Abstract Drought restricts plant growth and agricultural production. As an important medicinal plant, Blumea balsamifera is sensitive to water, but there is still a lack of systematic research on its drought response mechanism. In this study, four-month-old B. balsamifera seedlings were used as materials, and three groups were set up: normal irrigation (CK), drought stress (DS), and rewatering recovery (RW). The results showed that drought significantly inhibited the growth and photosynthesis of B. balsamifera. With the prolongation of stress time (day 12), the limiting factor of photosynthesis changed from initial stomatal limitation to non - stomatal limitation. In terms of physiology and biochemistry, B. balsamifera increased MDA content by actively reducing SPAD value and relative water content of leaves; and activates the antioxidant enzyme system to remove ROS, synergistically accumulates lignin, soluble sugar, proline and other osmotic adjustment substances, and jointly maintains cell water balance and membrane system stability. Through transcriptome and proteome analysis, 20,874 DEGs and 2770 DEPs were screened out, which were significantly enriched in terms related to ribosome, oxidoreductase activity, biosynthesis of unsaturated fatty acids and other pathways. A total of 55 drought - related DEGs - DEPs were identified by two - omics, and 18 key regulatory genes were screened. In summary, B. balsamifera formed a comprehensive drought resistance mechanism through photosynthesis, physiology and DEGs - DEPs network. This study provides theoretical support for the breeding and resource development of B. balsamifera, and also provides reference for the study of stress resistance of other medicinal plants. Supplementary Information The online version contains supplementary material available at 10.1186/s12870-025-06916-w. Keywords: Drought stress, Drought resistance mechanism, Transcriptomics, Proteomics, Blumea balsamifera L Introduction Plants are inevitably exposed to various environmental stresses during their growth, such as abiotic and biotic stresses [[44]1]. With global climate change intensifying, drought has become one of the major factors threatening agricultural production and the ecological environment [[45]2–[46]4]. Physiologically, plants increase the contents of proline, soluble sugars, soluble proteins to reduce the osmotic potential and enhance their water retention capacity [[47]5–[48]7]. At the molecular level, plants counter abiotic stress by activating drought-responsive genes and associated functional molecules (e.g., CmbZIP9 [[49]8], PIP aquaporins [[50]9], chitinase [[51]10]). Overexpression of the CmbZIP9 gene in Chrysanthemum mongolicum enhances drought tolerance by decreasing malondialdehyde (MDA) content [[52]8]. Furthermore, studies in other species have demonstrated that overexpression of genes such as GhMYB102 [[53]11], RcECP [[54]12], and TaERF87 [[55]13] improves plant drought adaptability. In recent years, with the deepening of research, a large number of drought-resistant genes and metabolic pathways have been elucidated, such as brassinosteroids (BRs) signaling [[56]14], Reactive Oxygen Species (ROS) signaling [[57]15], abscisic acid (ABA) induction [[58]16], and calcium signal regulation [[59]17]. In Arabidopsis thaliana (L.) Heynh., the HBI transcription factor can regulate the nitrate signaling pathway, effectively reducing the accumulation of ROS in the plant body and alleviating the damage caused by oxidative stress to the plants [[60]18]. In addition, the AtERF71/HRE2 transcription factor plays a crucial role in the response of A. thaliana to complex environmental stresses. It not only mediates the ABA signaling pathway to regulate the plant’s response to drought stress but also participates in the regulation of the hypoxia response mechanism in A. thaliana [[61]19]. Blumea balsamifera (L.) DC. is a perennial herb belonging to the Asteraceae family. It is the only source of traditional Chinese medicine Aipian and one of the top ten Miao medicine in Guizhou Province [[62]20] The fresh leaves are extracted and processed into Aipian, which has the effects of resuscitation, heat - clearing and pain - relieving. It can be used to treat febrile coma, convulsions, stroke and symptoms related to phlegm congestion [[63]21]. The study found that B. balsamifera showed strong adaptability under drought conditions, which was closely related to its unique biological characteristics: its roots were developed and extended up to 1.5 m, which were good pioneer plant species for soil and water conservation [[64]22–[65]24]. In addition, B.balsamifera showed higher photosynthetic efficiency and water use efficiency under drought conditions, which provided potential for its cultivation and ecological restoration in arid and semi - arid areas [[66]25]. However, there is still a lack of systematic research on the physiological adaptation mechanism and molecular regulation network of B.balsamifera in response to drought stress. With the advancements in sequencing technology and plant genome assembly technology, researchers have delved into and revealed the mechanisms underlying plants’ responses to drought stress at the molecular level. For example, investigations into the transcriptome and proteome of Hippophae rhamnoides have indicated that the peroxisomal channel gene (PEX19) and plant hormone signal transduction genes (PYR/PYL) are linked to drought resistance [[67]26]. In tomatoes, a combined analysis of differentially expressed genes (DEGs) and proteins (DEPs) has demonstrated that activating AREB1 expression can regulate TAS14, GSH-Px1, and Hsp, thereby enhancing the plant’s drought resistance [[68]27]. Moreover, through co - expression network analysis of the transcriptome and proteome of Eleusine coracana (L.) Gaertn., two candidate DEGs-DEPs associated with drought resistance have been uncovered [[69]28]. Collectively, these multi-omics studies have significantly broadened our understanding of the molecular mechanisms of plant drought resistance. In this work, the growth indexes, as well as the physiological and biochemical indexes of B. balsamifera leaves under drought stress were determined, and the molecular regulatory network of drought-stress response was comprehensively analyzed by integrating transcriptome and proteome analysis. The research results can not only enrich the genetic information related to B. balsamifera’s drought-stress response, but also provide theoretical support for breeding new varieties with strong drought resistance, and contribute to its ecological restoration and medicinal resource development in arid and semi-arid areas. Materials and methods Test materials and drought In April 2024, B. balsamifera seeds were collected from Hainan Tropical Botanical Garden (19°30’36"N, 109°34’12"E) and subsequently identified by Prof. Pang Yuxin of Guizhou University of Traditional Chinese Medicine following the classification criteria of Flora Reipublicae Popularis Sinicae [[70]29]. These seeds were then sown and raised in the No. 2 greenhouse (26°22’20"N, 106°27’34"E) of the Medicinal Plant Cultivation Test Nursery at Guizhou University of Traditional Chinese Medicine. In this study, plants were planted in a non-thermostatic greenhouse with natural ventilation and light transmission. The universal seedling substrate used was composed of peat, coconut coir, vermiculite, and perlite, produced by Xiangzhengnongke. The planting containers were flowerpots with a height of 85 mm and a diameter of 100 mm, with single-plant per pot planting.​ In August 2024, four-month-old B. balsamifera seedlings with uniform growth (average plant height: 18.76 cm) were selected for drought stress research. Three treatments were applied: control (CK) with normal irrigation; drought stress (DS); and rewatering recovery (RW). Each treatment comprised 3 biological replicates. In the greenhouse, treatments were arranged using a completely randomized block design, with pot spacing exceeding 15 cm to minimize inter-plant interference. Pots were evenly arranged in rows and columns on the seedbed. In the CK group, plants were watered daily in the evening until water exuded from the bottom of the pots. The DS group underwent continuous drought stress treatment for 16 days following uniform pre-experimental irrigation. For the RW group, one-time saturating rewatering treatment was conducted on the 8th, 10th, 12th, and 14th days of drought stress, respectively. Sampling was performed 2 days after rewatering, corresponding to the 10th, 12th, 14th, and 16th days. For the CK group and DS group, sampling time points were set as the 0th, 5th, 8th, 10th, 12th, 14th, and 16th days. At each time point, three plants per treatment were randomly selected to measure growth indices and photosynthetic parameters. The 4th − 6th middle leaves of each plant were sampled, avoiding the main vein. Leaf relative water content was measured immediately, and the remaining leaves were placed in labeled sterile cryotubes, snap-frozen in liquid nitrogen, and stored at −80 °C for subsequent analysis. Measurement of growth indicators Plant height (H), leaf length (LL), and leaf width (LW) were measured using a tape measure. Leaf thickness (LT) and basal stem diameter (BS) were measured with a vernier caliper. Subsequently, the plant height growth (ΔH) was calculated. Assessment of photosynthetic indices A portable LCI photosynthetic instrument (Beijing Aozuo Ecological Instrument Co., Ltd.) was used to measure photosynthetic - related indicators before sampling (from 9:00–11:00 a.m., with a fixed light intensity of 800 µmol m⁻²s⁻¹). The measured indicators included net photosynthetic rate (Pn), stomatal conductance (Gs), transpiration rate (Tr), and intercellular CO₂ concentration (Ci). Subsequently, the instantaneous water use efficiency (WUEi), stomatal limitation (Ls), and carboxylation efficiency (CE) were calculated [[71]30]. The calculation formulas are as follows: graphic file with name d33e563.gif graphic file with name d33e571.gif graphic file with name d33e579.gif In the formula: Ca is atmospheric CO[2] concentration. Determination of physiological and biochemical indicators The relative chlorophyll content (SPAD value) of B. balsamifera seedlings’ leaves was measured using a SPAD − 502PLUS plant chlorophyll content meter. The relative water content (RWC) of leaves was determined by the saturated water weight method [[72]31]. The calculation formulas are as follows: graphic file with name d33e601.gif For the determination of malondialdehyde (MDA) content and the assessment of the antioxidant system in B. balsamifera seedlings’ leaves, the MDA content, superoxide dismutase (SOD) activity, catalase (CAT) activity, and peroxidase (POD) activity were measured using the thiobarbituric acid colorimetric method [[73]32], WST − 8 method [[74]33], ultraviolet spectrophotometry [[75]34], and guaiacol colorimetric method [[76]35], respectively. Regarding lignin (LIG) and osmotic adjustment substances, the contents of lignin, soluble sugar (SS), soluble protein (SP), and proline (PRO) in the leaves of B. balsamifera were determined by the acetylation method [[77]36], anthrone colorimetry [[78]37], coomassie brilliant blue method [[79]38], and ninhydrin colorimetry, respectively [[80]39]. Comprehensive phenotypic traits, photosynthetic parameters, and physiological-biochemical indices revealed significant intergroup variations at day 10 of treatment. The net photosynthetic rate (Pn) and soil plant analysis development (SPAD) values in the DS group exhibited inflection points, indicating that plants had entered a critical transition phase from stress adaptation to damage. Hypothesizing that differential expression of drought-responsive genes mediates this regulatory switch, this time point was selected for subsequent transcriptomic sequencing analysis. Transcriptome analysis On the 10th day, leaf tissue samples of CK and DS (3 samples each) were selected for transcriptome sequencing analysis. RNA was extracted from the 6 tissue samples using the MJZol total RNA extraction kit (Shanghai Meiji Biomedical Technology Co., Ltd.). Using 1 µg of total RNA, the RNA-seq transcriptome library was prepared according to the Illumina^® Stranded mRNA Prep, Ligation kit (San Diego, CA). The concentration, purity, and integrity of the extracted RNA were detected using Nanodrop2000, agarose gel electrophoresis, and Agilent5300, respectively, with the RQN value determined by the latter. Sequencing was carried out on the Illumina Miseq platform (Shanghai Meiji Biomedical Technology Co., Ltd.). After basic quality control and filtration, the raw paired-end reads were trimmed and quality-controlled by the fastp tool [[81]40]. The criteria included removing reads containing adapter sequences, reads with more than 5% of unknown bases (N), and reads where more than 20% of the bases had a quality score ≤ 20. Clean data with high quality were then used for subsequent analysis. For the procedure of eliminating redundancy, filtering, and optimization, the Trinity software was first employed to perform de novo assembly on all clean data to generate primary Unigene. Subsequently, the CD-HIT program was used to cluster the primary Unigene sequences to eliminate redundancy, with a sequence similarity threshold of 0.95. Meanwhile, sequences shorter than 200 bp were filtered out to obtain non-redundant Unigene. The expression levels of genes and transcripts were quantified using the RSEM software (default parameters, [82]http://deweylab.github.io/RSEM/). Differential gene expression was analyzed with DESeq2 software, and the screening threshold was set as: |log2FC| >= 1 and padjust < 0.05 [[83]41, [84]42]. Based on the gene expression information in different samples, the gene expression patterns were clustered. The DEGs in the gene set were then subjected to functional annotation and enrichment analysis in the GO (Gene Ontology, [85]http://www.geneontology.org/) and KEGG (Kyoto Encyclopedia of Genes and Genomes, [86]http://www.genome.jp/kegg/) databases. Proteome analyses On the 10th day, 3 tissue samples each of CK and DS were ground into powder in liquid nitrogen. The powder was suspended in the protein lysate (8 M urea + 1% SDS, containing protease inhibitor cocktail) to extract the protein supernatant. The supernatant was divided into two parts: one for on - line analysis and the other for protein detection via SDS-PAGE electrophoresis. TMT labeling and mass spectrometry analysis were performed on the tissue proteins (Shanghai Meiji Biotechnology Co., Ltd.). Reverse - phase liquid chromatography (RPLC) was used for one - dimensional separation, and nano - upgrade liquid chromatography tandem mass spectrometry (EasynLC 1200 combined with Q Exactive mass spectrometer) was employed for the second-dimensional separation. Mass spectrometry was then used to identify all proteins and their sequences, which were compared against seven databases (EggNOG, GO, KEGG, NR, Pfam, String, Uniprot). Based on the expression quantification results, differentially expressed proteins between the two groups were analyzed. The R Project for Statistical Computing ([87]https://www.r-project.org/) was used to analyze the differentially expressed proteins (DEPs). A fold - change (FC) of ≥ 2 for up-regulated proteins, ≤ 0.5 for down-regulated proteins, and a P-value < 0.05 were set as the default criteria to indicate significant differences [[88]43]. Association analysis between transcriptome and proteome Shanghai Meiji Biotechnology Co., Ltd. provided a platform to correlate and compare the transcriptome and proteome sequencing data [[89]44]. The molecular expression of B. balsamifera tissues under drought stress was compared and studied at the RNA and protein levels. Key molecules were further explored to reveal the plant’s drought-stress response mechanism. When DEPs are associated with corresponding DEGs at the transcriptional level, the resulting combinations are called gene-protein pairs (DEGs/DEPs). We used Venn diagrams and cluster heat maps to present the association data between differential mRNAs and differential proteins among groups, and performed quantitative correlation analysis using the Pearson correlation coefficient. Function annotation and GO and KEGG enrichment analyses were then conducted by leveraging the blastp function in Diamond software([90]https://github.com/bbuchfink/diamond). The protein-protein interaction (PPI) network of drought-responsive DEGs and DEPs was constructed using the STRING database ([91]https://string-db.org/). The parameters were configured as follows: Helianthus annuus was designated as the reference species, the default comprehensive score threshold of 0.4 was applied, and the top 50 proteins with the highest comprehensive scores were selected. Subsequently, the gene co-regulatory relationships were visualized using Cytoscape software ([92]https://cytoscape.org/), an open-source platform for complex network analysis and visualization. Determination of gene expression levels (RT-qPCR) To verify the accuracy of mRNA differential expression identified by high - throughput sequencing, nine transcripts were selected from the DEGs/DEPs that showed associations in the integrated transcriptome - proteome analysis for RT - qPCR verification. Subsequently, the RT - qPCR results were compared with those of RNA - Seq analysis. RNA was extracted using the MJZol total RNA extraction kit (Shanghai Meiji Biomedical Technology Co., Ltd.). The qualified RNA was reverse - transcribed into cDNA using ExonScript RTMix (with dsDNase), with 18 S rRNA serving as the internal reference gene (forward primer: CGGCTACCACATCCAAGGAA, reverse primer: GCTGGAATTACCGCGGCT). The experimental procedures were carried out following the instructions of the SYBR Prime qPCR kit (FastHS, product code BG0014). For RT - qPCR, three biological replicates (each from independent plants of the same treatment group) were used at each time point, with three technical replicates per biological replicate. Table [93]S1 lists the genes and corresponding primers used in the qPCR test. The relative expression of the target gene was calculated using the CT method (2^−△△Ct). Data statistics and analysis SPSS22.0 and Excel2021 were used for the statistical analysis of the experimental data, and GraphPad Prism9.5 and OriginPro2024 software were used for graphing. All data were presented as the average of three replicates. All data were presented as the mean ± standard deviation (mean ± SD). Duncan’s new multi - range test was used to determine whether there were significant differences among groups. Different lowercase letters indicate significant differences among different treatments at the same drought time point (P < 0.05). Results Phenotypic response Under drought stress treatments, the well-watered CK group maintained robust growth. Its leaves were plump, without wilting or drooping, and the stems were erect. In the DS group, the leaves started to wilt and droop from the 8th day, and these symptoms gradually worsened as the drought time extended. In the RW group, the stems and most leaves could recover after timely rehydration. However, as the drought time prolonged, some leaves still showed wilting and drooping, or even yellowing and withering (Fig. [94]1A). The growth indexes of B. balsamifera, such as H, ΔH, and BS, at certain time points, CK exhibited a trend of being higher than RW, which in turn was higher than DS (Fig. [95]1B and G). After 8 days of treatment, significant differences were observed in H, ΔH, and other relevant indexes between the DS and RW groups compared with the CK group (Fig. [96]1A and B). In the first 10 days, there was no significant difference in LL and LT among the groups (Fig. [97]1E and G). After 12 days, significant differences emerged in BS and LW (Fig. [98]1D and F). Fig. 1. [99]Fig. 1 [100]Open in a new tab Effect of drought stress on growth and morphology of B. balsamifera. The first three time points of the RW group showed the phenotype and data of drought treatment, and the subsequent four time points showed the change of phenotype data after rehydration. A Phenotypic characters. B, C, D, E, F and G Plant height, plant height growth, basal stem diameter, leaf length, leaf width and leaf thickness. Different lower case letters in the graphs indicate significant differences (P < 0.05), and same lower case letters indicate no significant differences (P > 0.05) Photosynthetic response The Pn and Tr of the DS group decreased to the lowest value at 16 d and were significantly lower than those of the CK and RW groups. The Pn and Tr of the RW group first increased, then decreased, and finally gradually declined (Fig. [101]2B and C). The Ci of the DS group dropped to its lowest at 12 d, while the Ci of the RW group was significantly lower than that of the CK and DS groups after 14 d (Fig. [102]2D). The trend of Gs was similar to that of Tr. Gs decreased to its lowest at the end of the drought treatment, but there was no significant difference compared with the Gs of the RW group (Fig. [103]2E). With the aggravation of drought stress, the CE of the DS group continuously decreased, while the CE of the RW group peaked on the 10th day (Fig. [104]2F). Ls gradually increased, showing an opposite trend to Ci. In the DS group, Ci gradually increased after reaching its valley value at 12 d and peaked at 16 d, indicating that 12 d might be the key node for stomatal limitation (Fig. [105]2G and D). Meanwhile, comparison of growth status across groups on the 12th day of treatment revealed significantly curled and wilted leaves in the DS group (Fig. [106]2A). It was speculated that in the DS group, damage to the photosynthetic structure of the leaves led to stomatal closure. Additionally, the WUEi of the DS group reached its maximum at the end of the drought stress on day 16 and was significantly higher than that of the CK and RW groups (Fig. [107]2H). Fig. 2. [108]Fig. 2 [109]Open in a new tab Effect of drought stress on photosynthesis-related parameters of B. balsamifera.A Comparison of the growth status of each group at the 12th day of treatment. B, C, D, E, F, G and H Pn, Tr, Ci, Gs, CE, Ls and WUEi. Different lowercase letters in the graphs indicate significant differences (P < 0.05), and the same lowercase letters indicate no significant differences (P > 0.05) Physiological and biochemical responses The SPAD value of B. balsamifera leaves in the DS group initially trended upward, yet showed no significant difference compared to the CK group. Subsequently, it decreased sharply and became significantly lower than that of the CK and RW groups (Fig. [110]3A). The RWC decreased continuously and was significantly lower than that of the CK group starting from the 8th day. The RWC of the RW group could return to the CK-group level on the 10th day, but then it was significantly lower than that of the CK group, though still higher than that of the DS group (Fig. [111]3B). Further investigations revealed that drought stress significantly increased the MDA content, suggesting aggravated membrane lipid peroxidation (Fig. [112]3C). Meanwhile, the antioxidant system (SOD, POD, and CAT) was activated to alleviate oxidative damage. The activities of SOD and POD in the DS group peaked at 16 d, while the CAT activity increased first and then decreased (Fig. [113]3D and F). Additionally, drought stress significantly elevated the LIG content (Fig. [114]3G). The contents of SS and SP in the DS group peaked on the 12th day and then declined, being significantly higher than those in the CK group. The PRO content kept increasing, reaching 282.68 µg·g⁻¹ at the end of the treatment, which was 20.92 times that of the CK group and significantly higher than that of the CK and RW groups (Fig. [115]3H and J). Fig. 3. [116]Fig. 3 [117]Open in a new tab Effect of drought stress on leaf physiology and biochemistry of B. balsamifera.A SPAD comparison. B RWC comparison. C, D, E and F MDA content, SOD, POD and CAT activities. G, H, I and J LIG, SS, SP and PRO contents. Different lowercase letters in the graphs indicate significant differences (P < 0.05), and the same lowercase letters indicate no significant differences (P > 0.05) Comprehensive phenotypic traits, photosynthetic parameters, and physiological-biochemical indices revealed significant intergroup variations at day 10 of treatment. The net photosynthetic rate (Pn) and soil plant analysis development (SPAD) values in the DS group exhibited inflection points, indicating that plants had entered a critical transition phase from stress adaptation to damage. Hypothesizing that differential expression of drought-responsive genes mediates this regulatory switch, this time point was selected for subsequent transcriptomic sequencing analysis. Transcriptome sequencing analysis For each treatment, three groups of biologically replicated sequences were generated, and six cDNA libraries (Control-1, Control-2, Control-3, Drought-1, Drought-2, and Drought-3) were constructed. A total of 79.85 Gb of clean data was obtained. The clean data of each sample exceeded 5.96 Gb, and the percentage of Q30 bases was over 95.66%. The clean data from all samples were assembled using Trinity, and the assembly results were optimized and evaluated. The number of assembled unigenes was 101,973, the number of transcripts was 183,661, the average length was 1,046.38 bp, and the N50 length was 1,739 bp. The statistical table of sequencing data is shown in Table [118]S2, and The comparison of sequencing data and assembly results is shown in Table [119]S3. Principal component analysis was conducted on the samples based on gene expression levels, and the results showed a significant difference between the DS and CK groups (Fig. [120]4A). DESeq2 software was used to analyze differential gene expression, with a screening threshold of |log2FC| >= 1 and P < 0.05. Compared with the CK group, a total of 20,874 differentially expressed genes (DEGs) were identified in the DS group, including 7,569 up - regulated and 13,305 down - regulated genes (Fig. [121]4B). The DEGs were highly enriched in Gene Ontology (GO) pathways such as the amide metabolic process, oxidoreductase activity, and organonitrogen compound biosynthetic process (Fig. [122]4C). Further Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis revealed that the DEGs were significantly enriched in key pathways such as ribosome, oxidative phosphorylation, and biosynthesis of unsaturated fatty acids (Fig. [123]4D).In summary, B. balsamifera may respond to drought stress by regulating metabolic synthesis and substance transport, and synergistically activating multiple key metabolic pathways. Fig. 4. [124]Fig. 4 [125]Open in a new tab Transcriptome expression analysis. A Principal component analysis. Each dot represents a replicate in a grouped experiment, and different colors distinguish different groups. B Volcano plot of identified genes. Purple color in the graph indicates down-regulated differentially expressed genes, red dots indicate up-regulated genes, and gray dots indicate insignificant differentially expressed genes. C GO functional enrichment of DEGs. D KEGG functional enrichment of DEGs Proteome sequencing analysis In the quantitative proteomics analysis, principal component analysis results indicated a significant difference between the two groups of samples (Fig. [126]5A). A total of 68,063 peptides and 8,615 proteins were identified, and 2,770 differentially expressed proteins (DEPs) were screened. Among them, 2,016 were up - regulated and 754 were down - regulated (Fig. [127]5B). Fig. 5. [128]Fig. 5 [129]Open in a new tab A Principal component analysis. Each dot represents a replicate in a grouped experiment, and different colors distinguish different groups. B Volcano plot of identified proteins. Purple dots in the graph indicate down-regulated differentially expressed proteins, red dots indicate up-regulated proteins, and gray dots indicate non-significant differentially expressed proteins. C GO functional enrichment of DEPs. D KEGG functional enrichment of DEPs For the GO enrichment analysis, within the biological process category of GO, DEPs were significantly enriched in the carboxylic acid biosynthetic process and organic acid biosynthetic process. In the molecular function category, significant enrichments were observed in organic cyclic compound binding and heterocyclic compound binding (Fig. [130]5C). KEGG pathway analysis revealed that in the genetic information processing category, DEPs were highly enriched in pathways such as the mRNA surveillance pathway and spliceosome. In the metabolic category, phenylalanine, tyrosine and tryptophan biosynthesis, as well as photosynthesis-antenna proteins, were significantly enriched. Additionally, in the biological system category, DEPs were significantly enriched in the plant-pathogen interaction pathway (Fig. [131]5D). Combined analysis of transcriptome and proteome Analysis of correlation The proteomics and transcriptomics data generated in this study were comprehensively analyzed to determine the correlation between the two different data types. By comparing Figs. [132]6 A and B and 2410 DEGs were identified in the transcriptome and 2770 DEPs in the proteome. Among them, 796 genes were detected in both analyses.Compared with the CK group, in the DS group, a total of 754 DEPs were up - regulated and 2016 were down - regulated. Also, compared with the CK group, 1245 DEGs were up - regulated and 1165 were down - regulated in the DS group.Simultaneously, 320 DEPs and their corresponding DEGs were down - regulated, and 221 DEPs and their corresponding DEGs were up - regulated (these are related DEGs - DEPs). Moreover, 166 DEPs were down - regulated while their corresponding DEGs were up - regulated, and 89 DEPs were up - regulated while their corresponding DEGs were down - regulated.In addition, the correlation between the proteome and transcriptome data was evaluated. Pearson correlation analysis results indicated a moderate correlation between mRNA and protein expression abundance. The correlation coefficient was 69.22%, and among these, 40.58% of proteins were significantly positively correlated with transcripts (Fig. [133]6C). Fig. 6. [134]Fig. 6 [135]Open in a new tab Joint transcriptome and proteome analysis. A Statistical data on the distribution of DEGs-DEPs in the control and drought groups. B Comparative analysis of DEGs-DEPs in control and drought groups. C Correlation analysis of mRNAs with proteins. The horizontal coordinate is the range of values of the correlation coefficients and the vertical coordinate is the density distribution of the correlation coefficients. D GO enrichment analysis of DEGs-DEPs. E KEGG pathway analysis of DEGs-DEPs GO and KEGG pathway enrichment analysis of DEGs-DEPs GO and KEGG pathway enrichment analyses were performed on DEGs and DEPs. GO analysis indicated that most correlations between DEGs and DEPs were positive and were significantly enriched in catalytic activity, biological processes, and multiple metabolic pathways (Fig. [136]6D). KEGG pathway enrichment analysis showed that DEGs - DEPs were significantly enriched in SNARE interactions in vesicular transport, motor proteins, and biosynthesis of unsaturated fatty acids (Fig. [137]6E). Thus, it is speculated that B. balsamifera responds to drought stress in terms of material transport, metabolic regulation, and maintenance of cell homeostasis through the coordinated operation of these pathways. DEPs and DEGs associated with drought tolerance and PPI network analysis Based on the review of the literature, GO and KEGG enrichment analyses were performed to evaluate key annotations related to drought stress, such as ribosomal metabolic pathways, oxidoreductase activity, and biosynthesis of unsaturated fatty acids, etc. In the control (CK) and drought stress (DS) groups, 55 genes and proteins showing co-differential expression (DEGs-DEPs) were enriched in drought stress-related GO/KEGG categories. Among these, genes exhibited consistent or inconsistent regulation patterns between transcriptional and protein levels. In the drought treatment comparison group, most of the DEPs and DEGs were up-regulated, suggesting that these up-regulated DEPs and DEGs may play certain roles in drought response (Fig. [138]7A). Fig. 7. [139]Fig. 7 [140]Open in a new tab Co-analysis of transcriptome and proteome. A Expression of drought tolerance-associated DEGs-DEPs. B Protein-protein interaction network, the size of the diamond or circle represents the number of proteins interacting with the protein To explore the molecular mechanism of B. balsamifera’s response to drought stress, we constructed a drought - related DEGs - DEPs protein interaction network with the assistance of the STRING database ([141]https://string-db.org/) and visualized the gene co - regulation relationship using Cytoscape software ([142]https://cytoscape.org/). The constructed drought-tolerant co-regulation network contained 18 nodes, which were selected from the above-mentioned 55 DEGs-DEPs and participated in the protein interaction network construction (Fig. [143]7B). Based on the characteristics of network topology, nodes with more edges were more crucial in the regulation network. The degree of connectedness for TRINITY_DN394_c0_g1 (Trinity transcript ID), TRINITY_DN55119_c0_g1 (Trinity transcript ID), and TRINITY_DN10879_c0_g2 (Trinity transcript ID) were 10, 9, and 8 respectively. These were likely to be the central nodes of B. balsamifera’s drought response regulation network. Among them, TRINITY_DN394_c0_g1 had the largest degree of connectedness. As a network hub, it was likely to play a central role in B. balsamifera’s drought response process.The function of these 18 genes is shown in Table [144]S4. Hierarchical clustering analysis indicated that most of these 18-node genes (i.e., the 18 genes involved in the constructed drought-related DEGs-DEPs protein interaction network) were significantly up-regulated under drought stress, suggesting that these genes played an active regulatory role in B. balsamifera’s response to drought stress (Fig. [145]7A and B). Validation of RNA-Seq data The RT– qPCR and RNA-Seq data for the nine selected genes exhibited a strong correlation (R = 0.76), which validated the reliability of the transcriptome analysis (Fig. [146]8). This result strongly demonstrates the reliability of the transcriptome analysis data. Additionally, the differential expression of these genes at both the transcriptome and proteome levels implies that they play a crucial role in regulating the drought response of B. balsamifera. Fig. 8. [147]Fig. 8 [148]Open in a new tab Comparative analysis of RNA-seq and RT-qPCR Discussion According to phenotypic observation, the growth of B. balsamifera was inhibited under drought stress. After 10 days of drought, the six growth indexes were significantly lower than those of the CK group. After rehydration, its growth was alleviated to some extent. Additionally, with the aggravation of drought stress, the Pn, Tr, Gs, and CE of B. balsamifera generally showed a downward trend, which was consistent with the results of Bhuma [[149]45] on the drought resistance of chrysanthemum. Studies have indicated that the inhibition of plant photosynthesis is mainly due to stomatal and non - stomatal limitations [[150]46]. In this study, within the first 12 days of treatment, Ci, Pn, Tr, and Gs decreased synchronously while Ls increased, suggesting that stomatal limitation was the main factor at this time. After 12 days, Ci increased, while Pn, Tr, and Gs decreased, and Ls decreased significantly, indicating that non - stomatal limitation became the main cause of the decline in the photosynthetic rate. In terms of physiology and biochemistry, the RWC and SPAD values of B. balsamifera leaves decreased, which were consistent with the changes in maize under drought stress [[151]47]. The generation and removal of ROS in plants are generally in a dynamic balance, but when plants encounter adversity, excessive ROS will be produced [[152]48, [153]49]. ROS attacks the unsaturated fatty acids of the cell membrane, resulting in an increase in MDA content, which is consistent with the results of this experiment [[154]49–[155]51]. Moreover, the activities of POD and SOD increased significantly, and the activity of CAT decreased after rising until 12 days, indicating that there was a threshold for CAT’s repair ability. Under drought stress, B. balsamifera enhances antioxidant capacity through the synergistic effect of antioxidant enzyme activities such as SOD, POD, and CAT, which may help alleviate oxidative stress, as reflected by MDA content changes and enzyme activity dynamics. Simultaneously, it enhances drought resistance by accumulating LIG and various osmotic pressure regulators. Among these, the contents of SS and SP increased first and then decreased, reaching the maximum at 12 days of stress. Notably, the PRO content continued to rise, suggesting that B. balsamifera may mainly rely on PRO to maintain cellular water balance during extreme drought. Phenotypic and physiological changes alone cannot fully elucidate the response of B. balsamifera to drought. Plant drought resistance involves a complex regulatory network of multi - level interactions. Transcriptome profiling uncovered a substantial number of DEGs, with 20,874 genes showing altered expression. These DEGs were enriched in GO pathways such as the amide metabolic process and oxidoreductase activity, and in KEGG pathways such as ribosome and oxidative phosphorylation. This is similar to the findings of studies showing that plants maintain cell homeostasis by regulating metabolic pathways and redox reactions under drought stress. Proteome analysis identified a notable set of DEPs. The GO functions, including the carboxylic acid biosynthetic process and organic cyclic compound binding, and the KEGG pathways, such as the mRNA surveillance pathway and biosynthesis of unsaturated fatty acids, were enriched. Among these, changes in mRNA - surveillance - pathway - related proteins may affect the stability and translation efficiency of mRNA, thereby regulating the expression of drought - responsive genes [[156]52]. The activation of the unsaturated fatty acid biosynthesis pathway may be involved in plant membrane reconstruction and enhance plant tolerance to drought. Co - regulation analysis of the identified DEGs - DEPs revealed that there were more drought - related up - regulated genes than down - regulated ones, and the up - regulation extent of DEGs was greater than that of DEPs. Through integrated analysis of the transcriptome and proteome data, GO and KEGG analyses demonstrated that drought - related DEGs - DEPs, involved in processes such as ribosome function, oxidoreductase activity, and biosynthesis of unsaturated fatty acids, were significantly activated in metabolic and physiological pathways. In the study of the drought - related protein interaction network, analysis of 18 nodes led to the identification of three DEGs - DEPs involved in drought stress response pathways: 40 S ribosomal protein S7 (RPS7), numbered TRINITY _ DN2896 _ c0 _ g1; 40 S ribosomal protein S25–4 (RPS25–4), numbered TRINITY _ DN30026 _ c0 _ g3; and chlorophyll a - b binding protein 6 (CAB6), TRINITY _ DN13120 _ c0 _ g1. Under drought stress, B. balsamifera up - regulates the expression of CAB6 to enhance the efficiency of light - energy capture and transmission, thereby maintaining normal photosynthesis. This mechanism is highly congruent with the expression pattern of CAB6 in barley under drought stress [[157]53].Moreover, previous studies have confirmed that 40 S ribosomal proteins can respond to drought stress via an overexpression mechanism in A. thaliana, significantly improving the plant’s drought tolerance [[158]54]. In this study, two 40 S ribosomal proteins (RPS) were identified. Given the conclusions of previous (A) thaliana studies, they are highly likely to play a similar regulatory role in (B) balsamifera’s response to drought stress. It is inferred that B. balsamifera likely responds to drought stress by precisely regulating these hub genes or proteins. In subsequent studies, further functional identification and verification of these genes are required. Conclusion This study systematically analyzed the drought - stress response mechanism of B. balsamifera. Drought significantly inhibited its growth, which was alleviated after re - watering. At the photosynthetic level, Pn and Tr decreased, and the Ci concentration fluctuated. The 12 - day mark was the key node for the regulation of stomatal opening and closing, and the photosynthetic - limiting factor shifted from stomatal to non - stomatal. Physiologically and biochemically, the SPAD value and RWC of the leaves decreased, the MDA content increased, and antioxidant enzymes like SOD worked synergistically to scavenge ROS. Meanwhile, the plant accumulated LIG, SS, SP, and PRO to maintain cell water balance.Transcriptome and proteome analyses identified 20,874 DEGs and 2,770 DEPs, with significant enrichments in metabolic, material - transport, and redox pathways. Joint analysis pinpointed 55 drought - related DEGs - DEPs, constructed an interaction network, and identified 18 potential core genes such as TRINITY _ DN394 _ c0 _ g1, as well as 3 DEGs - DEPs involved in response pathways, including CAB6. Subsequently, the functions of these genes need further verification.In summary, B. balsamifera developed a comprehensive drought - resistance mechanism through phenotypic adjustments, regulation of photosynthetic characteristics, physiological and biochemical adaptations, and the synergy of the DEGs - DEPs network (Fig. [159]9). This study provides a crucial basis for revealing the drought - resistance mechanism of B. balsamifera and also offers a reference for other plant drought - resistance research. Fig. 9. [160]Fig. 9 [161]Open in a new tab Network of drought mechanisms in B. balsamifera. The green upward arrow indicates that the activity or content of the substance increases with time; the red downward arrow indicates that the activity or content of the substance decreases with time; both upward and downward arrows indicate that the activity or content of the substance increases first and then decreases with time. A phenotypic response. B photosynthetic response. C physiological and biochemical response. D molecular response, possibly regulated transcription factors Supplementary Information [162]12870_2025_6916_MOESM1_ESM.xlsx^ (11.1KB, xlsx) Supplementary Material 1. Table S1.Genes and corresponding primers used in the qPCR test. [163]12870_2025_6916_MOESM2_ESM.xlsx^ (14.8KB, xlsx) Supplementary Material 2. Table S2. Statistical table of sequencing data of six samples. [164]12870_2025_6916_MOESM3_ESM.xlsx^ (9.5KB, xlsx) Supplementary Material 3. Table S3. The comparison results of the sequences of six samples. [165]Supplementary Material 4.^ (9.8KB, xlsx) Acknowledgements