Identification of candidate genes that regulate the trade-off between seedling cold tolerance and fruit quality in melon (Cucumis melo L.)

Abstract

Trade-offs between survival and growth are widely observed in plants. Melon is an annual, trailing herb that produces economically valuable fruits that are traditionally cultivated in early spring in China. Melon seedlings are sensitive to low temperatures, and thus usually suffer from cold stress during the early growth period. However, little is known about the mechanism behind the trade-offs between seedling cold tolerance and fruit quality in melon. In this study, a total of 31 primary metabolites were detected from the mature fruits of eight melon lines that differ with respect to seedling cold tolerance; these included 12 amino acids, 10 organic acids, and 9 soluble sugars. Our results showed that concentrations of most of the primary metabolites in the cold-resistant melons were generally lower than in the cold-sensitive melons; the greatest difference in metabolite levels was observed between the cold-resistant line H581 and the moderately cold-resistant line HH09. The metabolite and transcriptome data for these two lines were then subjected to weighted correlation network analysis, resulting in the identification of five key candidate genes underlying the balancing between seedling cold tolerance and fruit quality. Among these genes, CmEAF7 might play multiple roles in regulating chloroplast development, photosynthesis, and the ABA pathway. Furthermore, multi-method functional analysis showed that CmEAF7 can certainly improve both seedling cold tolerance and fruit quality in melon. Our study identified an agriculturally important gene, CmEAF7, and provides a new insight into breeding methods to develop melon cultivars with seedling cold tolerance and high fruit quality.

Figure 1

The cold tolerance of eight inbred lines of melon. (a-b) Cold response phenotypes in seedlings (a) and leaves (b). (c) Chlorophyll fluorescence images of the first true leaves. (d) Chilling injury index. There were three biological replicates for each line with 15 seedlings in each replicate. (e) Chlorophyll fluorescence (Fv/Fm), n = 5. Values are means ± SD. Statistical difference were determined by one-way analysis of variance (ANOVA). The same letter above a column indicates no significant difference at p < 0.05.

Figure 2

Metabolite analysis was performed by gas chromatography–mass spectrometry (GC–MS) in extracts of melon fruit flesh. (a) Heat map of metabolite levels in fresh fruits from eight inbred melon lines. There were four biological replicates for each inbred line. (b) Comparison of metabolite levels in inbred lines H581, HH94, and HH09. (c–e) The contents of organic acids, AsA, GABA, and soluble sugars in H581 and HH09. (f–g) Visualization of metabolite-metabolite correlations in melon lines H581(f) and HH09 (g).

Figure 3

Gene expression analysis in H581 and HH09 fruits. (a and b) GO and KEGG gene enrichment analysis of differentially expressed genes (DEGs). The three main GO categories (biological process, cellular component, and molecular function) are indicated on the right side of the figure in (a). (c) The expression levels of genes involved in the biosynthesis of organic acids and sugars. There were three biological replicates for each line. Left represent H581 and right represent HH09. Grids represent the expression levels of genes, which are shown as FPKM values.

Figure 4

Construction of co-expression regulatory networks in the cold-resistant melon lines H581 and HH09. (a) The cluster dendrogram and network heatmap of genes involved in coexpression module calculation. (b) Sample dendrogram and module trait heatmap for the inbred lines H518 and HH09.

Figure 5

Gene networks and key candidate genes involved in cold tolerance and primary metabolite trade-offs in melon as identified by WGCNA. (a) Clustering dendrogram of the gene coexpression modules. (b) Module-trait relationships based on Pearson correlation coefficients. (c) Gene networks for four modules; from left to right MEbrown3, MEhoneydew, MEorangered3, and MEpink3 (r² > 0.90, p < 0.01).

Figure 6

Sequence variation in candidate genes for cold tolerance and fruit quality (Primary metabolite) trade-offs. (a) The relative changes in expression levels of candidate genes with sequence variations. (b and c) Statistical analysis of sequence variations in genes in H581 and HH09. (d and e) Plant phenotypes and chlorophyll fluorescence (Fv/Fm) in wild-type Arabidopsis thaliana Col-0 and the eaf7 mutant (SALK_020366C) grown under control and cold stress conditions, n = 9. Values are means ± SD, Student's t test. ^** indicates significant differences at p < 0.01. (f) Changes in plant growth in 4-week-old seedlings of Col-0 and the eaf7 mutant under control and cold stress conditions.

Figure 7

Cold tolerance in CmEAF7-silenced seedlings of melon lines H581, HH09, and HH14. (a) Phenotypes of CmEAF7-silenced plants. (b–c) Phenotypes and chlorophyll fluorescence in the first true leaves before and after cold treatment. (d and g) Statistical analysis of chlorophyll fluorescence (Fv/Fm) in the first true leaves before and after cold treatment, n = 6. (e and h) Statistical analysis of chlorophyll content in the first true leaves before and after cold treatment, n = 6. (f and i) Statistical analysis of ion leakage in the first true leaves before and after cold treatment (4°C for 16 h), n = 3. Seedlings of H581 and HH09 were treated as described in the materials and methods section. Values are means ± SD, Student's t test. ^* and ^** indicate significant differences at p < 0.05 and p < 0.01, respectively. Scale bars = 2 cm.

Figure 8

Transient expression of CmEAF7 affects the accumulation of soluble sugars in melon fruits. (a) The contents of sucrose, fructose, and glucose after CmEAF7 transient overexpression and silencing at 1.5, 3.5, 5.5, and 7.5 days after injection in H581 fruits, and expression of the CmEAF7 gene at 5.5 and 7.5 days after injection. (b) The contents of sucrose, fructose, and glucose after CmEAF7 transient overexpression and silencing at 1.5, 3.5, 5.5, and 7.5 days after injection in HH14 fruits, and expression of the CmEAF7 gene at 1.5 and 5.5 d after injection. CmActin was used as a reference gene for normalization. OE-LUC: Transient overexpression of the empty vector; OE-CmEAF7: Transient overexpression of CmEAF7; AS-GFP: Transient silencing of the empty vector; AS-CmEAF7: Transient silencing of CmEAF7; n = 3–6; Values are means ± SD, Student's t test. ^* and ^** indicate significant differences at p < 0.05 and p < 0.01, respectively.