Walnut fruits (Juglans regia cv. Chandler) were harvested from fully mature walnut orchard from Karaj, Alborz province, Iran. After the harvest, the fruits were immediately carried to the postharvest physiology of horticultural crops laboratory at the University of Tehran, Karaj, Iran. Then, the green husk and the hard shell of walnut fruits were manually removed. In a preliminary experiment, unshelled nuts, without being dried, were coated with 0, 5, 10, 15 and 20% (m/V) gum arabic alone or enriched with γ-aminobutyric acid (GABA) at a concentration of 0, 0.01, 0.1, 1 and 10 mM. Gum arabic solution was prepared according to Ali et al. (11) by dissolving 0, 5, 10, 15 and 20 g of gum arabic powder in 100 mL distilled water. The solutions were stirred at low heat (40 °C) for 60 min, then filtered to remove any undissolved impurities. The pH of the solutions was adjusted to 5.6 by 1 M NaOH.

GABA and gum arabic powder, food grade (kibble size: KB 120, i.e. <125 micron), was supplied by Labortecnic SA (Almeria, Spain). All other chemicals were purchased from Sigma-Aldrich, Merck (Madrid, Spain) and were of analytical grade. Different concentrations (0, 0.01, 0.1, 1 and 10 mM) of GABA were added to gum arabic solution according to Khaliq et al. (12). For coating treatments, healthy walnut kernels, macroscopically free from disorders and disease symptoms were placed in a mesh container and immersed in the coating solutions for 1 min. The coated walnut kernels were then dried for 3 h at room temperature and 100 g of kernel halves were kept at 20 ºC for 6, 12 and 18 weeks (10).

In the treated kernels, peroxide value (PV) and malondialdehyde (MDA) content (as oxidative rancidity markers) increased and kernel surface whiteness index (WI), as oxidative browning marker, decreased during storage at 20 °C for 18 weeks (data not shown). Coating resulted in the lowest kernel oxidative rancidity and browning. Based on these results, the solutions of gum arabic 5 and 10% with the addition of 0.1 and 1 mM GABA, respectively, were selected for further study. PV, MDA accumulation and WI were evaluated every 6 weeks during storage at 20 °C. Phenylalanine ammonia-lyase (PAL) and polyphenol oxidase (PPO) enzyme activity, total phenolic accumulation, DPPH^• scavenging capacity and LOX enzyme activity were also investigated.

Kernel colour was measured on the upper part of the outer surface on each half by chromatometer (CR-400; Minolta, Tokyo, Japan) using CIE colour parameters L*, a* and b*, and WI was calculated according to the following equation (13):

Lipid extraction was measured according to Lee et al. (14). PV was measured by iodometric titration with 0.01 N Na₂S₂O₃ according to AOAC method No. 965.33 (15) using 1 g of lipids separated from walnut kernels and expressed in mmol peroxide per kg kernel oil. MDA content was measured by the thiobarbituric acid (TBA) method described by Nepote et al. (16) and expressed in nmol TBA per g dry mass.

Total phenolic content was determined according to the Folin-Ciocalteu procedure (17). The absorbance of the samples was measured by UV/Vis spectrophotometer (PC-1650; Shimadzu Kyoto, Japan) and the content was expressed in mg gallic acid equivalents (GAE) per g dry mass (dm). Free radical DPPH^• scavenging activity was traced according to Brand-Williams et al. (18). The reduction percentage of DPPH^• was calculated according to the following equation:

where A_control is the absorbance of DPPH^• solution without extract.

Fatty acids were quantified by forming fatty acid methyl esters according to Arena et al. (19), using gas chromatograph equipped with flame ionization detector (GC-FID model 6890 N; Agilent Technologies, Santa Clara, CA, USA). Fatty acid content was expressed in percentage as mass fraction in oil. The unsaturated/saturated fatty acid ratio was calculated by the formula:

where according to C:D number 16:0 is palmitic, 18:0 stearic, 18:1 oleic, 18:2 linoleic and 18:3 linolenic acid.

PAL was assayed based on cinnamic acid production according to Nguyen et al. (20), and was expressed in μmol cinnamic acid per mg protein per h. PPO was assayed by measuring the oxidation of catechol as substrate according to Nguyen et al. (20), and was expressed in U/mg protein.

LOX activity was measured by linoleic acid as substrate according to Phetsirikoon et al. (21), and was expressed in U/mg protein. The H₂O₂ content was measured according to Patterson et al. (22) and expressed in μmol/g dm.

The experiment was conducted as split plot in time based on completely randomized design with three replications (100 g of kernels for each replication). Data analysis was performed using IBM SPSS software v. 21 (23). Differences were assessed by Tukey's test at p<0.05.

During storage at 20 °C for 18 weeks, we recorded a significant increase in PV and MDA accumulation and a decrease in WI of kernels (Table 1), indicating oxidative rancidity and browning incidence manifested in kernels by irreversible UNSFA peroxidation via ROS/LOX and phenolic oxidation by PPO. Kernels coated with gum arabic (5%) enriched with GABA (0.1 mM) exhibited lower PV and MDA accumulation and higher WI than other treatments during storage at 20 °C for 18 weeks (p<0.01; Table 1). Coating with 5% gum arabic containing 0.1 mM GABA resulted in the lowest kernel oxidative rancidity and browning. Thus, it can be clearly indicated that gum arabic coating enriched with GABA could strongly protect kernel UNSFA peroxidation and phenolic oxidation, attenuating kernel oxidative rancidity and browning.

As shown in Table 2, linoleic, linolenic and oleic acid contents decreased during storage at 20 °C for 18 weeks, while palmitic and stearic acid contents increased. As a consequence, unsaturated/saturated fatty acid (UNSFA/SFA) ratio decreased. Meanwhile, coated kernels with lower palmitic and stearic acid contents (p<0.01) and higher linoleic, linolenic and oleic acid contents (p<0.01) had higher UNSFA/SFA ratio during storage at 20 °C for 18 weeks (Table 2). Furthermore, kernels coated with 5% gum arabic and 0.1 mM GABA had lower LOX activity (p<0.01) and H₂O₂ accumulation (p<0.01) during storage (Table 3). Oxidative rancidity of walnut kernels is a consequence of the peroxidation of UNSFA by ROS/LOX, and a decline in UNSFA/SFA ratio was concurrent with the higher PV and MDA accumulation, as oxidative rancidity markers. In addition to the oxidation of kernels, UNSFA and LOX can participate in O₂^•− accumulation (24). Therefore, overcoming the kernel oxidative rancidity is achievable by decreasing O₂ availability to prevent ROS production using edible coatings or by scavenging ROS radicals (7-10). Kang et al. (7) reported that the walnut kernels coated with soy protein enriched with antioxidant catechin exhibited significantly lower UNSFA oxidation with the lower PV and MDA accumulation during storage at 35 ºC for 21 days, which was the result of the synergistic action of O₂ barrier function of soy protein and antioxidant activity of catechin. The authors suggested that the edible soy protein coating enriched with catechin delayed UNSFA peroxidation of walnut kernels by protecting them from O₂ exposure during storage. Another study carried out by Vidrih et al. (25) showed that the unstable linolenic acid content decreased due to its oxidation during storage, while the other fatty acids remained largely unaffected. Sabaghi et al. (10) reported that walnut kernels coated with chitosan enriched with antioxidant green tea exhibited a lower UNSFA peroxidation, confirmed by lower PV and MDA accumulation and higher sensory quality observable in lower kernel surface browning during storage at room temperature for 18 weeks, which was a consequence of the synergistic action of O₂ barrier function of chitosan and the antioxidant activity of green tea. Christopoulos and Tsantili (3) reported that during storage at 20 ºC for 9 months, walnut kernel UNSFA/SFA ratio decreased and the oleic, linoleic and linolenic acid peroxidation resulted in kernel rancidity assayed by the higher PV, as oxidative kernel rancidity marker. Additionally, they reported that walnut kernels stored at low temperature (1 ºC) and packaged under N₂ or CO₂ exhibited lower PV, as oxidative kernel rancidity marker, which may be a consequence of the lower oleic, linoleic and linolenic acid peroxidation, resulting in higher kernel UNSFA/SFA ratio. Our results suggest that the debilitation of oxidative rancidity in walnut kernels coated with gum arabic (5%) enriched with GABA (0.1 mM) may be caused by a lower LOX enzyme activity and lower H₂O₂ accumulation. Eventually, higher UNSFA/SFA ratio may be an outcome of the synergistic action of gum arabic, which lowers O₂ availability, with GABA, which scavenges the ROS (26). This suggests that the gum arabic coatings enriched by GABA delayed lipid oxidation of walnuts by protecting them from oxygen exposure during storage period.

As shown in Table 4, walnut kernels coated with 5% gum arabic enriched with GABA (0.1 mM) exhibited higher total phenolic accumulation (p<0.01) and DPPH scavenging capacity (p<0.01) during storage at 20 °C for 18 weeks. Moreover, kernels had lower PPO (p<0.01) and higher phenylalanine ammonia-lyase (PAL) enzyme activity (p<0.01).

In walnuts, light kernel colour is crucial for sensory quality attributes leading to consumer preference. According to Christopoulos and Tsantili (27), phenolics are accumulated in walnut kernels by the activation of phenylpropanoid pathway through PAL enzyme activity, which is responsible for non-oxidative deamination of phenylalanine to trans-cinnamic acid. Walnut kernel phenolics have a high antioxidant capacity, which contributes to a higher DPPH or FRAP scavenging capacity. Phenolics are crucial for the human health and play a crucial role in attenuating oxidative UNSFA rancidity in the walnut kernels by obstructing ROS accumulation (28, 29). However, under unfavourable postharvest handlings and due to membrane integrity damage caused by MDA accumulation, higher phenolic accumulation in walnut kernels may contribute to surface browning in kernels, as a result of the increased PPO (enzyme responsible for browning) activity, which leads to a lower consumer acceptability and reduced sensory quality (3, 10, 30). In line with oxidative rancidity, kernel surface browning as a consequence of phenolic oxidation by PPO results in a huge reduction in kernel antioxidant capacity (31). Fuentealba et al. (32) reported that the walnut kernel light colour was accompanied with higher phenolic accumulation resulting from lower phenolic oxidation by PPO activity and seemingly attributed higher DPPH scavenging capacity. Likewise, light colour kernels exhibit a higher arbutin accumulation and antioxidant activity, which is crucial for attenuating the browning incidence by hampering tyrosinase enzyme activity. Along with higher phenolic accumulation, light walnut colour was accompanied with higher linolenic acid accumulation, revealing the lessening of oxidative rancidity via antioxidant phenolics (32). Moreover, Wang et al. (4) reported that the green hull of fresh walnuts in the hull packed in modified atmosphere packaging during storage at 1 °C for 60 days exhibited higher phenolic and flavonoid accumulation, leading to higher FRAP scavenging capacity derived from higher PAL/PPO enzyme activity, which diminished the decay of fresh walnut in the hull and maintained kernel quality with lower PV (4). Christopoulos and Tsantili (31) reported that the walnut kernels stored at 20 ºC for 9 months exhibited a higher surface browning, reflected by the lower L*, hº and WI, along with lower FRAP and DPPH˙ scavenging capacity due to phenolic oxidation. In comparison, walnut kernels stored at low temperature (1 °C) and packed under N₂ or CO₂ exhibited a lower kernel surface browning and higher FRAP and DPPH˙ scavenging capacity owing to the increased phenolic accumulation, as the consequence of reduced phenolic oxidation by the low temperature and/or low O₂ availability (30). Christopoulos and Tsantili, (31) suggested that walnut kernel surface browning at a higher temperature and O₂ availability may be attributed to lower FRAP and DPPH˙ scavenging capacity and the lower phenolic accumulation caused by the higher phenolic oxidation. Based on the data from the present experiment, to overcome the oxidative browning in walnut kernels, coating with gum arabic (5%) enriched with GABA (0.1 mM) would be an applicable method due to higher PAL/PPO enzyme activity concurrent with higher phenolic accumulation and higher DPPH˙ scavenging capacity, mainly due to the synergistic behaviour of gum arabic coating function through lowering O₂ availability with the ROS scavenging capacity of GABA. In addition to phenolics, GABA itself exhibits a strong ROS scavenging activity and may contribute to reduction of oxidative browning of walnut kernels (25).