Citric and malic acids determine the sourness and sweetness of banana pulp, which are the two main determinants of consumer preferences. The present work focused on the physiological processes controlling the accumulation of citric and malic acids in banana pulp (Musa sp. AA) using experimental analysis and ecophysiological modeling. We chose an integrative approach linking the pre and post-harvest stages, and focused on the effect of genotype and fruit growing conditions. Experiments were conducted to study the effect of fruit load, potassium fertilization and fruit age at harvest on the accumulation of citrate and malate in banana pulp. To account for genotypic variability, three genotypes with contrasting acidity at the eating stage were studied. Major differences in the pattern of citrate and malate accumulation were found in the three cultivars both during growth and post-harvest ripening and were shown to be the result of metabolic changes. The harvest stage had a significant effect on the concentrations of acids during post-harvest ripening. Fruit load and potassium fertilization had no effect. Ecophysiological models were developed to predict several banana acidity criteria during the pre and post harvest stages. pH and titratable acidity were predicted by a model of acid-base reactions; malate content by a model of vacuolar storage; and citrate content by a model of the TCA cycle. These models led to the identification of the key physiological processes that control banana acidity. Genotypic parameters were identified, which were related to the activity of the mitochondrial malic enzyme and of the malate mitochondrial carriers in the citrate model, as well as to the activity of the vacuolar proton pump, ATPase, in the malate model. The two models were also used to analyze the effects of fruit growth conditions on banana acidity. Combining the three models in a global model of banana acidity, and the possible use of this model for varietal improvement are discussed.
