Genome-wide identification and functional characterization of bHLH transcription factors in Herpetospermum pedunculosum under salt stress

Abstract

The widely conserved basic helix-loop-helix (bHLH) transcription factors, which constitute a major transcription factor family in plants, play crucial roles in regulating plant growth, development, and environmental stress responses. Despite their importance, the genomic composition and functional roles of bHLH transcription factors in Herpetospermum pedunculosum remain uncharacterized. In this study, we identified 98 HpbHLH genes in H. pedunculosum, which were grouped into 23 subfamilies, each exhibiting distinct conserved motifs, domains, and gene structures. Promoter cis-element analysis revealed that HpbHLH genes are enriched in phytohormone- and stress-response-responsive elements. Transcriptomic analysis of salt-stressed H. pedunculosum highlighted HpbHLH28, HpbHLH45, and HpbHLH51 as candidates for further functional analysis. Subsequent quantitative real-time PCR confirmed their expression patterns under salt stress and demonstrated tissue-specific expression differences. Subcellular localization studies revealed that all three proteins localize to the nucleus. Transcriptional activation assays further demonstrated that HpbHLH28 and HpbHLH51 possess transcriptional activity, whereas HpbHLH45 remains inactive. Additionally, dual-luciferase reporter assays showed that HpbHLH51 binds to and activates the promoter of its downstream target gene HpTIFY6B, which encodes a TIFY family protein. This study presents the first comprehensive characterization of the bHLH gene family in H. pedunculosum, establishing a molecular foundation for elucidating its stress resistance mechanisms and supporting future variety improvement efforts.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12870-026-08945-5.

Introduction

The basic helix-loop-helix (bHLH) family represents a widely distributed superfamily of transcription factors in the plant kingdom. The bHLH domain consists of approximately 60 amino acids and contains two distinct regions that differ in structure and function [1]. The basic region is located at the N-terminus of the bHLH domain, and is primarily composed of approximately 15 to 20 amino acids, most of which are basic. This region functions as the DNA-binding domain, enabling it to specifically recognize and bind to the G-box and E-box sequences of target genes [2]. On the other hand, the helix-loop-helix region is located at the C-terminal end of the bHLH domain, and is made up of two antiparallel hydrophilic α-helices that are separated by a variable ring structure [3, 4]. This region is intrinsically linked to the dimerization of bHLH proteins, which is crucial for their binding to target DNA sequences [3, 5, 6].

Members of the bHLH gene family exhibit notable diversity in both gene number and classification across different species. For example, Arabidopsis thaliana contains 177 bHLH genes divided into 25 subfamilies [1], wheat (Triticum aestivum L.) has 225 bHLH genes classified into 23 subfamilies [7], maize (Zea mays) possesses 208 bHLH genes grouped into 18 subfamilies [8], while tomato (Solanum lycopersicum) has 159 SlbHLH genes grouped into 21 subfamilies [9]. Recent years have seen significant advances in the study of the bHLH gene family in Cucurbitaceae. An early study identified 118 bHLH genes in melon (Cucumis melo L.) [10]. Subsequently, a comprehensive comparative genomic analysis of seven cucurbit crops, including cucumber (Cucumis sativus), melon (C. melo), watermelon (Citrullus lanatus), bottle gourd (Lagenaria siceraria), waxgourd (Benincasa hispida), bitter gourd (Momordica charantia), and pumpkin (Cucurbita pepo), identified 1,160 bHLH genes, which were grouped into 28 subfamilies, revealing their overall distribution and evolutionary relationships within Cucurbitaceae [11]. More recent research further expanded this knowledge base, with 222 bHLH genes identified in pumpkin (Cucurbita moschata), and classified into 19 subfamilies [12].

The bHLH transcription factors play crucial roles in regulating plant secondary metabolism, plant growth and development, and response to stress [1, 13, 14]. For instance, the first bHLH transcription factor identified in plants was R (Lc), which regulates anthocyanin biosynthesis in maize [15]. Additionally, other bHLH transcription factors, such as SlbHLH96, down-regulate the abscisic acid (ABA) catabolic gene SlCYP707A2, resulting in elevated ABA levels and enhanced tomato drought tolerance [16]. Overexpression of CabHLH035 improves drought tolerance in A. thaliana [17], while AtbHLH122 acts as a positive regulator in response to drought, NaCl, and osmotic signals [18]. In silver birch, BpbHLH1 promotes salt tolerance by enhancing reactive oxygen species scavenging capacity and maintaining cellular homeostasis [19]. Similarly, while CsbHLH041 enhances the adaptability and survival capacity of transgenic A. thaliana and C. sativus seedlings under salt and ABA stress conditions [20]. Furthermore, AabHLH112 was identified as a positive regulator in the biosynthesis pathways of β-caryophyllene, epi-cedrol, and β-farnesene in Artemisia annua [21].

In chrysanthemums, CmbHLH110 interacts with CmERF110 to regulate flowering [22]. While in apple, the OE-MdbHLH4 strain exhibits longer stems, more extensive root systems, and larger leaves compared to the wild type. Furthermore, ectopic expression of MdbHLH4 in tomatoes enlarges fruits, alters fruit colouration, reduces accumulation of soluble solids, decreases flesh firmness, and increases seed size [23]. Therefore, these findings show that, beyond their roles in development and metabolism, bHLH transcription factors are also critical mediators of abiotic stress tolerance, such as drought and salinity, by modulating phytohormone signaling, enhancing antioxidant capacity, and maintaining cellular homeostasis. This consolidated understanding underscores their significant potential as targets for the genetic improvement of stress resilience in plants.

Herpetospermum pedunculosum is an annual dicotyledonous medicinal plant in the Cucurbitaceae family, primarily distributed in high-altitude regions from 2,300 to 3,500 m above sea level in southwestern China, Nepal, and northeastern India [24]. Its dried mature seeds are used in traditional Tibetan medicine to treat liver diseases, biliary disorders, indigestion, and other ailments [25]. H. pedunculosum contains medicinal constituents such as lignans [26], coumarins [27], and triterpenoids [28], with lignans being the primary active medicinal component [29]. Ecologically, H. pedunculosum exhibits strong adaptability to the soil conditions in the main growing regions. However, globally, fertile land resources are diminishing, with soil degradation often manifesting as organic matter depletion, nutrient exhaustion, and salinization [30, 31]. Moreover, severe soil salinization may lead to reduced crop yields and lower medicinal value. Therefore, understanding how H. pedunculosum adapts to soil salinization is both a necessary and scientifically significant pursuit.

In this study, we identified and characterized 98 bHLH family genes in H. pedunculosum, and comprehensively analyzed their phylogenetic relationships, conserved motifs, domains, gene structures, chromosomal localization, cis-acting elements, and interspecies collinearity. Based on previous salt stress data for H. pedunculosum roots, we identified three genes that were significantly up-regulated and further investigated their expression patterns. The subcellular localization of these genes was validated using a tobacco system, whilst their transcriptional activation abilities were confirmed in both tobacco and yeast systems. Additionally, the regulatory effects of these genes on the target genes were assessed through a dual luciferase reporter assay. Collectively, these results advance our understanding of the molecular mechanisms underlying the salt stress response in H. pedunculosum and provide a molecular foundation for future varietal improvements.

Materials and methods

Identification of the bHLH gene from H. pedunculosum

The complete genomic data of the H. pedunculosum were downloaded from Figshare (10.6084/m9.figshare.21626153.v2) [24], while the conserved domain of bHLH (PF00010) was obtained from the Pfam database (http://pfam.xfam.org/). Similarly, HMMER 3.0 was used to obtain the HpbHLH candidate sequences from the genomic database of the H. pedunculosum, which were then screened on the NCBI database using NCBI Batch CD-Search (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi) to obtain HpbHLH sequences with a complete bHLH conserved domain.

Analysis of the physical and chemical properties of HpbHLH

The physicochemical properties of the candidate proteins of the bHLH family of the H. pedunculosum, including the number of amino acids, molecular weight, theoretical isoelectric point (pI), instability index, aliphatic index and grand average of hydropathicity, were analyzed using the “The Protein Parameter Calc” module in the TBtools version 2.400 [32]. Similarly, the subcellular localization of HpbHLH was predicted using the online website WOLF PSORT (https://www.genscript.com/wolf-psort.html).

The phylogenetic analysis of HpbHLH

Gene sequences of A. thaliana and C. sativus were obtained from the Arabidopsis Information Resource (TAIR; http://www.arabidopsis.org) and CuGenDB (http://cucurbitgenomics.org), respectively, and used to systematically identify gene family members using consistent screening parameters across all species. Multiple sequence alignment was conducted with MAFFT using default settings, and subsequently trimmed with TrimAL under the -automated1 parameter. The alignment was used to perform phylogenetic reconstruction using the maximum likelihood method in MEGA software X with 1,000 bootstrap replicates. The resulting tree was visualized and annotated using iTOL (http://itol.embl.de/).

Conserved motifs and domains, and gene structure analysis

The online website MEME (version 5.5.8) (https://meme-suite.org/meme/tools/meme) was used to conduct a conservative motif analysis of HpbHLH, with the number of motifs set at 20. The results of the HpbHLH gene were obtained using the Visualize Gene Structure module in the TBtools software. The obtained structures were visualized using the Gene Structure View (Advanced) module of the TBtools software.

Chromosome localization, gene duplication, and collinearity analysis of HpbHLHs

The position of the chromosome information of HpbHLH was determined using the gene annotation file of the H. pedunculosum and visualized using the Gene location Visualize in the GTF/GFF tool in the TBtools software, while its tandem repeat analysis was conducted using the MCScanX tool in the TBtools software. Subsequently, we conducted segmental duplication analysis using the MCScanX tool and visualization of HpbHLH using the Circos module in TBtools software. The MCScanX tool in the TBtools software was also used to perform the inter-species collinearity analysis of the three species, including H. pedunculosum, A. thaliana, and C. sativus.

Analysis of the cis-acting elements of HpbHLH

The TBtools was used to extract the 2000 bp sequence upstream of the coding region of the HpbHLH gene and used to identify and analyze its cis-acting elements using the online website PlantCARE (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/). These analysis results were processed in Excel and visualized using TBtools.

Plant materials and salt stress

The seeds of the H. pedunculosum were sampled from the Tibet Rhodiola Pharmaceutical Holding Company in Lhasa (latitude 29°79’N, longitude 94°09’E). These seeds (No. H-250405) were morphologically identified by Professor Qi Zhao from Chengdu University and stored at the Sichuan-Xizang Traditional Medicinal Plant Engineering Research Center in Chengdu, China. The plants used to verify the transcriptome results of salt stress in the H. pedunculosum root and to verify the function of the HpbHLH gene were 20-day-old seedlings grown in Hoagland nutrient solution at 25℃ with 16 h of light and 8 h of darkness for 20 days and exposed to 200 mM NaCl.

RNA extraction and qRT-PCR analysis

The total RNA was extracted using the Plant Total RNA Extraction Kit (Foregene Biotechnology, Chengdu, China) from the H. pedunculosum root, following the manufacturer’s instructions, and subsequently reverse-transcribed using the HiFiScript All-in-one RT Master Mix for qPCR kit (Cwbio, Jiangsu, China). The expression level of HpbHLH was detected using the SuperStar Universal SYBR Master Mix kit (Cwbio, Jiangsu, China) on the Bio-Rad CFX96 Real-time PCR system. The results were standardized using the actin gene of the H. pedunculosum as the internal reference gene. The primer sequences used for the analysis are presented in Supplementary Table 1.

Identification and subcellular localization analysis of key HpbHLH genes

We used publicly available transcriptomic data of H. pedunculosum (PRJNA1301034) to identify salt stress-responsive genes in the roots of H. pedunculosum. The data originated from an experiment in which roots of 20-day-old H. pedunculosum seedlings were exposed to 0 (control) or 200 mM NaCl, and then collected at 1, 3, 24, 48, and 96 h. The expression data for HpbHLH gene family members were extracted and analyzed with TBtools, followed by the generation of heatmaps to represent their expression profiles. The expression patterns of significantly up-regulated genes under salt stress were validated using quantitative real-time PCR (qRT-PCR). Primers containing BamH I and Kpn I sites were used to amplify the open reading frames (ORFs) of HpbHLH28, HpbHLH45, and HpbHLH51 without stop codons, which were cloned into the pCAMBIA1300-35 S-YFP vector (Supplementary Table 1), and subsequently transformed into GV3101-P19 competent cells (Weidi, Shanghai, China). RFP-NLS_SV40 was used to label the cell nucleus [33]. The recombinant and the empty vector bacterial solutions were mixed with the RFP-NLS_SV40 bacterial solution in the ratio of 1:1, and then injected into the leaves of Nicotiana benthamiana. The empty vector was used as the negative control. After two days of cultivation, fluorescence was observed using laser confocal microscopy (Nikon Ti2-E AXR + NSPARC, Tokyo, Japan).

Transcriptional activation analysis

The transcriptional activation experiment was carried out using the tobacco and yeast systems. The ORFs of HpbHLH28, HpbHLH45, and HpbHLH51 without stop codons were amplified using primers containing Sal Ⅰ and Pst Ⅰ sites, and then cloned into the pCAMBIA2300-GAL-GFP vector (Supplementary Table 1), and subsequently transformed into the GV3101-P19 competent cells (Weidi, Shanghai, China). These transformed cells were mixed with the pGreenII 0800-LUC-reporter bacterial solution at the same proportion, cultivated in the leaves of N. benthamiana, and observed for fluorescence after two days. A pCAMBIA2300-GAL-GFP empty vector bacterial solution and the pGreenII 0800-LUC-reporter bacterial solution were mixed in equal proportions and used as the negative control. For the yeast system, the gene was amplified using primers containing EcoR Ⅰ and BamH Ⅰ sites and then cloned into the pGBKT7 vector (Supplementary Table 1). The pGADT7 and pGBKT7 vectors were co-transfected into the Yeast AH109 competent cells (Weidi, Shanghai, China), with the co-transfected pGADT7-T and pGBKT7-53 vectors as positive controls, the co-transfected pGADT7-T and pGBKT7-Lam vectors as negative controls, and the pGADT7 and pGBKT7 empty vectors as the control. The transformed yeasts were cultured in the SD/-Leu/-Trp (DDO) liquid medium for 1 day, and then in the DDO solid medium and SD/-Ade/-His/-Leu/-Trp (QDO) solid medium for 3 days, and subsequently stained with the chromogenic agent X-α-Gal (Coolaber, China).

Dual luciferase transient expression assay

The Pearson correlation analysis was used to identify genes exhibiting highly synchronized expression patterns with the target bHLH protein using the transcriptome count data. The protein interaction network of the target bHLH was retrieved from the STRING database (https://cn.string-db.org/). Thereafter, the upstream 2 kb promoter sequences of the candidate genes were predicted and then scanned using FIMO software, and the resulting data were combined to predict high-confidence downstream target genes of the bHLH. Furthermore, primers containing the Hind Ⅲ and BamH Ⅰ sites were used to amplify the promoter sequence of the target gene and then cloned into the pGreenII 0800-LUC vector as a Reporter gene (Supplementary Table 1). Subsequently, the pGreenII 0800-LUC vector was transformed into the GV3101-P19 competent cells (Weidi, Shanghai, China). The ORF of HpbHLH was cloned into pCAMBIA1300-35 S-YFP as an Effector, while an empty vector was used as the negative control. Both vectors were mixed in equal volumes with the Reporter gene and injected into N. benthamiana. The cultures were then incubated for two days, followed by the application of a 1 mM solution of D-luciferin potassium salt (Coolaber, China) to the abaxial surface of the N. benthamiana leaves. Visualization was performed using an iBright 1500 imaging system (Thermo Fisher Scientific, Waltham, MA, USA), while the enzyme activities of firefly luciferase (LUC) and sea urchin luciferase (REN) were determined using dual luciferase reporter assay reagents (Beyotime, Shanghai, China), following the manufacturer’s instructions. Each assay was performed with three biological replicates and three technical replicates.

Results

Identification and physicochemical properties of bHLH gene family in H. pedunculosum

In this study, a hidden Markov model (HMM) was used to identify the PF00010 conserved domain of the bHLH family. Through screening with NCBI-CDD in the existing H. pedunculosum protein database, 98 of 109 sequences with intact bHLH domains were selected. The sequences were designated as HpbHLH01-HpbHLH98 according to their chromosomal locations (Supplementary Table 2). To further characterize the 98 HpbHLHs, we predicted their physicochemical properties (Supplementary Table 2). The amino acid lengths of HpbHLHs ranged from 108 aa in HpbHLH47 to 722 aa in HpbHLH83, with an average length of 358.72 aa. Their molecular weight (MW) spanned from 12,679.81 Da in HpbHLH47 to 76,995.87 Da in HpbHLH83, with an average MW of 39698.76 Da. The pI of these proteins ranged from 4.37 in HpbHLH63 to 10.16 in HpbHLH47. Instability coefficient analysis revealed that 94 HpbHLH proteins had a score greater than 40, while only four were below this threshold, suggesting that the majority are stable in vivo. Additionally, 39 HpbHLH proteins had aliphatic indices ranging from 50 to 70, while 59 ranged from 70.61 in HpbHLH61 to 100.46 in HpbHLH15. Notably, the average hydrophilicity of HpbHLHs was negative, suggesting they are likely thermostable and hydrophilic proteins. Subcellular localization predictions revealed that 90 HpbHLHs are localized in the nucleus. Additional localizations included chloroplast (HpbHLH48), cytoplasm (HpbHLH15, HpbHLH49, HpbHLH65, HpbHLH72), Golgi apparatus (HpbHLH80, HpbHLH86), and both nucleus and cytoplasm (HpbHLH47). These findings suggest that members of the HpbHLH transcription factor family may constitute a complex and finely tuned regulatory network, which functions across different cellular compartments.

Analysis of phylogenetic, conserved motifs, conserved domains, and gene structure of the HpbHLH gene family

A phylogenetic tree was constructed using the maximum likelihood method to elucidate the evolutionary relationships among the 98 HpbHLHs, 136 CsbHLHs, and 146 AtbHLHs (Fig. 1). According to the classification system established for A. thaliana [1], the 380 bHLH genes identified in this study were systematically categorized into 28 subfamilies. Of these, the 98 HpbHLHs were unevenly distributed across 23 subfamilies. The largest number of HpbHLHs was found in subgroup XII, which contained 14 members, followed by subgroup VII with 10 members. Notably, no HpbHLH members were identified in subgroups IIIf, XIII, XIV, VI, and X. The lineage-specific retention and loss of HpbHLH subfamilies may explain this distribution pattern, potentially indicating functional specialization among these genes.

Fig. 1.

A phylogenetic tree of bHLH proteins constructed using 98 HpbHLH proteins, 136 CsbHLH proteins, and 146 AtbHLH proteins. Green pentagrams represent HpbHLH proteins, blue circles represent CsbHLH proteins, and orange squares indicate AtbHLH proteins. Different colors denote distinct subfamilies

To elucidate the structure and function of the HpbHLH gene family, we analyzed the conserved motifs, conserved domains and gene structures of the 98 HpbHLH proteins, thereby presenting the results in the context of phylogenetic analysis (Fig. 2A). Using the MEME program, we identified 20 conserved motifs within the HpbHLH proteins (Fig. 2B). Among these, HpbHLH45, HpbHLH50, HpbHLH51, HpbHLH58, and HpbHLH56 possessed the highest number of motifs, with eight each, whereas HpbHLH09, HpbHLH24, HpbHLH32, HpbHLH47, HpbHLH57, HpbHLH59, HpbHLH63, HpbHLH64, HpbHLH70, HpbHLH78, HpbHLH83 and HpbHLH88 contained the fewest, with only two each. Further, the results showed that members within the same subfamily of HpbHLHs have similar motif compositions. Notably, Motif 1 and 2 were present in all the HpbHLH family members, while certain motifs were unique to specific subfamilies, potentially contributing to the functional diversity of these genes.

Fig. 2.

Conserved motifs, conserved domains, and gene structure of HpbHLHs are illustrated as follows: A A phylogenetic tree constructed from 98 HpbHLH sequences. B Conserved motifs in HpbHLHs are represented by distinct colored squares, indicating different conserved motifs. The x-axis indicated the number of amino acids. C Conservative domains of HpbHLHs, with differently colored squares, representing unique domains. The x-axis shows the amino acid number. D Gene structure of HpbHLHs, with the x-axis denoting base pair length (bp)

Analysis of the conserved domains revealed the presence of 10 distinct types (Fig. 2C), two of which belonged to the characteristic bHLH domain. MYC transcription factors are recognized as core regulators of the jasmonic acid (JA) signaling pathway and play a central role in plant stress responses [34]. Notably, we observed that certain members of the group IIIa, IIIc, IIId, and IIIe subfamilies possess a MYC domain, suggesting their involvement in mediating the JA signaling pathway and its associated physiological processes in H. pedunculosum. Additionally, analysis of the intron–exon structures of all 98 HpbHLH genes revealed considerable variation in exon and intron numbers (Fig. 2D). HpbHLH69 and HpbHLH89 exhibited the greatest structural complexity, with 11 and 11 exons and introns, respectively. In contrast, the gene structures of HpbHLH09, HpbHLH32, HpbHLH45, HpbHLH47, HpbHLH58, HpbHLH56, HpbHLH51, and HpbHLH63 contained only a single exon and lacked introns. Overall, most HpbHLH genes in the same subfamily share similarities in the composition of conserved motifs, gene structure, and distribution of conserved domains. These structural features closely reflect the evolutionary relationships within the HpbHLH gene family.

Chromosome localization, gene replication, and collinearity analysis of HpbHLHs

A detailed analysis of HpbHLHs was conducted, focusing on their chromosomal localization. Among the 98 HpbHLHs, an uneven distribution was observed across the 10 chromosomes. For instance, chromosome (Chr) 8 harbored the highest number of HpbHLH family members (16, approximately 16.3%), whereas Chr10 had the fewest (5, approximately 5.1%). Thirty-three duplicate gene pairs were identified in the HpbHLH family, comprising three tandem duplicate pairs of HpbHLH35/36, HpbHLH52/53, and HpbHLH53/54 (Fig. 3A). The first pair was located on Chr4, while the latter two pairs were found on Chr6. In comparison, 30 segmental duplicate gene pairs were identified (Fig. 3, Supplementary Table 3), greatly outnumbering the tandem duplicates. These findings suggest that segmental duplication events may have played an important role in the expansion of the HpbHLH gene family.

Fig. 3.

Chromosomal localization, gene duplication, and collinearity analysis of HpbHLH genes. A Chromosomal localization and tandem duplication analysis of HpbHLHs, with tandem duplicate genes indicated by red curves. The scale on the left represents the chromosome length (Mb) of H. pedunculosum. B Segmental duplication analysis of HpbHLHs, with segmental duplicate genes shown by red curves. C Collinearity analysis of HpbHLHs, CsbHLHs, and AtbHLHs, where genes exhibiting collinearity are shown by red curves

In genetics, the Ka/Ks ratio represents the rate of the non-synonymous (Ka) to the synonymous (Ks) substitutions between two protein-coding genes. This metric is widely used to infer selective pressure acting on the protein-coding genes [35]. Therefore, we calculated the Ka/Ks values for 33 pairs of HpbHLH duplicate genes (Supplementary Table 4). Among these, seven pairs of genes showed NaN values, potentially resulting from substantial sequence divergence or alignment issues. All the remaining 26 pairs had Ka/Ks values less than 1, suggesting that HpbHLH gene pairs have experienced natural selection pressure during evolution.

To further elucidate the origin and phylogenetic evolution of the HpbHLH family, we performed an interspecific collinearity analysis among C. sativus, A. thaliana, and H. pedunculosum (Fig. 3C). The analysis showed that 63 HpbHLH genes were collinear with 74 genes in A. thaliana, while 84 HpbHLH genes were collinear with 87 genes in C. sativus (Supplementary Table 5). Notably, 61 HpbHLH genes were collinear with both species, suggesting the presence of a conserved core set that has been maintained since the divergence of these lineages.

Analysis of cis-acting elements of HpbHLHs

To investigate the potential transcriptional regulatory elements involved in the salt stress response, we extracted 2,000 bp sequences upstream of the promoter regions of the 98 HpbHLH genes and analyzed them using the PlantCARE online tool to predict cis-acting elements. Our analysis revealed that the promoter regions of HpbHLHs contain a large number of cis-acting elements (Fig. 4), which can be grouped into four major categories, including light-responsive elements, phytohormone-responsive elements, abiotic and biotic stress-related elements, and growth and development-related elements.

Fig. 4.

Analysis of cis-acting elements in HpbHLHs. A The position distribution of cis-acting elements in HpbHLHs is illustrated, with ellipses of different colors representing distinct elements. The x-axis indicates the number of bp. B A heatmap depicting the abundance of cis-acting elements in HpbHLHs. The data in the heatmap were log₂-transformed with a baseline offset of 1.0 (i.e., log₂(value + 1.0)) to stabilize variance and visually enhance the representation of fold-change differences in gene expression. C A histogram showing the frequencies of cis-acting element frequencies in HpbHLHs. The legend indicates cis-acting elements associated with various responses, including abiotic and biotic stresses in yellow, light responsiveness in red, phytohormone responsiveness in blue, and plant growth and development in green

Cis-acting element profiling of the HpbHLH gene family reveals its members’ broad capacity to respond to light and phytohormone signals. In this study, photoresponsive elements were the most abundant category, totaling 1,121 (Fig. 4, Supplementary Table 6). Collectively, all members harbored 986 phytohormone-responsive elements, which mediate responses to JA, gibberellin, abscisic acid, salicylic acid, and auxin (Fig. 4, Supplementary Table 6). Additionally, 333 abiotic and biotic stress-responsive elements were identified in all but four HpbHLHs (Fig. 4, Supplementary Table 6), conferring responses to stresses, including drought, low temperature, anoxia, and wounding. In contrast, elements associated with growth and developmental regulation were the least prevalent, with only 133 such elements identified across 66 members (Fig. 4, Supplementary Table 6). These findings indicate that the HpbHLH gene family is not only intricately involved in light and phytohormone signaling networks but may also broadly mediate responses to abiotic and biotic stresses and contribute to the regulation of growth and developmental processes.

Expression patterns of three HpbHLH genes

Building on previous transcriptome analyses of salt-stressed H. pedunculosum roots (Supplementary Fig. 1), we identified three HpbHLH genes, including HpbHLH28, HpbHLH45, and HpbHLH51, that were up-regulated under salt stress and subsequently validated their expression patterns using qRT-PCR (Fig. 5A). The qRT-PCR results demonstrated that all three genes exhibited peak transcript accumulation in the roots at 24 h following salt stress treatment. Notably, the HpbHLH28 expression was significantly elevated at all time points except at 1 h compared with the control. HpbHLH45 showed significantly higher expression at 24 h and 48 h, but a marked decrease at 96 h. HpbHLH51 was strongly up-regulated at 24, 48, and 96 h, with a > 32‑fold increase at 24 h relative to the control. These qRT-PCR results were largely consistent with the previous transcriptome data of salt-stressed H. pedunculosum roots. To further characterize the expression profiles and potential roles of these three genes in response to salt stress, we used qRT-PCR to quantify their transcriptional expression levels across five tissues of the H. pedunculosum. HpbHLH28 exhibited the highest expression in roots and the lowest in flowers, HpbHLH45 showed maximal expression in leaves and minimal in roots, while HpbHLH51 was most abundant in stems and least in leaves (Fig. 5B). These results indicate that HpbHLH28, HpbHLH45, and HpbHLH51 display pronounced tissue-specific expression patterns, suggesting potential functional specialization in the organs, where they were most prominently expressed.

Fig. 5.

Expression patterns of HpbHLH28, HpbHLH45, and HpbHLH51 genes. A Relative expression levels of the three HpbHLH genes in H. pedunculosum under salt stress, as detected by qRT-PCR. B Relative expression levels of the three HpbHLH genes in various tissues of H. pedunculosum, as determined by qRT-PCR. Data represent the mean ± standard deviation from three independent biological replicates. Significance was determined by one-way analysis of variance followed by Tukey’s multiple comparison test. Different letters indicate significant differences at P < 0.05

Characterization of the HpbHLH28, HpbHLH45, and HpbHLH51 transcription factors and transcriptional activation of the target gene HpTIFY6B by HpbHLH51

To determine whether HpbHLH28, HpbHLH45, and HpbHLH51 function as transcriptional activators, we performed subcellular localization and transcriptional activation assays. Our results demonstrated that all three proteins were localized in the nucleus (Fig. 6A), which is consistent with their roles as transcription factors. To verify the transcriptional activation ability of HpbHLH28, HpbHLH45, and HpbHLH51, we employed both tobacco and yeast systems. Transient expression in N. benthamiana leaves revealed that the fluorescence intensity of HpbHLH28-YFP and HpbHLH51-YFP was higher than that of the empty vector group, whereas HpbHLH45-YFP exhibited lower fluorescence intensity (Supplementary Fig. 2). In yeast assays, all yeast cells grew well in DDO, but only the positive control and those containing HpbHLH28 or HpbHLH51 sequences grew well in QOD. Upon addition of the chromogenic substrate X-α-Gal, the colonies of the positive control and yeast cells containing HpbHLH28 and HpbHLH51 sequences turned blue (Fig. 6B). These results demonstrate that HpbHLH28 and HpbHLH51 possess transcriptional activation ability, while HpbHLH45 does not. Therefore, subsequent experimental sections will focus exclusively on HpbHLH28 and HpbHLH51.

Fig. 6.

Subcellular localization and transcription activation capacity validation of HpbHLH28, HpbHLH45, and HpbHLH51, and promoter regulation of HpbHLH51. A Subcellular localization of HpbHLH28, HpbHLH45, and HpbHLH51 proteins in N. benthamiana leaves. The YFP empty vector was used as a negative control, while an RFP-NLSSV₄₀ fusion protein served as a nuclear marker. Scale bar: 50 μm. B The transcriptional activation capacity of HpbHLH28, HpbHLH45, and HpbHLH51 was verified using a yeast system. Yeast cells were diluted in three different ratios (10⁰, 10^− 1, and 10^− 2) and simultaneously inoculated onto SD/-Leu/-Trp (DDO) and SD/-Leu/-Trp/-His/-Ade (QDO) plates. After three days of cultivation, X-α-gal was added to the QDO plates (QDO/X). C Validation of HpbHLH51 regulation of the downstream target gene HpTIFY6B and detection of relative LUC/REN activity. A dual-luciferase assay was conducted using the HpTIFY6B promoter driving the luciferase (LUC) gene as the reporter, combined with the HpbHLH51 effector. The YFP empty vector was used as the negative control. Data represent the mean ± standard deviation of three independent replicates. Statistical significance of treatment compared to the control is indicated (**P < 0.01; Student’s t-test; n = three biological replicates)

To further elucidate the role of HpbHLH28/51 in response to salt stress, we first identified their putative target genes. Through an integrated analysis, we revealed that, among the factors examined, only HpbHLH51 had a specific downstream target gene Hsped.01g05650.1 ( predicted to be HpTIFY6B), as supported by transcriptome correlations, STRING protein interaction networks, and promoter cis-elements. To confirm the transcriptional activation effect of HpbHLH51 on the target gene promoter, we performed a dual luciferase reporter assay. The experimental group co-transfected with the HpbHLH51 overexpression plasmid showed a significant increase in firefly luciferase activity driven by the target gene promoter compared to the control group co-transfected with an empty vector plasmid, with approximately 1.7-fold higher activity (Fig. 6C). These results demonstrate that HpbHLH51 specifically activates the promoter of its target gene HpTIFY6B.

Discussion

Genomic characterization and evolutionary dynamics of the HpbHLH family

The bHLH transcription factor family is one of the largest transcription factor families in plants [36], which plays a pivotal role in regulating plant growth and development, secondary metabolism, and stress responses [37–39]. As an important medicinal plant in the high-altitude regions of Tibet, H. pedunculosum has drawn considerable attention due to its unique environmental adaptability and the mechanisms underlying the synthesis of its medicinal constituents. Therefore, in this study, we used a genome-wide systematic analysis and identified 98 HpbHLH gene members in H. pedunculosum. This number is lower than that observed in other Cucurbitaceae species, such as C. sativus with 142 [20], C. moschata with 222 [12], and C. melo L. with 118 members [10]. This disparity likely reflects the expansion, reduction, and functional diversification of the bHLH family during evolution across different plant lineages [40, 41]. We further conducted physicochemical analyses of the HpbHLH gene family members (Supplementary Table 2) and revealed that the average pI of HpbHLH proteins was 6.59, with 71 HpbHLH proteins exhibiting a pI below 7. These findings are consistent with observations in C. sativus [20] and Gynostemma pentaphyllum [42], indicating that most HpbHLH proteins are acidic [43]. Additionally, the instability coefficient, aliphatic index, and average hydrophilicity of HpbHLH proteins fell within ranges comparable to those of bHLH proteins from C. moschata [12]. Together, these findings suggest that the HpbHLH protein family in H. pedunculosum are evolutionary conserved.

In this study, 380 bHLH proteins from H. pedunculosum, A. thaliana, and C. sativus were used to construct a phylogenetic tree (Fig. 1). Following the established classification of AtbHLHs [1], all HpbHLHs were grouped into 28 subfamilies, with most distributed in the subfamily XII, a pattern similar to that observed in Wintersweet (Chimonanthus praecox (L.) Link, Calycanthaceae) [44] and Carthamus tinctorius [45]. Notably, no HpbHLH proteins were detected in subfamilies IIIf, XIII, XIV, VI, and X, suggesting possible adaptive gene loss events in H. pedunculosum. This pattern suggests that H. pedunculosum may have streamlined and optimized its transcription factor repertoire during evolution to better suit its unique habitat and developmental strategy [46].

Gene duplication events contribute significantly to the expansion and functional diversification of gene families. Using the genome and annotation data of H. pedunculosum, this study analyzed tandem and segmental repeats within the HpbHLH gene family, as well as their interspecific collinearity between A. thaliana and C. sativus (Fig. 3). The analysis revealed that HpbHLH possesses only two clusters of tandem repeat genes (Fig. 3A), located on Chr4 (two genes) and Chr6 (three genes), which is slightly fewer than the numbers reported for C. sativus, with five clusters [20], C. lanatus with five clusters [11], and M. charantia with four clusters [11]. The HpbHLH segmental repeat genes comprise 30 pairs (Fig. 3B, Supplementary Table 3), exceeding the number of tandem repeat genes, a trend consistent with reports from C. sativus [20]. These results indicate that segmental duplication may be the primary evolutionary driver of the HpbHLH gene family in H. pedunculosum. Furthermore, collinearity analysis revealed 63 and 84 collinear gene pairs between HpbHLH genes and their homologues in A. thaliana and C. sativus, respectively (Fig. 3C, Supplementary Table 5). This conserved homology suggests that these HpbHLH genes likely perform core, indispensable biological functions, whereas HpbHLH genes unique to the H. pedunculosum lineage may be associated with adaptive traits specific to this species.

Structural features and functional divergence of HpbHLH subfamilies

Genes belonging to the same subfamily generally possess similar structures, suggesting that they have conserved functions. To date, numerous AtbHLHs have been functionally characterized, providing valuable insights for inferring the functions of HpbHLHs. For instance, the Members of the AtbHLH IIIf subfamily participate in anthocyanin synthesis [47], while members of the AtbHLH IIIe subfamily, such as MYC2/3/4, are inhibitors of primary root elongation under salt stress [48]. Notably, the IIIe subfamily often exhibits functional antagonism with members of the IIId subfamily [49, 50], while members of the AtbHLH IX subfamily, such as AKS, play a key role in stomatal opening [51, 52]. In summary, the functional differentiation of bHLH genes across different subfamilies in A. thaliana provides a systematic reference framework for deciphering the potential functions of HpbHLH genes.

We further conducted a comprehensive bioinformatics analysis of the HpbHLH transcription factor family, highlighting its key characteristics in motif composition, domain evolution, and gene structure, which are instrumental for its functional inference (Fig. 2). Consistent with the previous results for Ginkgo biloba [53] and Prunus mume [54], all HpbHLH gene family members contain Motifs 1 and 2 (Fig. 2B). This indicates that their conserved core structure consists of a basic region responsible for DNA binding and a HLH region facilitating dimerization, which constitutes the structural foundation for their transcriptional regulatory function. In addition to these conserved motifs, the distribution of motifs shows systematic differences between subfamilies. The analysis of conserved domains revealed that, consistent with the findings for G. pentaphyllum [42], all HpbHLH gene family members possess a complete bHLH domain. The distribution of these conserved domains across subfamilies is intimately associated with their functional roles (Fig. 2C). Notably, all members of the HpbHLH subfamilies IIId and IIIe harbor the bHLH-MYC-N domain, which is analogous to observations in Cyclocarya paliurus [55]. Genes in these two subfamilies typically function as key regulatory factors within the JA signaling pathway [56, 57]. Therefore, these results may provide important reference for future functional studies of corresponding subfamily genes in H. pedunculosum.

Analysis of cis-acting elements within the promoter regions of HpbHLH provides crucial insights into the functional diversity of this transcription factor family and its transcriptional regulatory networks (Fig. 4, Supplementary Table 6). HpbHLH genes harbor a significant number of cis-acting elements associated with ABA and methyl jasmonate (MeJA) (Fig. 4B, Supplementary Table 6), indicating that HpbHLHs may act as key transcription factors within the regulatory pathways of ABA and MeJA. Consistent with findings in C. sativus [20] and C. moschata [12], many HpbHLH gene promoters contain elements responsive to abiotic and biotic stresses as well as phytohormones, although elements related to growth and development are notably fewer compared to those in these species (Fig. 4C, Supplementary Table 6). These findings imply that the HpbHLH gene family likely plays a pivotal role in coordinating plant growth and development with responses to abiotic and biotic stresses through phytohormone signaling pathways. Interestingly, while CsbHLH [20] and CmobHLH [12] possess similar numbers of light, stress, and phytohormone-responsive elements, HpbHLH exhibits a significantly more light-responsive elements compared to the other three cis-acting element classes (Fig. 4B). Furthermore, HpbHLH contains the highest number of anaerobic regulatory elements among the stress response element classifications (Fig. 4B). This may reflect an adaptation of H. pedunculosum to its alpine environment, which is characterized by prolonged exposure to intense ultraviolet radiation and hypoxic conditions.

Functional prediction and identification of HpbHLH genes in salt stress

Transcriptome sequencing analysis of H. pedunculosum roots under salt stress revealed the role of HpbHLH genes in the plant’s response to salinity. Under salt stress, the majority of HpbHLH genes exhibited down-regulated expression (Supplementary Fig. 1). Therefore, we screened and validated three up-regulated HpbHLH genes, including HpbHLH28, 45 and 51 (Fig. 5A). These three HpbHLH proteins were classified into three distinct subfamilies as follows: HpbHLH28 (subfamily IX), HpbHLH45 (subfamily IIId), and HpbHLH51 (subfamily IIIe) (Fig. 1). HpbHLH28 exhibits no orthologs in A. thaliana but shows collinearity with CsbHLH122, 130, and 129 (Supplementary Table 5), while HpbHLH45 exhibits collinearity with AtbHLH013 and 017 and CsbHLH017. Furthermore, HpbHLH51 shows collinearity with AtMYC2, 3, 4, and 5 and CsbHLH004 and 005 (Supplementary Table 5). Studies indicate that AtbHLH013 and 017, CsbHLH004, 005, and 017, and AtMYC2, 3, 4, and 5 are associated with plant stress resistance and phytohormone regulation [20, 34, 58–61]. In this study, the analysis of cis-acting elements revealed that all HpbHLH28, 45, and 51 contain elements associated with stress- and phytohormone-responsive (Fig. 4). Taken together, the up-regulation under salt stress, conserved association with stress-related orthologs, and possession of stress-responsive promoter elements all indicate that HpbHLH28, HpbHLH45, and HpbHLH51 may participate in regulating salt stress responses in H. pedunculosum.

The distinct tissue-specific expression patterns of HpbHLH28, HpbHLH45, and HpbHLH51 suggest functional divergence (Fig. 5B). Functional characterization showed that only HpbHLH28 and HpbHLH51 possess transcriptional activation activity (Fig. 6B). Notably, integrated analysis identified HpTIFY6B as a specific downstream target of HpbHLH51. Conserved domain analyses revealed that HpbHLH51 contains both a complete bHLH domain and a bHLH-MYC-N domain (Fig. 2), clustering with AtMYC2/3/4 on the phylogenetic tree (Fig. 1). Thus, as a bHLH transcription factor, HpbHLH51 may possess MYC-like properties. As negative regulators of the JA signaling pathway, JAZ proteins precisely control the output intensity of JA signaling by repressing the activity of transcription factors such as MYC2 [62]. In this study, the target gene HpTIFY6B was found to encode a protein assigned to the JAZ subfamily yet lacking the canonical Jas domain (Supplementary Fig. 3) [63]. Consequently, HpTIFY6B cannot function as a typical JAZ repressor in JA signaling. However, studies have shown that JAZ10.4, a splicing variant of JAZ10 lacking the Jas motif in A. thaliana, employs a cryptic domain to bind MYC2/MYC3/MYC4 and suppress JA responses, thereby preventing excessive defense activation under stress [64]. Based on this mechanism, we hypothesize that under salt stress, HpbHLH51 activates HpTIFY6B transcription to generate atypical JAZ proteins that bind MYC2/MYC3/MYC4, thereby restraining excessive JA responses in H. pedunculosum (Fig. 7).

Fig. 7.

Hypothetical regulatory mechanism of HpbHLH51 in response to salt stress. Under salt stress conditions, HpbHLH51 acts as an upstream transcription factor by directly binding to the promoter region of its specific target gene HpTIFY6B, thereby activating its expression. This activation subsequently participates in the regulation of the JA signaling pathway, ultimately mediating the adaptive response of H. pedunculosum to salt stress. This figure was created using BioRender (https://www.biorender.com/)

Conclusion

In this study, 98 HpbHLH genes were identified in H. pedunculosum for the first time. Comprehensive analyses were performed to evaluate their phylogenetic relationships, conserved motifs and domains, gene structures, chromosomal localizations, gene duplication events, synteny, and cis-acting elements. These genes were classified into 23 subfamilies, with segmental duplication events playing a significant role in the expansion of the HpbHLH gene family. Transcriptome data and qRT-PCR analysis revealed the up-regulation of three candidate genes, namely HpbHLH28, 45, and 51, under salt stress. Transcriptional activation assays revealed that only HpbHLH28 and HpbHLH51 exhibited transcriptional activation activity. Further analysis identified HpTIFY6B as a downstream target gene specifically regulated by HpbHLH51, which was confirmed by dual-luciferase assays demonstrating that HpbHLH51 activates HpTIFY6B transcription. We further hypothesize that the HpTIFY6B protein may interact with MYC2/MYC3/MYC4 to modulate the excessive activation of the jasmonic acid (JA) signaling pathway under salt stress. Collectively, these findings deepen our understanding of the complexity of the HpbHLH gene family in H. pedunculosum and highlight the potential roles of HpbHLH genes in the salt stress response.

Authors’ contributions

HL: Writing–original draft, Conceptualization, Methodology. QL: Writing–original draft, Investigation. MS: Conceptualization, Methodology. TY: Resources, Methodology. YY: Resources. YX: Writing–review and editing, Supervision. QZ: Resources, Supervision. ZZ: Writing–review and editing, Conceptualization, Supervision. RL: Writing–review and editing, Conceptualization, Supervision.

Funding

This work was supported by funding from the following sources: the Natural Science Foundation of Sichuan (Grant No. 2025ZNSFSC1125), the Open Project Program (Grant No. SKL-KF202402) of State Key Laboratory of Crop Gene Exploration and Utilization in Southwest China, and the Sichuan Province “three districts” talent support project.

Data availability

The RNA-seq raw data is available in NCBI (https://www.ncbi.nlm.nih.gov/) with the ID ‘PRJNA1301034’. The complete genome sequence of *H. pedunculosum* was obtained from Figshare (https://doi.org/10.6084/m9.figshare.21626153.v2). All data supporting the findings of this study are available within the article and its supplementary materials.

Declarations

Ethics approval and consent to participate

All experimental procedures involving plants were carried out in strict accordance with pertinent institutional, national, and international guidelines and legislation.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.