ANTIOXIDANT CAPACITY OF PINOTAGE WINE AS AFFECTED BY VITICULTURAL AND ENOLOGICAL PRACTICES DALENE DE BEER DOCTOR OF PHILOSOPHY IN FOOD SCIENCE

ANTIOXIDANT CAPACITY OF PINOTAGE WINE AS AFFECTED BY VITICULTURAL AND ENOLOGICAL PRACTICES DALENE DE BEER Disserttion presented in prtil fulfilment of the requirements for the degree of DOCTOR OF PHILOSOPHY IN FOOD SCIENCE in the Deprtment of Food Science, Fculty of AgriSciences, Stellenbosch University Promotor: Co-promotors: Dr E. Joubert Dr J. Mris Dr M. Mnley December 2006

ii DECLARATION I, the undersigned hereby declre tht the work contined in this disserttion is my own originl work nd hs not been previously in its entirety or in prt submitted t ny university for degree. Dlene de Beer Dte

ABSTRACT iii The im of the study ws to provide the South Africn wine industry with guidelines for the production of Pinotge wines with optiml totl ntioxidnt cpcity (TAC), while retining sensory qulity. The contribution of individul phenolic compounds to the wine TAC is importnt in this regrd. The wine TAC ws mesured with the 2,2 -zino-di(3-ethylbenzo-thizoline-sulphonic cid rdicl ction) (ABTS + ) scvenging ssy. The contributions of individul phenolic compounds to the wine TAC were clculted from their content in the wines nd the Trolox equivlent ntioxidnt cpcity (TEAC) of pure phenolic stndrds. The effects of climte region, vine structure, enologicl techniques (pre-fermenttion mcertion, juice/skin mixing, ddition of commercil tnnins, extended mcertion) nd mturtion (ok brrels, lterntive ok products, oxygention) on the phenolic composition, TAC nd sensory qulity of Pinotge wines were lso investigted. The TEAC vlues of quercetin-3-glctoside, isorhmnetin nd peonidin-3-glucoside were reported for the first time. TEAC vlues observed for most compounds were much lower thn those reported previously, lthough TEAC vlues for gllic cid, cftric cid, cffeic cid nd kempferol were consistent with some previous reports. Cftric cid nd mlvidin-3-glucoside were the lrgest contributors to the wine TAC. The contents of monomeric phenolic compounds nd procynidin B1, however, only explined smll mount (between 11 nd 24%) of the wine TAC, with the remining TAC ttributed to oligomeric nd polymeric phenolic compounds nd other unknown compounds. Some synergy between different monomeric phenolic compounds ws lso demonstrted. All the viticulturl nd enologicl fctors investigted ffected the phenolic composition of Pinotge wines, while the wine TAC ws only ffected by some tretments. Chnges in wine TAC could not lwys be explined by chnges in phenolic composition s the contribution of oligomeric, polymeric nd unknown compounds could not be ssessed, but could ply lrge role. Differences in wine colour were lso difficult to explin due to the lrge number of fctors involved nd the drk wine colour, which mde objective mesurements difficult. The concentrtion of vitisin A, n ornge-red pyrnonthocynin, ws incresed consistently s result of prefermenttion mcertion tretments nd ffected the wine colour of oxygented wines. Incresed wine TAC ws observed when cultivting Pinotge grpes on bush vines nd in cooler climtic regions, compred to cultivtion on trellised vines in wrmer climtic regions. All the climtic regions nd vine structure tretments, however, resulted in wines with good sensory qulity. In terms of enologicl techniques, pumping-over, s opposed to punching-down nd rotor tretments, is not recommended s juice/skin mixing technique, due to reduced wine TAC, colour nd sensory qulity. Pre-fermenttion mcertion, ddition of commercil tnnin preprtions, nd ok mturtion using trditionl nd lterntive tretments, resulted in improved sensory qulity, but with no chnge in wine TAC. However, optimistion of the tnnin ddition protocol my result in incresed wine TAC if dditions re mde fter fermenttion or higher dosges re used. Oxygention of Pinotge wine needs further investigtion to optimise the protocol, s improvements to the wine colour nd fullness were observed for some tretments, but loss of sensory qulity nd TAC were observed in most cses.

UITTREKSEL iv Die doel vn die studie ws om riglyne n die Suid-Afriknse wynbedryf te verskf vir die produksie vn Pinotge wyne met optimle totle ntioksidntkpsiteit (TAK), mr met die behoud vn sensoriese kwliteit. Die bydre wt individuele fenoliese verbindings tot die wyn-tak lewer, is belngrik in hierdie opsig. Die wyn- TAK is gemeet met die 2,2 -sino-di(3-etielbensotisoliensulfoonsuur)-rdiklktioon (ABTS + ) blussingstoets. Die bydre wt individuele fenoliese verbindings tot die wyn-tak lewer, is bereken deur hul konsentrsies in die wyn en die Trolox ekwivlente ntioksidntkpsiteit (TEAK) vn suiwer fenoliese stndrde in g te neem. Die effek vn klimtre, stokontwikkeling, wynmktegnieke (dopkontk voor gisting, sp/dop vermenging, byvoeging vn kommersiële tnniene, verlengde dopkontk) en veroudering (eikevte, lterntiewe eikeprodukte, oksigensie) op die fenoliese smestelling, TAK en sensoriese kwliteit vn Pinotge wyne, is gevolglik ondersoek. Die TEAK-wrdes vn kwersitien-3-glktosied, isorhmnetien en peonidien-3-glukosied is vir die eerste keer bepl. Die TEAK-wrdes vir meeste verbindings ws heelwt ler s gepubliseerde wrdes, mr die TEAK-wrdes vir gllussuur, kftrsuur, kfeësuur en kempferol ws soortgelyk n dié wt voorheen gerpporteer is. Kftrsuur en mlvidien-3-glukosied het die grootste bydres tot die wyn-tak gelewer. Die inhoud vn monomeriese fenoliese verbindings en prosinidien B1 het egter slegs n klein hoeveelheid (tussen 11 en 24%) vn die wyn-tak verklr, terwyl die oorblywende TAK n oligomeriese en polimeriese fenoliese verbindings en onbekende verbindings toegeskryf kn word. n Mte vn sinergie tussen verskillende monomeriese fenoliese verbindings is ook ngetoon. Al die wingerd- en wynboukundige fktore wt ondersoek is, het die fenoliese smestelling vn Pinotge wyne beïnvloed, terwyl die wyn-tak slegs deur sommige behndelings beïnvloed is. Vernderinge in wyn-tak kon nie ltyd deur verskille in fenoliese smestelling verklr word nie, omdt die bydre vn oligomeriese en polimeriese fenoliese verbindings en onbekende verbindings lewer nie bepl kon word nie, mr n groot rol kn speel. Verskille in wynkleur ws ook moeilik om te verklr weens die groot ntl fktore wt n rol speel, en die donker wynkleur wt objektiewe meting bemoeilik. Vitisien A, n ornje-rooi pirno-ntosinien, se vlkke is deurgns verhoog deur middel vn dopkontk voor gisting, en het die wynkleur vn geoksigeneerde wyne beïnvloed. n Hoër wyn-tak is wrgeneem wr Pinotge druiwe op bosstokke in koeler klimtsres verbou is teenoor verbouing op opgeleide stokke in wrmer res. Al die klimtsres en stokontwikkelingsbehndelings het egter wyne met goeie sensoriese kwliteit gelewer. Betreffende wynmktegnieke, word oorpomp- teenoor deurdruk- en rotorbehndelings, nie nbeveel sp/dop mengtegniek nie, ngesien dit die TAK, kleur en sensoriese kwliteit vn die wyn verlg het. Dopkontk voor gisting, byvoeging vn kommersiële tnniene, en veroudering met behulp vn eikehout (trdisionele en lterntiewe eikeprodukte) het wyne met verbeterde sensoriese kwliteit gelewer, mr geen verndering in wyn-tak is wrgeneem nie. Optimering vn die tnnienbyvoegingsprotokol behoort verhoogde wyn-tak tot gevolg te hê s die byvoegings n gisting of in hoër dosisse gemk word. Die oksigensieprotokol vir Pinotge wyn benodig verdere optimering, ngesien verbeterings in die wynkleur en volheid vir sommige behndelings wrgeneem is, mr n verlies n sensoriese kwliteit in meeste gevlle ngetref is.

ACKNOWLEDGEMENTS My sincere thnks nd pprecition to the following people nd institutions: v Dr Elizbeth Joubert of the Division Post-Hrvest nd Processing Technology, ARC Infruitec- Nietvoorbij, s study leder, for her excellent guidnce nd inspirtion during the course of my studies, s well s her support in the execution of this study nd her help during the preprtion of this mnuscript; Dr Johnn Mris of the Division Post-Hrvest nd Processing Technology, ARC Infruitec- Nietvoorbij, s study leder, for his excellent guidnce nd positive criticism during the execution of this study nd the preprtion of this mnuscript; Dr Mren Mnley of the Deprtment of Food Science, University of Stellenbosch, for her willing ssistnce nd her positive criticism during the preprtion of this mnuscript; ARC Infruitec-Nietvoorbij, Winetech nd Technology nd Humn Resources for Industry Progrmme (THRIP) for funding; The Ntionl Reserch Foundtion (NRF) for reserch funding. Opinions expressed nd conclusions rrived t, re those of the uthor nd re not necessrily to be ttributed to the NRF; Frikkie Clitz of the Division Agrimetry, ARC Infruitec- Nietvoorbij, for sttisticl nlyses of dt nd interprettion of results; Dr Johnn Mris, Mnr Dnie Vn Schlkwyk nd Mnr Bertus Fourie for kindly helping to obtin experimentl smples of Pinotge wines; Adele Louw nd the stff t the experimentl winery of ARC Infruitec-Nietvoorbij, for preprtion of experimentl wines for this study; Members of the Pinotge tsting pnel for expert ssistnce with sensory evlution of wines; Hester Redelinghuys, Crin De Wet, Dr Chris Hnsmnn, George Dico nd the rest of the stff t the Division Post-Hrvest nd Processing Technology, ARC Infruitec-Nietvoorbij, for friendly ssistnce with experimentl work nd equipment; Frncois October for help with sensory evlution nd oxygention tretments; Mnd Rossouw for ssistnce with HPLC methodology; Stff of the Deprtment Food Science t the University of Stellenbosch for dvice nd ssistnce; Steve Kroon, fmily nd friends for morl support during the execution of this reserch study nd preprtion of this mnuscript.

CONTENTS vi Abstrct Uittreksel Acknowledgements Chpter 1: Introduction References Chpter 2: Phenolic Compounds in Red Wine - Role in Antioxidnt Cpcity nd Qulity, Fctors Influencing Composition, nd Quntittion Introduction Phenolic compounds Non-flvonoids Flvonoids Chemicl Rections of Phenolic Compounds in Red Wines Enzymtic nd Non-enzymtic Oxidtion Rections Formtion of Pyrnonthocynins Aldehyde-medited Condenstion Rections Involving Anthocynins nd Flvn-3-ols Direct Condenstion Rections Involving Anthocynins nd Flvn-3-ols Co-pigmenttion Role of Phenolic Compounds in Red Wine Qulity Colour Sensory Perception Influence of Viticulturl Fctors on the Phenolic Composition Climtic Conditions Sunlight Exposure Cnopy Mngement Influence of Vinifiction Techniques on Phenolic Composition Pre-fermenttion Mcertion t Low Tempertures Juice/Skin Mixing Techniques During Mcertion Other Fctors nd Techniques Affecting Extrction of Phenolic Compounds During Mcertion Influence of Mturtion on Phenolic Composition Ok Mturtion Oxygention of Wine Phenolic Composition Determintion Spectrophotometric Assys HPLC Methods Antioxidnt Cpcity Determintion Antioxidnt Cpcity Assys Methodologicl Aspects of ABTS Rdicl Ction Scvenging Assy Structure-ctivity Reltionships of Phenolic Compounds iii iv v 1 3 6 6 6 7 8 10 10 13 14 14 16 17 17 18 19 19 20 21 22 22 24 24 26 26 28 29 29 31 31 31 33 35

vii Conclusions References Chpter 3: Unrvelling the Totl Antioxidnt Cpcity of Pinotge Wines - Contribution of Phenolic Compounds Abstrct Introduction Mterils nd Methods Wines Chemicls nd Phenolic Reference Stndrds HPLC Anlysis of Phenolic Composition ABTS Rdicl Ction Scvenging Assy Ultrfiltrtion of Wines Synergy between Phenolic Compounds Sttisticl Anlysis Results nd Discussion Phenolic Composition of Wines TEAC of Phenolic Reference Stndrds Contribution of Individul Phenolic Compounds to TAC of Wine Ultrfiltrtion Synergy nd Unknown Compounds Affecting TAC Conclusions References Chpter 4: Climtic Region nd Vine Structure - Effect on Pinotge Wine Phenolic Composition, Totl Antioxidnt Cpcity nd Colour Abstrct Introduction Mterils nd Methods Viticulturl Tretments nd Wine-mking Procedure Chemicls nd Phenolic Reference Stndrds Spectrophotometric Anlysis of Phenolic Composition HPLC Anlysis of Phenolic Composition ABTS Rdicl Ction Scvenging Assy Objective Colour Prmeters Sttisticl Anlysis Results Vintge-relted Vritions Climtic Region x Vine Structure Tretment Interction Climtic Region: Grpe Sugr Content Climtic Region: Effect on Phenolic Composition Climtic Region: Effect on Antioxidnt Cpcity 35 37 51 51 51 53 53 53 54 55 56 56 57 57 57 57 62 62 67 69 70 73 73 73 74 74 75 76 76 77 77 77 78 78 82 84 84 89

viii Climtic Region: Effect on Objective Colour Prmeters Climtic Region: Discriminnt Anlysis Vine Structure: Grpe Sugr Content Vine Structure: Effect on Phenolic Composition Vine Structure: Effect on Antioxidnt Cpcity Vine Structure: Effect on Objective Colour Prmeters Vine Structure: Discriminnt Anlysis Discussion Effect of Grpe Mturity Vintge-relted Vritions Effect of Climtic Region Effect of Vine Structure Conclusions References 90 92 92 92 95 95 95 96 98 98 100 102 105 105 Chpter 5: Mcertion Before nd During Fermenttion - Effect on Pinotge Wine Phenolic Composition, Totl Antioxidnt Cpcity nd Objective Colour Prmeters Abstrct Introduction Mterils nd Methods Pre-fermenttion Mcertion Juice/Skin Mixing during Mcertion Chemicls nd Phenolic Reference Stndrds Spectrophotometric Anlysis of Phenolic Composition HPLC Anlysis of Phenolic Composition ABTS Rdicl Ction Scvenging Assy Objective Colour Prmeters Sttisticl Anlysis Results Vintge-relted Vrition Pre-fermenttion Mcertion: Effect on Phenolic Composition Pre-fermenttion Mcertion: Effect on Antioxidnt Cpcity Pre-fermenttion Mcertion: Effect on Objective Colour Prmeters Juice/Skin Mixing during Mcertion: Effect on Phenolic Composition Juice/Skin Mixing during Mcertion: Effect on Antioxidnt Cpcity Juice/Skin Mixing during Mcertion: Objective Colour Prmeters Discussion Vintge-relted Vrition Pre-fermenttion Mcertion Juice/Skin Mixing during Mcertion 108 108 108 110 110 110 111 111 111 111 111 112 112 112 115 118 120 120 126 126 129 129 130 132 134 134

ix Conclusions References Chpter 6: Use of Ok nd Oxygention during Mturtion - Effect on Pinotge Wine Phenolic Composition, Totl Antioxidnt Cpcity nd Colour Abstrct Introduction Mterils nd Methods Ok Mturtion Tretments Oxygention Tretments Chemicls nd Phenolic Reference Stndrds Spectrophotometric Anlysis of Phenolic Content HPLC Anlysis of Phenolic Composition ABTS Rdicl Ction Scvenging Assy Objective Colour Prmeters Sensory Anlysis Sttisticl Anlysis Results Ok Mturtion: Effect on Phenolic Composition Ok Mturtion: Effect on Antioxidnt Cpcity Ok Mturtion: Effect on Objective Colour Prmeters Oxygention: Effect on Phenolic Composition Oxygention: Effect on Antioxidnt Cpcity Oxygention: Effect on Objective Colour Prmeters nd Sensory Qulity Discussion Ok Mturtion Oxygention Conclusions References Chpter 7: Effect of Vrious Tretments During nd After Fermenttion on Pinotge Wine Phenolic Composition, Totl Antioxidnt Cpcity, Objective Colour nd Sensory Qulity Abstrct Introduction Mterils nd Methods Wine-mking Procedures Chemicls nd Phenolic Reference Stndrds Spectrophotometric Anlysis of Phenolic Content HPLC Anlysis of Phenolic Composition ABTS Rdicl Ction Scvenging Assy Objective Colour Prmeters 134 134 137 137 137 139 139 139 141 141 141 141 142 142 142 142 142 148 148 149 158 158 160 160 165 168 168 172 172 172 174 174 176 176 176 176 176

x Sensory Anlysis Sttisticl Anlysis Results Phenolic Composition Antioxidnt Cpcity Objective Colour Prmeters nd Sensory Qulity Discussion Conclusions References Chpter 8: Generl Discussion nd Conclusions References Addendum A: Trolox Equivlent Antioxidnt Cpcity (TEAC) of Phenolic Compounds - Effect of Solvent Composition Addendum B: Ultrfiltrtion Protocol Testing Addendum C: Chpter 4 Dt Addendum D: Chpter 5 Dt Addendum E: Chpter 6 Dt Addenddum F: Chpter 7 Dt 176 177 177 177 180 180 180 185 185 187 194 196 201 204 218 226 238 The lnguge nd style in this disserttion re in ccordnce with the requirements of the Journl of Agriculturl nd Food Chemistry. This disserttion represents compiltion of mnuscripts where ech chpter is n individul entity nd some repetition between chpters hs, therefore, been unvoidble.

LIST OF TABLES xi Chpter 2 Tble 1. Spectrl properties of phenolic groups found in red grpes nd wine. Tble 2. HPLC methods for quntifiction of phenolic compounds in red wines using reversed-phse HPLC with UV/visible diode rry detection. Tble 3. ABTS rdicl scvenging ssy protocols used for the determintion of Trolox equivlent ntioxidnt cpcity of phenolic compounds. Tble 4. Trolox equivlent ntioxidnt cpcity (TEAC) vlues of wine phenolic compounds. Chpter 3 Tble 1. Content nd totl ntioxidnt cpcity contribution of individul phenolic compounds in Pinotge wines. Tble 2. Phenolic composition nd TAC vlues of ultrfiltrtion permetes nd retenttes of Pinotge wine A. Tble 3. Phenolic composition nd TAC vlues of ultrfiltrtion permetes nd retenttes of Pinotge wine B. Tble 4. Averge phenolic composition nd totl ntioxidnt cpcity of ultrfiltrtion permetes nd retenttes of two Pinotge wines. Tble 5. Mixtures of phenolic compounds in typicl concentrtions found in Pinotge wines tested for synergy. Chpter 4 Tble 1. Vintge-relted vrition in sugr content of grpes, s well s the phenolic composition (mesured spectrophotometriclly), ntioxidnt cpcity nd objective colour prmeters of the 2001, 2002 nd 2003 Pinotge wines. Tble 2. Vintge-relted vrition in phenolic composition (mesured by HPLC) of the 2002 nd 2003 Pinotge wines. Tble 3. Correltions between phenolic group content nd totl ntioxidnt cpcity of the 2001, 2002 nd 2003 Pinotge wines. Tble 4. Interction of climtic region nd vine structure system with regrd to phenolic composition of the 2002 nd 2003 Pinotge wines. Tble 5. Sugr content of grpes nd phenolic composition (mesured spectrophotometriclly) of the 2001, 2002 nd 2003 Pinotge wines. Tble 6. Anthocynin content of the 2002 nd 2003 Pinotge wines. Tble 7. Flvonol content of the 2002 nd 2003 Pinotge wines. Tble 8. Phenolic cid, flvn-3-ol nd polymer content of the 2002 nd 2003 Pinotge wines. Tble 9. Antioxidnt cpcity nd objective colour prmeters of the 2001, 2002 nd 2003 Pinotge wines. Tble 10. Effect of cordon height on the phenolic composition of the 2002 nd 2003 Pinotge wines. 30 32 34 36 60 63 64 65 68 79 80 81 83 85 86 87 88 91 94

xii Chpter 5 Tble 1. Vintge-relted vrition in phenolic composition (mesured spectrophotometriclly), ntioxidnt cpcity nd objective colour prmeters of the 2000, 2001 nd 2002 Pinotge wines. Tble 2. Vintge-relted vrition in phenolic composition (mesured by HPLC) of the 2001 nd 2002 Pinotge wines. Tble 3. Phenolic composition (mesured spectrophotometriclly) of the 2000, 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols nd mixing frequencies. Tble 4. Anthocynin nd flvonol content of the 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols nd mixing frequencies. Tble 5. Phenolic cid, flvn-3-ol, non-coloured polymer nd totl monomer contents of the 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols nd mixing frequencies. Tble 6. Antioxidnt cpcity nd objective colour prmeters of the 2000, 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols nd mixing frequencies. Chpter 7 Tble 1. Totl ntioxidnt cpcity (TAC) nd totl phenol content (TP) of ok nd grpe tnnin extrct nd their respective theoreticl contribution to the wine TAC nd TP content. Addendum A Tble 1. Trolox equivlent ntioxidnt cpcity vlues of phenolic compounds with test solution prepred using different solvents. Addendum B Tble 1. Phenolic composition nd tnnin content of ultrfiltrtion permetes nd retenttes using protocol 1. Tble 2. Phenolic composition nd tnnin content of ultrfiltrtion permetes nd retenttes using protocol 2. Tble 3. Phenolic composition nd tnnin content of ultrfiltrtion permetes nd retenttes using protocol 3. Addendum C Tble 1. Phenolic composition (mesured using spectrophotometric ssys), ntioxidnt cpcity (mesured) nd objective colour prmeters of the 2001, 2002 nd 2003 individul Pinotge wines from different climtic regions nd vine structure tretments. Tble 2. Anthocynin nd flvonol content (mesured using HPLC) of the 2001, 2002 nd 2003 individul Pinotge wines from different climtic regions nd vine structure tretments. Tble 3. Phenolic cid, flvn-3-ol nd totl monomer content (mesured using HPLC), nd ntioxidnt cpcity (clculted nd remining) of the 2001, 2002 nd 2003 individul Pinotge wines from different climtic regions nd vine structure tretments. Addendum D Tble 1. Phenolic composition (mesured spectrophotometriclly) of the 2000, 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Tble 2. Anthocynin content of the 2001 nd 2002 Pinotge wines produced ccording to different prefermenttion mcertion protocols. 113 114 122 123 124 127 180 199 202 202 202 205 210 214 219 220

xiii Tble 3. Flvonol content of the 2001 nd 2002 Pinotge wines produced ccording to different prefermenttion mcertion protocols. Tble 4. Phenolic cid, flvn-3-ol, non-coloured polymer nd totl monomer content of the 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Tble 5. Antioxidnt cpcity of the 2000, 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Tble 6. Objective colour prmeters of the 2000, 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Tble 7. Anthocynin content of the 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols nd frequencies. Addendum E Tble 1. Effect of ok mturtion on the nthocynin content of Pinotge wines. Tble 2. Effect of ok mturtion on the flvonol content of Pinotge wines. Tble 3. Effect of ok mturtion on the phenolic cid content of Pinotge wines. Tble 4. Effect of ok mturtion on the flvn-3-ol, totl monomer nd totl phenol content of Pinotge wines. Tble 5. Effect of ok mturtion on the ntioxidnt cpcity nd objective colour prmeters of Pinotge wines. Tble 6. Effect of oxygention on the phenolic composition (mesured spectrophotometriclly) of Pinotge wines. Tble 7. Effect of oxygention on the nthocynin content of Pinotge wines. Tble 8. Effect of oxygention on the flvonol content of Pinotge wines. Tble 9. Effect of oxygention on the phenolic cid, flvn-3-ol, non-coloured polymer nd totl monomer content of Pinotge wines. Tble 10. Effect of oxygention on the ntioxidnt cpcity of Pinotge wines. Tble 11. Effect of oxygention on the objective colour prmeters nd sensory qulity of Pinotge wines. Addendum F Tble 1. Phenolic composition (mesured spectrophotometriclly), ntioxidnt cpcity, objective colour prmeters nd sensory qulity of Pinotge wines produced ccording to different enologicl protocols. Tble 2. Anthocynin nd flvonol content of Pinotge wines produced ccording to different enologicl protocols. Tble 3. Phenolic cid, flvn-3-ol nd non-coloured polymer content of Pinotge wines produced ccording to different enologicl protocols. 221 222 223 224 225 226 228 229 230 231 232 233 234 235 236 237 239 240 241

LIST OF FIGURES xiv Chpter 2 Figure 1. Structures of the most common non-flvonoids found in red grpes nd wine. Figure 2. Structures of the most common flvonoids found in red grpes nd wine. Figure 3. Equilibri mong the vrious forms of nthocynins in wine s ffected by ph nd sulphur dioxide. Figure 4. Flvonoid rective sites (dpted from Fulcrnd et l., 2004). Figure 5. Coupled oxidtion of cffeoyltrtric cid nd glutthione producing 2-S-glutthionylcffeoyltrtric cid. Figure 6. Formtion of pyrnonthocynins. Figure 7. Vrious types of ldehyde-medited condenstion products. Figure 8. Vrious types of direct condenstion products. Figure 9. Schemtic of co-pigmenttion complex for delphinidin-3-glucoside nd quercetin. Figure 10. Division of Western Cpe Pinotge cultivtion res into climtic regions on the bsis of men Februry tempertures (MFT) s described by De Villiers et l. (1996). Figure 11. Flow digrm of the most common vinifiction processes in the production of red wine. Chpter 3 Figure 1. Typicl HPLC chromtogrms of Pinotge wine recorded t 280, 316, 360 nd 520 nm. Figure 2. Structures of phenolic compounds identified nd quntified in Pinotge wines. Figure 3. Scheme of totl ntioxidnt cpcity (TAC) contribution of different wine frctions to the totl ntioxidnt cpcity (TAC) of wine. Chpter 4 Figure 1. Correltion of totl phenol content with mesured totl ntioxidnt cpcity (TAC M ) for Pinotge wines. Figure 2. Crtesin plot of L* vlues ginst C* (chrom), * (red/green) nd b* (yellow/blue) vlues for ll Pinotge wines. Figure 3. Phenolic group contributions to the clculted totl ntioxidnt cpcity (TAC CAL ) of wines from different climtic regions (s described in the mterils nd methods). Figure 4. Distribution of the 2002 Pinotge wines in the plne defined by the first two discriminnt functions ccording to climtic regions (s described in the mterils nd methods). Figure 5. Distribution of the 2003 Pinotge wines in the plne defined by the first two discriminnt functions ccording to climtic regions (s described in the mterils nd methods). Figure 6. Phenolic group contributions to the clculted totl ntioxidnt cpcity (TAC CAL ) of wines from different vine structure tretments. Figure 7. Distribution of the 2002 Pinotge wines in the plne defined by the first two discriminnt functions ccording to vine structure tretment. Figure 8. Distribution of the 2003 Pinotge wines in the plne defined by the first two discriminnt functions ccording to vine structure tretment. 7 8 9 11 12 13 15 16 17 20 23 58 59 68 81 83 90 93 93 96 97 97

xv Chpter 5 Figure 1. The percentge differences in phenolic content s mesured spectrophotometriclly, totl ntioxidnt cpcity nd objective colour between the 2000, 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols compred to the control wine. Figure 2. The percentge differences in phenolic cid, flvn-3-ol, non-coloured polymer nd totl monomer content between the 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols nd the control wine. Figure 3. The percentge differences in monomeric nthocynin nd flvonol content between the 2001 nd 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols nd the control wine. Figure 4. Clculted totl ntioxidnt cpcity contributions of phenolic groups for the 2001 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Figure 5. Clculted totl ntioxidnt cpcity contributions of phenolic groups for the 2002 Pinotge wines produced ccording to different pre-fermenttion mcertion protocols. Figure 6. Clculted totl ntioxidnt cpcity contributions of phenolic groups for the 2001 nd 2002 Pinotge wines produced ccording to different juice/skin mixing protocols. Figure 7. Clculted totl ntioxidnt cpcity contributions of phenolic groups for the 2001 nd 2002 Pinotge wines produced with different juice/skin mixing tretment frequencies. Chpter 6 Figure 1. Scheme of oxygention tretments of Pinotge wine. Figure 2. The percentge differences in monomeric nthocynin content between Pinotge wines mtured for 28 weeks using different ok tretments nd the control wine. Figure 3. Effect of ok mturtion on the monomeric nthocynin content (HPLC) of Pinotge wines. Figure 4. The percentge differences in flvonol nd phenolic cid content between Pinotge wines mtured for 28 weeks using different ok tretments nd the control wine. Figure 5. The percentge differences in flvn-3-ol, totl monomer nd totl phenol content between Pinotge wines mtured for 28 weeks using different ok tretments nd the control wine. Figure 6. Effect of ok mturtion on the mesured totl ntioxidnt cpcity (TAC M ) of Pinotge wines. Figure 7. The percentge differences in totl ntioxidnt cpcity nd colour, between Pinotge wines mtured for 28 weeks using different ok tretments nd the control wine. Figure 8. Clculted totl ntioxidnt cpcity contributions of phenolic groups for different ok mturtion tretments. Figure 9. Effect of ok mturtion on the chrom (C*) of Pinotge wines. Figure 10. Effect of ok mturtion on the * (red/green chromticity) of Pinotge wines. Figure 11. Effect of ok mturtion on the hue ngle (h*) of Pinotge wines. Figure 12. Effect of ok mturtion on the lightness (L*) vlue of Pinotge wines. Figure 13. Effect of oxygention on the phenolic composition of Pinotge wine, mesured with spectrophotometric ssys nd HPLC. Figure 14. Effect of oxygention on the totl ntioxidnt cpcity nd colour of Pinotge wine. Figure 15. Clculted totl ntioxidnt cpcity contributions of phenolic groups for different oxygention tretments 116 117 119 121 121 128 128 140 143 145 146 147 149 150 151 151 152 152 153 154 159 160

xvi Figure 16. Effect of oxygention on the sensory qulity of Pinotge wine. 161 Chpter 7 Figure 1. Digrm of enologicl tretments pplied during nd fter fermenttion. Figure 2. The percentge differences in nthocynin content between Pinotge wines produced ccording to different enologicl protocols compred to the control wine. Figure 3. The percentge differences in flvonol, flvn-3-ol, polymer nd totl phenol content between Pinotge wines produced ccording to different enologicl protocols compred to the control wine. Figure 4. The percentge differences in totl ntioxidnt cpcity, colour nd sensory qulity between Pinotge wines produced ccording to different enologicl protocols compred to the control wine. Addendum A Figure 1. Dose-response curves of ABTS rdicl ction scvenging ctivity of Trolox test solutions in different solvents. Figure 2. Dose-response curves of ABTS rdicl ction scvenging ctivity of gllic cid test solutions in different solvents. 175 178 179 181 198 198

1 Chpter 1: Introduction The ntioxidnt cpcity of foods nd wine is receiving considerble scientific nd commercil interest. As consumers hve become more wre of the helth benefits of ntioxidnts vi the minstrem medi, the food nd beverge industries hve recognised new mrketing opportunities for their products. The ntioxidnt cpcity of foods nd beverges my therefore become n importnt qulity prmeter, especilly in niche mrkets concerned with helth benefits. In this regrd, the South Africn wine industry by mens of Winetech, their reserch funding orgnistion, hs expressed interest in the potentil of wine ntioxidnt cpcity in mrketing. Red wine received prominent press coverge in recent yers s result of the French Prdox. This phenomenon is the reltively low level of coronry hert disese incidence, despite the high intke of sturted ft observed in the French popultion. This phenomenon is ssocited with the consumption of red wine (Renud & De Lorgeril, 1992). Phenolic compounds in red wine exhibit brod spectrum of beneficil phrmcologicl properties, believed to be relted to their ntioxidtive properties (Kinsell et l., 1993). Anti-therogenic ctivity (Stocker & O Hllorn, 2004), nti-tumour ctivity (Clifford et l., 1996), nti-ulcer ctivity (Sito et l., 1998), regultion of pltelet ggregtion (Keevil et l., 2000) nd nti-inflmmtory ctivity (Estruch et l., 2004) hve ll been demonstrted by the consumption of red wine nd/or red wine phenolic compounds. These properties re thought to contribute to the prevention or llevition of coronry hert disese, cncer nd geing. French scientists recently produced n ntioxidnt-enriched Chrdonny wine, which showed ntioxidnt cpcity levels much higher thn tht of other French white wines, nd closer to tht of French red wines (Lndrult et l., 2003). The purpose of their study ws to obtin white wine with similr helth benefits thn those of red wines. Enhncement of the phenolic content, nd hence the ntioxidnt cpcity, of South Africn wines, while retining sensory qulity, will provide locl winemkers with n opportunity to ensure greter interntionl mrket shre (Beyers Truter, Beyerskloof Wines, personl communiction, 2004). A first step in this direction ws preliminry study of the free rdicl scvenging nd lipid peroxidtion inhibitory ctivity of the mjor commercil South Africn red nd white cultivr wines, ws crried out to identify cultivr wines with the highest ntioxidnt potentil (De Beer et l., 2003; 2005). The results demonstrted tht the unique South Africn cultivr, Pinotge, produces wines with good free rdicl scvenging nd lipid peroxidtion inhibitory ctivity, with similr ctivity to Cbernet Suvignon wines. Since Pinotge is the focus of incresing interest in interntionl mrkets, further study of its ntioxidnt cpcity nd the role of different phenolic compounds in these wines, ws merited.

Pinotge is cross between Vitis vinifer L. cv. Pinot noir nd Cinsut (Hermitge), developed by Professor Abrhm Perold in 1924 in South Afric. Wines of this cultivr re typiclly fruity, with berry, plum nd bnn chrcters, nd cn be mde for erly drinking or geing. Pinotge grpes represented 20% of the totl red grpes crushed for the production of South Africn red wine in 2004, long with roughly the sme mount of Cbernet Suvignon nd Shirz grpes (Anonymous, 2005). Reserch on Pinotge wine hs minly focused on determining cultivr impct rom compounds (Vn Wyk et l., 1979; Wldner & Mris, 2002) nd improving grpe nd wine qulity (Mris, 2003, 2003b; Mris, 2004). The unique phenolic composition of Pinotge hs received ttention recently: Schwrz et l. (2003) isolted new nthocynin-derived pigment similr to vitisin A, which they nmed pinotin A (fter the Pinotge cultivr). The high cffeic nd/or cffeoyltrtric cid content of Pinotge wine further highlight its unique phenolic profile (Rossouw & Mris, 2003, 2004; Schwrz et l., 2004). Mny viticulturl fctors cn ffect the phenolic content of red wines. Mcro-climtic fctors, such s temperture, sunlight rdition, rinfll nd wind, hve gret effect on grpevine physiology nd the biosynthesis of phenolic compounds (Mullins et l., 1992). In the vineyrd, the micro-climte of the grpevine cn be djusted by cnopy mngement techniques, which im to optimise berry temperture nd sunlight exposure (Smrt & Robinson, 1991). The effect of mcroclimtic fctors nd soil chrcteristics on phenolic content nd qulity of vrious wines hs been studied (Kliewer, 1970; Mteus et l., 2001). Interctions between these fctors re very complex culminting in the concept of terroir, which hs been studied (e.g. Vivs de Gulejc et l., 2001; Vn Leewen et l., 2004), lthough mny spects still need further elucidtion. In ddition, the effects of grpevine trining systems on Shirz nd Cbernet Suvignon wines (Wolf et l., 2003; Vnden Heuvel et l., 2004) nd micro-climte on Shirz wines (Bergqvist et l., 2001; Downey et l., 2004) hve lso been studied. However, the effects of climtic fctors nd trining systems on wine ntioxidnt cpcity hve not yet been studied to the best of the uthor s knowledge. Mny developments hve tken plce in terms of enologicl prctices in recent yers. Enologicl techniques such s enzyme tretments (Prdo et l., 1999; Butist-Ortín et l., 2005), use of commercil tnnins (Butist-Ortín et l., 2005), use of lterntive ok sources (Del Almo Snz & Domínguez, 2006), nd micro-oxygention (Cstellri et l., 1998), hve become commonplce in modern wineries. The use of some of these techniques is minly bsed on trilnd error, nd more reserch is needed to clrify the effects of these tretments on the phenolic composition nd qulity of red wines. Although very few studies hve ddressed the effect of these tretments on the ntioxidnt cpcity of red wines thus fr, Del Álmo et l. (2006) observed n initil increse in the redox potentil of red wines over the first three months of ok mturtion, followed by decrese for t lest the next eight months. 2

Antioxidnt ctivity cn be mesured by lrge vriety of ssys. Rdicl scvenging ctivity is n importnt spect of ntioxidnt ctivity, lthough the in vitro rdicl scvenging ctivity of wine components does not necessrily coincide with in vivo ntioxidnt ctivity, s biovilbility, metl chelting properties, lipid phse prtitioning, nd metbolism of individul wine components my differ considerbly (Astley, 2003). On the other hnd, for development purposes, n esy nd rpid screening method is required. The 2,2 -zino-di(3-ethylbenzo-thizoline-sulphonic cid rdicl ction (ABTS + ) scvenging ssy is such method (Re et l., 1999) nd hs lredy been used extensively for nlysis of food, beverges nd plnt extrcts. Furthermore, the use of the ABTS + scvenging ssy will lso permit comprison of results with those obtined in previous study on South Africn wines, including Pinotge wines (De Beer et l., 2003). These resons determined the choice of the ABTS + scvenging ssy for use during this study. Prior et l. (2005) lso proposed tht the ABTS + scvenging ssy be stndrdised for high-throughput screening of smples, long with the oxygen rdicl ntioxidnt cpcity (ORAC) nd Folin-Cioclteu ssys. The first objective of this study ws to investigte the contributions of individul phenolic compounds to the totl ntioxidnt cpcity of Pinotge wines. In this regrd, Trolox equivlent ntioxidnt cpcity vlues of individul phenolic compounds nd the Pinotge wines were determined by the ABTS + scvenging ssy. The phenolic composition of the wines ws determined by HPLC. The second objective ws to determine the effect of climtic region, vine structure, enologicl techniques (pre-fermenttion mcertion, juice/skin mixing, ddition of commercil tnnins, extended mcertion) nd mturtion (ok brrels, lterntive ok products, oxygention) on the phenolic composition, totl ntioxidnt cpcity nd sensory qulity of Pinotge wines. The knowledge gined in this study will provide the South Africn wine industry with guidelines for the production of Pinotge wines with optiml totl ntioxidnt cpcity, while retining sensory qulity. 3 References Anonymous. South Africn Wine Industry Sttistics nr. 29; SAWIS, Prl, South Afric, 2005. Astley, S. B. Dietry ntioxidnts - Pst, present nd future? Trends Food Sci. Technol. 2003, 14, 93-98. Butist-Ortín, A. B.; Mrtinez-Cutills, A.; Ros-Grci, J. M.; Lopez-Roc, J. M.; Gomez-Plz, E. Improving colour extrction nd stbility in red wines: The use of mcertion enzymes nd enologicl tnnins. Int. J. Food Sci. Technol. 2005, 40, 1-12. Bergqvist, J.; Dokoozlin, N.; Ebisud, N. Sunlight exposure nd temperture effects on berry growth nd composition of Cbernet Suvignon nd Grenche in the Centrl Sn Joquin Vlley of Cliforni. Am. J. Enol. Vitic. 2001, 52, 1-7. Cstellri, M.; Arfelli, G.; Amti, A. Evolution of phenolic compounds in red winemking s ffected by must oxygention. Am. J. Enol. Vitic. 1998, 49, 91-94.

4 Clifford, A. J.; Ebeler, S. E.; Ebeler, J. D.; Bills, N. D.; Hinrichs, S. H.; Teissedre, P.-L.; Wterhouse, A. L. Delyed tumor onset in trnsgenic mice fed n mino cid-bsed diet supplemented with red wine solids. Am. J. Clin. Nutr. 1996, 64, 748-756. De Beer, D.; Joubert, E.; Gelderblom, W. C. A.; Mnley, M. Antioxidnt ctivity of South Africn red nd white cultivr wines: Free rdicl scvenging. J. Agric. Food Chem. 2003, 51, 902-909. De Beer, D.; Joubert, E.; Gelderblom, W. C. A.; Mnley, M. Antioxidnt ctivity of South Africn red nd white cultivr wines nd selected phenolic compounds: In vitro inhibition of microsoml lipid peroxidtion. Food Chem. 2005, 90, 869-577. Del Álmo, M.; Nevres, I.; Cárcel, L. M. Redox potentil evolution during red wine ging in lterntive systems. Anl. Chim. Act 2006, 563, 223-228. Del Almo Snz, M.; Domínguez, I. N. Wine ging in bottle from rtificil systems (stves nd chips) nd ok woods. Anthocynin composition. Anl. Chim. Act 2006, 563, 255-263. Downey, M. O.; Hrvey, J. S.; Robinson, S. P. The effect of bunch shding on berry development nd flvonoid ccumultion in Shirz grpes. Aust. J. Grpe Wine Res. 2004, 10, 55-73. Estruch, R.; Scnell, E.; Bdi, E.; Antúnez, E.; Nicolás, J. M.; Fernández-Solá, J.; Rotilio, D.; de Getno, G.; Rubin, E.; Urbno-Márquez, A. Different effects of red wine nd gin consumption on inflmmtory biomrkers of therosclerosis: A prospective rndomized crossover tril. Effects of wine on inflmmtory biomrkers. Atherosclerosis 2004, 175, 117-123. Keevil, J. G.; Osmn, H. E.; Reed, J. D.; Folts, J. D. Grpe juice, but not ornge juice or grpefruit juice, inhibits humn pltelet ggregtion. J. Nutr. 2000, 130, 53-56. Kinsell, J. E.; Frnkel, E.; Germn, B.; Knner, J. Possible mechnisms for the protective role of ntioxidnts in wine nd plnt foods. Food Technol. 1993, 47, 85-89. Kliewer, W. M. Effect of dy temperture nd light intensity on colortion of Vitis vinifer L. grpes. J. Am. Soc. Hort. Sci. 1970, 95, 693-697. Lndrult, N.; Poucheret, P.; Azy, J.; Krosnik, M.; Gsc, F.; Jenin, C.; Cros, G.; Teissedre, P.-L. Effect of polyphenol-enriched Chrdonny white wine in dibetic rts. J. Agric. Food Chem. 2003, 51, 311-318. Mris, J. Effect of different wine-mking techniques on the composition nd qulity of Pinotge wine. I. Lowtemperture skin contct prior to fermenttion. S. Afr. J. Enol. Vitic. 2003, 24, 70-75. Mris, J. Effect of different wine-mking techniques on the composition nd qulity of Pinotge wine. II. Juice/skin mixing prctices. S. Afr. J. Enol. Vitic. 2003b, 24, 76-79. Mris, J. The contribution of b-dmscenone to the berry-like rom of red wines. Wynboer 2004, 180, 14-16. Mteus, N.; Mrques, S.; Gonçlves, A. C.; Mchdo, J. M.; De Freits, V. Pronthocynidin composition of red Vitis vinifer vrieties from the Douro vlley during ripening: Influence of cultivtion ltitude. Am. J. Enol. Vitic. 2001, 52, 115-121. Mullins, M. G.; Bouquet, A.; Willims, L. E. Biology of the grpevine. Cmbridge University Press: Cmbridge, UK, 1992. Prdo, F.; Slins, M. R.; Alonso, G. L.; Nvrro, G.; Huert, M. D. Effect of diverse enzyme preprtions on the extrction nd evolution of phenolic compounds in red wines. Food Chem. 1999, 67, 135-142. Prior, R. L.; Wu, X.; Schich, K. Stndrdized methods for the determintion of ntioxidnt cpcity nd phenolics in foods nd dietry supplements. J. Agric. Food Chem. 2005, 53, 4290-4302. Re, R.; Pellegrini, N.; Proteggente, A.; Pnnl, A.; Yng, M.; Rice-Evns, C. A. Antioxidnt ctivity pplying n improved ABTS rdicl ction decoloriztion ssy. Free Rd. Biol. Med. 1999, 26, 1231-1237.

5 Renud, S.; De Lorgeril, M. Wine, lcohol, pltelets, nd the French Prdox for coronry hert disese. Lncet 1992, 339, 1523-1526. Rossouw, M.; Mris, J. Phenolic compounds in South Africn red wines: A preliminry study. Wynboer 2003, 172, 9-11. Rossouw, M.; Mris, J. The phenolic composition of South Africn Pinotge, Shirz nd Cbernet Suvignon wines. S. Afr. J. Enol. Vitic. 2004, 25, 94-104. Sito, M.; Hosoym, H.; Arig, T.; Ktok, S.; Ymji, N. Antiulcer ctivity of grpe seed extrct nd procynidins. J. Agric. Food Chem. 1998, 46, 1460-1464. Schwrz, M.; Jerz, G.; Winterhlter, P. Isoltion nd structure of pinotin A, new nthocynin derivtive from Pinotge wine. Vitis 2003, 42, 105-106. Schwrz, M.; Hofmnn, G.; Winterhlter, P. Investigtions on nthocynins in wines from Vitis vinifer cv. Pinotge: Fctors influencing the formtion of pinotin A nd its correltion with wine ge. J. Agric. Food Chem. 2004, 52, 498-504. Smrt, R.; Robinson, M. Sunlight into wine: hndbook for winegrpe cnopy mngement. Winetitles: Underdle, Austrli, 1991. Stocker, R.; O'Hllorn, R. A. Delcoholized red wine decreses therosclerosis in polipoprotein E gene-deficient mice independently of inhibition of lipid peroxidtion in the rtery wll. Am. J. Clin. Nutr. 2004, 79, 123-130. Vnden Heuvel, J. E.; Proctor, J. T. A.; Sullivn, J. A.; Fischer, K. H. Influence of trining/trellising system nd rootstock selection on productivity nd fruit composition of Chrdonny nd Cbernet frnc grpevines in Ontrio, Cnd. Am. J. Enol. Vitic. 2004, 55, 253-264. Vn Leeuwen, C.; Frint, P.; Choné, X.; Tregot, O.; Koundours, S.; Dubourdieu, D. Influence of climte, soil, nd cultivr on terroir. Am. J. Enol. Vitic. 2004, 55, 207-217. Vn Wyk, C. J.; Augustyn, O. P. H.; De Wet, P.; Joubert, W. A. Isomyl cette - A key fermenttion voltile of wines of Vitis vinifer cv Pinotge. Am. J. Enol. Vitic. 1979, 30, 167-173. Vivs de Gulejc, N.; Nonier, M.-F.; Guerr, C.; Vivs, N. Anthocynin in grpe skins during mturtion of Vitis vinifer L. cv. Cbernet Suvignon nd Merlot Noir from different Bordeux terroirs. J. Int. Sci. Vigne Vin 2001, 35, 149-156. Wldner, M.; Mris, J. Impct rom components in South Africn red wines: A preliminry study. Wynboer 2002, 161, 15-17. Wolf, T. K.; Dry, P. R.; Ilnd, P. G.; Botting, D.; Dick, J.; Kennedy, U.; Ristic, R. Response of Shirz grpevines to five different trining systems in the Bross vlley, Austrli. Aust. J. Grpe Wine Res. 2003, 9, 82-95.

6 Chpter 2: Phenolic Compounds in Red Wine - Role in Antioxidnt Cpcity nd Qulity, Fctors Influencing Composition, nd Quntittion Introduction Phenolic compounds in red wines re responsible for colour nd contribute to stringency nd mouthfeel. This very importnt group of compounds in wine lso hs excellent ntioxidnt properties. In recent yers myrid of studies on the possible helth benefits hs cused greter wreness of ntioxidnts in generl nd phenolic compounds in prticulr mong consumers. The ntioxidnt cpcity of foods nd beverges, especilly red wines, my become n importnt new qulity prmeter. In the light of this phenomenon, the South Africn wine industry is interested in reserch to enhnce the phenolic content nd subsequently the ntioxidnt cpcity of wines with retention of sensory qulity for greter competitiveness in the interntionl nd locl mrket. This review ims to give overview of phenolic compounds found in red wines nd their role in the ntioxidnt nd qulity chrcteristics of red wine. Methods for determining phenolic composition, ntioxidnt cpcity nd qulity of red wines will be discussed. A discussion on the structure-ctivity reltionships of the phenolic compounds in red wines using the 2,2 -zinobis(3- ethylbenzothizoline-6-sulphonic cid (ABTS) rdicl ction scvenging ssy will follow. Finlly, fctors ffecting the phenolic composition nd content of red wines, s investigted in this study, will be covered. Emphsis, in terms of viticulturl spects, will be on climtic conditions nd viticulturl tretments ffecting the micro-climtic conditions. In terms of enologicl fctors the focus will be on mcertion techniques nd fctors ffecting mcertion, s well s mturtion using vrious ok products nd the effect of oxygen ppliction. Phenolic Compounds Phenolic compounds occur in ll fruits s lrge nd complex group of compounds with prticulr importnce regrding sensory chrcteristics (colour, stringency, bitterness nd rom) nd bio-ctivity (ntimicrobil, nti-inflmmtory nd ntioxidnt ctivity) of red wines. An romtic ring bering one or more hydroxyl groups is common structurl feture of this group of compounds. The phenolic compounds in red wines re produced by yest metbolism nd extrcted from the grpe pulp, skin, seeds nd stems during fermenttion, s well s from ok cooperge fter fermenttion (Mcheix et l., 1990). Two distinct groups of phenolic compounds occur in red grpes nd wine, nmely non-flvonoids nd flvonoids. The mjor phenolic compounds occurring

7 O HO R 1 R 3 R 2 Benzoic cids p-hydroxybenzoic cid (R 2 = ; R 1, R 3 = H) Gentisic cid (R 1 = ; R 2, R 3 = H) Protoctechuic cid (R 1, R 2 = ; R 3 = H) Vnillic cid (R 2 = ; R 1 = OCH 3 ; R 3 = H) Syringic cid (R 2 = ; R 1, R 3 = OCH 3 ) Gllic cid (R 1, R 2, R 3 = ) HO Stilbenes trns-resvertrol HO HO HO Tyrosol O O O O Ellgic cid HO HO O R 1 R 2 R 3 Hydroxycinnmic cids p-coumric cid (R 2 = ; R 1, R 3 = H) Cffeic cid (R 1, R 2 = ; R 3 = H) Ferulic cid (R 2 = ; R 1 = OCH 3 ; R 3 = H) Sinpic cid (R 2 = ; R 1, R 3 = OCH 3 ) O O O O R 1 R 2 R 3 Hydroxycinnmic cid derivtives p-coumroyltrtric cid (R 2 = ; R 1, R 3 = H) Cffeoyltrtric cid (R 1, R 2 = ; R 3 = H) Feruloyltrtric cid (R 2 = ; R 1 = OCH 3 ; R 3 = H) Figure 1. Structures of the most common non-flvonoids found in red grpes nd wine. in grpes nd wine will be discussed briefly, focussing specificlly on those in red wine. A detiled review on this subject hs been published recently (Mongs et l., 2005). Non-flvonoids The non-flvonoids occurring in red grpes nd wines comprise the hydroxycinnmic cids, hydroxybenzoic cids, stilbenes, s well s vrious derivtives (Figure 1). The most common nonflvonoids in red wine re the trns-hydroxycinnmic cid derivtives, nmely the trns-isomers of the trtric cid esters of cffeic, ferulic nd p-coumric cid (Ribéreu-Gyon, 1965), which re extrcted from the grpe pulp nd skins. The free forms of these cids generlly do not occur in grpes, but cn be relesed by enzymtic hydrolysis fter crushing of the grpes (Mcheix et l., 1990). The cis-forms of the hydroxycinnmic cids nd their derivtives re only found in smll quntities (Singleton et l., 1978). Benzoic cids usully occur in low quntities in red wine, except for gllic cid, which occurs in lrger quntities. Of the benzoic cids, only gllic cid is extrcted from the grpe pulp. Other benzoic cids found in wines, nmely p-hydroxybenzoic, protoctechuic, vnillic, syringic nd gentisic cid (Mcheix et l., 1990), re extrcted from ok wood or formed during hydrolysis of ok wood hydrolysble tnnins (Singleton et l., 1971; Puech, 1987). Ellgitnnins extrcted from ok wood re hydrolysble tnnins consisting of ellgic cid nd glucose moieties in polymerised form. The stilbenes, of which trns-resvertrol is the most common, lso occur in dimeric nd glycosylted forms nd re extrcted from the grpe skins.

8 HO O H R 1 R 2 Flvn-3-ol monomers (+)-Ctechin (R 1, R 3 = H; R 2 = ) R 3 (-)-Epictechin (R 1 = ; R 2, R 3 = H) (+)-Glloctechin (R 1 = H; R 2, R 3 = ) (-)-Epiglloctechin (R 1, R 3 = ; R 2 = H) HO O O O R 3 R 1 Flvonols Quercetin (R 1 = ; R 2, R 3 = H) R 2 Quercetin-3-glc (R 1 = ; R 2 = H; R 3 = glc) Quercetin-3-gl (R 1 = ; R 2 = H; R 3 = gl) Quercetin-3-rhm (R 1 = ; R 2 = H; R 3 = rhm) Myricetin (R 1, R 2 = ; R 3 = H) Kempferol (R 1, R 2, R 3 = H) Isorhmnetin (R 1 = OCH 3, R 2, R 3 = H) 7 6 8 A 5 O C 4 Bsic flvonoid skeleton with numbering 2' 3 3' B 6' 4' 5' HO HO H H O O O O + R 1 R 2 O + O R 1 Glc R3 R 2 Anthocynins Delphinidin -3-glc (R 1, R 2 = ; R 3 = H) Petunidin-3-glc (R 1 = OCH 3 ; R 2 = H; R 3 = H) Peonidin-3-glc (R 1 = OCH 3 ; R 2 = H; R 3 = H) Mlvidin-3-glc (R 1, R 2 = OCH 3 ; R 3 = H) Flvn-3-ol dimers Procynidin B1 (R 1, R 4 = ; R 2, R 3 = H) For cette esters nd coumrte esters R 3 = cetyl nd coumroyl groups, respectively O R 3 R 4 Procynidin B2 (R 1, R 3 = ; R 2, R 4 = H) Procynidin B3 (R 2, R 4 = ; R 1, R 3 = H) Procynidin B4 (R 2, R 3 = ; R 1, R 4 = H) Figure 2. Structures of the most common flvonoids found in red grpes nd wine. Tyrosol is n exmple of yest-derived non-flvonoid in red wine, which cn contribute to bitterness (Thorngte, 1997). Flvonoids The flvonoids consist of two phenols bridged by three-crbon chin (C 6 -C 3 -C 6 flvone skeleton). The three-crbon bridge is commonly cyclised with oxygen. The most common flvonoids in red wines comprise the nthocynins, flvn-3-ols nd flvonols, which re differentited by the degree of unsturtion nd oxidtion of the three-crbon bridge (Mcheix et l., 1990) (Figure 2). Flvonoids occur in free, polymerised, glycosylted or cylted forms. The