced by RDA rearrangement of5,7-dihydroxy-flavon-3-ols such quercetin, isorhamnetin andmyricetin using deuterium labelling experiments (McNab, Ferreira,Hulme, & Quye, 2009). Peaks 16, 18 (Fig 3c) and 22 showed similar

UV spectra and [M�H]� ions at m/z 449, 479 and 463 respectivelyall yielding a MS2 daugther ion at m/z 317 (myricetin) and weretentatively identified as myricetin 3-O-pentose, 3-O-hexose and3-O-rhamnose (Michodjehoun-Mestres et al., 2009). Peak 24 witha MW 494 (full ESI–MS main peak: 493 U, Fig. 3c) was tentativelyidentified as a methyl-myricetin hexoside derivative (laricitrinderivative). A flavonol derivative with a [M�H]� ion of 493 Uwas identified as quercetin 3-methoxy-hexoside in blueberries(Cho, Howard, Prior, & Clark, 2005). However, MS–MS data of com-pound 27 is in agreement with laricitrin 3-O-hexose (Castillo-Muñoz et al., 2009) or myricetin 30 methyl ether-3-O-hexose(Min et al., 2010). Further fragmentation of the ion at m/z 493yielded an ion at m/z 331 (laricitrin, or myricetin 50 methyl ether),which in turn, yielded an ion at m/z 316 (myricetin-2 H). In thesame manner, compound 21 with UV spectral data of 255, 293,358 nm and a [M�H]� ion at 645 (Fig. 3c, MSn ions at 493 and331 U) was tentatively identified as myricetin 50 methyl ether-(600

galloyl) 3-O-hexose (Min et al., 2010).Full scan MS main peak for compound 28 (m/z 477 [M�H]�)

was consistent with the molecular formula C22H22O12. MS–MSfragmentation pattern (Fig 3c, MS2 315 U and MS3 300 U) was coin-cident with that reported for isorhamnetin-3-O-glucoside (Gutzeit,Wray, Winterhalter, & Jerz, 2007). In the same manner, peak 25was characterised as isorhamnetin-3-O-(600-O-galloyl)-glucose(Fig. 3c) and peak 31 was characterised as the aglycon isorhamne-tin since it showed the characteristic UV spectra and has an[M�H]� ion at m/z 315 consistent with a molecular anion of

Table 2Identification of phenolic compounds in Luma fruits and aerial parts by LC–DAD, LC–MS and MS/MS data.

Peak Rt (min) HPLC–DAD kmax (nm) ESI mode [M�H]�

(m/z)[2M�H]�

(m/z)MS–MS ions (m/z) Tentative identification Species/plant

part

1 2.0 275 – 481 301, 257, 229 HHDP-glucose Cf2 2.3 290 – 533 191 Unknown quinic acid derivative Ap, Cp3 2.9 270 – 783 633, 481, 301 Bis-HHDP hexose Cf4 2,9 262 – 933 631, 451, 301 Castalagin or Vescalagin Af5 3.0 275 – 633 615, 481, 301 Galloyl HHDP hexose (pedunculagin I) Ap, Cp6 3.6 270 – 815 481, 301, 257, 229 Furosinin Cf7 4.1 280 – 577 407, 289 Procyanidin B1 Cp8 4.7 275 – 357 715 169, 125 Unknown gallotannin Cf9 6.8 275, 341sh, 512 + 465 303, 257 Delphinidin 3-O-hexose Af, Cf10 10.0 275 – 509 357, 169 Unknown Gallotannin Cf11 9.7 275 – 457 331, 169 Epigallocatechin gallate Af, Ap12 10.0 275 – 509 357, 169 Unknown Gallotannin Cp13 10.1 278, 503 + 449 287, 213, 147 Cyanidin-3-O-glucose Af, Cf14 11.3 275, 343sh, 512 + 479 317, 302 Petunidin 3-O-glucose Af,Cf15 12.1 257, 292sh-361 – 631 479, 317 My-3-O-(600-O-galloyl)-hexose Af, Cp, Ap16 13.5 257, 362 – 449 317, 179 My-3-O-ribose Ap17 14.1 268, 357sh, 503 + 463 301, 286 Peonidin 3-O-glucose Af,Cf18 14.4 254–362 – 479 959 317, 179 My-3-O-galactose (myricitrin) Af,Cf, Cp19 15.0 275, 343sh, 512 + 493 331, 299, 179 Malvidin 3-O-glucose Af, Cf20 15.9 255, 290sh, 356 – 615 1231 463, 301 Q-3-O-(600-O-galloyl)-hexose Af, Cp, Ap21 16.7 255, 293sh, 358 – 645 493, 331, 315, 179, 151 L-(600-O-galloyl)3-O-hexose Cp

22 17.7 254, 360 – 463 927 317, 179, 151 My 3-O-rhamnose Af, Cp, Ap23 18.8 254, 354 – 463 927 301, 257, 179, 151 Q-3-O-glucose (isoquercitrin) Af, Cp, Ap24 19.4 266, 361 – 493 331-316, 179 L3-O-hexose Cp

25 20.3 264, 292sh, 357 – 629 477, 315, 300 IRh-3-O-(600-O-galloyl)-hexose Ap, Cp26 21.3 255, 355 – 433 301, 179, 151 Q-3-O-ribose Ap27 22.8 254, 355 – 447 895 301, 179, 151 Q-3-O-rhamnose (quercitrin) Ap28 23.0 265, 354 477 955 315, 300 IRh-3-O-glucose Cp29 23.3 265, 357 – 507 1015 344, 345, 315 Syringetin-3-O-glucose Cp30 25.1 265, 357 – 345 330, 315 Syringetin Ap, Cp31 25.5 265, 354 315 300, 179, 151 IRh Cp

Abbreviations. Compounds: Q: quercetin; My: myricetin; IRh: isorhamnetin; L: Laricitrin; Species and plant parts: L. apiculata: (Ap) aerial parts (Af) fruits; L. chequén: (Cp)aerial parts (Cf) fruits.

296 M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299

C16H11O7-, which in turn, produced an MS2 ion at m/z 300,

([M�15�H]�), due to loss of a methyl group and MS3 ions typicalof 5,7 dihydroxy-flavonols (Justesen, 2001; McNab et al., 2009;Simirgiotis & Schmeda-Hirschmann, 2010a). Peak 29 was identi-fied as syringetin-3-O-glucoside with a prominent peak at 507[M�H]� in the full scan mass spectra (Fig. 3c) and MS/MS ions at345 [M�hexose�H]� and 315 (Gutzeit et al., 2007), while com-pound 30 eluting 2 min later was identified as the flavonol aglyconmyricetin 30,50-di-O-methyl ether (syringetin, ESI–MS–MS data:315, 300, 179, 151 U).

The full scan mass spectra of compound 20 showed mainly anintense ion at m/z 615, which yielded an MS2 ion at m/z 463 (quer-cetin 3-O-glucoside: isoquercitrin) (Gutzeit et al., 2007) corre-sponding to the loss of a gallic acid moiety ([M�galloylmoiety�H]�) (Sannomiya et al., 2007) which in turn, fragmentedto an MS3 ion at m/z 301 (deprotonated quercetin, MS4 ions at m/z 179, 151) by loss of an hexose unit (162 U). UV spectral data ofthis compound is consistent with the typical UV overlap of a gallicacid moiety (UV shoulder at kmax 290 nm) and a flavonol structure(kmax band I: 354 nm, band II: 254 nm)(Djoukeng, Arbona, Argam-asilla, & Gomez-Cadenas, 2008). Thus, this compound was identi-fied as a quercetin-3-O-(600 galloyl) glucoside (Barakat, Souleman,Hussein, Ibrahiem, & Nawwar, 1999). In the same manner, peak15 (kmax: 257, 292 and 361 nm) was identified as myricetin-3-O-(600 galloyl)glucoside/galactoside (molecular anion at m/z 631 andMS–MS ions at 479 and 317 U as reported (Romani, Campo, &Pinelli, 2012).

Peak 27 was identified as the quercetin 3-O-rhamnoside:quercitrin by comparison of retention time and MS and UV prop-erties with authentic sample. This compound is common in Myrt-aceae plants, and was identified in U. molinae leaves (Rubilar

et al., 2006). Compound 26 was identified as a quercetin-3-O-pentoside (ribose, loss of 132 U from the molecular anion at m/z433) (Simirgiotis, Theoduloz, Caligari, & Schmeda-Hirschmann,2009). The identity of flavonoids in the extracts was corroboratedby comparing DAD (at 280, 320 and 520 nm) chromatograms andMS data before and after acid and alkaline hydrolysis. Aftersaponification peaks 15, 20, 21 and 25 were not detected, andafter acid hydrolysis the branched flavonoids described in thiswork gave glucose, ribose, galactose and rhamnose as sugar moi-eties (by TLC) and myricetin, isorhamnetin, syringetin, quercetin,cyanidin, petunidin, peonidin, delphinidin and malvidin (byHPLC–DAD and ESI–MS comparison with authentic standards)as aglycones. After extraction and SPE method the recovery ofan external standard flavonoid compound (quercetin) and a phe-nolic acid (gallic acid) were 88 ± 5%, and 99 ± 7% respectively byHPLC.

3.6. Unknown phenolic derivatives

Peak 2 has a [M�H]� ion at m/z 533 and UV max at 320 nm. Acompound with these UV and similar MS characteristics (plus afragment at m/z 179 [caffeic acid�H]�) was identified as a caffeicacid derivative occurring in leaves of Helichrysum melaleucum(Gouveia & Castilho, 2011). Because of its main MS2 fragment atm/z 191 (quinic acid, MS3 127), it was partially identified as an un-known quinic acid derivative. Peak 8 with a [M�H]� ion at 357 aswell as peaks 10 and 12 both with UV data (kmax 275 nm) andpseudomolecular ions at m/z 509 and MS-MS ions at m/z 357[M�152�H]� and 169 were tentatively identified as gallotanninisomers or gallic acid derivatives, since detection of the fragment

Fig. 3. Full MS and MS–MS spectra of some selected compounds identified in Luma berries. (a) Tannins (peaks 3, 4 and 7); (b) anthocyanins (peaks 9, 13, 14 and 17); (c)flavonol derivatives (peaks 18, 21 24, 25, 28 and 29).

M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299 297

MS3 ion at m/z 169 led to the identification of a gallic acid residue(MW: 170).

4. Conclusions

The LC–DAD and ESI–MS–MS system used in this work allowedthe detection of 31 phenolic compounds and the identification ortentative identification of 27 of them in fruits and aerial parts ofL. apiculata and L. chequen. The results obtained pointed out that

the methodology developed is appropriate for rapid analysis andidentification of phenolic substances in extracts from native Lumaspecies and can be potentially used for other edible South Ameri-can myrtaceae fruits. The anthocyanin pattern for both Luma spe-cies was similar but higher content was found in L. apiculata fruits.

The highest total phenolic content was found in Luma chequenaerial parts while L. apiculata aerial parts showed the highest totalflavonoid content. The antioxidant properties and high content ofphenolics and flavonoids found in the aerial parts can explain, at

Fig. 3. (continued)

298 M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299

least in part, the traditional use of both native plants and the re-puted health benefits of the infusions. Further research is neededto explore other biological activities of Luma fruits to support theirpotential in the human diet.

Acknowledgment

Financial support from FONDECYT (Project 1110068) and CODEI(University of Antofagasta grant 5383) is gratefully acknowledged.

References

Aliferis, K. A., & Chrysayi-Tokousbalides, M. (2011). Metabolomics in pesticideresearch and development: Review and future perspectives. Metabolomics, 7(1),35–53.

Arancibia-Avila, P., Toledo, F., Werner, E., Suhaj, M., Leontowicz, H., Leontowicz, M.,et al. (2011). Partial characterization of a new kind of Chilean murtilla-likeberries. Food Research International, 44(7), 2054–2062.

Barakat, H. H., Souleman, A. M., Hussein, S. A. M., Ibrahiem, O. A., & Nawwar, M. A.M. (1999). Flavonoid galloyl glucosides from the pods of Acacia farnesiana.Phytochemistry, 51(1), 139–142.

Barnes, J. S., Nguyen, H. P., Shen, S., & Schug, K. A. (2009). General method forextraction of blueberry anthocyanins and identification using high performanceliquid chromatography–electrospray ionization–ion trap-time of flight-massspectrometry. Journal of Chromatography A, 1216(23), 4728–4735.

Benzie, I. F. F., & Strain, J. J. (1996). The ferric reducing ability of plasma (FRAP) as ameasure of ‘‘Antioxidant Power’’: The FRAP assay. Analytical Biochemistry, 239,70–76.

Cai, Y. Z., Xing, J., Sun, M., Zhan, Z. Q., & Corke, H. (2005). Phenolic antioxidants(hydrolyzable tannins, flavonols, and anthocyanins) Identified by LC–ESI–MSand MALDI–QIT–TOF MS from Rosa chinensis flowers. Journal of Agricultural andFood Chemistry, 53, 9940–9948.

Castillo-Muñoz, N., Gómez-Alonso, S., García-Romero, E., Victoria Gómez, M.,Velders, A. H., & Hermosín-Gutiérrez, I. (2009). Flavonol 3-O-glycosides seriesof Vitis vinifera Cv. Petit Verdot red wine grapes. Journal of Agricultural and FoodChemistry, 57(1), 209–219.

Cho, M. J., Howard, L. R., Prior, R. L., & Clark, J. R. (2005). Flavonol glycosides andantioxidant capacity of various blackberry and blueberry genotypes determinedby high-performance liquid chromatography/mass spectrometry. Journal of theScience of Food and Agriculture, 85, 2149–2158.

M.J. Simirgiotis et al. / Food Chemistry 139 (2013) 289–299 299

de Mösbach, E. W. (1991). Botánica Indígena de Chile. In M. Chileno (Ed.), De ArtePrecolombino, Fundación Andes y Editorial Andrés Bello (pp. 95–96). Santiago deChile: C. Aldunate and C. Villagrán.

Djoukeng, J. D., Arbona, V., Argamasilla, R., & Gomez-Cadenas, A. (2008). Flavonoidprofiling in leaves of Citrus genotypes under different environmental situations.Journal of Agricultural and Food Chemistry, 56(23), 11087–11097.

Ferreres, F., Valentao, P., Pereira, J. A., Bento, A., Noites, A., Seabra, R. M., et al. (2008).HPLC–DAD–MS/MS–ESI screening of phenolic compounds in Pieris brassicae L.reared on Brassica rapa var. rapa L. Journal of Agricultural and Food Chemistry, 56,844–853.

Figueirinha, A., Paranhos, A., Pérez-Alonso, J. J., Santos-Buelga, C., & Batista, M. T.(2008). Cymbopogon citratus leaves: Characterization of flavonoids by HPLC–PDA–ESI/MS/MS and an approach to their potential as a source of bioactivepolyphenols. Food Chemistry, 110(3), 718–728.

Fischer, U. A., Carle, R., & Kammerer, D. R. (2011). Identification and quantificationof phenolic compounds from pomegranate (Punica granatum L.) peel, mesocarp,aril and differently produced juices by HPLC-DAD–ESI/MS(n). Food Chemistry,127(2), 807–821.

Gouveia, S., & Castilho, P. C. (2011). Characterisation of phenolic acid derivativesand flavonoids from different morphological parts of Helichrysum obconicum bya RP-HPLC–DAD–(�)-ESI–MS(n) method. Food Chemistry, 129(2), 333–344.

Gutzeit, D., Wray, V., Winterhalter, P., & Jerz, G. (2007). Preparative isolation andpurification of flavonoids and protocatechuic acid from Sea Buckthorn juiceconcentrate (Hippophae rhamnoides L. ssp. rhamnoides) by high-speed counter-current chromatography. Chromatographia, 65(1/2), 1–7.

Hao, C., Zhao, X., & Yang, P. (2007). GC–MS and HPLC–MS analysis of bioactivepharmaceuticals and personal-care products in environmental matrices. TrAC –Trends in Analytical Chemistry, 26(6), 569–580.

He, D. X., Shan, Y., Wu, Y. H., Liu, G. Z., Chen, B., & Yao, S. Z. (2011). Simultaneousdetermination of flavanones, hydroxycinnamic acids and alkaloids in citrusfruits by HPLC–DAD–ESI/MS. Food Chemistry, 127(2), 880–885.

Hoffmann, A. E. (1995). Flora silvestre de Chile (3rd ed.). Santiago de Chile: EdicionesFundacion Claudio Gay.

Justesen, U. (2001). Collision-induced fragmentation of deprotonated methoxylatedflavonoids, obtained by electrospray ionization mass spectrometry. Journal ofMass Spectrometry, 36, 169–178.

Kang, H. J., Yang, H. J., Kim, M. J., Han, E. S., Kim, H. J., & Kwon, D. Y. (2011).Metabolomic analysis of meju during fermentation by ultra performance liquidchromatography–quadrupole-time of flight mass spectrometry (UPLC–Q-TOFMS). Food Chemistry, 127(3), 1056–1064.

Labbe, C., Coll, J., Connolly, J. D., Faini, F., Farrugia, L. J., & Rycroft, D. F. (1992).Bioactive flavanones from Luma chequen. Collection of Czechoslovak ChemicalCommunication, 67, 115–123.

Lee, J. H., & Choung, M.-G. (2011). Identification and characterisation ofanthocyanins in the antioxidant capacity-containing fraction of Liriopeplatyphylla fruits. Food Chemistry, 127, 1686–1693.

Lee, J., Durst, R. W., & Wrolstad, R. E. (2005). Determination of total monomericanthocyanin pigment content of fruit juices, beverages, natural colorants, andwines by the pH differential method: Collaborative study. Journal of AOACInternational, 88, 1269–1278.

Li, R., Wang, P., Guo, Q.-q., & Wang, Z.-y. (2011). Anthocyanin composition andcontent of the Vaccinium uliginosum berry. Food Chemistry, 125(1), 116–120.

Luximon-Ramma, A., Bahorun, T., Crozier, A., Zbarsky, V., Datla, K. P., Dexter, D. T.,et al. (2005). Characterization of the antioxidant functions of flavonoids andproanthocyanidins in Mauritian black teas. Food Research International, 38(4),357–367.

Mabry, T. J., Markham, K. R., & Thomas, M. B. (1970). The systematic identification offlavonoids, Part II. Berlin: Springer-Verlag.

Maoka, T. (2009). Recent progress in structural studies of carotenoids in animalsand plants. Archives of Biochemistry and Biophysics, 483(2), 191–195.

Markowicz Bastos, D. H., Saldanha, L. A., Catharino, R. R., Sawaya, A. C. H. F., Cunha, I.B. S., Carvalho, P. O., et al. (2007). Phenolic antioxidants identified by ESI-MSfrom yerba maté (Ilex paraguariensis) and green tea (Camelia sinensis) extracts.Molecules, 12, 423–432.

Mattoli, L., Cangi, F., Ghiara, C., Burico, M., Maidecchi, A., Bianchi, E., et al. (2011). Ametabolite fingerprinting for the characterization of commercial botanicaldietary supplements. Metabolomics, 7(3), 437–445.

McNab, H., Ferreira, E. S. B., Hulme, A. N., & Quye, A. (2009). Negative ion ESI-MSanalysis of natural yellow dye flavonoids – An isotopic labelling study.International Journal of Mass Spectrometry, 284(1–3), 57–65.

Michodjehoun-Mestres, L., Souquet, J. M., Fulcrand, H., Bouchut, C., Reynes, M., &Brillouet, J. M. (2009). Monomeric phenols of cashew apple (Anacardiumoccidentale L.). Food Chemistry, 112, 851–857.

Min, B. S., Cuong, T. D., Lee, J. S., Shin, B. S., Woo, M. H., & Hung, T. M. (2010).Cholinesterase inhibitors from Cleistocalyx operculatus buds. Archives ofPharmacal Research, 33(10), 1665–1670.

Muñoz, M., Barrera, E., & Meza, I. (1981). El uso medicinal y alimenticio de plantasnativas y naturalizadas en Chile. Santiago de Chile: Museo de Historia Natural.Publicación Ocasional No. 33.

Murillo, A. (1889). Plantes médicinales du Chili (Vol. 80). Paris: Exposition Universellede Paris Section Chilienne.

Ou, B., Huang, D., Hampsch-Woodill, M., Flanagan, J. A., & Deemer, E. K. (2002).Analysis of antioxidant activities of common vegetables employing oxygenradical absorbance capacity (ORAC) and ferric reducing antioxidant power(FRAP): A comparative study. Journal of Agricultural and Food Chemistry, 50,3122–3128.

Pantelidis, G. E., Vasilakakis, M., Manganaris, G. A., & Diamantidis, G. (2007).Antioxidant capacity, phenol, anthocyanin and ascorbic acid contents inraspberries, blackberries, red currants, gooseberries and Cornelian cherries.Food Chemistry, 102(3), 777–783.

Piccinelli, A. L., De Simone, F., Passi, S., & Rastrelli, L. (2004). Phenolic constituentsand antioxidant capacity of Wendita calysina leaves (Burrito), a folk Paraguayantea. Journal of Agricultural and Food Chemistry, 52, 5863–5868.

Reynertson, K. A., Yang, H., Jiang, B., Basile, M. J., & Kennelly, E. J. (2008).Quantitative analysis of antiradical phenolic constituents from fourteen edibleMyrtaceae fruits. Food Chemistry, 109(4), 883–890.

Romani, A., Campo, M., & Pinelli, P. (2012). HPLC/DAD/ESI–MS analyses and anti-radical activity of hydrolyzable tannins from different vegetal species. FoodChemistry, 130(1), 214–221.

Rubilar, M., Pinelo, M., Ihl, M., Scheuermann, E., Sineiro, J., & Nuñez, M. J. (2006).Murta leaves (Ugni molinae Turcz) as a source of antioxidant polyphenols.Journal of Agricultural and Food Chemistry, 54, 59–64.

Ruiz, A., Hermosin-Gutierrez, I., Mardones, C., Vergara, C., Herlitz, E., Vega, M., et al.(2010). Polyphenols and antioxidant capacity of calafate (Berberis microphylla)fruits and other native berries from southern Chile. Journal of Agricultural andFood Chemistry, 58(10), 6081–6089.

Salminena, J. P., Ossipova, V., & Pihlajaa, K. (2002). Distribution of hydrolysabletannins in the foliage of Finnish birch species. Zeitschrift für Naturforschung C,57c, 248–256.

Sannomiya, M., Campaner dos Santos, L., Carbone, V., Napolitano, A., Piacente, S.,et al. (2007). Liquid chromatography/electrospray ionization tandem massspectrometry profiling of compounds from the infusion of Byrsonima fagifoliaNiedenzu. Rapid Communications in Mass Spectrometry, 21, 1393–1400.

Simirgiotis, M. J., Adachi, S., To, S., Yang, H., Reynertson, K. A., Basile, M. J., et al.(2008). Cytotoxic chalcones and antioxidants from the fruits of Syzygiumsamarangense (Wax Jambu). Food Chemistry, 107, 813–819.

Simirgiotis, M. J., Caligari, P. D. S., & Schmeda-Hirschmann, G. (2009). Identificationof phenolic compounds from the fruits of the mountain papaya Vasconcelleapubescens A. DC. grown in Chile by liquid chromatography–UV detection–massspectrometry. Food Chemistry, 115(2), 775–784.

Simirgiotis, M. J., & Schmeda-Hirschmann, G. (2010a). Determination of phenoliccomposition and antioxidant capacity in fruits, rhizomes and leaves of thewhite strawberry (Fragaria chiloensis spp. chiloensis form chiloensis) usingHPLC–DAD–ESI-MS and free radical quenching techniques. Journal of FoodComposition and Analysis, 23(6), 545–553.

Simirgiotis, M. J., & Schmeda-Hirschmann, G. (2010b). Direct identification ofphenolic constituents in Boldo Folium (Peumus boldus Mol.) infusions by high-performance liquid chromatography with diode array detection andelectrospray ionization tandem mass spectrometry. Journal of ChromatographyA, 1217(4), 443–449.

Simirgiotis, M. J., Theoduloz, C., Caligari, P. D. S., & Schmeda-Hirschmann, G. (2009).Comparison of phenolic composition and antioxidant properties of two nativeChilean and one domestic strawberry genotypes. Food Chemistry, 113(2), 377–385.

Steinmann, D., & Ganzera, M. (2011). Recent advances on HPLC/MS in medicinalplant analysis. Journal of Pharmaceutical and Biomedical Analysis, 55(4), 744–757.

Stoggl, W. M., Huck, C. W., & Bonn, G. n. K. (2004). Structural elucidation of catechinand epicatechin in sorrel leaf extracts using liquid-chromatography coupled todiode array-, fluorescence-, and mass spectrometric detection. Journal ofSeparation Science, 27, 524–528.

Sun, J., Liang, F., Bin, Y., Li, P., & Duan, C. (2007). Screening non-colored phenolics inred wines using liquid chromatography/ultraviolet and mass spectrometry/mass spectrometry libraries. Molecules, 12, 679–693.

Takuo, O. (2005). Systematics and health effects of chemically distinct tannins inmedicinal plants. Phytochemistry, 66(17), 2012–2031.

Theoduloz, C., Franco, L., Ferro, E., & Schmeda Hirschmann, G. (1988). Xanthineoxidase inhibitory activity of Paraguayan Myrtaceae. Journal ofEthnopharmacology, 24, 179–183.

Theoduloz, C., Pacheco, P., & Schmeda Hirschmann, G. (1991). Xanthine oxidaseinhibitory activity of Chilean Myrtaceae. Journal of Ethnopharmacology, 33(3),253–255.

Wright, P. (2011). Metabolite identification by mass spectrometry: Forty years ofevolution. Xenobiotica, 41(8), 670–686.