Comparative transcriptome analysis of genes related to tuberization in late and early maturing potato (Solanum tuberosum) cultivars

Abstract

Potato maturity is a complex and vital agronomic trait directly impacting yield, quality, and economic value. Despite its importance, the molecular regulatory networks governing potato maturity remain largely unexplored. In this study, the factors underlying maturity differences between early-maturing (KX23) and late-maturing (DN310) tetraploid potato cultivars were investigated for the first time under field conditions. Tuber yields, starch content, flowering time, maturity and the number of morphologically modified stolons and initiated tubers were measured, revealing that maturity differences between KX23 and DN310 are linked to the timing of tuberization processes. To explore the molecular basis of these differences, RNA-seq analysis was performed on hooked stolons, swollen stolons, and initiated tubers, identifying key pathways involved in tuberization. WGCNA and qRT-PCR further pinpointed critical genes contributing to these pathways. DEGs between the two cultivars were primarily enriched in secondary metabolite pathways. Notably, a significant overlap of DEGs between KX23 and DN310 across the three developmental stages was identified, with enrichment in carbohydrate metabolism pathways. In total, 28 candidate genes consistently involved in tuber induction and formation were identified. Based on their functions, a model explaining how KX23 achieves faster tuberization, shortening its maturity period, was proposed. These findings provide valuable insights into the molecular regulatory mechanisms of tuberization and maturity in potatoes.

Keywords: Maturity; Potato; RNA-seq; Tuberization; WGCNA.

Fig. 1

Comparison of metamorphic stolon and tuber numbers between KX23 and DN310. (A) Statistical comparison of the number of hooked stolons at 36 and 38 days post-planting. (B) Statistical comparison of the number of swollen stolons at 40 and 42 days post-planting. (C) Statistical comparison of the number of tubers at 44 and 46 days post-planting. Data are shown as the mean values from three biological replicates (n = 10 plants per replicate). Statistical significance was assessed using a t-test, with ** indicating a highly significant difference (P < 0.01).

Fig. 2

DEGs in stolons and initiated tubers of DN310 compared to KX23. (A-C) Volcano plots illustrating DEGs that are upregulated (red dots) or downregulated (green dots) in DN310 relative to KX23. The plots represent hooked stolons (A), swollen stolons (B), and initiated tubers (C), respectively.

Fig. 3

Functional analysis of DEGs in stolons and initiated tubers of DN310 compared to KX23. (A–C) KEGG enrichment analysis of DEGs in hooked stolons (A), swollen stolons (B), and initiated tubers (C) of DN310 relative to KX23, Qvalue < 0.1. (D–F) GO annotation analysis of DEGs in hooked stolons (D), swollen stolons (E), and initiated tubers (F) of DN310 relative to KX23, Qvalue < 0.05.

Fig. 4

Analysis of common DEGs in hooked stolons, swollen stolons, and initiated tubers. (A) Venn diagram showing the overlap of common DEGs. (B) KEGG pathway annotation of the common DEGs. (C) GO term annotation of the common DEGs.

Fig. 5

Transcriptomic WCGNA analysis. (A) Module composition analysis showing the distribution of included genes. (B) Heatmap of the gene co-expression network. (C) Gene count for each module. (D) Expression patterns of the genes across the modules.

Fig. 6

Expression analysis of hub genes in the midnight-blue module. (A) Venn diagram showing the overlap between hub genes and DEGs. (B) Heatmap of differentially expressed hub genes. (C) qRT-PCR validation of differentially expressed hub genes. Data are presented as means ± standard deviations. Statistical significance was determined using a t-test, with ** indicating a highly significant difference (P < 0.01).

Fig. 7

Expression analysis of hub genes in the brown module. (A) Venn diagram showing the overlap between hub genes and DEGs. (B) Heatmap of the differentially expressed hub genes. (C) qRT-PCR validation of differentially expressed hub genes. Relative expression was calculated using the 2^−ΔCT method. Data are presented as means ± standard deviations. Statistical significance was determined using a t-test, with ** indicating a highly significant difference (P < 0.01).

Fig. 8

Based on initial modeling derived from this study’s findings, we hypothesize that during tuber induction in KX23, the coordinated up-regulation of StiPGAM1, StE2Fa, and StQKY coupled with the down-regulation of StSBP and StLOB may promote glycolytic flux and radial growth while concurrently suppressing vertical elongation. It is proposed that this specific gene expression signature collectively contributes to accelerated tuber maturation in KX23. Conversely, we hypothesize that the reciprocal expression pattern of these genes in DN310 underlies its contrasting phenotype, characterized by reduced glycolytic activity, diminished radial expansion, sustained vertical growth, and prolonged maturation time.