| In present study,’Hayward’ kiwifruit (Actinidia deliciosa [A. Chev.] CF Liang et AR Ferguson var. Deliciosa cv. Hayward) were treated with ethylene to fasten ripening process and low temperature to extend fruit shelf-life, electronic nose (E-nose) and headspace solid phase microextraction coupled with chromatography-mass (HS-SPME GC-MS) were chosen to investigate changes in volaitile compounds during postharvest storage. The main results are as follows:1. E-nose equipped18metal oxide sensors had different values in analyzing Rlwifruit samples, in which LY2/Gã€LY2/AAã€LY2/GH. LY2/gCTã€LY2/gCT1showed positive values while the othter types of sensor produced negative values. Response valuses of PA/2, P40/2, P30/1and T40/2sensors tended to increase with kiwifruit ripening and senescence. Differences in E-nose generated volatile fingerprints indicated significant changes in volatile compounds during kiwifurit postharvest ripening and sescence and in response to different treatments. Approximately90%of total variance was explained in a DFA model generated by E-nose based on sensors values, in which fruit sample mainly distributed from the positive axis to the negative axis of DF1according to fruit ripening stages. In the case of kiwifruit treated with low temperature followed by subsequent shelf-life, samples were mainly distributed on the DF1negative axis and DF2positive axis, clearly beding separated from the controls and ethylene treated fruit. These results indicated that cold storage greatly influenced kiwifruit volatiles during postharvest storage, and could be used as a reference for further GC-MS analysis.2. More than40volatiles were identified in kiwifruit by HS-SPME GC-MS technique, belonging to C6compounds, aldehydes, alcohols, esters, terpenes, furans, ketones and benzenoid compounds. Based on changes in contents of volatile compounds, the fruit samples at different ripening stages could be separated and grouped by principal component analysis (PCA) and hierarchical cluster analysis (HCA), respectively. On kiwifruit harvest day, C6compounds were the main volatile compounds, accounting for about30%content of the total volatile compounds. With furit postharvest ripening and senescence, esters levels increased significantly and promoted by ethylene treatment. There were no significant differences between aldehydes, alcohols and ketons between ethylene treated fruit and the controls. Content of total volatiles tended to increase during postharvest storage, being similar to pattern of esters. After cold storage for120d plus shelf-life, kiwifruit produced lower contents of (E)-2-hexenol,(E)-2-hexenal and ethyl butanoate, being approximately0.5%,2%and30%of that in controls, respectively. Contents of total esters in cold stored fruit were approximately30%of ethylene treated kiwifruit. The above results showed that cold storage significantly regulated formation of volatile compounds, consisting with results from the E-nose analysis.3. Volatile compounds that were significantly correlated to fruit firmness and TSS were identified based on correlation analysis. The highest positive correlation with fruit firmness was observed for (Z)-3-hexenol, being r=0.85and P<0.05, respectively.(E)-2-hexenal,(E,E)-2,4-heptadienal and (E)-2-heptenal also had significant positive correlations, while3-methyl furans showed significant negative correlation with kiwifruit firmess. TSS showed the highest significant positive correlation methyl butanoate, with a correlation coefficient of0.96. Similar significant positive correlation were also found for ethyl hexanoate, methyl benzoate and (E)-2-hexenol.In summary, kiwifruit samples could be clearly separated between samples with different ripening stages and among different treatments using E-nose and volatiles detected by GC-MS. And volatiles that had significant correlations with fruit firmeness and TSS during postharvest storage were identified. DFA models prouded by E-nose and statistical results of volatiles contents indicated that cold storage significantly changed kiwifruit volatile esters levels, which may be explained the separation between treatement in the DFA model. |