Combined transcriptomic and metabolomic analysis of the molecular mechanism of seed dormancy release in Notopterygium incisum

Abstract

Background

Notopterygium incisum seeds have both morphological and physiological dormancy characteristics and require stratification to break seed dormancy, but the mechanism of seed dormancy release during stratification is still unclear. In this study, different stages of N. incisum seed stratification were employed as experimenta objects, and the dynamic changes during seed dormancy release were studied through embryo morphology, physiological index determination, transcriptome, and metabolome.

Results

(1) Stratification treatment reduced the content of stored nutrients in N. incisum seeds, significantly changed enzyme activity, reduced ABA content, and increased GA₃ and IAA contents. (2) A total of 110,539 differentially expressed genes (DEGs) and 1656 metabolites (DAMs) were identified during dormancy release. Transcriptome analysis showed that after the dormancy of N. incisum seeds was released, the expression of genes in the abscisic acid signaling pathway (ABI1, PP2CA, ABI5 and ABF4) and the gibberellin signaling pathway (GAI, GAI1 and RGL1) were significantly down-regulated, and there were significant changes in the differentially expressed genes in the auxin, cytokinin and ethylene signaling pathways. The genes related to starch and sucrose metabolism were up-regulated during dormancy release. The genes related to phenylpropanoid and flavonoid biosynthesis were significantly up-regulated after dormancy release. (3) Combined transcriptomics and metabolomics analysis showed that phenylpropanoid biosynthesis and flavonoid biosynthesis are the key pathways for the dormancy release of N. incisum seeds. (4) Metabolomics analysis showed that the accumulation of metabolites of the phenylpropanoid biosynthesis pathway (p-coumaric acid, coniferyl aldehyde, coniferyl glycoside, 5-caffeoylshikimic acid and sinapinic acid) decreased during and after the dormancy release of N. incisum seeds, while the accumulation of flavonoids such as quercetin, rutin, delphinidin and naringenin chalcone increased significantly after dormancy release.

Conclusion

Dormancy release in N. incisum seeds involves differential regulation of hormones, carbohydrates, phenylpropanoids, and flavonoid metabolites. Our results provide important insights into the molecular regulatory network of dormancy release in N. incisum seeds.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12864-025-12021-x.

Introduction

Notopterygium incisum Ting ex H.T. Chang is a medicinal plant of the Apiaceae family. N. incisum is an endangered highland medicinal material in China, mainly produced in Gansu, Sichuan, and Qinghai provinces. N. incisum has anti-inflammatory, analgesic, antipyretic, antiarrhythmic, anti-myocardial ischemic, antibacterial, and gastrointestinal function-improving pharmacological effects [1]. In recent years, N. incisum has been over-exploited, driven by economic interests, causing wild N. incisum resources to be on the verge of depletion, and Industrialization cultivation of N. incisum is imminent [2]. N. incisum can be propagated either by rhizome transplanting or by seed cultivation. Because of the low coefficient and high cost of rhizome propagation, N. incisum is mainly propagated by seeds. However, N. incisum seeds have problems with morphological and physiological dormancy and low natural germination rate, which seriously affect the production of high-quality N. incisum seedlings and their large-scale planting. At present, there has been no breakthrough in the technology of artificially breaking the dormancy of N. incisum seeds, which is one of the primary problems restricting the large-scale cultivation of N. incisum [3]. Stratification can soften the seed coat and enhance the metabolic activity inside the seeds, thereby promoting seed ripening [4]. Germination in many species is stimulated by low temperatures [5]. The variable temperature stratification treatment is a stratification method that uses alternating high and low temperatures to accelerate seed germination. Some studies have reported that after the seeds of Cardiocrinum giganteum (Wall.) Makino and Schisandra chinensis (Turcz.) Baill were stratified at variable temperatures, and the embryos completed morphological after-ripening and germinated [6, 7]. Previous studies by our group have shown that after 180 days of temperature stratification, the embryos of N. incisum are fully developed, and morphological dormancy is released. Then, after 210 days of temperature stratification, the seeds’ morphological and physiological dormancy was released, and they began to germinate [8].

The induction, maintenance, and release of seed dormancy is a continuous process that is mainly regulated by plant hormones [9]. ABA induces seed dormancy, while GA relieves dormancy and promotes seed germination [10]. The ABA content in deep dormant potato tubers is the highest. As the potato tubers gradually germinate, the ABA content gradually decreases [11]. Studies have reported that the GA content of Populus tremuloides and Vitis vinifera cv decreases during dormancy and gradually increases as dormancy is released [12, 13]. To date, many genes have been identified as contributors to the hormone-signaling pathways that regulate plant germination. Studies have shown that temperature changes can inhibit the expression of NCED and ABI5, decrease the content and sensitivity of ABA, and thus enhance the GA/ABA ratio in Cynara cardunculus seeds [14]. IAA and ABA signaling cross-regulate and affect seed germination [15]. CTK plays a part in various phases of plant growth and development [16]. ETH negatively affects ABA biosynthesis and signal transduction [17, 18]. Transcription factors (TFs), such as ABI3, ABI5, WRKY, bZIP, and bHLH, also greatly impact the regulation of seed dormancy and germination [19, 20]. The storage substances in seeds provide energy for seed vitality maintenance, germination, and other life activities. Starch and soluble sugars are important carbohydrate reserves in plant seeds [21, 22]. In addition, studies have elucidated gene expression and metabolic changes during seed germination and found that phenylpropanoid and flavonoid biosynthesis may have a positive effect on seed germination [23].

As a result of recent developments in omics technologies, there has been an increasing number of studies on metabolic and transcriptional changes during plant seed germination [24, 25]. However, the combined metabolome and transcriptome analysis of dormancy release in N. incisum seeds has not been reported. This project uses N. incisum seeds as experimental materials, conducts stratification treatments on the seeds using different methods, and analyzes the embryo morphology and physiological and biochemical parameters during the dormancy release process. At the same time, transcriptome technology and broadly targeted metabolomics were used to screen differentially expressed genes and differential metabolites (DAMs) in the five stages of N. incisum seed dormancy release. During dormancy release in N. incisum seeds, phytohormone signaling, starch and sucrose metabolism, and flavonoid and phenylpropanoid biosynthesis genes were studied. The potential regulatory mechanism in the process of N. incisum seed dormancy release was revealed, providing a certain theoretical basis for N. incisum breeding.

Materials and methods

Plant materials and treatment

The experimental material N. incisum seeds were collected in Rushu Village, Nanhe Town, Tanchang County, Longnan City, Gansu Province (104 °E, 34 °N) in mid-September 2023. They were identified as Notopterygium incisum Ting ex H. T. Chang seeds by Professor Jin Ling of Gansu University of Chinese Medicine and placed in a 4 °C low-temperature seed storage box. Outdoor natural stratification was carried out at the experimental material collection site in mid-September 2023. The seeds were sown directly into the ground with a stratification depth of 2–3 cm. Indoor stratification was carried out in late September 2023, and mature, plump, and uniformly sized seeds were selected and placed in mesh bags. At the same time, the stratification matrix was also sterilized at high temperature, and three stratification matrix treatments were set up: A: river sand; B: river sand: vermiculite = 1:1; C: river sand: perlite = 1:1. The humidity of all stratification matrices was maintained at around 60%. Indoor seeds with different treatments were placed in a light incubator with temperature (20 ℃/4 ℃), light (3000 lx) and humidity (60%) for stratification for 90 days, and then moved into a seed low-temperature storage box with temperature (4 ℃), light (3000 lx) and humidity (60%) for stratification for 120 days. Each treatment was replicated 3 times, with 4 g of seeds per replicate.

Samples were taken every 30 days for morphological observation. At the same time, some seeds were stored at −80 ℃ for omics analysis and physiological index determination. After 210 days of stratification, the morphological and physiological dormancy of N. incisum seeds was released and they germinated. The germination sign was when the radicle elongated more than 2 cm. The number of germinated seeds in different stratification treatments outdoors and indoors was recorded, and each treatment was repeated 3 times. The sampling times were stratification day 0 (morphological dormancy period, MD), stratification day 90 (morphological dormancy period, MD), stratification day 180 (physiological dormancy period, PD), stratification day 210 (physiological dormancy period, PD), and stratification day 210 (germination period, G), and each period was repeated three times. In May 2024, the samples were sent to Shanghai Baiqu Biomedical Technology Co., Ltd. for targeted metabolome analysis and transcriptome sequencing. We declare that the research programme complies with relevant institutional, national and international guidelines and legislation, and we have permission to collect N. incisum seeds.

Determination of physiological and biochemical parameters of N. incisum

The protein content of N. incisum seeds was determined by the Coomassie brilliant blue method, the soluble sugar content was determined by the anthrone colorimetric method, and the starch content was determined by the anthrone colorimetric method [26, 27]. The specific operations were referred to the kits of Nanjing Jiancheng Biological Company. An enzyme-linked immunosorbent method was utilized to quantify the levels of gibberellins, auxins, and abscisic acid [28]. The specific operation was referred to the kit instructions of Shanghai Qincheng Biotechnology Co., Ltd. Each indicator was biologically replicated 3 times.

Transcriptome analysis

cDNA library construction and transcriptome sequencing

The Total sample of RNA was extracted and purified using TRIzol. NanoDrop ND-1000 controlled total RNA amount and purity, while Bioanalyzer 2100 assessed RNA integrity. When the RNA purity and integrity were qualified, the library was constructed. After the library quality inspection was qualified, Illumina NovaseqTM 6000 sequenced it with a double-end 2*150 bp (PE150) read length.

Sequencing quality control and sequence assembly

Cutadapt is used to remove the sequencing adapters, and then fqtrim is used to filter out unqualified sequences to obtain valid data (Clean Data). Currently, there is no reference genome for N. incisum. Therefore, Trinity software is used to perform de novo assembly of the reads obtained from sequencing, and the obtained assembled sequences are optimized, filtered, and evaluated again. Finally, unigenes are obtained as reference sequences.

Unigenes functional annotation and differentially expressed gene analysis

Unigenes were compared and analyzed in databases such as NR, GO, Swiss-Prot, COG, PFAM, and KEGG to obtain the protein function annotation information of Unigene, and then the number of Unigenes with annotation information was counted. The screening criteria for differentially expressed genes were FDR < 0.05 and |log2 FC| ≥ 1.

Metabolome analysis

Sample extraction

The sample was freeze-dried and then ground (60 Hz, 30 s). Weigh 0.025 g of sample and add 1000 µL of extraction solution (methanol: water = 3:1, volume ratio, including internal standard). Vortex for 30 s, homogenize at 40 Hz for 4 min and sonicate in an ice-water bath for 5 min. The homogenate was sonicated three times and centrifuged again, and 2 mL of the supernatant was used for LC–MS/MS analysis.

Chromatographic and mass spectrometric conditions

In this project, ultra-high performance liquid chromatography was used to separate the target compounds using a UPLC Kinetex C18 (2.1 mm × 100 mm, 2.6 μm) column. The HPLC phase A was an aqueous solution containing 0.01% acetic acid, and the phase B was 50% acetonitrile/isopropanol. The column oven temperature was 25 °C, the autosampler temperature was 4 °C, the injection volume was 2 µL, and the flow rate was 0.3 mL/min. Mass spectrometry analysis in multiple reaction monitoring (MRM) mode. The ion spray voltage was 5500 V/−4500 V, the ion curtain gas was 35 psi, the ion source gas was 50 psi, and the temperature was 400 °C.

Data preprocessing

Biobud software was used for mass spectrometry data gathering and target compound quantification. Peak extraction and annotation were completed in combination with the self-built database. The differential metabolites in different dormancy release stages were screened, and the screening criteria were VIP > 1 and P-value < 0.05.

qRT-PCR analysis

The internal reference gene was ACT11, and fluorescence quantitative PCR primers were designed (Additional file Table S1). 1% gel was prepared for electrophoresis to detect RNA quality; RNA concentration, A260/A230, and A260/A280 were determined using a full-wavelength microplate reader. Reverse transcription into cDNA was performed using the Beijing Quanshi Jin Biological Reverse Transcription Kit (AE311-02). The experimental conditions were set as follows: 95℃ pre-denaturation for 10 min, 1 cycle; 95℃ denaturation for 15 s, 40 cycles; 60℃ annealing/extension for 30 s, 40 cycles. Each sample was biologically replicated 3 times (each replicate contained 3 technical replicates), and the relative expression of the selected genes was calculated using 2^−△△Ct.

Data statistics and analysis

Excel 2016 was used to organize and count the data, IBM SPSS Statistics 26.0 software was used to perform one-way analysis of variance, multiple comparisons, and correlation analysis on the measured data, and Origin 2021 software was used to draw the graphs.

Results

Determination of germination rate and physiological and biochemical parameters of N. incisum seeds

Basic morphological characteristics of N. incisum seeds

The seeds of N. incisum are double-hanging fruits, flat and oval in shape; the surface is brown, with 3 to 4 longitudinal ridges on the back, often extending into light brown wings, the ventral side is slightly concave, and there is often a thin linear hanging fruit stalk in the center, with a protruding stylopodium at the top. The seed structure of N. incisum is the same as most seeds, consisting of three parts: seed coat, embryo, and endosperm.

Embryo morphology and germination rate during dormancy release of N. incisum seeds

Throughout the stratification process, the embryo of N. incisum seeds continued to develop. The results showed that: at 0 day of stratification, the embryo had not yet developed and was in a dormant state(Fig. 1a); During stratification for 30 to 90 days, the embryo developed slowly(Fig. 1b, c,d); At 120 days of stratification, the seed embryo developed into a heart-shaped embryo (Fig. 1e); At 150 days of stratification, the embryo differentiated into a complete plumule, hypocotyl, and radicle (Fig. 1f); After stratification for 180 days, the embryo developed into a cotyledon-type embryo (Fig. 1g). The embryo’s after-ripening process was completed, its morphological dormancy was released, and it entered a physiological dormancy period. The results of N. incisum seeds after different stratification treatments are shown (Fig. 2). After the stratification with warm temperature followed by low temperature (20℃/4℃/12 h for 90 d + 4℃ for 120 d) was completed, the germination rate of river sand (A) in the indoor stratification treatment was the highest at 53.49%; River sand: vermiculite = 1:1 (B) germination rate is 50.07%; River sand: perlite = 1:1 (C) germination rate is 47.15%; The germination rate of outdoor natural stratification (D) was 60.43%.

Fig. 1.

Morphological changes of embryo during dormancy release of N. incisum seeds. A unstratified (stratified for 0 days); B stratified for 30 days; C stratified for 60 days; D stratified for 90 days; E stratified for 120 days; F stratified for 150 days; G stratified for 180 days; Scale bar = 0.2 mm

Fig. 2.

Germination of N. incisum seeds. a (A) River sand; (B) River sand: vermiculite = 1:1; (C) River sand: perlite = 1:1; (D) Outdoor natural stratification. b Statistics of germination rates under different treatments. Note: Different lowercase letters indicate significant differences between treatment times at the 0.05 level (P < 0.05)

Determination of physiological and biochemical parameters during dormancy release of N. incisum seeds

We explored the dynamic changes of stored substances during seed dormancy release. The soluble protein content first increased and then decreased from 0 to 60 days, gradually increased from 60 to 180 days, and reached the maximum value at 180 days. At 210 days, the protein may be decomposed to provide energy and nutrition to the seeds, and its content has dropped significantly(Fig. 3a). The starch content showed an overall downward trend from 0 to 210 days, reaching the maximum value at 0 days and the minimum value at G210 (Fig. 3b). The soluble sugar content showed a “W” trend from 0 to 210 d (PD), with the highest content at 210 d (PD), reaching the maximum value (Fig. 3c).

Fig. 3.

Dynamic changes of physiological and biochemical indices during dormancy release of N. incisum seeds. a Soluble protein content; b Starch content; c Soluble sugar content; d ABA content; e GA₃ content; f IAA content; g ABA/GA₃; h ABA/IAA. The letters in the figure indicate significant differences between different stratification stages (p < 0.05)

To study the impacts of endogenous hormones on dormancy release in N. incisum seeds, the contents of abscisic acid, gibberellins, and auxins were determined. The ABA content initially declined and subsequently climbed during both the 0-120d and 120-210d (PD), reaching its maximum value at 0 d and the minimum value at 210 d (G) (Fig. 3d). The trend of gibberellin content in 0-120d was “W,” first decreasing and then increasing, and it showed a decreasing trend in 120-150d, and a continuous increasing trend in 150-210d (G), and reached the maximum value at 210 d (G) (Fig. 3e). The auxin content increased first and then decreased at 0-60d, 60-120d, 120-180d, and 180-210d (G), reaching the maximum value at 210 d (PD) and slightly decreased at 210 d (G) (Fig. 3f). The ABA/GA₃ ratio initially declined and subsequently climbed during both 0-90d and 90-150d, and a continuous and significant decreasing trend in 150-210d (G), with the lowest content at 210 d (G) (Fig. 3g). The ABA/IAA ratio first decreased and then increased during 0-120 d and continued to decrease significantly during 120–210 d (G), with the lowest content at 210 d (G) (Fig. 3h).

Transcriptome analysis

Sequencing data evaluation and gene function annotation

After filtering low-quality sequencing reads, 84.95 Gb of clean data was retrieved. Each sample had homogeneous GC content, clean data larger than 4.64 Gb, and Q30 base percentage greater than 91.02% (Table 1). Comparing DEGs to the database annotated 605,517 differentially expressed genes (Table S2). Among them, 115,216 (55.06%) unigenes were obtained from the GO database, 46,564 (22.25%) unigenes were obtained from the KEGG database, 108,074 (51.65%) unigenes were obtained from the Pfam database, 92,904 (44.40%) unigenes were obtained from the Swiss-Prot database, 130,914 (62.56%) unigenes were obtained from the eggNOG database, and 111,845 (53.45%) unigenes were obtained from the NR database.

Table 1.

Statistics of transcriptome data quality of each sample

Sample	Raw-Reads	Raw-Bases	Valid-Reads	Valid-Basess	Valid(%)	Q20(%)	Q30(%)	GC(%)
MD0-1	38,472,148	5.77G	37,103,724	5.45G	96.44	97.69	93.50	46.89
MD0-2	50,741,534	7.61G	49,516,488	7.32G	97.59	97.91	94.00	47.22
MD0-3	43,673,164	6.55G	42,520,068	6.27G	97.36	97.78	93.71	46.65
MD90-1	39,752,484	5.96G	38,931,106	5.76G	97.93	97.80	93.73	45.56
MD90-2	35,105,454	5.27G	34,345,994	5.08G	97.84	97.94	94.11	45.09
MD90-3	34,185,802	5.13G	33,389,808	4.94G	97.67	97.87	94.05	45.25
PD180-1	34,505,802	5.18G	33,765,766	5.00G	97.86	98.01	94.35	44.99
PD180-2	32,960,408	4.94G	32,233,954	4.77G	97.80	98.04	94.37	46.06
PD180-3	55,421,780	8.31G	54,245,172	7.96G	97.88	96.82	91.02	46.81
PD210-1	34,199,604	5.13G	33,466,576	4.95G	97.86	98.13	94.66	45.86
PD210-2	32,301,950	4.85G	31,436,292	4.64G	97.32	97.93	94.18	45.81
PD210-3	40,320,376	6.05G	39,313,614	5.79G	97.50	97.81	93.83	45.54
G210-1	36,164,018	5.42G	35,355,432	5.21G	97.76	97.35	92.43	44.50
G210-2	37,631,938	5.64G	36,932,210	5.48G	98.14	98.13	94.56	45.48
G210-3	43,533,692	6.53G	42,723,786	6.33G	98.14	97.99	94.22	44.52

Multivariate statistical analysis of transcriptome data

The gene expression correlation diagram showed that there was a low correlation between the gene expression levels of seeds at different stages, while the average correlation coefficient of samples at the same stage was higher than 0.9, indicating that the three repeated samples at the same stage were highly correlated (Fig. 4a). Principal component analysis showed that N. incisum seeds at different stages were clustered into one category. After dimensionality reduction, the distances within each group of samples were close or overlapped, indicating good repeatability within each group; the samples in each group were separated from each other and had a certain interval, indicating good discrimination between groups. However, the three groups of MD90, PD180, and PD210 were relatively close, probably because they were all in a dormant state (MD or PD) and their gene expression patterns were similar, especially the expression of core genes related to maintaining dormancy tended to be consistent. The MD0 and G210 groups were farther apart than the other groups, probably because MD0 was in the initial morphological dormancy stage and its gene expression pattern had not yet undergone significant changes induced by stratification; whereas G210 had completed dormancy release and initiated the germination process, with significant differences in gene expression (Fig. 4b). During the dormancy release process, 3253 differentially expressed genes were up-regulated, 13,933 were down-regulated in MD0vsMD90, 5920 differentially expressed genes were up-regulated, and 22,379 were down-regulated in MD0vsPD180, 6071 differentially expressed genes were up-regulated, and 14,629 were down-regulated in MD0vsPD210, and 7526 differentially expressed genes were up-regulated, and 36,828 were down-regulated in MD0vsG210 (Fig. 4c). The Venn diagram illustrates the DEGs in the four comparison groups, with 2075 intersection genes (Fig. 4d).

Fig. 4.

Multivariate statistical analysis of transcriptome data. a Gene expression correlation diagram; b Principal component analysis (PCA) diagram; c Statistics of the number of DEGs in different comparison groups; d Venn diagram of DEGs between the four comparison groups

GO enrichment analysis of DEGs

GO results showed (Additional file Fig. S1) that there were 24,178 DEGs, 41,198 DEGs, 30,196 and 23,110 DEGs in MD0vsMD90, MD0vsPD180, MD0vsPD210 and MD0vsG210, respectively. In the biological process, compared with MD0, DEGs were significantly enriched in biological process, translation, redox process, transcriptional regulation, DNA template and other items at MD90, PD180, PD210 and G210. Translation and transcriptional regulation were always the key biological processes, which indicated that protein synthesis and gene expression regulation played an important role in dormancy release and germination. In molecular function, compared with MD0, DEGs were significantly enriched in molecular function, structural components of ribosomes, ATP binding, and protein binding at MD90, PD180, and PD210, and DEGs were significantly enriched in structural components of ribosomes, protein binding, DNA-binding transcription factor activity, sequence-specific DNA binding, and DNA-binding transcription factor activity at G210, indicating that ribosome function and protein binding remain active during seed dormancy release and germination, and the enhanced transcription factor activity during germination makes gene expression regulation more precise. Among the cellular components, compared with MD0, at MD90, PD180 and PD210, the cellular components were nucleus, cytoplasm, and plasma membrane, and at G210, the plasma membrane, cytoplasmic large ribosomal subunits, cytoplasmic small ribosomal subunits, chloroplasts, and extracellular regions. As the stratification time increases, the activity of ribosomes, chloroplasts, and extracellular regions in seeds during the dormancy release and germination period indicates that processes such as translation, photosynthesis, and cell wall modification are activated to support seed germination.

KEGG enrichment of differentially expressed genes

KEGG annotation showed (Fig. 5) that 134 pathways were enriched between the MD0vsMD90 groups, and 2845 DEGs were annotated; 139 pathways were enriched between the MD0vsPD180 groups, and 3989 DEGs were annotated; 138 pathways were enriched between the MD0vsPD210 groups, and 3400 DEGs were annotated; 141 pathways were enriched between the MD0vsG210 groups, and 5053 DEGs were annotated. Among them, plant hormone signal transduction (map04075), starch and sucrose metabolism (map00500), tryptophan metabolism (map00380), glycerolipid metabolism (map00561) and phenylpropanoid biosynthesis (map00940) were significantly enriched between the MD0vsMD90 groups, indicating that during the morphological dormancy period, seeds began to initiate plant hormone signal transduction, activating starch, sucrose, lipid and phenylpropanoid metabolism, and providing energy and material basis for dormancy release and germination. Ribosome (map03010), protein processing in the endoplasmic reticulum (mapmap04141), and RNA transport (map03013) were significantly enriched between the MD0vsPD180 groups, indicating that during the physiological dormancy period, seeds prepare for dormancy release and germination by enhancing ribosome function, protein processing, and RNA transport, promoting protein synthesis and gene expression regulation. Plant hormone signal transduction (map04075), phenylpropanoid biosynthesis (map00940), starch and sucrose metabolism (map00500), pentose and gluconic acid interconversion (map00040), and glycerolipid metabolism (map00561) were significantly enriched between the MD0vsPD210 groups, indicating that seeds continued to maintain dormancy through plant hormone signal transduction and metabolic pathways (starch, sucrose, lipid and phenylpropanoid metabolism) and provide energy and material basis for germination. In the MD0vsG210 group, DEGs were significantly enriched in the ribosome (map03010), RNA transport (map03013), oxidative phosphorylation (map00190), plant hormone transduction (map 04075), and phenylpropanoid biosynthesis (map00940) pathways, indicating that after seed dormancy is released, protein synthesis, gene expression regulation, and energy supply are promoted by enhancing ribosome function, RNA transport, and oxidative phosphorylation, while plant hormone signal transduction and phenylpropanoid metabolism continue to regulate seedling growth. Among the top 20 KEGG pathways, plant hormone signaling pathway, starch and sucrose metabolism, phenylpropanoid biosynthesis, and flavonoid biosynthesis pathways were significantly enriched in the four comparison groups, which may play an important role in the release of seed dormancy.

Fig. 5.

KEGG pathway enrichment scatter plot of differentially expressed genes in different comparison groups. a–d KEGG enriched pathways of DEGs in MD0vsMD90, MD0vsPD180, MD0vsPD210, and MD0vsG210

DEGs involved in multiple plant hormone signal transduction

We focused on the abscisic acid, gibberellin, auxin, cytokinin, and ethylene signaling pathways (Fig. 6). The expression of ABA receptors (PYR/PYL) involved in ABA signal transduction was down-regulated at MD0, MD90, and PD180 stages, and was significantly up-regulated at PD210 and G210 stages. Serine/threonine protein kinase SRK2 (SnRK2) is significantly differentially expressed during dormancy release. ABA response element binding factor (ABF) and Protein phosphatase 2 C (PP2C) genes were mostly up-regulated during stratification and significantly down-regulated at the G210 germination stage. The suppressive influence of ABA signal transduction on N. incisum seed germination may be lessened with stratification treatment. Two gibberellin receptor GID1 genes associated with the GA signal transduction pathway were identified: GID1B was significantly up-regulated at the G210 stage, and GID1C was significantly down-regulated at the G210 germination stage. DELLA proteins (DELLA) are negative regulators of the gibberellin regulatory pathway. Their expression is mostly up-regulated during stratification and markedly down-regulated at the G210 germination stage. Stratification treatment enhanced the hormone signal transduction of GA, thereby promoting the dormancy release of N. incisum seeds. In the auxin signal transduction pathway, auxin response protein IAA (AUX/IAA/LAX) and SAUR family protein (SAUR) gradually increased during seed dormancy release and were mostly significantly up-regulated at the G210 germination stage. Auxin response factor (ARF) and auxin response GH3 gene family (GH3) were mostly significantly down-regulated at the germination stage of G210. DEGs related to CTK signaling, histidine-containing phosphotransfer protein (AHP) was mostly up-regulated during stratification and down-regulated at the G210 germination stage. The expression of two-component response regulators ARR-A family (A-ARR) was significantly up-regulated at PD180, PD210, and G210 germination stages, indicating that CTK may actively participate in the dormancy release process of N. incisum seeds. Some DEGs related to ETH signal transduction, ethylene insensitive protein 2 (EIN2), ethylene insensitive protein 3 (EIN3), transcription factor EBF family (EBF2), and ethylene response factor (ERF) were related to ethylene signal transduction, and most of them were down-regulated at the germination stage of G210. It is noteworthy that ABA signaling may be primarily triggered during seed dormancy, and its accumulation decreases after dormancy is released. In contrast, GA signaling pathways were enhanced after dormancy was released. In addition, DEGs in the IAA, CTK, and ETH signaling pathways were also differentially expressed, indicating that they might have an important effect on the release of N. incisum seed dormancy.

Fig. 6.

Expression of DEGs in the plant hormone signal transduction pathway during seed dormancy release in N. incisum. DEG expression was enriched in ABA, GA, IAA, CTK, and ETH signaling pathways. The heat map shows the five stages of MD0-G210 from left to right, and the changes in red and blue indicate an increase or decrease in FPKM values, respectively

DEGs involved in sucrose and starch metabolism

Carbohydrate metabolism provides energy for plant growth and is linked to dormancy release and germination [29]. We identified DEGs in starch and sucrose metabolic pathways that were associated with N. incisum seed dormancy release (Fig. 7). DEGs related to starch synthesis, including starch branching enzyme (SBE), starch synthase (SS), and alkylglycerol phosphate synthase (AGPS1), were mostly up-regulated during stratification (MD90, PD180, and PD210); however, they were mostly down-regulated at the G210 germination stage. These indicated that starch accumulation gradually decreased. α-Amylase (AMY) and β-amylase (BAM) are key enzymes that catalyze starch degradation [30]. BAM was down-regulated in the G210 germination stage, while AMY was markedly up-regulated in the G210 germination stage. Sucrose invertase (INV) and fructokinase (scrK) were markedly up-regulated at the G210 germination stage. Sucrose synthase (SUS), sucrose phosphate synthase (SPS), and sucrose phosphatase (SPP) were mostly up-regulated during stratification. These indicate that stratification accelerates the buildup of soluble sugars. Meanwhile, enzymes involved in the interconversion of starch and sucrose, such as phosphoglucomutase (PGM) and UDP-glucose pyrophosphorylase (UGP), were mostly up-regulated during stratification.

Fig. 7.

Expression of DEGs in starch and sucrose metabolic pathways during dormancy release in N. incisum seeds. The heat map shows the five stages of MD0-G210 from left to right, and the changes in red and blue indicate an increase or decrease in FPKM values, respectively

DEGs involved in phenylpropanoid and flavonoid biosynthesis

In this research, DEGs involved in phenylpropanoid biosynthesis and flavonoid biosynthesis were screened to determine their relationship with dormancy release in N. incisum seeds. DEGs in this pathway included phenylalanine ammonia lyase (PAL), 4-coumarate: CoA ligase (4CL), cinnamate 4-monooxygenase (CYP73A), cinnamoyl-CoA reductase (CCR), cinnamoyl alcohol dehydrogenase (CAD), caffeic acid O-methyltransferase (COMT), 5-O-(4-coumaryl)-D-quinone 3′-monooxygenase (CYP98A), peroxidase (PER), caffeoyl-shikimate esterase (CSE), and cinnamoyl-CoA 4-hydroxylase (CYP84A). Among them, PAL, CAD, COMT, CSE, CYP84A, CYP98A, and PER were mostly down-regulated during the dormancy release process (MD0, MD90, PD180, and PD210) but were markedly up-regulated in the G210 germination stage. The expression of 4CL and CCR was mostly down-regulated at the G210 germination stage. This suggests different metabolic requirements during seed cryostratification and after dormancy release. The key genes screened in the flavonoid biosynthesis pathway include chalcone isomerase gene (CHI), chalcone synthase gene (CHS), flavonol synthase gene (FLS), and flavonoid 3’-hydroxylase (CYP75B2). The expression of CHI, CHS, and CYP75B was mostly down-regulated during the stratification process and markedly up-regulated at the G210 germination stage. FLS was markedly up-regulated in the MD0 stage and continued to be down-regulated during the dormancy release process and the 210 germination stage. These findings indicate that gene expression of phenylpropanoid and flavonoid biosynthetic pathways changes significantly during seed dormancy release, suggesting that their metabolic dynamics may accompany or respond to the transition to dormancy.

Validation of RNA-Seq data by qRT-PCR

To validate the accuracy of RNA-Seq data, we selected eight differentially expressed genes (PYL2, GAI, ARF11, SPS2, AGPS1, PER42, CYP73A10, and CCR1) for qRT-PCR analysis. Our results indicated that the levels of expression of the chosen genes were largely congruent with the expression levels in the transcriptome, indicating that the results of the transcriptome sequencing analysis were reliable (Fig. 8).

Fig. 8.

qRT-PCR validation of 8 DEGs

Metabolome analysis

Quantitative statistics and principal component analysis of differential metabolites

PCA analysis was performed on the metabolite detection results at different stratification stages. The results showed that the biological replicates of each treatment were basically clustered together, indicating that the experimental repeatability was good, the intra-group variation was small, and the data quality was reliable. Except for PD180 and PD210, which were in the same quadrant, the other treatment groups were not in the same quadrant. This shows that the metabolome data at each stage of seed dormancy release are significantly different, and the PCA model can effectively explain the metabolic differences between different germination stages (Fig. 9,a). During the process of dormancy release, MD0vsMD90 had 66 up-regulated differential metabolites and 325 down-regulated, MD0vsPD180 had 104 up-regulated differential metabolites and 307 down-regulated, MD0vsPD210 had 109 up-regulated differential metabolites and 317 down-regulated, and MD0vsG210 had 174 up-regulated differential metabolites and 254 down-regulated (Fig. 9,b).

Fig. 9.

Multivariate statistical analysis of metabolomics data. a PCA score scatter plot; b Statistics of the number of DAMs in different comparison groups; c–f KEGG pathway analysis of DAMs in MD0vsMD90, MD0vsPD180, MD0vsPD210, and MD0vsG210

KEGG enrichment of differentially metabolites

The findings indicated that the differential metabolites in the four comparison groups, MD0vsMD90, MD0vsPD180, MD0vsPD210, and MD0vsG210, were annotated to 26, 29, 27, and 28 metabolic pathways, respectively (Fig. 9c, d, e, f). The differential metabolic pathways were markedly enriched in metabolic pathways such as flavonoid and flavonol biosynthesis (ath00944), flavonoid biosynthesis (ath00941), phenylpropanoid biosynthesis (ath00940), pantothenic acid and CoA biosynthesis (ath00770), histidine metabolism (ath00340) and taurine and oligotaurine metabolism (ath00430). Among them, flavonoid and flavonol biosynthesis, flavonoid biosynthesis, and phenylpropanoid biosynthesis pathways were significantly enriched in the four comparison groups. This indicates that the activity of these metabolic pathways may promote seed dormancy release and germination through multiple mechanisms such as anti-oxidation, energy metabolism, signal transduction and cell structure maintenance. Among the top 20 KEGG pathways, flavonoid and flavonol biosynthesis, flavonoid biosynthesis, and phenylpropanoid biosynthesis pathways were significantly enriched in the four comparison groups. These results provide important clues for inferring the regulatory mechanism of seed dormancy release.

DAMs involved in the biosynthesis of phenylpropanoids and flavonoids

This study identified 796 metabolites in N. incisum seeds by LC-MS, including flavonoids, phenylpropanoids, alkaloids, and lipids. We screened 17 DAMs (Fig. 10,b). Among them, metabolites in the phenylpropanoid biosynthesis pathway include p-coumaric acid, coniferaldehyde, coniferin, 5-caffeoylshikimic acid, sinapinic acid, and phenylalanine. The accumulation of most metabolites increased markedly in the MD0 stage and declined markedly during the stratification process (MD90, PD180, and PD210) and after the dormancy was released at G210. The results suggested that compounds in the phenylpropanoid biosynthesis pathway were reduced during seed dormancy release, which might be beneficial for dormancy release in N. incisum seeds. The metabolites of the biosynthesis pathway of flavonoids, including quercetin, rutin, delphinidin, and naringenin chalcone, decreased continuously during the stratification process (MD0, MD90, PD180, and PD210) but increased significantly during the G210 germination stage. Interestingly, other metabolites identified in this pathway were all decreased. Among them, the accumulation of apigenin, luteolin, eriodictyol, isoquercitrin, trifolin, astragalin, and kaempferide increased markedly in the MD0 stage, followed by a notable drop during the stratification process (MD90, PD180, and PD210) and upon the germination of G210 seeds. The significant changes in the content of flavonoids in different stagesmight be crucial to the dormancy release of N. incisum seeds.

Fig. 10.

a Heat map constructed based on the FPKM values of differentially expressed genes associated with the phenylpropanoid biosynthesis and flavonoid biosynthesis pathways. Changes in red and blue in the heat map indicate an increase or decrease in the FPKM value, respectively; b Heat map constructed based on the relative quantitative values of differentially expressed metabolites associated with the phenylpropanoid biosynthesis and flavonoid biosynthesis pathways. Changes in yellow and blue in the heat map indicate an increase or decrease in the relative quantitative value, respectively; c DEGs and DAMs regulatory mechanism diagram for phenylpropanoid and flavonoid biosynthesis pathways

Combined analysis of transcriptome and metabolome

In order to further study the molecular mechanism of dormancy release in N. incisum seeds, the metabolic-transcriptional association analysis was focused on the commonly enriched metabolic pathways: phenylpropanoid and flavonoid active ingredient biosynthesis pathways. Based on the KEGG pathways related to the biosynthesis of phenylpropanoids and flavonoid active ingredients (map00940, map00941, and mapo00944), this study obtained the regulatory network diagram of differential metabolites and differential genes (Fig. 10,c). The changing trend of the differential metabolite phenylalanine was largely congruent with that of the phenylalanine ammonia lyase (PAL) it encoded, and it showed an increasing trend during the germination stage of G210 seeds. The changing trend of p-coumaric acid was basically consistent with that of 4-coumaric acid: CoA ligase (4CL) encoded by it, showing a downward trend during the germination stage of G210 seeds. There are also differentially expressed genes that show similar changing trends to downstream metabolites. For example, the chalcone synthase gene (CHS) has basically the same regulatory trend as the downstream metabolite naringenin chalcone and is significantly upregulated during the germination stage of G210 seeds. 5-Hydroxyferulic acid generates sinapic acid under the action of caffeic acid oxygen methyltransferase (COMT), and the accumulation trend of sinapic acid is opposite to the expression change of the COMT gene. Caffeic aldehyde is converted into coniferyl aldehyde by the action of caffeic acid oxygen methyltransferase (COMT), and the accumulation trend of coniferyl aldehyde is opposite to the expression change of COMT gene. Dormancy release in N. incisum seeds is a complex regulatory network closely linked by the interaction of multiple substances and genes, so it is not just a one-to-one correspondence between metabolites and genes. The correlation between differential metabolites and differential genes showed that most of the differential genes were positively correlated with the accumulation of metabolites (Fig. 11).

Fig. 11.

Correlation heat map of differentially expressed genes and differentially expressed metabolites. The columns of the heat map represent DEGs, and the rows represent DAMs. Red indicates that DEG and DAM are positively correlated, blue indicates that DEG and DAM are negatively correlated, and “*” indicates that there is a significant correlation between DEG and DAM (p < 0.05)

Discussion

Seed dormancy is a normal response to adverse conditions, but prolonged dormancy can delay seed germination and seedling establishment [31]. Our results showed that at 20℃/4℃ for 90 d and 4℃ for 90 d, the embryos of N. incisum developed completely, and their morphological dormancy was released, showing the characteristics of germination. After the temperature-variable stratification at 20℃/4℃ for 90 d and 4℃ for 120 d, the seeds of N. incisum were released from the double morphological and physiological dormancy and germinated. However, the mechanism by which stratification affects dormancy release in N. incisum seeds is not yet fully understood. Therefore, we explored the morphological changes and physiological basis of seed dormancy release. Then, five key stages were selected in the dormancy release process for transcriptome sequencing and metabolomics analysis, which provided a profound comprehension of the molecular mechanism of dormancy release in N. incisum seeds.

DEGs involved in multiple plant hormone signal transduction during seed dormancy release in N. incisum

Hormones are one of the important intrinsic factors affecting seed dormancy and germination [32]. Research indicates that ABA is crucial in seed dormancy. A decrease in ABA content promotes the germination of plant seeds and is an important sign of breaking seed dormancy [33]. During low-temperature stratification at 4℃, the ABA content of Callery Pear seeds declined sharply [34]. This agrees with the results of this experiment that the content of seeds decreased significantly during the stratification process and the G210 germination stage. ABA signal transduction begins with the perception of ABA signals by ABA receptors [35]. PtPYRL1 and PtPYRL5 positively regulate ABA responses during seed germination [36]. The ABA-induced NiSnRK2 gene plays a positive regulatory role in ABA signal transduction processes, including stress tolerance and seed dormancy [37, 38]. In this research, the expression levels of ABA receptors PYR/PYL and SnRK2.6 were significantly higher in the germination stage of G210 than in dormant seeds, showing an up-regulated expression. The PP2C proteins ABI1 and ABI2 bind to ABA receptors to inhibit signal transduction [39]. Studies have shown that both ThABF and ThPP2C are down-regulated during the germination stage of Tamarix hispida seeds [25]. In this study, the negative regulatory factors PP2C (ABI1, PP2CA) and ABF (ABI5, ABF4) were down-regulated at the G210 germination stage. Among them, ABI5 was markedly down-regulated at the germination stage of G210, which agrees with the findings of earlier research on N. incisum seeds after dormancy was released after stratification treatment [40]. Furthermore, research has indicated that temperature changes reduced the expression of the ABI5 gene in Euphorbia esula seeds [41]. In summary, stratification reduced ABA content and inhibited the ABA signaling pathway, thereby promoting seed dormancy release and germination.

Gibberellic acid has been shown in many plants to promote the release of dormancy and seed germination [42]. Gibberellins bind to the GID1 receptor and induce the degradation of DELLA proteins in plants [43]. DELLAs can also promote the expression of ABI5 and enhance ABA-mediated inhibition of seed germination [44]. In this study, the DELLA protein (GAI) gene was significantly down-regulated at the G210 germination phase, which agrees with the previous study on the germination of N. incisum seeds after stratification treatment [40]. At the same time, the expression of DELLA proteins (GAI1, RGL1) was also significantly down-regulated in the G210 germination stage. Gibberellin receptor 1 (GID1) is a positive regulator of GA [45]. In this study, GID1B in the GA signaling pathway was markedly up-regulated at the G210 germination phase. The expression trend of GID1C was the opposite. Therefore, the results suggest that stratification can increase GA accumulation by reducing DELLA abundance, thereby releasing dormancy and promoting seed germination. This result agrees with the increase in GA content measured at the PD210 and G210 stages in this experiment. In addition, a higher ABA/GA ratio helps maintain seed dormancy, while the release of dormancy involves increased GA biosynthesis and ABA degradation, resulting in a lower ABA/GA ratio [46]. In this study, the changing trends of the ABA/GA ratio and ABA content were basically consistent. These findings agree with those in red bayberry and ginkgo seeds [47, 48].

Studies have shown that the IAA levels in arabidopsis and wheat increased during seed imbibition and after maturity [49, 50]. The increase in IAA content measured in this experiment at the PD210 and G210 stages is consistent with their findings. Research indicates that GH3 family genes may inhibit auxin signals during the initial phases of seed germination. Auxin can induce the expression of SAUR genes, causing them to rapidly up-regulate in a short period of time, thereby regulating the growth and development of plants [51]. In our study, Aux/IAA transcriptional repressors and SAUR were mostly significantly up-regulated at the G210 germination stage, while ARF and GH3 family genes were down-regulated at the G210 germination stage. Other phytohormones, including CTK and ETH, positively regulate seed dormancy release in dicotyledonous plants [52, 53]. This study showed different expression patterns of CTK and ETH during seed stratification. Their functions in the release of seed dormancy require additional examination. In conclusion, our results imply that a variety of hormones are involved in the regulation of seed dormancy release, facilitating germination.

DEGs involved in sucrose and starch metabolism during dormancy release in N. incisum seeds

The process of seed dormancy release is an energy-intensive process [54]. Starch is a polysaccharide stored in plants. It is converted into reducing sugars by starch reductase and used for seed growth and development [29, 55]. This study indicated that the starch content of seeds decreased with the extension of stratification time. This agrees with the finding on Callery Pear seeds [31]. Research has demonstrated that low- temperature stratification causes the soluble sugar content of seeds to first increase and then decrease [56]. This agrees with the results of soluble sugar content in the MD60-PD180 and PD180-G210 stages in this experiment. α-Amylase is the main enzyme that hydrolyzes starch into glucose [57]. Studies have found that AMY activity is positively correlated with rice seed germination rate. In this research, the expression of the AMY gene was significantly up-regulated in G210 germinated seeds. During starch biosynthesis, SS, GBSS, and SBE enzyme genes play crucial roles [58]. In this research, SS genes and SBE genes were down-regulated in G210. This is consistent with the findings during the germination of Polygonatum cyrtonema Hua seeds [59]. In addition, some significantly expressed DEGs were involved in sucrose and starch metabolic pathways, including SUS, SPS, SPP, UGP2, PGM, AGPS, and BAM, which changed significantly during stratification, and most of the DEGs were up-regulated during stratification but down-regulated in G210 germinated seeds. Sucrose invertase is central to providing carbon nutrients to plants and is important for sugar signaling [54]. In our study, INV and scrK were significantly up-regulated at the G210 germination stage. In summary, stratification treatment accelerated the hydrolysis of starch into sugars, providing energy for N. incisum seeds to break dormancy and germinate, thereby promoting seed germination.

DEGs and dams involved in the biosynthesis of phenylpropanoids and flavonoids during dormancy release in N. incisum seeds

Plant secondary metabolites are important for plant growth, development, and stress resistance [60]. The germination of seeds of many plants (e.g., Zelkova schneideriana and Chinese fir) involves the phenylpropanoid pathway [61, 62]. The phenylpropanoid pathway begins with phenylalanine, which is converted from photosynthetic sugars via the manganate pathway [63]. Subsequently, cinnamic acid, p-hydroxycinnamic acid, and p-coumaroyl-CoA are generated in sequence through the catalytic reactions of PAL, cinnamic acid 4-hydroxylase (C4H), and 4CL, respectively [64]. These compounds are ultimately converted into various phenylpropanoid compounds, such as flavonoids, flavonols, lignin, and alkaloids [65].

4CL is an important gene in the biosynthesis of phenylpropanoids and affects the biosynthesis of flavonoids [66, 67]. In this study, most of the DEGs related to phenylpropanoid biosynthesis (4CL, CCR) were downregulated at the G210 germination stage, while most of the other DEGs (PAL, CAD, COMT, CSE, PER, CYP98A) were significantly upregulated at the G210 germination stage. PAL is considered to be the rate-limiting enzyme in the biosynthesis of phenolic compounds, and the transcription level of PAL reflects the intensity of phenylpropanoid biosynthesis [68]. Some studies have reported that PAL is significantly up-regulated after germination of Zelkova schneideriana seeds [61]. According to our research, the PAL3 gene was significantly up-regulated during the germination phase of G210 seeds. This is consistent with findings reported in studies on Zelkova schneideriana seeds. Interestingly, most of the metabolites associated with this pathway were significantly down-regulated at the G210 germination stage. Its content was highest at MD0 and continued to decrease during the stratification process (MD90, PD180, PD210) and the G210 germination stage, including p-coumaric acid, coniferaldehyde, coniferin, 5-caffeoylshikimic acid, and sinapinic acid. The results showed that DEGs involved in phenylpropanoid biosynthesis undergo complex regulation in the dormancy release of N. incisum seeds, and the significant reduction of metabolites in this pathway may be beneficial for N. incisum seeds to break through the seed coat and release dormancy.

Flavonoids are widely distributed in the secondary metabolites of plants [69]. CHS and CHI are key genes in the flavonoid biosynthesis pathway [70, 71]. Research has indicated that CHS is highly expressed during the germination stage of Tamarix hispida seeds [25]. In this research, the expression levels of CHS and CHI gradually increased at the PD210 stage compared with the first three stages, and their expression was significantly up-regulated at the G210 germination stage. This is consistent with the findings from Tamarix hispida seeds. The expression trend of the FLS gene was opposite to that of the MD0 stage, which was significantly up-regulated and then continued to decrease at the MD90, PD180, PD210, and G210 stages. In conclusion, most of the DEGs in the flavonoid biosynthesis pathway were markedly up-regulated at the germination stage of G210. However, in this research, most of the metabolites related to the biosynthesis pathway of flavonoids were highest in the MD0 stage of seeds, and their contents gradually decreased during the dormancy release process (MD90, PD180, PD210) and the G210 germination stage, including apigenin, luteolin, eriodictyol, isoquercitrin, trifolin, astragalin, and kaempferide. At the same time, studies have shown that during the germination of Chinese fir seeds, the majority of metabolites in the flavonoid biosynthesis pathway decreased [62]. Their content was highest during the seed swelling period and gradually decreased in the later stages of development, such as apigenin and luteolin. This finding is consistent with our study. In addition, the compounds that accumulated significantly more in the germination stage of G210 in this study included quercetin, rutin, delphinidin, and naringenin chalcone. Flavonoids may be the key substances affecting the dormancy release of N. incisum seeds. These findings imply that the flavonoid monomers required during seed stratification and after dormancy release may be different. In conclusion, this study showed at the transcriptome and metabolome levels that the gene expression and corresponding metabolite abundance of the phenylpropanoid and flavonoid biosynthesis pathways changed significantly with the process of dormancy release, revealing that this pathway passively responds to the transition of dormancy state rather than directly promoting seed dormancy release.

Conclusion

In conclusion, variable temperature stratification enabled the embryo development of N. incisum seeds to be complete, and the key stage for the dormancy release of N. incisum seeds was preliminarily determined. During the stratification process, endogenous hormone levels and carbohydrate metabolism changed significantly. We focused on plant hormone signal transduction, starch and sucrose metabolism, phenylpropanoid biosynthesis, and flavonoid biosynthesis pathways. Genes involved in the plant hormone signal transduction pathway were enriched in different stratification stages and showed differential regulation during and after seed dormancy release. ABA and GA-related genes were identified as primary targets for maintaining and releasing seed dormancy in N. incisum. During the stratification of N. incisum seeds, the conversion of starch to soluble sugars is initiated to supply the energy required for the alleviation of seed dormancy. Comprehensive analysis of transcriptome and metabolome data showed that phenylpropanoid biosynthesis and flavonoid biosynthesis pathways are crucial for the dormancy release of N. incisum seeds. N. incisum seeds affect germination by regulating flavonoid and phenylpropanoid metabolites. In addition, we analyzed the association between DEGs and DAMs, and some of the genes and metabolites encode enzymes potentially implicated in seed dormancy release. The findings indicate that the regulation mechanism of N. incisum seeds is likely not attributable to a singular gene or metabolite but rather a multifaceted regulatory system. This study enriched the information on the deregulation of seed dormancy in N. incisum.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Z.Q.W. and W.L.Z. performed bioinformatics data analysis and wrote the manuscript. H.G.C., J.B.Z., and H.J.D. prepared plant materials and performed physiological experiments. L.J. conceived the study and supervised the entire process. All authors read and approved the final manuscript.

Funding

This research was funded by the Gansu Provincial Higher Education Young Doctoral Fund Project (2022QB-093), the Gansu Provincial Natural Science Foundation Project (23JRRA1208), the Gansu Provincial Science and Technology Major Project (23ZDFA013-1), and the Central Guidance of Local Science and Technology Development Fund Project (24ZYQA041).

Data availability

The data supporting the findings of this study are available from the corresponding author on reasonable request. The raw data of RNA-seq were deposited in the NCBI SRA database (accession: PRJNA1230347).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.