Abstract The walnut (Juglans regia) is an important oilseed tree species characterized by its extensive distribution, high oil yield, and nutrient-dense kernels, which provide substantial economic benefits. However, the rising incidence of late-spring frosts, exacerbated by global climate change, has adversely affected walnut yields. A comprehensive understanding of the regulatory mechanisms involved in bud dormancy, germination, and development is essential for developing strategies to mitigate the effects of late-spring frosts and for breeding frost-resistant cultivars. This study focused on W13, a protogynous walnut variety with early germination of dormant buds in spring, employing a combination of transcriptomic and hormone metabolomic analyses. Our results emphasized four key biological processes—cellular response to ethylene stimulus, phenylpropanoid metabolic process, ethylene-activated signaling pathway, and monooxygenase activity—along with several relevant pathways, including plant hormone signal transduction, flavone and flavonol biosynthesis, biosynthesis of secondary metabolites, and MAPK signaling pathway, all crucial for walnut bud germination. Additionally, bud germination is closely associated with alterations in various hormone signaling pathways, including abscisic acid, auxin, cytokinin, ethylene, gibberellins, jasmonic acid, and salicylic acid. By assessing hormone levels and gene expression at different developmental stages, we pinpointed potential regulatory genes and critical hormones associated with bud germination. Furthermore, through weighted correlation network analysis, we constructed a co-expression network, identifying gene modules specifically expressed during dormancy, germination, budding, and leafing phases. The hub genes within these modules are likely pivotal in regulating walnut bud germination. Our analysis also revealed that genes from various transcription factor families are central within the co-expression network, indicating their significant roles in the bud germination process. Correlation network analysis of hormone and gene further illuminated the mechanisms through which genes and hormones jointly influence walnut bud germination. These findings establish a crucial molecular basis for a more comprehensive understanding of the mechanisms governing germination and development in dormant walnut buds. Supplementary Information The online version contains supplementary material available at 10.1186/s12864-025-11272-y. Keywords: Walnut, Bud germination, Transcriptome, Hormone Introduction Late-spring frosts (LSFs), which occur following the germination of herbaceous plants and the budburst of woody species, have significant ecological and economic impacts on agriculture and forestry in temperate and frigid regions [[42]1]. Walnuts, a key temperate oilseed tree species found throughout Eurasia and the Americas, are especially vulnerable to LSFs during leaf emergence, flowering, and early fruit development, potentially resulting in reduced yield and quality. In conjunction with global climate change, the climate crisis may intensify both the frequency and severity of LSFs [[43]1]. Therefore, regulating walnut bud germination and modifying its developmental processes to protect delicate buds, floral structures, and young stems from LSFs can help mitigate the adverse effects of LSFs on walnuts, especially for protogynous cultivar. Bud dormancy is characterized by a decrease in meristem activity, leading to the absence of visible growth. This adaptive mechanism is crucial for trees facing extreme environmental conditions and seasonal climate variations, significantly contributing to species survival and reproduction [[44]2]. Researchers classify dormancy into three types according to their inducing factors: paradormancy, endodormancy, and ecodormancy. Paradormancy and ecodormancy are initiated by external stimuli, whereas endodormancy is regulated by environmental conditions and internal signals within the dormant structure [[45]3]. Recent studies have extensively examined the release of endodormancy, demonstrating that a period of accumulated low temperatures is essential for breaking this dormancy [[46]4]. Transcription factors are critical regulators of plant growth, development, and environmental adaptation, playing a significant role in the regulation of plant dormancy. MADS-box transcription factors are implicated in dormancy regulation across various plant species. The DORMANCY-ASSOCIATED MADS-BOX (DAM) genes, along with orthologs of SHORT VEGETATIVE PHASE (SVP) genes, serve as potential regulatory elements of dormancy in multiple species, including almond, apple, apricot, Chinese cherry, hybrid aspen, Japanese apricot, kiwifruit, leafy spurge, pear, and sweet cherry [[47]5]. The bZIP transcription factor PtoHY5a in poplar governs the initiation of winter dormancy through modulation of the PtoPHYB2–PtoHY5a–PtoFT2 module and influences the bud-break process by regulating gibberellin (GA) levels [[48]6]. The AP2/ERF family transcription factor EBB1 in poplar plays a crucial role in reactivating cell division in buds and leaf primordia following winter dormancy, and it regulates the timing of bud sprouting in woody perennial species [[49]7]. EBB3 undergoes epigenetic modulation at low temperatures, facilitating bud germination through the positive regulation of genes linked to cell proliferation [[50]8]. The grapevine WRKY transcription factor gene VvWRKY37 is markedly expressed in dormant buds and orchestrates the bud germination process via ABA-mediated signaling pathways [[51]9]. In apple trees, the ectopic expression of the peach CBF gene PpCBF1 can postpone spring bud germination [[52]10]. The tree peony PsMYB306 negatively regulates the release of cold-induced bud endodormancy by modulating the production of abscisic acid (ABA) [[53]11]. Plant hormones are essential for growth, development, and adaptive responses to environmental changes. Each hormone has unique and significant functions in the induction, maintenance, and release of bud dormancy. Notably, ABA and GA are the primary hormones that regulate these processes antagonistically [[54]12]. ABA is the key regulator in the initiation and maintenance of bud dormancy [[55]4]. Numerous observations demonstrate that endogenous levels of ABA rise during the establishment of bud dormancy and decline upon its release [[56]13]. The exogenous application of ABA can facilitate the induction of bud dormancy in woody plants and postpone bud germination [[57]14]. In contrast to ABA, GA may be linked to the release of endodormancy [[58]4]. Research indicates that in numerous woody species, the dynamic levels of GA are typically inversely correlated with ABA; GA levels decline during dormancy induction and rise again upon dormancy release. Exogenous applications of GA, including GA3 and GA4, expedite the release of bud dormancy and enhance bud germination [[59]13]. Furthermore, studies have demonstrated an interaction between ABA and GA in the regulation of the bud dormancy cycle [[60]15]. Additional plant hormones have also been implicated in the bud dormancy process. Notably, in the later stages of bud dormancy release in poplar, genes associated with auxin and cytokinin synthesis show upregulated expression, indicating their potential significance in bud break [[61]16]. In pears, leaf abscission triggers auxin release from the buds, which in turn promotes premature bud release. Conversely, the exogenous application of high concentrations of auxin analogs may inhibit this auxin outflow, thereby impeding flower bud germination [[62]17]. In poplar, the heterologous expression of AaRVE1 may modulate bud dormancy and germination by elevating cytokinin levels [[63]18]. Ethylene signaling plays a vital role in the initiation of bud dormancy release in grapevines [[64]19]. Brassinosteroids (BRs) and jasmonic acid (JA) facilitate the release of dormancy in pear buds. Under conditions of inadequate low-temperature accumulation, both 24-epibrassinolide (EBR) and methyl jasmonic acid (MeJA) can enhance pear bud break [[65]20]. In grape buds exposed to cold treatment, the expression of the salicylic acid (SA) biosynthesis gene (ICS2) and the endogenous SA concentration both increase. Furthermore, this expression correlates with the percentage of bud germination, indicating a potential role for the SA signaling pathway in the release of dormancy in grapes [[66]21]. Walnuts are extensively cultivated woody oilseed species of considerable economic importance [[67]22]. Investigating the germination mechanisms of dormant walnut buds and identifying the key factors regulating bud break and development are essential for developing effective strategies to mitigate LSFs damage and for breeding new LSFs-resistant varieties. However, current research on walnuts predominantly emphasizes nut quality, processing, nutritional components, stress resistance, and overall growth and development, while studies specifically examining the break and growth of dormant buds remain scarce. This study utilizes the protogynous walnut variety W13, characterized by an earlier dormancy break in spring, as the research material. It investigates four critical developmental stages of walnut buds: dormancy, germination, budding, and leafing, employing transcriptome and hormone metabolism sequencing analyses. The objective is to clarify the key biological processes and signaling pathways involved in the various phases of bud germination by analyzing gene expression modules, differentially expressed genes (DEGs), and differentially accumulated hormones (DAHs) at distinct developmental stages of walnut buds. Additionally, it seeks to identify potential regulatory genes and key hormones at each transition stage, thereby providing a foundation for a more comprehensive understanding of the mechanisms governing dormancy and development in walnut buds. Materials and methods Plant materials W-13, a second-generation walnut variety of Jinboxiang 1, was selected from the walnut germplasm resource nursery at the Research Institute of Pomology, Shanxi Agricultural University. This nursery is situated in the hilly region of the Loess Plateau (112° 32’ E, 37° 32’ N) at an altitude of 820–900 m. The area experiences an annual average temperature of 10.6 °C, with extreme high and low temperatures of 38.5 °C and − 23.6 °C, respectively. The frost-free period lasts between 160 and 180 days, while the region receives approximately 2300 h of sunshine annually. Annual precipitation ranges from 400 to 600 mm, and the soil is characterized as alkaline sandy loam. In March 2023, six healthy 10-year-old walnut trees, exhibiting uniform growth and free from diseases and pests, were selected for sampling. Walnut buds from the current year’s branches located on the south, east, north, and west sides of the trees were collected at four critical developmental stages: dormancy (W13-1, March 4), germination (W13-2, March 11), budding (W13-3, March 17), and leafing (W13-4, March 24). After removing the outermost scales, the buds were homogenized and divided into three portions, which were then preserved in liquid nitrogen for rapid freezing. A total of 12 samples were subsequently transported on dry ice to Metware Company (China) for transcriptomic and hormone metabolomic analyses. Transcriptome sequencing RNA was extracted from walnut bud samples, and its integrity, along with potential DNA contamination, was evaluated via agarose gel electrophoresis. RNA purity was assessed using a NanoPhotometer spectrophotometer (IMPLEN, CA, USA), while concentration was precisely quantified with a Qubit 4.0 fluorometer (Life Technologies, CA, USA). The completeness of RNA was further analyzed with a Qsep400 bioanalyzer (BiOptic, Jiangsu, China). Following confirmation of high-quality RNA, mRNA enrichment was conducted, and fragmentation buffer was introduced to cleave the RNA into shorter fragments. These fragments served as templates for the synthesis of first-strand cDNA using random hexamers. Subsequently, a buffer, dNTPs, and DNA polymerase I were added to facilitate the synthesis of second-strand cDNA, which was purified using magnetic beads designed for DNA purification. The resulting double-stranded cDNA underwent end repair, A-tailing, and ligation of sequencing adapters, followed by size selection using magnetic beads. PCR enrichment was then performed to generate the final cDNA library. Upon completion of library construction, preliminary quantification was conducted using the Qubit dye method, and the insert size was assessed with a fragment analyzer. Following confirmation that the insert size met specifications, accurate quantification of the effective library concentration was performed using Q-PCR. A library was deemed qualified if its effective concentration exceeded 2 nM, after which it was sequenced on the Illumina platform. The sequencing data were processed to eliminate adapter sequences and low-quality reads. The resulting clean reads were then aligned to the reference genome Juglans regia Walnut 2.0 using HISAT2 (version 2.2.1) [[68]23, [69]24]. Gene expression levels were normalized using fragments per kilobase of transcript per million mapped reads (FPKM) values, with featureCounts (version 2.0.3) utilized to compute these FPKM values for the genes [[70]25]. The correlation among biological replicates was assessed using pearson correlation coefficient and principal component analysis (PCA). Differential gene expression analysis between sample groups was conducted using DESeq2 (version 1.22.1) [[71]26], yielding a list of DEGs across the two biological conditions. The selection criteria included a minimum fold change of 2 and a false discovery rate (FDR) of less than 0.05. Gene clustering and enrichment analysis K-means gene clustering analysis was performed on standardized FPKM values of gene expression in R to identify groups of genes exhibiting similar expression patterns. Gene Ontology (GO) enrichment analysis was conducted using the clusterProfiler (version 4.6.0) and GO.db (version 3.16.0) packages to elucidate the biological processes, molecular functions, and cellular components associated with the genes [[72]27]. Furthermore, KEGG enrichment analysis was performed to ascertain the pathways in which the genes demonstrate significant enrichment. A scatter plot was employed to visualize the outcomes of GO and KEGG enrichment analyses, assessing term or pathway enrichment based on the rich factor, q-value, and the number of genes associated with each term or pathway. The rich factor is defined as the ratio of DEGs within a term or pathway to the total number of annotated genes. A higher rich factor indicates greater enrichment, whereas a lower q-value reflects enhanced significance of enrichment. The top 20 terms or pathways with the most pronounced enrichment were highlighted. Gene co-expression analysis Weighted correlation network analysis (WGCNA) was conducted utilizing the WGCNA package (version 1.71) to analyze gene expression data. The varFilter function from the genefilter package in R was employed to eliminate low-expressed genes and those with stable expression across all samples, thereby improving the precision of network construction. The optimal weight parameter, powerEstimate, was established at 18, the minimum gene count for modules, minModuleSize, was set to 50, and the merging cutoff height for modules, mergeCutHeight, was configured to 0.25. Cytoscape was employed to assess gene connectivity (degree values) within the identified modules, facilitating the identification of hub genes that may serve critical regulatory functions [[73]28]. The co-expression network was subsequently visualized to illustrate the interrelations among genes, reflecting the magnitude of their degree values. Hormone metabolome analysis Walnut bud samples were ground into a powder using liquid nitrogen. A 50 mg aliquot of the plant material was placed in a 2 mL plastic microtube, frozen in liquid nitrogen, and then dissolved in 1 mL of a methanol/water/formic acid mixture (15:4:1, V/V/V). An internal standard solution (10 µL at 100 ng/mL) was added to the extract for quantification purposes. The mixture was vortexed for 10 min and subsequently centrifuged for 5 min at 12,000 rpm and 4 °C. The supernatant was transferred to clean plastic microtubes, evaporated to dryness, and reconstituted in 100 µL of 80% methanol (V/V). The resulting solution was filtered through a 0.22 μm membrane filter prior to LC-MS/MS analysis (QTRAP 6500+, SCIEX, Shanghai, China). The walnut sample extracts were analyzed using Ultra Performance Liquid Chromatography (UPLC) coupled with Tandem Mass Spectrometry (MS/MS) for data collection. Qualitative analysis of the mass spectrometry data was performed using a database generated from standard compounds. Quantitative hormone analysis was carried out using Multiple Reaction Monitoring on a triple quadrupole mass spectrometer. Data acquisition utilized Analyst 1.6.3 software, while Multiquant 3.0.3 software facilitated the quantification of all metabolites. The correlation among biological replicates was assessed using pearson correlation coefficient and PCA. The criteria for identifying DAHs included a minimum fold change of 2 and a P-value of less than 0.05. These DAHs were subsequently mapped to KEGG metabolic pathways. Illustrate the interplay between metabolites and genes through a network graph, pinpointing DEGs and DAMs with an absolute pearson correlation coefficient surpassing 0.8 and a p-value below 0.05 within each pathway for depicting the correlation outcomes. Results Transcriptome sequencing analysis of walnut bud germination process To elucidate the alterations in gene expression and the associated biological processes and signaling pathways during walnut bud development, we performed transcriptome sequencing on bud samples from the early-germinating, protogynous walnut variety W13 across various developmental stages. This developmental timeline was categorized into three stages: Stage One, Stage Two, and Stage Three (Fig. [74]1A). A total of 12 samples produced 83.39 G of high-quality clean data, with individual samples ranging from 6.11 G to 7.57 G and a sequencing depth exceeding 10× (Fig. [75]1B). The read mapping to the reference genome (Juglans regia) varied from 96.22 to 97.18%, with unique mapped reads constituting 92.46–93.35% (Fig. [76]1B). The sequencing data exhibited a Q20 value of approximately 98% and a Q30 value exceeding 94%, with GC content between 45.47% and 45.89% (Fig. [77]1B). Correlation analysis among the samples revealed that the correlation coefficient for different biological replicates was greater than 0.95 (Fig. [78]1C). PCA demonstrated that biological replicates clustered together (Fig. [79]1D). Collectively, these findings confirm the acquisition of high-quality sequencing data, facilitating subsequent data analysis and investigation. Fig. 1. [80]Fig. 1 [81]Open in a new tab Schematic representation of walnut bud sample collection across various developmental stages, alongside quality control metrics for transcriptome sequencing data. (A) Walnut bud samples categorized by developmental stage. (B) Transcriptome sequencing data statistical outcomes. (C) Correlation analysis among transcriptome sequencing samples. (D) Principal component analysis of transcriptome sequencing results. W13-1, W13-2, W13-3, and W13-4 denote the bud samples of the W13 walnut variety at the dormancy, germination, budding, and leafing stages, respectively, each comprising three biological replicates To elucidate the expression patterns of walnut genes across various samples, K-means clustering analysis was performed alongside gene expression data. As illustrated in Figure [82]S1, the walnut genes were distinctly classified into four groups based on their expression profiles. Subclass 4 comprised the largest group, containing 9,304 genes that showed peak expression levels in leafing buds. This was followed by Subclass 2 with 8,418 genes, Subclass 3 with 8,031 genes, and Subclass 1 with 6,859 genes, each exhibiting the highest expression levels in budding buds, dormant buds, and germinating buds, respectively. The maintenance of elevated expression levels within these groups during specific developmental stages indicates their potential roles as critical components in the regulatory networks of these phases. Consequently, further investigation was conducted into the biological processes and pathways associated with these genes. The analysis revealed that the Subclass 3 genes, highly expressed in dormant buds, was significantly enriched in biological processes such as aromatic amino acid family biosynthetic process, dicarboxylic acid biosynthetic process, aromatic amino acid family metabolic process, salicylic acid metabolic process, salicylic acid catabolic process, and response to heat, as well as pathways related to the biosynthesis of secondary metabolites, alpha-Linolenic acid metabolism, phenylpropanoid biosynthesis, and galactose metabolism (Figure [83]S2). The Subclass 1 genes, predominantly expressed in germinating buds, demonstrated significant enrichment in processes related to cytoskeleton organization, microtubule cytoskeleton organization, microtubule-based process, embryonic meristem initiation, microtubule binding, and regulation of anatomical structure size, in addition to the ATP-dependent chromatin remodeling pathway (Figure [84]S2). The Subclass 2 genes, with high expression in budding buds, was significantly enriched in processes related to the response to reactive oxygen species, response to oxygen levels, cellular response to oxygen levels, phenylpropanoid biosynthetic process, and secondary metabolite biosynthetic process, alongside pathways involved in biosynthesis of secondary metabolites and phenylpropanoid biosynthesis (Figure [85]S2). Lastly, Subclass 4, which showed high expression in leafing buds, was significantly enriched in biological processes linked to shoot system morphogenesis, cytoplasmic translation, and polysome, as well as pathways involving ribosome, proteasome, and ribosome biogenesis in eukaryotes (Figure [86]S2). Identification and analysis of DEGs at different stages of walnut bud germination To elucidate the gene expression alterations associated with various stages of walnut bud germination, we conducted a comprehensive identification and analysis of DEGs. We examined three developmental stages, uncovering 2,066 DEGs in Stage One (981 upregulated and 1,085 downregulated), 2,369 DEGs in Stage Two (1,560 upregulated and 809 downregulated), and 1,770 DEGs in Stage Three (488 upregulated and 1,282 downregulated) (Fig. [87]2A, Data [88]S1). Gene enrichment analysis indicated that DEGs implicated in the transition from dormancy to germination are associated with biological processes such as cellular response to ethylene stimulus, ethylene-activated signaling pathway, shoot system morphogenesis, hormone binding, and secondary metabolite biosynthetic process (Fig. [89]2B). Key pathways identified include plant hormone signal transduction, phenylpropanoid biosynthesis, biosynthesis of secondary metabolites, and MAPK signaling pathway-plant (Fig. [90]2C). During the transition from germination to budding, DEGs are linked to the ethylene-activated signaling pathway, cellular response to ethylene stimulus, phosphorelay signal transduction system, and response to jasmonic acid, alongside pathways such as plant hormone signal transduction, biosynthesis of secondary metabolites, cutin, suberine and wax biosynthesis, and phenylpropanoid biosynthesis (Fig. [91]2B, C). In the transition from greening to leafing, DEGs are associated with processes including cellular response to ethylene stimulus, ethylene-activated signaling pathway, response to oxygen levels, secondary metabolite biosynthetic process, and phosphorelay signal transduction system, as well as pathways like MAPK signaling pathway-plant, plant hormone signal transduction, biosynthesis of secondary metabolites, flavone and flavonol biosynthesis (Fig. [92]2B, C). Importantly, DEGs across all three stages were found to participate in four biological processes—cellular response to ethylene stimulus, phenylpropanoid metabolic process, ethylene-activated signaling pathway, and monooxygenase activity; and seven pathways, including plant hormone signal transduction, flavone and flavonol biosynthesis, biosynthesis of secondary metabolites, and MAPK signaling pathway. This underscores the significance of these biological processes and pathways in walnut bud germination. A comparative analysis of DEGs revealed that 461 genes were consistently differentially expressed across all stages (Fig. [93]2D), encompassing biological processes such as phenylpropanoid metabolic process, phenylpropanoid biosynthetic process, secondary metabolite biosynthetic process, and response to oxygen levels, along with pathways like MAPK signaling pathway-plant, cutin, suberine and wax biosynthesis, and biosynthesis of secondary metabolites (Fig. [94]2E, F). These DEGs are likely critical regulatory and functional elements in walnut bud development and environmental adaptation, providing insight into the internal mechanisms governing bud germination. Fig. 2. [95]Fig. 2 [96]Open in a new tab Differentially expressed genes (DEGs) analysis among walnut buds at diverse developmental stages. (A) Identification of DEGs, utilizing DESeq2 with a fold change threshold of ≥ 2 and a false discovery rate (FDR) < 0.05. (B) GO enrichment analysis for DEGs. (C) KEGG pathway enrichment analysis for DEGs. (D) Venn diagram illustrating the overlap of DEGs across developmental stages. (E) GO enrichment analysis for commonly expressed DEGs among the three stages. (F) KEGG enrichment analysis for commonly expressed DEGs across the three stages Identification of key genes at different developmental stages of walnut bud germination To elucidate the gene expression modules associated with distinct developmental stages during walnut bud germination, we constructed a gene co-expression network and identified key regulatory factors using sequencing data from 12 samples. Our analysis identified a total of 18 gene expression modules, with a particular emphasis on six (Fig. [97]3). Notably, genes within the Green, Pink, Blue, and Yellow modules displayed significantly elevated expression during the dormancy, germination, budding, and leafing stages, respectively (Fig. [98]3C). In contrast, genes in the Red and Tan modules exhibited markedly lower expression during the dormancy and leafing stages (Fig. [99]3C). Functional enrichment analysis revealed that genes in the Green module were significantly associated with biological processes such as cellular response to ethylene stimulus, ethylene-activated signaling pathway, negative regulation of response to external stimulus, and phenylalanine ammonia-lyase activity (Figure [100]S3), suggesting a role in maintaining dormant buds. Similarly, genes in the Red module were enriched in biological processes related to auxin binding, tracheary element differentiation, xylem development, and hormone binding (Figure [101]S3), indicating their involvement in bud break. Genes within the Blue module were significantly enriched in processes including fatty acid biosynthetic process, photosynthesis, signaling receptor binding, and wax metabolic process (Figure [102]S3), which are likely pertinent to the budding of buds. The Yellow module genes were significantly enriched in biological processes related to leaf development, such as xyloglucan metabolic process, shoot system morphogenesis, positive regulation of auxin mediated signaling pathway, and leaf morphogenesis (Figure [103]S3). Finally, the Tan module genes were significantly enriched in biological processes such as the MAPK cascade, ubiquitin binding, regulation of abscisic acid-activated signaling pathway, negative regulation of long-day photoperiodism, flowering, and protein autophosphorylation (Figure [104]S3). Fig. 3. [105]Fig. 3 [106]Open in a new tab Weighted correlation network analysis (WGCNA) of walnut bud samples at different developmental stages. (A) Hierarchical clustering dendrogram of gene modules, where genes with analogous expression profiles are grouped, with distinct colors indicating different modules. (B) Heatmap illustrating gene clustering within modules, with darker colors indicating stronger correlations. (C) Gene expression patterns of modules specifically expressed at particular developmental stages, displayed via heatmaps and bar charts, where red signifies high expression and green low expression We calculated the degree values of gene connections within the modules and constructed a co-expression network for the top 50 genes (Fig. [107]4). Initially, we annotated the hub genes, those with the highest connectivity, across the six modules. In the Green module, which demonstrated significantly elevated expression during bud dormancy, the hub gene identified was JrDIR20. DIR proteins are integral to the biosynthesis of lignans and lignin in plants and are upregulated under various stress conditions [[108]29]. Conversely, the hub gene for the Red module, which exhibited significantly low expression during dormancy, was JrSUB1. Its Arabidopsis homolog, AtSUB1, participates in the cryptochrome signaling pathway and serves as a regulatory factor in the phytochrome signaling pathway. In the Pink module, characterized by significantly high expression during the germination stage, the hub gene was JrSEOR1, which encodes a protein essential for the formation of phloem filaments. The hub gene in the Blue module, which demonstrated significantly high expression during the greening stage, was JrSAUR14, potentially linked to adaptive growth and development in plants. The hub genes in the Yellow and Tan modules, which show high and low expression during leaf unfolding, respectively, were Jr09_07010 and Jr01_20560; the functions of their homologs remain to be characterized. Recognizing the pivotal role of transcription factors in plant growth and development, we identified transcription factor genes within each network (Fig. [109]4). The Green module encompassed nine transcription factor genes (four ERF, four WRKY, and one MYB gene). The Red module contained eight transcription factor genes (two GRAS, two MYB, one Dof, one WRKY, one C2H2, and one C3H). The Pink module included five transcription factors from various families (GRAS, TCP, bHLH, C3H, and ERF). The Blue module identified three transcription factors from different families (WOX, NAC, and C2H2). The Yellow module had the highest number of transcription factors, totaling twelve (three C2H2, two TCP, one MYB, one WRKY, one ZF-HD, one ERF, one C3H, one Dof, and one bHLH). The Tan module identified two transcription factors (one bZIP and one MYB-related). The substantial presence of transcription factor genes at the core of these co-expression networks underscores their critical regulatory roles during the germination process of buds. Fig. 4. [110]Fig. 4 [111]Open in a new tab Gene co-expression networks. The co-expression networks constructed from the top 50 genes based on degree values from each module, where darker colors and larger shapes represent higher degree values Variations in hormone concentrations throughout the various developmental phases of walnut bud germination Gene enrichment analysis results indicate that the bud germination process is associated with alterations in various hormone signaling pathways. Consequently, we performed hormone metabolomics analysis on samples from distinct developmental stages. As illustrated in Fig. [112]5, biological replicates clustered together in the PCA. Notably, the correlation among sample data in the second stage was 0.93, while correlations among the other replicates exceeded 0.99. We identified a total of 9 major categories and 109 subcategories of plant hormones, including 3 forms of ABA, 27 auxins, 40 cytokinins, 1 ethylene type, 18 GAs, 11 JA types, 1 melatonin, 6 SAs, and 2 strigolactones (SLs) (Table [113]S1). Ultimately, we recognized 8 major categories and 74 subcategories of hormones, comprising 2 ABA types, 16 auxins, 27 cytokinins, 1 ethylene type, 12 GAs, 9 JAs, 6 SAs, and 1 SL, while the other hormones were undetectable in all samples (Fig. [114]5C). In all samples, the auxins TRA and TRP exhibited the highest concentrations, averaging 43.02 and 21.15 µg/g, respectively. The average concentration of auxin Indole also reached 5 µg/g. SA and ABA concentrations were notably high, with salicylic acid levels recorded at 5.47 µg/g for Phe, 2.56 µg/g for SAG, and 1.76 µg/g for SA, while abscisic acid levels for ABA-GE and ABA were 3.79 and 0.43 µg/g, respectively. The jasmonic acids JA-ILE, JA, and OPDA, along with gibberellin GA29, demonstrated moderate concentrations of 0.9, 0.64, 0.63, and 0.34 µg/g, respectively (Data [115]S2). In contrast, the concentrations of the other identified hormones were relatively low. The patterns of hormone levels and clustering across developmental stages are depicted in Fig. [116]5C. Subsequently, we identified DAHs at various developmental stages (Data [117]S3). The most significant hormonal changes occurred between early dormancy and the germination stage, with 19 types identified. This was followed by 8 types from germination to budding and 4 types from budding to leafing (Fig. [118]6A-C). In stage one, the ethylene precursor ACC and three types of cytokinins were upregulated, while three auxins, six cytokinins, two GAs, three SAs, and one ABA were downregulated. In stage two, six cytokinins were upregulated, whereas the ethylene precursor ACC and JA were downregulated. In stage three, one cytokinin was downregulated, while three GAs were upregulated. A comparison of differential hormones across the three stages indicated that the cytokinin 2MeScZR displayed varying levels in each stage (Fig. [119]6D). Additionally, in the first and second stages, metabolites with differing levels included the ethylene precursor ACC and three cytokinins: DZ, tZR, and DHZROG (Fig. [120]6D). Fig. 5. [121]Fig. 5 [122]Open in a new tab Hormone metabolomic profiling of walnut bud samples at various developmental stages. (A) Principal component analysis of hormone metabolomic data. (B) Correlation analysis of hormone metabolomic samples. (C) Heatmap displaying hormone content across samples Fig. 6. [123]Fig. 6 [124]Open in a new tab Differentially accumulated hormone analysis across developmental stages. (A-C) Heatmap visualizations of hormone content variations among different samples. (D) Venn analysis of hormone types exhibiting varying concentrations across distinct developmental stages Moreover, we performed a correlation analysis between the gene expression modules identified via WGCNA and hormone levels. The analysis revealed that the Green gene module, which is markedly expressed during bud dormancy, exhibited significant negative correlations with three auxins (TRP, IAA-Leu, Indole), two CKs (iPRMP, tZRMP), and one SA (Phe). Conversely, it demonstrated significant positive correlations with three JAs (JA-Val, JA-ILE, JA) and one SA (SAG). The Red gene module, with low expression during bud dormancy, showed significant negative correlations with one CK (tZOG), two JAs (JA-Val, JA-ILE), and three SAs (2-Coumarate, SA, SAG), while exhibiting significant positive correlations with one auxin (TRP) and two SAs (t-CA, Phe). The Pink module, which is prominently expressed in germinating buds, displayed significant negative correlations with three CKs (tZR, DHZROG, DZ) and a significant positive correlation with one ethylene precursor (ACC). The Blue module, expressed abundantly in budding buds, was significantly positively correlated with one auxin (ICAld) and two CKs (cZR, IPR). The Yellow module, highly expressed in leafing buds, demonstrated significant negative correlations with one JA (JA) and significant positive correlations with two auxins (TRP, IAA-Leu), three CKs (iPRMP, tZRMP, IP), and two GAs (GA9, GA6). Lastly, the Tan module, characterized by low expression in leafing buds, exhibited significant negative correlations with three GAs (GA9, GA6, GA29) (Figure [125]S4). To elucidate the regulatory functions of hormones and pathway genes during walnut bud germination and leaf development, we analyzed the fluctuations in hormone levels and gene expression within plant hormone signaling pathways (Figs. [126]7 and [127]8, and [128]9). Our findings revealed that in the initial phase of walnut bud development, 24 gene groups encompassing 91 DEGs were implicated across three categories of differential hormones and eight hormone signaling pathways (Fig. [129]7). Notably, ABA, which is known to inhibit bud dormancy release in fruit trees, exhibited a decrease in content. Furthermore, cytokinin Tz levels increased while cytokinin DZ levels decreased, alongside a reduction in SA content. In terms of gene expression, DEGs within the auxin signaling pathway, such as AUX1, TIR1, and the GH3 module, were consistently upregulated. In contrast, DEGs within the AUX/IAA and SAUR modules displayed mixed regulation. The cytokinin signaling pathway showed predominant upregulation of DEGs such as AHP, B-ARRs, and A-ARR, with the exception of one B-ARR. In the gibberellin signaling pathway, three GID1s were downregulated, while one gene was upregulated; most DEGs in the DELLA and TF categories were upregulated. Within the ABA signaling pathway, all three PP2C genes were downregulated, while one ABF gene was upregulated. The ethylene signaling pathway revealed downregulation of CTR1 and upregulation of ERF1/2. In the brassinosteroid pathway, each category of DEGs, except for BAK1s, showed consistent trends: eight BRI1s, one BSK, and one CYCD3 were upregulated, while six TCH4s were downregulated. In the JA signaling pathway, one JAZ gene was downregulated, whereas four MYC2s were upregulated and two were downregulated. Lastly, in the SA signaling pathway, one TGA gene was downregulated. During the second stage of walnut bud development, the number of DEGs reached its peak, although only the level of DZ was upregulated, while JA content declined (Fig. [130]8). Importantly, during this stage, each class of DEGs in the auxin signaling pathway exhibited a consistent expression pattern, with one AUX1, one ARF, and three GH3 genes downregulated, while one TIR1, two AUX/IAAs, and nine SAURs were upregulated. All six downstream A-ARRs in the cytokinin signaling pathway were upregulated. In the gibberellin pathway, most GID1s and TFs were downregulated. The ABA signaling pathway exhibited consistent trends in gene sets, with one PYR/PYL gene downregulated, six PP2Cs upregulated, and three ABFs downregulated. In the ethylene pathway, one CTR1 was upregulated, while two ERF1/2 genes were downregulated. In the brassinosteroid pathway, downstream genes TCH4 and CYCD3 were both upregulated. The jasmonic acid pathway showed a predominance of downregulated JAZ and MYC2 genes. The third stage of walnut bud development involved the fewest DEGs, with no significant changes in hormone levels (Fig. [131]9). In the auxin signaling pathway, one AUX1 and three of four AUX/IAAs, one ARF, and one GH3 were upregulated, whereas one AUX/IAA and three SAURs were downregulated. The cytokinin signaling pathway included only two DEGs, with B-ARR upregulated and A-ARR downregulated. The gibberellin pathway encompassed expression changes in GID1s, DELLA, and TFs. In the ABA signaling pathway, one PYR/PYL was upregulated, four PP2Cs were downregulated, and one ABF was downregulated. In the ethylene signaling pathway, four CTR1 and two EBF1/2 genes were downregulated, while three ERF1/2 genes were upregulated. In the brassinosteroid pathway, all BAK1s and four of five BRI1s were downregulated, alongside four TCH4s, while two CYCD3s were upregulated. In the jasmonic acid pathway, one JAZ and two MYC2s were downregulated, whereas four MYC2s were upregulated. Finally, in the SA signaling pathway, only TGA changes were observed, with two genes downregulated and one upregulated. Fig. 7. [132]Fig. 7 [133]Open in a new tab Analysis of DEGs and DAHs associated with phytohormone signaling pathways in W13-1 and W13-2 samples. The gene modules and hormones involved are marked in blue, with heatmaps visualizing hormone content and gene expression Fig. 8. [134]Fig. 8 [135]Open in a new tab Analysis of DEGs and DAHs related to phytohormone signaling pathways in W13-2 and W13-3 samples. The gene modules and hormones involved are marked in blue, with heatmaps depicting hormone content and gene expression Fig. 9. [136]Fig. 9 [137]Open in a new tab Analysis of DEGs and DAHs pertinent to phytohormone signaling pathways in W13-3 and W13-4 samples. The gene modules and hormones involved are marked in blue, with heatmaps illustrating hormone content and gene expression To explore the expression profiles and potential functions of genes related to plant hormone signaling during walnut bud development, we performed an extensive analysis of DEGs across three developmental stages. Visualization and clustering of expression profiles revealed 21 DEGs distributed among nine distinct gene modules, which were classified into four groups based on their expression patterns (Fig. [138]10). The first group, consisting of three TCH4s and one GID1, displayed a trend of sequential downregulation, upregulation, and subsequent downregulation throughout development, with peak expression occurring in the budding stage and minimal expression in the germination stage. The second group, which includes three PP2Cs, two DELLAs, two MYC2s, and two BAK1s, exhibited a similar expression trend to the first group but with variations in expression abundance. The third cluster, comprising one GID1 and one ERF1/2, demonstrated a temporal expression profile characterized by initial upregulation followed by downregulation, and subsequently another upregulation, culminating in peak expression at the germination stage. The fourth cluster, containing two DELLAs, two MYC2s, one GH3, and one BRI1, exhibited a parallel expression trajectory to the third cluster, though it displayed lower expression levels in the dormant buds. Fig. 10. [139]Fig. 10 [140]Open in a new tab Expression patterns of genes associated with plant hormone signaling pathways that exhibit differential expression across various developmental stages of walnut bud. A cluster analysis was conducted to group genes with similar expression profiles. The corresponding gene modules for each gene are denoted in parentheses Correlation network between hormones and genes To investigate the relationship between hormones and genes during walnut bud germination, we calculated and analyzed the correlations among various hormones and pathway-related genes. As shown in Fig. [141]11, a total of 36 genes were identified as being highly correlated with five types of hormones. Notably, the majority (27 genes) were associated with SA, followed by two genes linked to 2-Coumarate. Additionally, seven genes were linked to tZOG, two genes to JA, and two genes to GA9. Notably, Jr14_14070 was positively correlated with SA and 2-Coumarate, but negatively correlated with GA9. Furthermore, Jr11_24600 showed positive correlations with tZOG, SA, and 2-Coumarate. The discovery of these genes can provide a basis for examining the mechanisms of synthesis and metabolism of related hormones. Fig. 11. [142]Fig. 11 [143]Open in a new tab Correlation networks of DEGs and DAHs. Hormones are represented by green squares and genes by red circles; solid lines indicate positive correlations while dashed lines denote negative correlations Discussion Walnuts are commercially significant woody oilseed trees, with their kernel-derived oil being nutritionally rich and economically valuable. However, LSFs can adversely affect flower buds, young shoots, floral structures, and immature fruits, potentially resulting in reduced yields. This study investigates the developmental stages of spring dormant buds in the protogynous walnut variety W13. We performed transcriptomic, hormone metabolomic, and multi-omics correlation analyses, uncovering distinct gene expression patterns and hormonal fluctuations linked to bud dormancy, germination, budding, and leafing. Our findings elucidate the biological processes and signaling pathways governing transitions between developmental stages and highlight key regulatory genes and hormones involved in walnut bud germination. Gene clustering analysis is a pivotal bioinformatics technique that categorizes genes according to similarities in their expression profiles, thereby elucidating their functions and regulatory networks. Identifying gene modules with elevated expression at particular developmental stages offers insights into the developmental processes and regulatory mechanisms of organisms. In this study, we utilized K-means clustering analysis alongside gene expression data from walnut buds at various stages, resulting in the identification of four gene sets specifically expressed during these stages, which were further characterized regarding their associated biological processes and pathways. Aromatic amino acids are vital biosynthetic precursors in plants, contributing to hormone synthesis and potentially influencing dormancy and growth [[144]30]. SA is a central molecule in plant stress responses and defense, regulating numerous physiological processes [[145]31]. It may play a role in the formation and release of dormant buds, aiding plants in maintaining physiological stability under adverse conditions. Heat stress can alter levels of plant hormones, such as ABA and GA, which are integral to bud dormancy and germination [[146]32]. Secondary metabolites play crucial roles in plant defense, adaptation, and growth. In dormant buds, these metabolites (including alkaloids, flavonoids, and phenolic compounds) assist in plant survival under unfavorable environmental conditions [[147]33]. The organization of the cytoskeleton is essential for cell morphology and division; during bud germination, cytoskeletal reorganization is necessary for the formation and expansion of new cells [[148]34]. The organization of the microtubule cytoskeleton is essential for cell division and morphological changes; during bud germination, the reorganization of microtubules can facilitate cell expansion and differentiation, ensuring normal bud development [[149]35]. During the budding process, reactive oxygen species (ROS) frequently act as signaling molecules regulating plant growth and development. These reactions can promote cell division and expansion, aiding plants in coping with environmental stressors [[150]36]. Secondary metabolites are fundamental to plant growth, development, and interactions with the environment. In the budding phase, the synthesis of secondary metabolites enhances plants’ ability to adapt to environmental changes and resist pests and diseases. Phenylpropanoid compounds are significant components of plant secondary metabolism and play a role in the plant defense mechanism; their synthesis during bud greening can bolster the plant’s resistance to environmental stress [[151]37]. Bud morphogenesis refers to the developmental process of plant buds, including differentiation, growth, and morphological establishment. This process is critical for the morphological structure and growth development of the plant. During leaf unfolding, the growth and differentiation of buds require precise spatiotemporal regulation to ensure that the leaves can unfold properly. By examining gene expression variations in walnut buds across distinct developmental stages, we identified DEGs essential for elucidating key regulatory factors, biological processes, and signaling pathways pertinent to developmental transitions. Our analysis revealed that four biological processes—cellular response to ethylene stimulus, phenylpropanoid metabolic process, ethylene-activated signaling pathway, and monooxygenase activity—alongside seven signaling pathways, including plant hormone signal transduction, flavone and flavonol biosynthesis, secondary metabolite biosynthesis, and the MAPK signaling pathway, were significantly enriched among these DEGs throughout various bud developmental stages. Current investigations into ethylene’s role in plant dormancy predominantly address seed dormancy and germination, highlighting ethylene’s critical regulatory function in seed germination and seedling establishment [[152]38]. Additionally, ethylene interacts with vital regulatory hormones such as ABA and GA, which are influential in bud dormancy and sprouting in woody species [[153]38]. Research on grapevines has underscored ethylene’s importance in breaking grape bud dormancy [[154]19, [155]39]. In young leaves, ethylene and downstream ERF proteins serve as principal regulators of leaf growth inhibition, coordinating cell division and expansion [[156]40]. Consequently, further investigation into ethylene’s impact on walnut dormant buds and leaf development may uncover novel regulatory mechanisms in walnut growth. The phenylpropanoid pathway constitutes a major route in plant secondary metabolism, facilitating the synthesis of various compounds vital for plant defense, cell signaling, and overall growth and development [[157]37]. Secondary metabolites, including flavonoids, are also critical for plant growth, defense mechanisms, and environmental interactions [[158]41]. The MAPK signaling pathway is integral to numerous biological processes in plants, regulating responses to environmental stimuli, cell cycle progression, and programmed cell death [[159]42]. Furthermore, a comparative analysis of DEGs across the three developmental stages revealed that 461 genes were consistently differentially expressed at all stages. These genes are likely pivotal in walnut bud development, primarily involved in biological processes and signaling pathways related to secondary metabolite biosynthesis, the MAPK signaling pathway, and the biosynthesis of leaf-associated compounds, including cutin, suberine, and wax. To investigate the key gene modules governing specific stages of tissue development, we utilized WGCNA to construct a gene co-expression network. This analysis revealed a total of 18 gene expression modules, from which we selected six that demonstrated stage-specific high or low expression during walnut bud development. We also concentrated on hub genes and highly connected transcription factor genes within these modules. Bud dormancy is a critical adaptive mechanism for plants facing adverse environmental conditions, and the hub gene JrDIR20 (a stress-related gene) in the Green module, which is markedly expressed during dormancy, is likely integral to this process. Light conditions are vital in initiating, maintaining, and releasing dormancy in perennial trees; for instance, short photoperiods can induce bud dormancy in certain species. Thus, the hub gene JrSUB1, which regulates the phytochrome signaling pathway within the Red module, may also significantly influence walnut bud dormancy. Current literature indicates that transcription factors from various families are implicated in breaking plant bud dormancy and facilitating shoot and leaf development. Our study identified a considerable number of transcription factor genes situated at key positions within the co-expression network across different modules, presenting promising candidates for future research aimed at breaking dormancy or promoting development in walnut buds. It is noteworthy that four ERF transcription factor genes were identified in the Green module. Considering the significant enrichment of genes in this module in the ethylene signaling pathway, these four ERF genes may play critical roles within the module. Additionally, two GRAS genes were positioned at the core of the network in the Red module; given their essential regulatory functions in plant growth and development [[160]43], these genes also represent valuable candidates for subsequent functional studies. Plant hormones are essential for regulating bud dormancy and germination, with current evidence suggesting that most hormones are implicated in these developmental processes. This study assessed hormone levels at various stages of walnut bud development, finding that the average concentrations of auxins (TRA, TRP, Indole), SAs (Phe, SAG, SA), ABAs (ABA-GE, ABA), JAs (JA-ILE, JA, OPDA), and GA (GA29) remained relatively high throughout these stages, indicating their significant roles in bud development. The study further identified DAHs at distinct developmental stages, underscoring the involvement of multiple hormonal changes and suggesting a synergistic interaction among various hormones during walnut bud development. Previous research on hormone effects on the bud development of woody plants has primarily concentrated on the induction, maintenance, and release of bud dormancy. ABA and GA have been identified as the principal hormones that oppositely regulate these processes [[161]12]. Elevated endogenous ABA levels are pivotal in sustaining bud dormancy, while GA is responsible for its release. The rise in GA levels in dormant buds often coincides with a decline in ABA, both contributing to bud dormancy release [[162]4]. Moreover, investigations have demonstrated that heightened endogenous cytokinin levels and external application of ethylene promote bud dormancy release [[163]18, [164]19]. In this investigation, an examination of hormonal shifts in walnut buds transitioning from dormancy to germination unveiled an escalation in the ethylene precursor ACC and three cytokinins (mT9G, tZ, iPRMP), alongside a reduction in three auxins, six cytokinins, two GAs, three SAs, and one ABA. By integrating alterations in gene expression involving DEGs associated with the cellular response to ethylene stimulus and ethylene-activated signaling pathway, it is hypothesized that ethylene may contribute to walnut bud dormancy release. The decline in ABA content during walnut bud dormancy release aligns with findings in other plant species. The divergent shifts in cytokinin levels during walnut bud dormancy release might be linked to their distinct physiological actions and specific effects. Additionally, the functions of auxins, GAs, and SAs in walnut bud dormancy release warrant further investigation. During the transition from germination to budding in walnut buds, six cytokinins were upregulated, while the ethylene precursor ACC and JA were downregulated, suggesting the pivotal role of cytokinins in early walnut bud growth, with ACC level alteration possibly signifying the conclusion of its dormancy release function. JA, essential for plant responses to environmental stress, aids in enhancing dormant bud resistance to adversity, potentially explaining its reduced presence during walnut bud transition from dormancy to normal growth. As walnut buds shift from budding to leafing, one cytokinin was downregulated, while three GAs were upregulated, highlighting the crucial involvement of GAs in cell elongation, leaf expansion, and leaf morphology in walnut buds. In addition, a comprehensive analysis of the regulatory network governing bud dormancy release and bud break in the woody model plant poplar has revealed that during the bud dormancy release phase, the expression of the ABA biosynthesis gene NCED3 diminishes, while genes associated with ABA catabolism such as CYP707A4 exhibit increased expression. In contrast, genes linked to GA biosynthesis demonstrate elevated expression during this process, whereas those involved in GA catabolism exhibit decreased expression [[165]16]. Therefore, a more in-depth investigation into hormone biosynthesis and catabolism-related genes in walnut buds, along with a clearer elucidation of their biological roles, will be instrumental in elucidating the developmental mechanism of walnut buds. Conclusion This study performed an integrative analysis of transcriptomics and hormone metabolomics on the buds of the protogynous walnut variety W13 at various developmental stages. We identified four biological processes—cellular response to ethylene stimulus, phenylpropanoid metabolic process, ethylene-activated signaling pathway, and monooxygenase activity—as well as several pathways, including plant hormone signal transduction, flavone and flavonol biosynthesis, biosynthesis of secondary metabolites, and MAPK signaling pathway, that play significant roles in the germination of walnut buds. Plant hormones such as abscisic acid, auxin, cytokinin, ethylene, gibberellin, jasmonic acid, and salicylic acid play pivotal roles in the transition of walnut buds from dormancy to leafing, modulating germination and growth through the regulation of gene expression. We also identified specific gene modules exhibiting distinct expression patterns at different developmental stages, with hub genes and transcription factors likely acting as central regulators of walnut bud germination. Furthermore, we mapped differentially expressed genes and differentially accumulated hormones across developmental stages onto the plant hormone signal transduction pathways and constructed a correlation network linking hormone levels to gene expression, thereby providing insights into the roles of genes and hormones in the germination process of walnut buds. Electronic supplementary material Below is the link to the electronic supplementary material. [166]12864_2025_11272_MOESM1_ESM.tif^ (1.7MB, tif) Supplementary Material 1. Figure 1: K-Means clustering analysis of gene expression data from walnut buds at different developmental stages. The x-axis represents the samples, while the y-axis denotes normalized expression levels. [167]12864_2025_11272_MOESM2_ESM.tif^ (2.6MB, tif) Supplementary Material 2. Figure 2: Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses of distinct gene clusters identified through K-Means clustering. (A) GO enrichment results for different gene clusters. (B) KEGG pathway enrichment results for various gene clusters. [168]12864_2025_11272_MOESM3_ESM.tif^ (3.3MB, tif) Supplementary Material 3. Figure 3: GO and KEGG pathway enrichment analyses for members of various gene modules. (A) GO enrichment results. (B) KEGG pathway enrichment results. [169]12864_2025_11272_MOESM4_ESM.tif^ (4.5MB, tif) Supplementary Material 4. Figure 4: Heatmap depicting correlations between gene modules and different hormones. The heatmap visualizes correlation results, with numbers in the squares indicating correlation strength and values in parentheses denoting p-values. [170]Supplementary Material 5^ (12.1KB, xlsx) [171]Supplementary Material 6^ (729.6KB, xlsx) [172]Supplementary Material 7^ (12.7KB, xlsx) [173]Supplementary Material 8^ (15.7KB, xlsx) Abbreviations LSFs Late-spring frosts GA Gibberellin ABA Abscisic acid BRs Brassinosteroids JA Jasmonic acid MeJA Methyl jasmonic acid SA Salicylic acid SL Strigolactone DEGs Differentially expressed genes DAHs Differentially accumulated hormones FPKM Fragments per kilobase of transcript per million mapped reads FDR False discovery rate GO Gene Ontology WGCNA Weighted correlation network analysis ROS Reactive oxygen species PCA Principal component analysis Author contributions XY designed the research and modified the manuscript. FW conducted data analysis and revised the manuscript. KZ conducted data analysis and wrote the manuscript. XW and BL conducted sample collection and data analysis. SG, JY, YB, and YW conducted data analysis. XT provided experimental materials. All authors read and approved the final manuscript. Funding This research was supported by Shanxi Provincial Science and Technology Major Special Programs “jiebangguashuai” Project (202201140601027-5). Data availability All data generated or analysed during this study are included in this published article and its supplementary information files. The raw sequencing data used in this study has been deposited in NCBI SRA with the accession number PRJNA1202353. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Fang Wang, Email: fangwangsxau@163.com. Xiuqing Yang, Email: xiuqingyang@sxau.edu.cn. References