A metabolomics approach to elucidate apple fruit responses to static and dynamic controlled atmosphere storage Stefano Brizzolara a, *, Claudio Santucci b , Leonardo Tenori c , Maarten Hertog d , Bart Nicolai d,e , Stefan Stürz f , Angelo Zanella f , Pietro Tonutti a a Istituto di Scienze della Vita, Scuola Superiore Sant’Anna, Pisa, Italy b CERM, University of Firenze, Firenze, Italy c Fondazione FiorGen ONLUS, Firenze, Italy d Division of Mechatronics, Biostatistics and Sensors (MeBioS), Department of Biosystems (BIOSYST), KU Leuven, Leuven, Belgium e Flanders Centre of Postharvest Technology (VCBT), Leuven, Belgium f Laimburg Research Centre for Agriculture and Forestry, Ora, Bolzano, Italy A R T I C L E I N F O Article history: Received 7 November 2016 Received in revised form 22 January 2017 Accepted 22 January 2017 Available online xxx Keywords: Hypoxic metabolism Ultra low oxygen (ULO) Dynamic controlled atmosphere (DCA) Malus domestica Postharvest Metabolic profiling A B S T R A C T The response of apple fruit to storage conditions based on low oxygen protocols depends on their genetic background. In order to elucidate common and divergent processes characterizing the metabolic changes under hypoxia, fruit of two apple (Malus domestica) varieties (‘Granny Smith’, GS, and ‘Red Delicious’, RD) were stored under two different low oxygen protocols (Ultra Low Oxygen, ULO, at 0.9 kPa oxygen, and Dynamic Controlled Atmosphere based on chlorophyll fluorescence, DCA-CF, between 0.2 and 0.55 kPa oxygen) for up to 200 and 214 days of storage for GS and RD samples, respectively. Through an integrated metabolomics approach ( 1 H NMR, GC–MS, HS-SPME-GC–MS analyses) a total of 130 metabolites (volatiles and non-volatiles) were identified. Most of them (117) were common to both cultivars; 95 were significantly different between both cultivars when comparing the whole set of data (ULO + DCA-CF), whereas 13 volatile compounds, identified via HS-SPME-GC–MS, were specific for either GS or RD. Multivariate analyses (PCA and PLS) of the whole dataset allowed to clearly discriminate between GS and RD samples. When storage condition was used as a categorical response variable, a lower percentage explained variance was obtained as this effect was overshadowed by the large effect of storage time. After 4 months of storage, RD underwent more pronounced metabolic compositional changes of the cortex, possibly associated with the evolution of ripening. Based on the accumulation pattern of pyruvate- derived metabolites (ethanol, acetaldehyde, lactate, alanine) it can be hypothesized that there are two main metabolic reconfiguration strategies in GS and RD to regenerate NAD + and cope with energy crisis under hypoxia. GS showed more pronounced responses through changes in the nitrogen metabolism and limited induction of the ethanol fermentation while the latter was highly induced in RD under both ULO and DCA-CF. Marked differences were detected between the VOC profiles of the two cultivars regardless storage conditions. Ethyl esters and 2-methylbutyl derivatives appeared finely modulated by the oxygen level in GS and RD apples, respectively. © 2017 Elsevier B.V. All rights reserved. 1. Introduction A decreased oxygen level, coupled with refrigeration and increased carbon dioxide concentration, is commonly applied in order to prolong the market life of fruits such as apples, kiwifruit, and winter pears in so-called controlled atmosphere (CA) systems. Since the earliest commercial applications, the benefits of CA technology in apple, as compared to regular atmosphere storage, are the delay of ripening and senescence and a better maintenance of quality due to the synergistic effects of low temperature, increased carbon dioxide concentration and reduced levels of oxygen (Yahia, 2009). The discovery that oxygen concentrations of around 1 kPa improve storability has led to the worldwide application of Ultra Low Oxygen (ULO, 0.8–1.2 kPa) protocols (Dilley, 2006). In general, these hypoxic conditions are maintained from the beginning until the end of storage (static CA). This static approach does not always provide optimal post-storage results * Corresponding author at: Istituto di Scienze della Vita, Scuola Superiore Sant’Anna, Piazza Martiri della Libertà 33, 56127 Pisa, Italy. E-mail address: s.brizzolara@sssup.it (S. Brizzolara). http://dx.doi.org/10.1016/j.postharvbio.2017.01.008 0925-5214/© 2017 Elsevier B.V. All rights reserved. Postharvest Biology and Technology 127 (2017) 76–87 Contents lists available at ScienceDirect Postharvest Biology and Technology journal home page: www.elsevier.com/locat e/postharvbio