Abstract
BACKGROUND: Paeonia lactiflora Pall., a perennial plant with medicinal and ornamental value, exhibits a typical "double dormancy" characteristic in its seeds, which significantly limits large-scale cultivation. This study combines metabolomics and transcriptomics to explore the molecular mechanisms of dormancy release and germination in Paeonia lactiflora seeds during warm-cold stratification, focusing on hormonal regulation, metabolic pathway alterations, and gene expression changes. METHODS: Paeonia lactiflora seeds were subjected to stratification for 0, 28, 55, and 80 days (T0, T1, T2, T3). Endogenous hormones (ABA, GA₃, IAA) and sugars (sucrose, glucose, fructose) were quantified using high-performance liquid chromatography coupled with mass spectrometry (HPLC–MS). Nutrient contents and enzyme activities were measured using commercial kits (Solarbio), following the instructions and using standard reagents for quantification. RNA sequencing was performed for transcriptomic analysis, with differential gene expression (DEG) analysis conducted using DESeq2. Gene co-expression networks were built using weighted gene co-expression network analysis (WGCNA) to identify key regulatory modules. RESULTS: Significant changes in hormone and nutrient contents were observed during stratification. During the warm stratification phase (T0–45 days, 20 °C), ABA (abscisic acid) levels were dominant, while during the cold stratification phase (45–80 days, 4 °C), the seed’s hormonal composition underwent significant changes. ABA levels decreased from 72.54 ng/g at T0 to 1.49 ng/g at T2, GA₃ increased from 0.45 ng/g at T0 to 1.41 ng/g at T1, and IAA levels significantly increased from 4.32 ng/g at T0 to 70.09 ng/g at T1. Sugar levels showed a downward trend, with fructose content decreasing from 22.34% at T0 to 7.31% at T3. Starch content significantly decreased from 40.13% at T0 to 15.34% at T3. Enzyme activities of α-amylase and β-amylase peaked at 0.2267 U/mg and 0.3410 U/mg at T2, respectively. Transcriptomic analysis yielded 83.82 GB of high-quality clean data, identifying 83,082 differentially expressed genes (DEGs). DEG analysis revealed 11,045 DEGs during embryo axis growth (T0–T3), 10,042 DEGs during epicotyl elongation, and 923 DEGs common across all stages. WGCNA analysis identified the black, cyan, and turquoise modules as key regulatory modules related to hormonal regulation and nutrient mobilization. Pathway enrichment analysis showed that DEGs were significantly involved in metabolic pathways, including starch and sucrose metabolism, hormone signaling pathways (IAA, GA, ABA), and oxidative phosphorylation. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12870-025-07636-x.