Seedless citrus has become one of the breeding goals due to its high edible ratio and convenience in fresh consumption and processing. In this study, the ‘Hongjiangcheng’ sweet orange (WT) and its seedless mutant (MT) after cobalt-60 radiation were selected to study the formation metabolism of citrus seedless phenotype. Compared with WT, the MT had altered primary metabolite contents, as indicated by GC-MS analysis. The mature pollen of the MT was mostly distorted and shrunken, and the orange mutant exhibited significantly lower fertility than the WT. Through pollination experiments and paraffin sectioning of the MT, we observed self-compatibility during pollen tube germination in situ, in combination with the absence of natural parthenocarpy and arrested zygotic embryo development at the fourth week after pollination. From transcriptomic analyses of ovules in the fourth week, 815 differentially expressed genes (DEGs) were identified. Furthermore, according to the annotation of gene function and qRT-PCR analysis, Cs4g10930, Cs5g21900 and orange1.1t02243 were identified as candidate genes that may govern the mechanism of seedlessness. Finally, Agrobacterium-mediated transformation verified that the overexpression of Cs4g10930 and Cs5g21900 in Newhall navel orange calli inhibited embryoid production. This study provides a better understanding of seedless formation in citrus and two key genes that may play an important role in the early selection of seedless lines in citrus breeding programs.

To study the profiles and relationships of terpenoids and phytohormones in leaves of red-flesh citrus, we examined four red-flesh mutants that accumulate lycopene in the fruit and their corresponding wild types. We compared their contents of monoterpenoid and sesquiterpenoid volatiles, triterpenoid limonoid aglycones, and tetraterpenoid carotenoids. The types and concentrations of these terpenoids in the leaves varied among genotypes, and the leaf terpenoid profiles differed between red-flesh mutants and their wild types. However, lycopene was not detected in citrus leaves of all eight varieties, including the four mutants. According to phytohormone accumulation in the leaves, the citrus varieties could be classified into a low-phytohormone group ('Red Anliu' sweet orange, 'Anliu' sweet orange, and 'Cara Cara' navel orange), a high-phytohormone group ('Seike' navel orange), and a high-jasmonic acid group ('Red-flesh Guanxi' pummelo, 'Guanxi' pummelo, 'Chuhong' pummelo, and 'Feicui' pummelo). The contents of terpenoid volatiles were very low in leaves of 'Red-flesh Guanxi' and 'Guanxi' pummelo; therefore, they are ideal materials for verifying the functions of genes related to terpenoid volatiles in overexpression analyses. Sesquiterpenoid volatiles were positively correlated with phytohormones of abscisic acid, jasmonic acid, and salicylic acid. Taken together, lycopene was undetected in the leaves of red-flesh citrus mutants with abundant lycopene in their juice sacs, which suggested a tissue-specific accumulation pattern of lycopene in citrus red-flesh mutants. Furthermore, leaf volatile profiles could be suitable to develop fingerprint chromatograms of citrus resources, since leaf volatile profiles were dominated by several compounds that varied among genotypes in their combinations and concentrations.

The interior atmosphere of storage facilities is very important for maintaining fruit quality and delaying senescence, so it's of significance to determine the volatile profiles in this environment for the postharvest industry. However, volatile profiles inside fruit storage facilities have been rarely reported. To study the volatile profiles of the atmosphere inside fruit storage facilities, a method comprising Monotrap^TM adsorbent and GC–MS analysis in the study was developed and tested. Based on that, the volatile profiles in the atmosphere inside fruit storage facilities at four locations in Shaanxi Province were monitored. Altogether thirty-six volatiles were detected, and most of them were identified as esters. An analysis of the similarities of volatile profiles showed that the three storage rooms at each location clustered together. The storage rooms at two locations in Baoji had the most similar volatile profiles, and both were similar to that at Xianyang, but different from that at Yan'an, which were consistent with their geographic distributions. On the basis of a principal component analysis, heatmap dendrogram, and correlation matrix analysis, these compounds clustered into five groups. Compounds in Group 1, which were abundant in the storage room at Yan'an, were branched-chain esters, whereas the compounds in the other four groups were mainly straight-chain esters which were abundant in the storage rooms at Xianyang and Baoji. The difference among volatile profiles inside fruit storage facilities at four locations might be caused by ultraviolet radiation. In this study, we demonstrated the Monotrap^TM adsorbent and GC–MS analysis were an efficient method for volatile compounds detection inside fruit storage facilities.