7. Opublikowane badania własne

|ourn,il uf Chrom jtogrjphy A. IJ GO-20 I-I; 101109

•V'.¹-1 .•>::

lilU

FI SFV1FR

Contents lists available at ScienceDirect

Journal of Chromatography A

journal homepage: www.elsevier.com/locatc/chioma

Modeling of the total antioxidant capacity of rooibos

[Aspalathus lineańs) tea infusions from chromatographic fingerprints

and Identification of potential antioxidant markers

Joanna Orzeł', Michał Daszykowski'Małgorzata Kazura', Dalene de Beer^b.

Elizabeth Joubert^{b c}. Alexandra E. Schulze⁰, Theresa Beelders^c, Andre de Villiers^d, Christiaan J. Malherbe^b, Beata Walczak'

‘ Insimur o/Tłumimy. The Unlvenlry o/Sitara. 9 Stolno Srrrei. 40-006 Katowice. Polar,d

*    Post Harvnt and Wine Technology Dimslon. Agriculiural Rrtcarrh Councll (AMC). Infrwiec Nieteoorbij. Pnvcit Bog X5026 Strllmbołr/i 7599. South A/nca

*    Depanmem o/ Food Science. Stellenbouh tlnleenlry. Prhtate Bag XI. Matieland (Siellmbntch) 7603. South Afriea

⁴ Deporrment of Chemisrry and PoJymrr Science, Slellenbauii L/niveniry. Pneate Bag XI. Marie land (Strllenbosch) 7602. Soulh Afnca

ARTICIE I N F O

AB5TRACT

Anicie hluory:

Hrmvcd 21 Junc 2014 Received in rwised form 12 Scplember 2014 Accepied 12 Srptember 2014 Ayjiljfale onlin¹ 22 Septrmber 2014

Keywordii

Annox dmt nurkers Aspalathus llnearij Polyphennlt Cjllbrjtion Chrmometrlct Qujhry contro)

Models to predict the total antiottidant capacity (TAC)of rooibos tea infusions from their chromatographic fingerprints and peak table data (content of indivłdual phenolic compounds). obtained using HPLC with diodę array detection. were developed in order to identify potential antioxidant markers. Peak table data inrluded the content of 12 compounds. namely phenylpyruwic acid-2 O-glucoside. aspalathin. nothofa¹ gin. jsoorientin. orientin. ferulic acid. quercetin-3-0-rohinobioside. vitexin. hyperoside. rutin. isovitexin and isoquercitrin. The TAC values. measured using the oxygen radical absorbance capacity (ORAĆ) and DPPH radical scavenging assays. could be predicted from the peak table data or the chromatographic fingerprints (prediction errors 9-12%) using partia! least squares (PLS] regression. Prediction models created from samples ofonly two production years could additionally br used to predict the TAC of samples from another production year (prediction enors < 131] indicating the robustness of the models in a quality conirol environment. Furthetmore. the uninformative variable elimination (UVE) PLS method was used to identify potential antioxidant markers for rooibos infusions. Ali indiwidual phenolic compounds that were quannfied were selected as informative variables. exccpt vitexin. while UVE- PLS models developed from chromatographic fingerprints indicated addirional antioxidant markers. namely (S)-enodictyol-6-C-glucoside. (R)-eriodictyol-6-f-glucoside. aspalalinin and rwo unidentified compounds The potential anlioxidanl markers should be va1idaied pnor to use in quality control of rooibos tea.

O 2014 Elsevier B.V. Ali rights reserwed

Iłllp ».'dx Uoi.orR.M0.l0t6/icluonUL2014.0y.030 007 1-9673/C 2014 Flievier 8 V. Ali ri|hts ret¹rved

Strona 108