From average to individual fruit, a paradigm shift for accurate analysis of water accumulation and primary metabolism in developing berries

Abstract

Context and purpose of the study ‐ Presentknowledge about grape development is mainly driven by the premise that a typical berry would follow the same kinetics as the population average, the principal challenge being to gather representative samples. In this frame, the elaboration of harvest quality directly reflects the impact of the GenotypexEnvironment interaction on fruit metabolism. Much energy is then being devoted to identifying the sites that regulate grape metabolism, upon screening more and more genes and metabolites, and developing virtual berry models simulating sugar and water accumulation in the future harvest. Nevertheless, successive physiological stages never coexist in a fruit and one may wonder whether the “average physiological stage” paradigm does not inevitably lead to a dead end. The strict foundations of berry developmental biology are critically revisited here.

Material and methods – Disparate literature data on the intensity and duration of the second growth period were re‐interpreted, validated and clarified, upon non‐destructive analysis of single berry firmness and growth, on different cultivars in the experimental vineyard of Supagro, as well as on microvines grown in greenhouses. Organic acids and sugars were measured by HPLC on thousands individual berries of Syrah, Pinot and Zinfandel.

Results ‐ Previously unsuspected sub‐periods emerged from the developmental patterns of sugar, water and malic acid flows on single berries, metabolic fluxes and kinetic data being noticeably stable among all investigated cultivars. Berries accumulated sugars at nearly constant volume during the first week following softening, indicating intense xylem back‐flow at this stage. This first period of ripening was also characterized by a net malic acid/4 hexoses exchange consistent with the operation of a sucrose/H+ exchanger at the tonoplast membrane, in usual conditions and genotypes. Aerobic fermentation and vacuolar proton pumps were induced later, while vacuolar malic acid was progressively exhausted, without compromising sugar and water accumulation. As a matter of fact, phloem unloading definitively stopped 28 days after softening. It clearly appeared that the individual fruit develops in a far more determined, reproducible and finally intelligible way than has been predicted so far, based on average samples.New phenotyping procedures were consequently designed for genetic studies, improving heritability and QTLs detection.Switching from fruit genomics and physiology to harvest quality requires a real change in scale, from the fruit to the population. The determinant role of berries asynchrony within the population can’t be ignored any longer, but the impact of the GxE interaction on the population structure essentially remains terra incognita.