Introduction

Rheum pumilum Maxim is a perennial herbaceous plant belonging to the genus Rheum in the Polygonaceae family1. This species is endemic to the Qinghai-Tibetan Plateau (QTP), and serves as a distinctive representative of the local flora on the QTP2. It thrives primarily in habitats such as sparse shrubbery forests, forest edges, and wet meadows at altitudes ranging from 2800 to 4700 m1. Its stem is typically short, measuring only 5–25 cm, an adaptation to the harsh alpine environment characterized by low temperatures, intense ultraviolet light, strong winds, and drought3,4,5. Rh. pumilum is also one of several fundamental varieties of Tibetan medicine Rhubarb, with its dried roots and rhizomes utilized in Tibetan medicine under the name Qumamei6. In recent years, escalating demand for traditional Chinese and Tibetan medicines has led to the overexploitation of wild rhubarb, resulting in a significant decline in its resources7. For instance, Lu and Lan (2016) noted an increase in the endangered status of Rh. pumilum from level 3 to level 2 during their assessment of endangered plant resources in Lhasa8. Similarly, Sun et al. (2018) found a decline in wild Rhubarb resources during their investigation of Tibetan medicinal plants in Qinghai Province9. Consequently, there is an urgent need for research on rhubarb germplasm resources and artificial domestication.

In natural environments, seed germination strategy is crucial for plant distribution and abundance10. Seed dormancy is an important strategy for controlling germination time in plants11. The current classification system categorizes seed dormancy into five types: morphological dormancy (MD), physiological dormancy (PD), morphophysiological dormancy (MPD), physical dormancy (PY), and a combinational dormancy (PY + PD)11,12. For species in alpine environments, seed dormancy is particularly significant as it prevents premature germination under extreme conditions and facilitates germination when environmental factors such as temperature and moisture are favorable. The non-deep physiological dormancy type of plant seeds in alpine environments is the most common12. Previous studies have identified typical non-deep physiological dormancy in Rheum species on the Tibetan Plateau13. Rapid seed germination is another effective adaptation strategy for alpine plants in extreme environments. Rapid germination enables seedlings to gain early growth advantages and quickly establish colonization, thereby enhancing their resilience against snowmelt and flood events caused by seasonal rains, as well as temperature fluctuations13,14. This germination strategy is often linked to specific environmental cues, such as increasing spring temperatures or changes in habitat conditions, as demonstrated in various alpine plant species13,14,15,16. Seed germination strategies tend to be conservative within plant families17, which is not surprising given that genetic information governing germination strategies becomes fixed during long-term adaptation to the environment, thereby becoming inherent genetic traits of the seeds.

Despite numerous studies on rapid seed germination in alpine plants, understanding the molecular mechanisms underlying this process remains elusive. Therefore, elucidating these molecular mechanisms is crucial for ecological and medicinal purposes. Recent advancements in genomic technologies have now made it possible to conduct comprehensive investigations into seed germination processes. Transcriptome analysis has become a widely utilized and effective method for studying seed germination in non-alpine plants. For instance, research has revealed the regulatory mechanisms of GA3 on seed germination rate, vigor, and water absorption in Fraxinus hupehensis seeds18. Studies on Camellia oleifera seeds have unveiled the roles of various plant hormones and transcription factors across different germination stages19. Combined transcriptome and metabolome analyses of Cunninghamia lanceolata have elucidated flavonoid and phenylpropanoid biosynthesis pathways20. Transcriptome analyses of Leymus chinensis seeds have identified crucial regulatory genes during germination21. Additionally, transcriptome analysis of upland rice seeds has highlighted the roles of GA and ABA in drought stress adaptation22. Together, transcriptome analysis has highlighted the pivotal roles of hormones and transcription factors in plant seed germination, and identifying numerous genes involved in key metabolic pathways. This wealth of data has significantly advanced our understanding of the molecular mechanisms governing seed germination. Despite these advances, a comprehensive understanding of transcriptional dynamics during seed germination in alpine plants remains limited.

Rh. pumilum exhibits distinct morphological adaptations to alpine environments, but the genetic basis governing its seed germination response remains poorly understood. This study employs transcriptome technology to achieve the following objectives: (1) explore gene expression variations in wild Rh. pumilum seeds across different germination stages; (2) identify key regulatory genes and gene expression patterns involved in seed germination; (3) explore the possible molecular mechanism responsible for rapid seed germination of Rh. pumilum in alpine environment. The results are expected to lay a foundation for improving seed germination rate, breeding practices to maintain sustainable development of Rh. pumilum.

Materials and methods

Sample collection and seed germination

Seeds of Rh. pumilum were collected from Maduo County in Qinghai Province (35° 52′ 21.64″ N, 99° 41′ 00.33″ E, 4255 m) and Dege County in Sichuan Province (32° 56′ 17.95″ N, 98° 20′ 29.05″ E, 4084 m), located on the QTP. Seeds from two regions with similar elevation were selected for analysis to increase representativeness. Intact seeds were chosen and stored at 4 °C with humidity of 55% for over three months. Subsequently, the chilled seeds underwent a 24 h soaking period in double-steamed water. These soaked seeds were then placed in a humid petri dish covered with double filter paper and cultured in an incubator set to a temperature of 25 °C, humidity of 60%, and a photoperiod of 12 h light and 12 h dark. Regular monitoring of seed condition was conducted, with timely addition of water to the petri dish to ensure normal seed germination. Ultimately, the overall germination rate of the seeds was 52.22%. On the 9th and 11th days of culture, seeds that broken in shells (LB) and seeds with 1 cm root growth (MY) were respectively collected. All seeds from these three different developmental stages were immediately frozen with liquid nitrogen and stored in a − 80 °C refrigerator for subsequent RNA extraction. Each sample was collected with three replicates, and each replicate comprised 3–5 seeds. Characteristics of the collected seeds at the three stages are illustrated in Fig. 1.

Figure 1
figure 1

Characteristics of the collected seeds at the three germination stages. DZ, dormant seeds; LB, seeds that broken in shells; MY, seeds with 1 cm root growth.

Full size image

RNA extraction and sequencing

Total RNA from Rh. pumilum samples, including three biological replicates, was extracted using Trizol (Takara, Dalian, China) reagent following the manufacturer's protocol. RNA purity and quantity were determined using a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA), while RNA integrity was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Subsequently, libraries were constructed using the VAHTS Universal V6 RNA-seq Library Prep Kit for Illumina as per the manufacturer’s instructions. These libraries were sequenced using the Illumina Novaseq 6000 platform, generating 150 bp paired-end reads. Transcriptome sequencing and analysis were performed by OE Biotech Co., Ltd. (Shanghai, China).

RNA sequencing data analysis

RNA sequencing data underwent initial processing using Trimmomatic23 to remove low-quality reads, resulting in obtaining clean reads. Clean reads were then de novo assembled into transcripts using Trinity 2.424, with paired-end method. The longest transcript was selected as a unigene based on both similarity and length for subsequent analyses. The functions of the unigenes were annotated by aligning them with the NCBI non-redundant (NR), Swiss-Prot, evolutionary genealogy of genes: Non-supervised Orthologous Groups (eggNOG), and Clusters of Orthologous Groups for Eukaryotic Complete Genomes (KOG) databases using Diamond25 with a threshold e < 1e−5. The proteins with the highest hits to the unigenes were utilized for assigning functional annotations. Additionally, the unigenes were mapped to the Kyoto Encyclopedia of Genes and Genomes (KEGG) database to annotate potential pathways26. Gene Ontology (GO) classification was performed by establishing mapping relationships between Swiss-Prot and GO terms. Following annotation, Bowtie227 was employed to determine the number of reads aligned to each unigene in every sample, and eXpress software28 was utilized to calculate the expression levels of unigenes (FPKM). Differentially expressed unigenes (DEGs) among different groups were identified using DESeq2 software29 to calculate the fold change and significance of difference using the negative binomial distribution test (NB), with default filter conditions for DEGs set at q < 0.05 and fold change > 2 or < 0.5. Rstudio (v2023.12.1–402)30 employed for generating column diagrams, bubble diagrams, and Venn diagrams in this study.

Co-expression network analysis

The co-expression network of Differentially Expressed Genes (DEGs) was constructed utilizing the Weighted Gene Co-expression Network Analysis (WGCNA) module within the Image GP platform31. A soft threshold was selected to generate networks exhibiting a scale-free topology, following methodologies outlined in prior studies32,33. Within this network, transcripts exhibiting similar expression profiles were clustered into modules denoted by same colors, with corresponding eigengenes calculated. Correlations between these eigengenes and seed germination stages were computed, with resulting correlation coefficients and p-values reported for each module. The top 20 genes demonstrating the strongest associations within each module were identified as hub genes, subsequently subject to further investigation. To visualize the correlation network of these hub genes, Cytoscape (v.3.10.1)34 was employed.

Quantitative real-time PCR

Eighteen DEGs were selected as representative genes for validation via quantitative real-time PCR (qRT-PCR) analysis, aimed at corroborating the results from RNA-seq analysis (Table S1). In view of the important role of hormones and transcription factors in seed germination, we detected the expression levels of six hormone-related genes and one transcription factor gene in the whole three stages of germination (Table S1). Additionally, 11 genes were randomly selected, showing significant differences in expression levels across various germination stages (| log 2 Ratio |> 1 and q < 0.05). The housekeeping gene GAPDH was selected as the internal reference gene according to previous studies35,36. qRT-PCR assays were conducted on a Mastercycler ep realplex4 Real-time Quantitative PCR System (Eppendorf, Hamburg, Germany) employing 2 × Universal SYBR Green Fast qPCR Mix (ABclonal, Wuhan, China) with the program as follows: 94 °C for 30 s; 40 cycles of 94 °C for 5 s, 59 °C for 34 s. The relative expression levels of DEGs were determined using the 2−ΔΔCT method. A comprehensive list of the representative genes and corresponding qRT-PCR primers is provided in Table S2.

Results

RNA-Seq analysis of developing seeds of Rh. pumilum and de novo assembly

The transcriptomes of Rh. pumilum seeds at three developmental stages (dormant seeds stored at 4 °C (DZ), seeds that broken in shells (LB), seeds with 1 cm root growth (MY)) were subjected to RNA sequencing using the Illumina NovaSeq 6000 platform. A total of 445,096,520 clean reads were obtained after quality control with 95.7% Q30 bases (Table S3). All clean reads were assembled into 64,836 unigenes, totaling 64,308,537 bp, with a mean length of 991.86 bp and an N50 of 1391 bp. The distribution of unigenes of different lengths is depicted in Fig. S1. Through homologous searches, all assembled unigenes were queried against the NR, KOG, GO, Swiss-Prot, eggNOG, and KEGG databases for functional annotations, with 69% of unigenes achieving hits in at least one database. NR had the highest proportion of successful annotations (68.28%), while KEGG had the lowest (17.72%) (Table S4).

Functional enrichment of unigenes

To gain insights into metabolic and gene function during seed germination, KEGG pathway and GO functional enrichment analyses were conducted. eggNOG analysis was employed to assess the completeness of the de novo transcriptome assembly. Results of all unigenes' functional annotations are provided in Table S5.

A total of 11,491 unigenes were assigned to 133 KEGG level 3 pathways, 20 KEGG level 2 pathways, and 6 KEGG level 1 pathways, encompassing “Cellular processes”, “Environmental information processing”, “Genetic information processing”, “Human diseases”, “Metabolism”, and “Organismal systems” (Fig. 2A). Among these, “Metabolism” featured the largest number of genes (7,864), primarily in pathways such as “Carbohydrate metabolism”, “Energy metabolism”, and “Amino acid metabolism”. “Genetic information processing” followed, with 4,525 genes, prominently in pathways like “Translation” and “Folding, sorting, and degradation”. “Human diseases” exhibited the fewest genes, with only 45. In general, genes are mainly enriched in two metabolic pathways: “Metabolism” and “Genetic information processing”.

Figure 2
figure 2

Functional enrichment of unigenes. (A) KEGG pathway; (B) GO classification; (C) eggNOG classification.

Full size image

A total of 29,462 unigenes were annotated in the GO database, revealing enrichment across 49 functional groups categorized under “Biological process”, “Cellular component”, and “Molecular function” (Fig. 2B). Within the biological process category, twenty-two functional categories were enriched, with the most prominent ones including “Cellular process”, “Metabolic process”, “Response to stimulus”, “Regulation of biological process”, “Biological regulation”, “Cellular component organization or biogenesis”, “Localization”, “Establishment of localization”, “Developmental process”, and “Multicellular organismal process”. In the cellular components category, fourteen functional groups demonstrated enrichment, with prominent ones such as “Cell”, “Cell part”, “Organelle, organelle part”, “Membrane, membrane part”, and “Macromolecular complex”. The molecular function category displayed enrichment in 13 functional groups, with “Binding and catalytic activity” being the most prevalent. Overall, the number of genes involved in biological process was the highest, while the number of genes involved in molecular function was the lowest.

The eggNOG classifications assigned to all 41,446 unigenes were distributed across 23 functional categories. The largest category was “Function unknown”, followed by “Posttranslational modification”, “Protein turnover, chaperones”, and “Translation, ribosomal structure and biogenesis”, while the smallest category was “Nuclear structure” (Fig. 2C).

Identification of DEGs

The nine samples were categorized into three groups corresponding to the three phases of seed germination (DZ, LB, MY) to assess gene expression dynamics during different developmental stages. Principal component analysis (PCA) revealed significant shifts in gene expression between the dormant seed state (DZ) and the latter two germination stages (LB, MY), with the latter two stages showing clustering and minimal variation between them (Fig. 3A). Notably, the number of differentially expressed genes (DEGs) varied significantly across the three groups, with 17,176 DEGs in LBvs.DZ, 19,972 DEGs in MYvs.DZ, and only 1,355 DEGs in MYvs.LB (Fig. 3B,C), underscoring the pivotal physiological transitions occurring at the broken in shells stage (LB). Moreover, a subset of 184 DEGs was identified as common among all three groups, implying their potential key roles in the overall germination process of Rh. pumilum (Fig. 3C).

Figure 3
figure 3

Comparison of differentially expressed genes (DEGs) in different stages of Rh. pumilum seed germination. (A) Principal component analysis of DEGs at different germination stages; (B) Comparison of DEGs quantity in different germination stages; (C) Venn diagram of DEGs quantity at different germination stages. DZ, dormant seeds; LB, seeds that broken in shells; MY, seeds with 1 cm root growth.

Full size image

To delve deeper into the metabolic activities associated with the identified DEGs, KEGG analysis was conducted. The DEGs from the three groups were found to be enriched in 74–129 metabolic pathways, with 129 pathways enriched in LBvs.DZ and MYvs.DZ, and 74 pathways enriched in MYvs.LB. Notably, minimal alterations in DEGs functions were observed between the MY and LB stages (Fig. S2). Further analysis revealed that the 184 common DEGs across all three groups were linked to 30 metabolic functions, including key pathways such as “Photosynthesis”, “Glycolysis/Gluconeogenesis”, “Photosynthesis-antenna proteins”, “Carbon fixation in photosynthetic organisms”, “Arginine biosynthesis”, “Porphyrin and chlorophyll metabolism”, and “Starch and sucrose metabolism” (Fig. S3). During Rh. pumilum seed germination, enhanced transcriptional activities, such as photosynthesis, plant hormone signal transduction pathways, carbohydrate synthesis, and energy metabolism pathways, were observed (Fig. S2). These enhancements may provide the necessary energy and nutrients for normal germination and establish comprehensive signal pathways crucial for adapting to the external environment.

Investigation of genes associated with hormone signaling pathways

In light of the pivotal role hormones play in seed germination, hormone-related genes and their expression levels were investigated. Through a comprehensive survey of KEGG functions, a total of 82 genes implicated in the metabolism of six key hormones were identified, namely Abscisic acid (ABA), Auxin (IAA), Ethylene (ETH), Brassinosteroid (BR), Cytokinin (CTK), and Gibberellin acid (GA). Among them, the highest number of genes was associated with IAA function, totaling 36, followed by ETH function with 13 genes, ABA function with 12 genes, CTK function with 9 genes, and GA function with 8 genes, while BR function exhibited the lowest count with only 4 genes (Fig. 4). Furthermore, these genes involved in plant hormone signal transduction were predominantly expressed during the LB stage (LBvs.DZ), as opposed to the MY stage (MYvs.LB) (Fig. 4). A comparison with dormant seeds revealed a further increase in gene expression levels during the MY stage (MYvs.DZ) compared to the LB stage (LBvs.DZ) (Fig. 4), indicating that the regulatory effects of hormones primarily manifest in the stage of seeds breaking in shells.

Figure 4
figure 4

Expression profiles of genes correlated with hormones at different stages of Rh. pumilum seed germination. The expression of genes was Z-score normalized and hierarchically clustered in heatmap. DZ, dormant seeds; LB, seeds that broken in shells; MY, seeds with 1 cm root growth.

Full size image

Analysis of transcription factors (TFs)

Transcription factors play a crucial role in governing gene expression during plant growth and development. A total of 594 unigenes representing 61 TF families exhibited differential expression patterns during seed germination. Among these, the bHLH family exhibited the highest enrichment of transcription factors, followed by bZIP, MYB, AP2/ERF-ERF, C2H2, MYB-related, C3H, WRKY, and NAC families (Fig. 5A). Regarding the fluctuation in TFs quantity, the transition from LB stage to MY stage showed subtle changes, with slightly higher TFs quantity observed during the MY stage compared to the LB stage (Fig. 5A). Notably, TFs displayed distinct expression patterns between the dormant seed stage and the two post-germination stages, with some exhibiting stage-specific patterns at LB and MY stages (Fig. 5B,C; Fig. S4).

Figure 5
figure 5

Transcription factors of Rh. pumilum seed germination. (A) The number of transcription factors in different germination stages; (B) Transcription factors with specific high expression at LB stage; (C) Transcription factors with specific high expression at MY stage. DZ, dormant seeds; LB, seeds that broken in shells; MY, seeds with 1 cm root growth.

Full size image

A total of 252 TFs exhibited high expression patterns in both LB and MY phases, with 206 TFs from 44 TF families highly expressed in the DZ stage, 66 TFs from 30 TF families highly expressed in the LB stage, and 57 TFs from 29 TF families highly expressed in the MY stage (Fig. 5B,C; Fig. S4). Notably, 15 TF families demonstrated specific high expression in the DZ stage, such as EIL, Whirly, VOZ, Trihelix, SBP, SAP, RWP-RK, NF-YB, NF-YA, NF-X1, MADS-M-type, FAR1, CAMTA, C3H, and AP2/ERF-AP2. Similarly, four TF families showed specific high expression in the LB stage, including C2C2-YABBY, CPP, ULT, and zf-HD, while six TF families exhibited specific high expression in the MY stage, namely C2C2-LSD, DBB, HB-BELL, HB-KNOX, HB-other, and LIM (Fig. S5, Table S6). Additionally, 14 TF families displayed high expression across all three developmental stages, namely LOB, GRAS, AP2/ERF-AP2, AP2/ERF-ERF, B3, bHLH, bZIP, C2C2-GATA, EIL, HB-HD-ZIP, MYB, MYB-related, NAC, TCP, and WRKY (Fig. S5, Table S6). Through KEGG pathway analysis, seventeen unigenes involved in signal transduction were identified, with sixteen related to plant hormone signal transduction and one involved in light signal transduction. Particularly noteworthy is that all seven TFs identified at the LB stage are associated with plant hormone regulation, with four related to ABA and the remaining three related to ETH, BR, and AUX, respectively, underscoring the pivotal role of plant hormones at this stage (Table S7).

Relationship between seed germination process and gene expression

To delve into the key gene information at different stages of seed germination, we conducted Weighted Gene Co-expression Network Analysis (WGCNA) based on all DEGs. Nine distinct expression modules were identified, with four displaying significant positive correlations (r ≥ 0.7, p < 0.05) with the seed germination process of Rh. pumilum, namely the MEbrown, MEgreen, MEyellow, and MEturquoise modules (Fig. 6). Here, we focus on further analyzing these four modules.

Figure 6
figure 6

Weighted Gene Co-expression Network Analysis of DEGs during Rh. pumilum seed germination. (A) Hierarchical cluster dendrogram of DEGs. The clustered branches represent different modules, and each line represents one DEG. (B) Module-trait associations. Each row corresponds a module characteristic gene (eigengene), each column corresponds to a germination stage. The color of each cell represents the correlation between the characteristic genes and the developmental stage, and the corresponding correlation coefficient and p-value are presented on the cell. A |correlation coefficient|≥ 0.7 and a p-value < 0.5 were considered to be significant. (C) The expression patterns of genes with significant correlations in different modules.

Full size image

The MEbrown and MEgreen modules exhibited significant associations with the DZ stage of seed germination. The MEbrown modules encompasses 538 unigenes, annotated across 63 KEGG pathways, predominantly linked to metabolic pathways including “Protein processing in the endoplasmic reticulum”, “Spliceosome”, “RNA transport”, “Glyoxylate and dicarboxylate metabolism”, “Starch and sucrose metabolism”, “mRNA surveillance pathway”, “Endocytosis”, “Alanine, aspartate, and glutamate metabolism”, “Ribosome biogenesis in eukaryotes”, and “Arginine biosynthesis”, etc. (Table S8). On the other hand, the MEgreen modules comprises 142 unigenes annotated across 25 KEGG pathways, primarily associated with “Protein processing in the endoplasmic reticulum”, “RNA transport”, “Plant hormone signal transduction”, “Ubiquitin-mediated proteolysis”, and “MAPK signaling pathway—plant”, etc. (Table S8).

The MEyellow modules, containing 264 genes, was found to be significantly correlated with the LB stage of seed germination. These genes are annotated across 40 KEGG pathways, with notable pathways including “Ribosome counting Oxidative phosphorylation”, “Flavonoid biosynthesis”, “Protein processing in the endoplasmic reticulum”, “N-Glycan biosynthesis”, “Various types of N-glycan biosynthesis”, and “Fatty acid biosynthesis”, etc. (Table S8).

The MEturquoise modules demonstrated significant correlation with the MY stage of seed germination. This module comprises 1033 unigenes annotated across 93 KEGG pathways, primarily associated with metabolic pathways such as “Photosynthesis”, “Carbon fixation in photosynthetic organisms”, “Phagosome”, “Amino sugar and nucleotide sugar metabolism”, “Glyoxylate and dicarboxylate metabolism”, “Plant-pathogen interaction”, “Plant hormone signal transduction”, “Oxidative phosphorylation”, “Glycolysis/Gluconeogenesis”, and “Photosynthesis—antenna proteins”, etc. (Table S8).

The 20 most highly connected genes (hub genes) in each module were utilized to construct the gene co-expression network (Fig. 7). In the DZ stage, the core hub genes were identified as TRINITY_DN25385_c0_g2 and TRINITY_DN18672_c0_g4. While the former is associated with cellular defense function, the function of the latter remains unknown (Fig. 7A,B; Table S9). Additionally, three transcription factor family genes, namely TRINITY_DN34140_c0_g1 (SBP), TRINITY_DN21285_c0_g2 (PLATZ), TRINITY_DN31577_c0_g1 (HSF), and two key genes, E2.4.1.13 (related to Starch and sucrose metabolism) and EIF1 (translation initiation factor 1), were also identified in this stage.

Figure 7
figure 7

Network relationships of hub genes in four modules with significant correlation with different germination stages. (A,B) Hub genes significantly associated with the DZ stage (MEbrown and MEgreen mudules); (C) Hub genes significantly associated with the LB stage (MEyellow mudules); (D) Hub genes significantly associated with the MY stage (MEturquoise mudules).

Full size image

In the LB stage, the core hub gene was WBP1 (TRINITY_DN26657_c0_g2), associated with N-Glycan biosynthesis (Fig. 7C). Furthermore, most hub genes at this stage are related to Ribosome metabolic pathways, indicating rapid seed development (Table S9). Additionally, two hub genes, ACACA and ATPeV0C, were found to be involved in fatty acid synthesis and oxidative phosphorylation, respectively, suggesting energy storage for seed development. In the MY stage, the core hub gene remained functionally unknown, identified as TRINITY_DN22567_c0_g1 (Fig. 7D). Another five hub genes were recognized: GAUT (TRINITY_DN29421_c0_g1), AUX1 (TRINITY_DN23956_c0_g2), ppa (TRINITY_DN31591_c0_g2), MSI (TRINITY_DN27878_c0_g1), and HEXA_B (TRINITY_DN24920_c0_g1). These genes are associated with functions such as amino sugar and nucleotide sugar metabolism, plant hormone signal transduction, oxidative phosphorylation, mRNA surveillance pathway, and various types of N-glycan biosynthesis (Table S9).

DEGs expression verified by qRT-PCR

To validate the reliability of the transcriptome data, eighteen DEGs from the LBvs.DZ and MYvs.LB comparison groups were selected for qRT-PCR expression analysis. These DEGs encompassed transcription factor genes, genes of unknown function, and genes involved in various metabolic pathways such as environmental adaptation, biosynthesis of secondary metabolites, lipid metabolism, and carbohydrate metabolism (Table S1). In the LBvs.DZ group, the qRT-PCR analysis revealed relative expression levels for six up-regulated DEGs and one down-regulated DEG. As expected, these seven DEGs exhibited expression patterns consistent with the RNA-Seq data (Fig. S6A). Similarly, in the MYvs.LB group, relative expression levels of four upregulated DEGs were detected, with their expression differences aligning with the RNA-Seq findings (Fig. S6B). These results underscore the accuracy and reliability of the RNA-seq results.

Furthermore, considering the pivotal roles of hormones and transcription factors, six key hormones and one transcription factor (Table S1) were selected to assess their expression dynamics throughout the entire process of seed germination. The results demonstrated that the expression trends of hormone-related genes and the transcription factor gene across the three stages of seed germination were in concordance with the RNA-seq data, with significant differences in expression observed at the LB stage and DZ stage (Fig. S6C). This consistency further supported the expression profiles of hormone-related genes depicted earlier (Fig. 4).

Discussion

Seed germination in plants involves a series of intricate physiological transformations and energy conversion processes, orchestrated by the cooperative action of key genes. In this study, transcriptome sequencing identified a range of pivotal genes associated with transcription factors, hormone transduction, and metabolic pathways during the germination of Rh. pumilum seeds, thus providing insights into the potential molecular mechanisms underlying the germination process.

Rapid seed germination strategies of Rh. pumilum

Comprehensive analysis of the transcriptome revealed a pronounced similarity in the germination process of Rh. pumilum seeds during the LB and MY stages. In the Principal Component Analysis of DEGs, there was substantial overlap between the DEGs in the LB and MY stages (Fig. 3A). Similarly, the functional enrichment analysis of DEGs indicated minimal alteration in functional categories between LB and MY stages, except for slight variations in DEGs quantities (Fig. S2). Similar germination patterns were observed in Meconopsis integrifolia37, also an alpine plant, and Cinnamomum migao, a traditional Chinese medicine38, suggesting they may employ comparable germination strategies.

Furthermore, the survey of Transcription Factors (TFs) revealed identical TF families between LB and MY stages, with no significant disparity in their numbers (Fig. 5A). While multiple genes associated with six hormones exhibited significant expression differences between LB and DZ stages, no such disparities were noted between LB and MY stages (Fig. 4). Hormones and transcription factors play pivotal roles in seed germination, and the expression patterns observed here may represent a unique strategy specific to alpine plants. These results underscore the pivotal role of the LB stage in the germination process of Rh. pumilum seeds. Analysis of DEGs functional enrichment during the LB stage revealed the initiation of crucial metabolic activities including energy metabolism, MAPK signal transduction, hormone signal transduction, lipid metabolism, and amino acid metabolism (Fig. S2). These metabolic processes furnish the substrates, energy, and informational foundation necessary for Rh. pumilum seed germination. Similar functional enrichment results were observed in M. integrifolia and C. migao, suggesting that both alpine and non-alpine plants exhibit complete germination from the LB stage. Notably, Rh. pumilum seeds ruptured after 9 days of cultivation, M. integrifolia after approximately 15 days of cultivation37, while C. migao after more than 24 days of cultivation38. This indicates that Rh. pumilum seeds may initiate rapid germination in a shorter time frame than non-alpine plants, potentially due to the involvement of key transcription factors.

Previous morphological and physiological studies have revealed that species within the Rheum genus employ rapid seed germination strategies to confront significant temperature variations, promote seedling establishment, and effectively evade flooding, thus facilitating adaptation to the distinctive high-altitude environment13,39. The outcomes of this investigation furnish robust genetic substantiation for these studies. The swift seed germination strategy of Rh. pumilum facilitates rapid dissemination and enhances adaptation to harsh natural conditions.

Important hormones and key genes associated with Rh. pumilum seed germination

Hormones play a pivotal role in modulating seed dormancy and germination. Abscisic acid (ABA) and Auxin (IAA) collaborate to uphold seed dormancy40. Gibberellin acid (GA), Brassinosteroids (BR), and Cytokinins (CTK) promote seed germination by counteracting ABA, whereas Ethylene (ETH) positively regulates seed germination by augmenting the levels of GA, BR, and CTK41. During Rh. pumilum seed germination, highly expressed genes in association with the above six hormones were concurrently observed (Fig. 4). Notably, genes related to ABA and IAA exhibited elevated expression levels, such as ABA-related PYL and CYP707A, and IAA-related AUX1 and IAA (Fig. 4). In other plants, CYP707A is a key gene for seed germination, which participates in regulating the catabolism of ABA, thereby promoting the breaking of seed dormancy21,42. ABA not only participates in seed dormancy but also regulates various biological processes, including responses to environmental stresses like drought, salt, and cold stresses43. It is hypothesized that PYL, AUX1, and IAA genes may contribute to the adaptability of Rh. pumilum seeds to environmental stress during germination by modulating the expression of these two hormones (ABA and IAA). Certain hormone-related genes have exhibited heightened expression, such as ETR and EIN3 genes associated with ETH function, GID1 and E1.14.11.13 genes related to GA function, and CYP85A1, BZR1_2, and BIN2 genes associated with BR, along with IPT and AHK2_3_4 genes related to CTK function (Fig. 4). ETR is implicated in regulating starch accumulation and seed germination44. EIN3, a transcription factor of the EIL family, modulates ethylene expression and is significantly linked to plant disease resistance and cold tolerance45,46,47. GID1 acts as a GA receptor, while E1.14.11.13 is involved in the GA biosynthesis pathway at low temperatures48. BZR family genes play crucial roles in enhancing plant resistance to abiotic stresses and act as positive regulators of BR synthesis49,50. In addition, BZR1 is a key gene involved in plant photomorphogenesis, enabling seeds to better adapt to the external environment at the early stage of germination51. Therefore, it is speculated that BZR1_2 may play a similar important role in seed germination of Rh. pumilum. CYP85A1 is a pivotal enzyme in the BR synthesis pathway, and BIN2 negatively regulates plant resistance to pathogens52. BRI1 is associated with rapid seed germination after cold stratification53. Both IPT and AHK genes serve as key enzymes in CTK synthesis.

In summary, six hormones (ABA, IAA, ETH, GAs, BR, CTK) play pivotal roles in breaking seed dormancy and facilitating rapid seed germination of Rh. pumilum. Additionally, certain hormones such as ABA, ETH, and BR may also contribute to Rh. pumilum's response to abiotic and biotic stresses as in other species22,47,54. Rh. pumilum thrives in high-altitude areas, characterized by typical alpine climate features: low average temperatures, significant temperature fluctuations, prolonged winters, water scarcity, short summers with rain, and easily triggered ice and snow melting leading to floods. This necessitates early germination and swift seedling establishment to ensure a sufficiently long growing period to cope with summer floods and extended winter temperatures13. In fact, we observed a significant increase in the expression of hormone-related genes during the LB stage, which was also confirmed by qPCR experiments (Fig. 4, S6). Among these, several key genes deserve attention: CYP707A may promote breaking seed dormancy, ETR facilitates seed germination by accumulating starch for energy, EIN3, BZR, and BIN2 contribute to seed resistance against low temperatures and pathogens, while BRI1 promotes rapid seed germination. Additionally, BZR1_2 possibly participates in seedling photomorphogenesis. It's plausible that the collective action of these genes enhances the adaptability of Rh. pumilum seeds to alpine climates.

In the breeding of medicinal herbs, enhancing seed germination rates can be approached from two perspectives based on our results. Firstly, by regulating the identified key hormone-related genes and optimizing the endogenous hormones of Rh. pumilum, we can potentially increase the germination rate. Secondly, this study suggests that besides the commonly studied ABA/GA hormone combination for germination of Rh. pumilum seeds, four other hormones may also enhance seed germination. Notably, GA3, commonly used in practice, has shown limited efficacy in improving the germination rate of Rheum species13. Therefore, exploring alternative exogenous hormones or combinations thereof could be a promising avenue for improving the seed germination of Rheum species. This represents a straightforward and expedient technique worthy of further exploration.

Transcription factors and other important genes related to Rh. pumilum seed germination

Transcription factors (TFs) play pivotal roles in various biological processes by modulating gene expression levels through binding to promoter regions. In our investigation, we identified 61 TF families involved in Rh. pumilum seed germination, with 14 TF families particularly crucial across all developmental stages (Table S6). Notably, eleven of these families, such as LOB, GRAS, B3, bHLH, bZIP, EIL, MYB, MYB-related, NAC, TCP, and WRKY, are implicated in plant responses to biotic and abiotic stresses, including pathogens and environmental factors like low nitrogen, drought, salt, or cold stress55,56,57,58,59,60,61,62,63,64,65,66.

Moreover, co-expression analysis via WGCNA revealed several significant genes at different germination stages. In the DZ phase (MEbrown and MEgreen modules), TRINITY_DN31577_c0_g1, TRINITY_DN21285_c0_g2, and TRINITY_DN34140_c0_g1, representing HSF, PLATZ, and SBP TF families respectively (Fig. 7A,B), have been implicated in responding to drought and cold stress67,68,69. The SBP gene family also plays an important role in regulating plant growth and development, transition at different developmental stages, secondary metabolism, fertility maintenance, plant hormone response, and other aspects, which deserves special attention70,71. Additionally, TRINITY_DN23818_c0_g2, encoding a sucrose synthase E2.4.1.13 (Fig. 7A), is associated with plant drought resistance72. This suggests that Rh. pumilum seeds begin responding to abiotic stress after after-ripening (DAR), facilitating seed germination. During the LB phase (MEyellow modules), ribosome-related genes predominate as hub genes, alongside ATPeV0C (TRINITY_DN3059_c0_g1) and ACACA (TRINITY_DN27432_c0_g1), involved in plant substances and energy metabolism, crucial for subsequent germination stages (Fig. 7C, Table S9). In the MY phase (MEturquoise modules), genes linked to the Golgi apparatus, like TRINITY_DN29421_c0_g1(GAUT) and TRINITY_DN26872_c0_g1, were identified, aiding protein transport for diverse physiological functions (Fig. 7D, Table S9). Additionally, a photosynthesis-related gene (TRINITY_DN26872_c0_g1) was identified, which was not annotated by KEGG, and may be potentially unique to Rh. pumilum. Another alternative splicing associated gene, TRINITY_DN27878_c0_g1 (MSI), was identified and plays an important role in a variety of plant physiological activities (Table S9).

These results could support to the Rh. pumilum's swift germination strategy in alpine environments. Post-DAR during the long winter, the genetic foundation for germination is laid under low temperatures. In spring, despite suboptimal conditions (low soil temperature and moisture), seeds initiate germination swiftly, achieving seedling colonization, and the physiological activities of resistance to drought and low temperature are always accompanied in this process. The identification of these pivotal genes also offers valuable insights for implementing artificial intervention strategies aimed at enhancing the seed germination rate during the cultivation of Rh. pumilum.

Conclusions

In this study, transcriptome sequencing was utilized to scrutinize the alterations in gene expression levels across three distinct stages of Rh. pumilum seed germination. Our results reveal the genetic mechanisms behind the swift seed germination strategy adopted by alpine plants to thrive in challenging environments.Furthermore, several key genes, including hormones and transcription factors, were identified as potentially influencing the rapid germination process of Rh. pumilum seeds. These results can provide data support for improving seed germination rate and breeding of Rh. pumilum, so as to promote agricultural planting, reduce the use of wild resources and maintain the sustainable development of Rh. pumilum germplasm resources. However, it should be noted that this study's findings are solely based on transcriptome analysis and do not validate the functions of these genes. Future research should focus on examining the response and regulation of these key genes under diverse conditions to better understand the underlying mechanisms of Rh. pumilum seed germination.