Thesis (MScAgric (Viticulture and Oenology))--University of Stellenbosch, 2010. / ENGLISH ABSTRACT: Previous research has shown that carotenoids are precursors of C13-norisoprenoid aroma compounds in wine. C13-norisoprenoids have low threshold values in wine with the most prominent C13-norisoprenoids being β-damascanone and β-ionone which contribute honey and floral like aroma to wine. Chlorophyll and its derivates have also been detected in wine with potential to be precursors to aroma compounds. Apart from the contribution of these pigments to wine aroma and quality they are vital role players in photosynthesis and are widely found in plants and plant products. The main functions of these pigments in plants are light collection and light-protection. Research has shown that environmental conditions, climate, light exposure of bunches and soil water deficit influence the carotenoid content of grape berries. Furthermore the concentration of carotenoids and chlorophylls has also been shown to differ between cultivars. No research in this regard has been done on Merlot grape berries. With this in mind, the aim of this study was to evaluate the effect of vigour and soil water content on the evolution of carotenoids and chlorophylls through ripening of grape berries from the cv. Merlot. However, when looking at methods to analyse carotenoids and chlorophylls in berry tissue, especially lyophilised tissue, there were no readily available methods. Thus, an extraction method to identify and quantify the carotenoid and chlorophyll profile of lyophilised tissue from unripe (green) to ripe (red) Merlot grape berries was needed. In this study the RPHPLC method of Taylor et al. (2006) for carotenoids and the extraction method of Mendes-Pinto et al. (2004) were adapted to analyse both carotenoids and chlorophylls in lyophilised grape tissue. The RP-HPLC method baseline separated all the carotenoids and chlorophylls and their derivatives. Recovery of standards from mock extractions was high, indicating that the extraction procedure was acceptable. However, extraction recovery tested in the matrix of the grape tissue showed less promising results due to the high acid content of grape tissue. Violaxanthin, neoxanthin and the chlorophylls were especially sensitive to low pH conditions which facilitated their degradation. The degradation products of these compounds under acidic conditions were identified as pheophytin a, b, chlorophillide a, pyropheophytin b, cisviolaxanthin, cis-neoxanthin, neochrome, mutatoxanthin and luteoxanthin. There is a possibility that some degradation products were already present in the tissue due to lyophilisation (since the water in the berry was then removed and the acid concentrated). More work is needed to investigate the effect of lyophilisation and storage on the composition of grape tissue of different maturity. The extraction method for grape berry tissue at different ripening stages should also be optimised further to effectively neutralise tissue acidity, without compromising the extraction of carotenoids significantly, in especially green berry tissue. The question as to whether cisisomers and chlorophyll degradation products are naturally present in grape berries or are formed during sampling and processing remains unanswered in the current study. This study confirmed that in general carotenoids and chlorophylls decrease on a per berry (μg/berry) and concentration (μg/g) basis from veraison to harvest. Furthermore, this study was inconclusive in showing that vigour differences have an effect on the rate of synthesis/degradation of carotenoids, chlorophyll and some other ripening parameters, namely malic acid, total glucose and fructose, total tannin and total anthocyanin, from pre-veraison (pea size) to harvest. Additionally, no significant effect of soil water content on carotenoids, chlorophylls and ripeness parameters was found in this study, most likely due the fact that high soil water capacity was found in lower soil layers which may have prevented significant differences in grapevine water status. Experimental plots selected for vigour differences based on normalised difference vegetation index (NDVI) images, pruning mass and soil water measurements by means of a neutron probe, showed significant differences in soil water content in only the first 30 cm of the soil for the ripening seasons studied. Predawn plant water potential measurements, however, indicated that none of the experimental vines experienced severe water stress which was previously shown to effect carotenoid content of grapes. The carotenoid 5,8-epoxy--carotene was quantified for the first time in grapes and represents a significant amount of the total carotenoids present at harvest. All the carotenoids and chlorophylls except -carotene appeared to be sensitive to seasonal variation in climatic conditions. Lutein and β-carotene were found to be the most abundant carotenoids present in Merlot grape berries together with chlorophyll a for both seasons studied. The values of these carotenoids also correlated well with previous research. However, chlorophyll a was found in much larger quantities in Merlot berries compared to reported data. This is possibly because in this study the chlorophyll degradation products were included in the calculation of chlorophyll a. Multivariate analysis showed promising preliminary prediction models (with correlation values of above 0.8 for both seasons analysed) for the prediction of the concentration of ripeness parameters (glucose, fructose, malic acid, total tannins and anthocyanins) with carotenoid and chlorophyll content. This result highlights the opportunity for the development of a rapid non-destructive method to measure carotenoids and chlorophylls in berries which in turn can predict optimal ripeness. Furthermore, since carotenoids are the precursors to C13- norisoprenoid aroma compounds in wine a preview of the potential contribution of these aromas to wine might be evaluated. Further research is necessary to investigate the possibility of building and validating such models. / AFRIKAANSE OPSOMMING: Vorige navorsing het getoon dat karotenoïede die voorlopers is van C13-norisoprenoïed aromaverbindings in wyn. C13-norisoprenoïede het lae drempelwaardes in wyn, met β- damassenoon en β-jonoon as die prominentste C13-norisoprenoïede wat ‘n bydrae tot die heuning en blomagtige aroma van die wyn maak. Chlorofil en sy derivate is ook reeds in wyn bespeur, met die potensiaal om voorlopers van aromaverbindings te wees. Buiten die bydrae van hierdie pigmente tot wynaroma en -kwaliteit is hulle ook belangrike rolspelers in fotosintese en kom hulle wydverspreid in plante en plantprodukte voor. Die vernaamste funksies van hierdie pigmente in plante is om lig te versamel en om as beskerming teen lig op te tree. Navorsing het getoon dat omgewingstoestande, klimaat, ligblootstelling van die trosse en grondwatertekorte die karotenoïedinhoud van druiwekorrels beïnvloed. Verder is ook getoon dat die konsentrasie van karotenoïede en chlorofille tussen kultivars verskil. Geen navorsing is al in hierdie opsig op Merlot-druiwekorrels gedoen nie. Met hierdie aspek in gedagte was die doelwit van hierdie studie om die effek van groeikrag en grondwaterinhoud op die evolusie van karotenoïede en chlorofille tydens die rypwording van druiwekorrels van die cv. Merlot te evalueer. Wanneer mens egter kyk na die metodes waarvolgens die karotenoïede en chlorofille in korrelweefsel geanaliseer word, is daar geen geredelik beskikbare metodes nie. ‘n Ekstraksiemetode om die karotenoïed- en chlorofilprofiel van geliofiliseerde weefsel van onryp (groen) tot ryp (rooi) Merlot-bessies te identifiseer en kwantifiseer was dus nodig. In hierdie studie is die RP-HPLC metode van Taylor et al. (2006) vir karotenoïede en die ekstraksiemetode van Mendes-Pinto et al. (2004) aangepas om beide karotenoïede en chlorofille in geliofiliseerde druiweweefsel te analiseer. Die basislyn van die RP-HPLC metode het all karotenoïede en chlorofille en hul derivate geskei. Herwinning van die standaarde vanaf skynekstraksies was hoog, wat aandui dat die ekstraksieprosedure aanvaarbaar was. Ekstraksieherwinning wat in die matriks van die druiweweefsel getoets is, het egter minder belowende resultate getoon as gevolg van die hoë suurinhoud van die druifweefsel. Violaxantien, neoxantien en die chlorofille was veral sensitief vir toestande van lae pH, wat hulle afbreking gefasiliteer het. Die afbrekingsprodukte van hierdie verbindings onder suurtoestande is geïdentifiseer as feofitien a en b, chlorofillied a, pirofeofitien b, cis-violaxantien, cis-neoxantien, neochroom, mutatoxantien en luteoxantien. Daar is ‘n moontlikheid dat sommige afbreekprodukte reeds in die weefsel teenwoordig was as gevolg van liofilisering (aangesien die water in die korrel reeds verwyder was en die suur gekonsentreerd was). Meer werk is nodig om die effek van liofilisering en berging op die samestelling van druifweefsel van verskillende rypheid te bepaal. Die ekstraksiemetode vir druifkorrelweefsel op verskillende stadia van rypwording moet ook verder geoptimaliseer word om weefselsuurheid doeltreffend te neutraliseer, sonder om die ekstraksie van karotenoïede noemenswaardig te kompromitteer, veral in groen korrelweefsel. Die vraag of cis-isomere en chlorofil afbreekprodukte natuurlik in die druifkorrels teenwoordig is en of hulle tydens monsterneming en prosessering gevorm word, kon nie in hierdie studie beantwoord word nie. Hierdie studie het bevestig dat karotenoïede en chlorofille oor die algemeen op ‘n korrel (μg/korrel) en konsentrasie (μg/g) basis afneem vanaf deurslaan tot oes. Hierdie studie het nie daarin geslaag om te toon dat groeikragverskille vanaf voor-deurslaan (ertjiekorrelgrootte) tot oes ‘n effek het op die tempo van sintese/afbreking van karotenoïede, chlorofil en ander rypwordingsparameters nie, naamlik op appelsuur, totale glukose en fruktose, totale tannien en totale antosianien. Daar is ook in hierdie studie geen noemenswaardige effek van grondwaterinhoud op karotenoïede, chlorofille en rypheidsparameters gevind nie, heel moontlik as gevolg van die feit dat hoë grondwaterkapasiteit in die laer grondlae gevind is, wat betekenisvolle verskille in wingerdwaterstatus kon verhoed het. Eksperimentele persele wat gekies is vir groeikragverskille op grond van genormaliseerde verskil plantegroei indeks (NDVI) beelde, snoeimassa en grondwatermetings met ‘n neutronvogmeter het net in die eerste 30 cm van die grond noemenswaardige verskille in grondwaterinhoud getoon vir die rypwordingseisoene wat bestudeer is. Voor-sonopkoms plantwaterpotensiaalmetings het egter aangedui dat geen van die eksperimentele wingerdstokke ernstige waterstres ervaar het nie. Sulke stres is voorheen aangedui om ‘n effek op die karotenoïedinhoud van druiwe te hê. Die karotenoïed 5,8-epoksi--karoteen is vir die eerste keer in druiwe gekwantifiseer en verteenwoordig ‘n noemenswaardige hoeveelheid van die totale karotenoïede wat met oes teenwoordig is. Al die karotenoïede en chlorofille behalwe -karoteen blyk sensitief vir seisoenale verskille in klimaatstoestande te wees. Luteïen en β-karoteen was die volopste karotenoïede in die Merlot-druifkorrels, tesame met chlorofil a, vir beide seisoene wat bestudeer is. Die waardes van hierdie karotenoïede was ook goed gekorreleer met vorige navorsing. Chlorofil a is egter in baie groter hoeveelhede in Merlot-korrels gevind in vergelyking met dít wat in die data gerapporteer is. Die rede hiervoor is moontlik dat die chlorofil-afbreekprodukte in hierdie studie in die berekening van chlorofil a ingesluit is. Meerveranderlikeontleding het belowende voorlopige voorspellingsmodelle getoon (met korrelasiewaardes van meer as 0.8 vir beide die seisoene wat geanaliseer is) vir die voorspelling van die konsentrasie van rypheidsparameters (glukose, fruktose, appelsuur, totale tanniene en antosianiene) met karotenoïed- en chlorofilinhoud. Hierdie resultaat beklemtoon die geleentheid vir die ontwikkeling van ‘n vinnige, nie-destruktiewe metode om karotenoïede en chlorofille in korrels te meet, wat op sy beurt optimate rypheid kan voorspel. Aangesien karotenoïede die voorlopers van C13-norisoprenoïed aromaverbindings in wyn is, kan ‘n voorskou van die potensiële bydrae van hierdie aromas tot wyn moontlik verder evalueer word. Verdere navorsing is nodig om die moontlikheid van die bou en geldigheidsbepaling van sulke modelle te ondersoek.