Identification of key gene networks controlling organic acid and sugar metabolism during watermelon fruit development by integrating metabolic phenotypes and gene expression profiles

Abstract

The organoleptic qualities of watermelon fruit are defined by the sugar and organic acid contents, which undergo considerable variations during development and maturation. The molecular mechanisms underlying these variations remain unclear. In this study, we used transcriptome profiles to investigate the coexpression patterns of gene networks associated with sugar and organic acid metabolism. We identified 3 gene networks/modules containing 2443 genes highly correlated with sugars and organic acids. Within these modules, based on intramodular significance and Reverse Transcription Quantitative polymerase chain reaction (RT-qPCR), we identified 7 genes involved in the metabolism of sugars and organic acids. Among these genes, Cla97C01G000640, Cla97C05G087120 and Cla97C01G018840 (r² = 0.83 with glucose content) were identified as sugar transporters (SWEET, EDR6 and STP) and Cla97C03G064990 (r² = 0.92 with sucrose content) was identified as a sucrose synthase from information available for other crops. Similarly, Cla97C07G128420, Cla97C03G068240 and Cla97C01G008870, having strong correlations with malic (r² = 0.75) and citric acid (r² = 0.85), were annotated as malate and citrate transporters (ALMT7, CS, and ICDH). The expression profiles of these 7 genes in diverse watermelon genotypes revealed consistent patterns of expression variation in various types of watermelon. These findings add significantly to our existing knowledge of sugar and organic acid metabolism in watermelon.

Fig. 1. Total soluble solids (Brix %), pH, soluble sugars (glucose, sucrose, and fructose), and organic acids (malic, oxalic, and citric acid) (mg/g FW) in watermelon fruit at 10, 18, 26, and 34 DAP.

Each value is the mean of three biological replicates

Fig. 2

a Cluster dendrogram and network heatmap of genes subjected to coexpression module calculation. b Sample dendrogram and module trait heatmap at each developmental stage

Fig. 3. Gene networks and key candidate genes involved in sugar and organic acid regulation during watermelon fruit development as identified by WGCNA.

a Hierarchical clustering presenting eleven modules having coexpressed genes. Each leaflet in the tree corresponds to an individual gene. b Module-trait associations based on Pearson correlations. The color key from green to red represents r² values from -1 to 1. c Gene network for the blue module, which positively correlated with the sucrose content (r² = 0.92, P = 0.001). d Gene network for the brown module, which positively correlated with the fructose (r² = 0.73, P = 0.04), malic acid (r² = 0.75, P = 0.03) and citric acid (r² = 0.85, P = 0.007) contents. e Gene network from the yellow module, which positively correlated with the fructose (r² = 0.79, P = 0.02) and glucose (r² = 0.83, P = 0.01) contents. Hub genes (key candidates) within each network are highlighted in red due to the highest weight within the module and coded for gene descriptors based on annotations

Fig. 4. The organic acid and sugar biosynthetic pathway.

Enzyme names are shown along with their expression patterns at various stages. Grids represent the expression patterns of genes shown as FPKM values: 10, 18, 26, and 34 DAP, left to right. The absolute expression is represented by the grids at 10, 18, 26, and 34 DAP, with FPKM values of 0–1, 1–2, 2–4, 4–8, 8–16, 16–32, 32–64, 64–128, 128–256, 256–512, 512–1200 and >1200

Fig. 5

Measurement of pH, TSS, and qRT-PCR expression of candidate genes linked to sugars and organic acids in eleven watermelon accessions at different developmental stages of watermelon fruit