The apple skin: colourful healthiness : developmental and environmental regulation of flavonoids and chlorogenic acid in apples

The ultimate objective of the production, handling and distribution of fresh fruits and vegetables is to satisfy consumers requirements. In general the attractiveness of fruits and vegetables to consumers is determined both by visible (e.g. colour) and invisible (e.g. healthiness) quality attributes. Flavonoids and hydroxycinnamic acid derivatives, secondary metabolites, contribute largely to both fruit colour and, through fruit consumption, to human health and it is, therefore, very useful to study factors that affect these substances with the aim of further improving the relevant fruit attributes. Flavonoids and hydroxycinnamic acid derivatives are widespread in the plant kingdom, comprise a large group of naturally occurring antioxidants that form part of the human diet. There is considerable evidence for the role of antioxidant constituents of fruits and vegetables in the maintenance of health and disease prevention. Recent studies have shown that the majority of the antioxidant activity of a fruit or vegetable may originate from the flavonoids and other phenolic compounds. Apple fruit are rich in flavonoids such as flavonols (quercetin 3-glycosides), flavanols (catechin, epicatechin, gallocatechin, procyanidins and its polymers), dihydrochalcone glycosides (phloritin glucoside (phloridzin) and phloritin xyloglucoside), and cyanidin 3-glycoside (anthocyanins). Apple fruits also contain considerable amounts of hydroxycinnamic acid derivatives mainly represented by chlorogenic acid. The red colour of apples is primarily a consequence of the flavonoid pigments anthocyanins which are located in the vacuole. Despite the importance of flavonoids for the intrinsic quality of apples very little is known of their regulation in fruit. The aim of the work described in this thesis was therefore, to obtain knowledge on the extent to which the contents of flavonoids and chlorogenic acid in the skin of apples varies, how they develop during fruit growth phase, ripening phase and post harvest phase and how they can be manipulated.Chapter 1 contains a review of the literature. It appears that the accumulation of flavonoids and phenolic acids in plants is under control of many internal and external factors.In Chapter 2 the extent of natural variation in flavonoids and chlorogenic acid concentration due to within fruit, within tree, between orchards, between cultivars and among mutants was determined. Considerable variation was observed among these variables. Individual flavonoids and chlorogenic acid concentrations were not equally distributed within the fruit. Quercetin 3-glycosides and anthocyanin were almost exclusively found in the skin. The sun-exposed skin of individual fruit had much higher cyanidin 3-galactoside (anthocyanin) and quercetin 3-glycoside concentrations than the shaded skin, while phloridzin, catechins and chlorogenic acid were similar in the skin of both sides (Chapter 2). Significant genotypic variation was observed for the concentration of flavonoids and chlorogenic acid. 'Jonagold' apples contain significant higher concentrations (about 30% higher) (Chapters 2 and 6) and amounts (about 2-fold higher) (Chapter 6) of the total flavonoids than 'Elstar' apples. Chlorogenic acid concentration was about 3-fold higher in 'Jonagold' than in 'Elstar' apples. However 'Elstar' apples contained significant higher concentrations of some quercetin glycosides types as quercetin 3-rhamnoglucoside (about 2-fold higher) and quercetin 3-glucosides (about 30% higher) than 'Jonagold' apples. This might be relevant with respect to differences in bio-activity and antioxidant capacity of various flavonoid compounds. As far as the potential maximum concentration of flavonoids in apple is genetically determined breeding would be an important tool for increasing healthiness of apples. The differences between basic forms and coloured mutants within a given cultivar (for example Jonagold-Jonaprince; Elstar-Elshof) show that the potential anthocyanin accumulation (but only that) may increase several fold without influencing the concentrations of other flavonoid classes (Chapter 2). Microscopic study showed that the most blushed mutants had a higher number of red cells per cell layer and more cell layers containing red cells than the standard cultivar and the less blushed mutants. It is striking to observe coloured and completely uncoloured cells as neighbours. Since the selection of coloured mutants is inherently based on the amount of red coloration, selection of mutants for higher levels of other potential healthy flavonoid classes e.g. quercetin 3-glycosides could be considered, providing that such characteristics can be relatively easily determined. The concentrations of anthocyanin, quercetin 3-glycosides and total flavonoids were highest in fruit borne in the top of the tree followed by fruit from the outer tree parts, whereas the lowest concentrations were found in fruit from the inner tree. Terminal fruit contained the highest concentrations of these compounds, including catechins, compared to lateral and spur fruit. Phloridzin and chlorogenic acid were not affected by the position of the fruit in the tree nor by the bearing wood type. The maximum possible difference in flavonoid concentrations, based on difference between top fruit (optimal light conditions) and inner fruit (minimal light conditions) may be 3-fold for quercetin 3-glycosides and 2-fold for total flavonoids (Chapters 2 and 3). There were significant differences in flavonoid and chlorogenic acid concentrations in 'Elstar' fruit between two normally productive orchards differing mainly in growth vigour and internal shading. All these results show that light conditions are a main regulatory factor in the biosynthesis of flavonoids in apples.In Chapter 3 the natural distribution of light within the tree canopy in relation to the concentration of flavonoids and chlorogenic acid in fruit skin was analysed. The concentrations of cyanidin 3-galactoside and quercetin 3-glycosides and the percentage of blush in the fruit skin were directly related to light level in the direct vicinity of the fruit. Light in the interior of the canopy was poorer in UV-A, blue, green and red (R) but richer in far-red (FR) light than at all other positions. Consequently, the FR/R ratio (with large influence on formative processes) was much larger at the interior of the canopy than at all other positions. There was a critical FR/R ratio of about 1 above which no anthocyanin and only low amounts of quercetin 3-glycosides were formed.In Chapter 4 the relationships between the fruit nutrients N, P, K, Ca and Mg and concentrations of flavonoids and chlorogenic acid in fruit skin were studied with two types of 'Elstar'. In an experiment with the mutant 'Elshof' with the 5 nutrients applied at 5 rates in 4 replications, only N and Ca applications resulted in higher concentration of these nutrients in the fruit, but sufficient variation was present among treatments to correlate the concentration of the other nutrients with those of flavonoids and chlorogenic acid. Negative correlations were frequently found between the concentration of N and Mg and the N/Ca ratio in fruit during growth, and anthocyanin and total flavonoids concentration at maturity in 1996, 1997 and 1998. In 1997, these correlations were weakest but still significant. In that season, P and K concentration were frequently negatively correlated with the concentration of anthocyanin and total flavonoids. The concentration of Ca was not related to the concentration of anthocyanin and total flavonoids. In a study in 1996 with standard 'Elstar', we used the variation in nutrient concentration due to differences in fruit position on tree. The concentrations of N and K and the N/Ca ratio in fruit at maturity were negatively and that of Ca was positively correlated with the concentration of anthocyanin and total flavonoids. Magnesium concentration was negatively correlated with anthocyanin concentration but not with total flavonoids. As a consequence of the relation with position of the fruit in the tree an interaction with the influence of light may, however, be expected. Multiple regression models mainly containing N as factor accounted for up to 40% and 30% of the variance in anthocyanin and total flavonoids concentration of 'Elshof' mutant apples, and for up to 70% and 65% of the variance in anthocyanin and total flavonoids concentration of standard 'Elstar' apples. The relationships between plant nutrients and chlorogenic acid concentration in apples were not consistent and further study is required. It is concluded that, in addition to improving light conditions, the concentration of flavonoids in fruit skin could be further increased by optimising fertilization especially that of N, directed at preventing excess N accumulation.In Chapter 5 we tested the concept that under condition of high carbon supply, plants may increase the formation of their secondary metabolites, like phenolic compounds. In field experiments crop load was manipulated by applying flower or fruit thinning at different stages of development and at different severity. At a low crop load, fruit weight, soluble solids, acidity and firmness were significantly higher than at high and moderate loads. However, the concentrations of flavonoid and chlorogenic acid were similar at the different levels of crop load. Time of thinning had no significant influence on the concentration of flavonoids and chlorogenic acid in fruit skin and had no further effect on fruit quality characteristics such as weight, soluble solids, acidity and firmness. Removal of only the interior fruits (about one-third of total fruit) at about 4 weeks before expected commercial harvest had no influence on the concentration of flavonoids and chlorogenic acid or on the quality characteristics of the remaining exterior fruits of either 'Elstar' or 'Jonagold'. The results indicate that, within the 'normal' range of conditions, assimilate availability is not a major regulatory factor in flavonoids and chlorogenic acid formation in apples. These results are in agreement with the lack of any influence of the supply of precursors in the orchard (Chapter 7).In Chapter 6 the changes that take place in the concentration and the amount of individual flavonoids and chlorogenic acid in the skin of 'Elstar' and 'Jonagold' apples during development and ripening were investigated. In both cultivars, the concentration on a dry weight basis of quercetin glycosides, phloridzin and chlorogenic acid was highest early in the season but decreased at different rates during fruit development to reach a steady level during maturation and ripening. Catechins (catechin plus epicatechin) concentration showed a similar pattern, but a temporary increase was observed in an early stage of development. The concentration of cyanidin 3-galactoside (anthocyanin) was relatively high early in the season, gradually decreased to a very low steady level during growth, but started to increase near maturation, especially in the outer fruit. On a fruit basis the amount of quercetin glycosides increased during development and was about two times higher in 'Jonagold' compared to 'Elstar', both in outer and inner fruit. These compounds were the most abundant flavonoids in the skin of both cultivars and their accumulation showed a strong dependency on fruit position on tree. The amount of the second most abundant flavonoid type, catechins, increased during development to a maximum and then showed some decrease by mid season which was independent of fruit position on tree. The amount of phloridzin increased only early in the season reaching a steady level during development and ripening, and was independent of fruit position on tree. The amount of chlorogenic acid in both cultivars initially increased, but subsequently decreased to reach a low steady level and was slightly higher in outer than in inner fruit. The latter phenomenon is the only direct evidence for (net) breakdown of any of the studied phenolics. Although anthocyanin concentration was relatively high at early stages of development, significant accumulation on a fruit basis only occurred during maturation and ripening. The accumulation of anthocyanin, similar to that of quercetin glycosides, showed a strong dependency on fruit position on tree. The results indicate that, in general, the overall production of total flavonoids, with the exception of anthocyanin, and chlorogenic acid in apple skin is completed during fruit development before the onset of maturation.Chapter 7 reports the influence of exogenous application of a number of chemicals that are precursors of flavonoids or are known to affect ripening on the accumulation of flavonoids and chlorogenic acid in 'Jonagold' apple skin with emphasis on anthocyanin. One aim was to identify a possible substrate limitation and another to separate the formation of anthocyanin from other related maturity/ripening events. Since the occurrence of the second peak in anthocyanin formation more or less parallels the maturation and ripening phase (like starch degradation and aroma production), anthocyanin formation itself is often considered as a ripening phenomenon triggered by ethylene. Our results suggest, however, that there is no simple relation to ripening and consequently to ethylene production (though we did not measure ethylene). This is concluded from the promotion of anthocyanin formation by ethephon (an ethylene releasing compound) and the retardation of anthocyanin formation by ABG and GA ₃ (known to lower or counteract endogenous ethylene), without significantly altering starch degradation and changes in streif index (combination of starch index, firmness and sugar concentration). Our results have also shown that the other flavonoid classes quercetin 3-glycosides, catechins and phloridzin and chlorogenic acid do not respond to any of the applied chemicals. It is concluded that anthocyanin formation is dependent on developmental signals and independent of both fruit maturity/ripening and of the synthesis of other flavonoid classes and responds in a complicated way to ethylene.In Chapter 8 the changes in individual flavonoids and chlorogenic acid during regular (RS) or ultra low oxygen (ULO) storage conditions at 1°C are reported in both 'Jonagold' and 'Elstar' apples. It could convincingly be shown that during storage of both 'Jonagold' (3, 6 and 8 months) and of 'Elstar' (2, 4 and 6 months) and during 1 or 2 weeks shelf life, the concentrations of cyanidin 3-galactoside and quercetin glycosides were relatively constant, while the concentrations of catechins, phloridzin and chlorogenic acid showed only minor decreases. Moreover there were no significant differences in the concentration of flavonoids and chlorogenic acid between fruits stored under ULO compared to RS conditions. It is concluded that, following harvest, flavonoids present in apples are stable. There is no direct or indirect proof for breakdown (net metabolic turnover) during storage and shelf life.In Chapter 9 the practical applications of the findings made in this study were discussed. Our results show that there is much room for increasing the level of potential health phytochemicals in apples. The first step would be cultivar selection either from already available genotypes or by developing new cultivars through classical breeding or molecular biology and gene technology. We showed that light has a significant impact on the final level of flavonoids in fruit. Therefore, the second and more proximate option would be the optimisation of light conditions within tree canopy by measures such as choice of root stocks, planting system, row orientation and training and pruning systems or covering the orchard floor with reflecting films (though the latter is not promoting the visual aspect of the orchard). A third step could be optimisation of the fertilization programme especially avoiding excess N and better timing of N-application. A further possibility is to sort fruit in healthiness classes. As long as a simple method to detect non-destructively quercetin 3-glycosides is lacking, sorting of fruit based on their blush might be a way to make healthiness classes, since blush is a good marker for exposure to light during growth and thus to some extent for the quercetin 3-glycosides level. Even when cultivar choice and cultivation methods succeed in getting high levels of flavonoids in fruit still the treatment by the consumer determines how much of these substances will be consumed. Many consumers still peel the fruit before consumption thereby removing almost all anthocyanin and quercetin 3-glycosides (Chapter 2). Promotion of fruit on the basis of healthiness is, in our opinion, however, only useful if it is accompanied with a guarantee of absence of pesticides, as is most credible, at least to the public, in organic farming.Because of the large influence of a number of factors at several steps of the production chain, a quantitative model e.g. integrated with light distribution models, would offer a practical and effective tool for estimating the effect of certain measures and to predict and maximise the final level of healthy compounds in apples enabling the development of more accurate intake data and dietary recommendations.