Genome-Wide Identification and Role of the bHLH Gene Family in Dendrocalamus latiflorus Flowering Regulation

Abstract

The basic helix–loop–helix (bHLH) gene family is a crucial regulator in plants, orchestrating various developmental processes, particularly flower formation, and mediating responses to hormonal signals. The molecular mechanism of bamboo flowering regulation remains unresolved, limiting bamboo breeding efforts. In this study, we identified 309 bHLH genes and divided them into 23 subfamilies. Structural analysis revealed that proteins in specific DlbHLH subfamilies are highly conserved. Collinearity analysis indicates that the amplification of the DlbHLH gene family primarily occurs through segmental duplications. The structural diversity of these duplicated genes may account for their functional variability. Many DlbHLHs are expressed during flower development, indicating the bHLH gene’s significant role in this process. In the promoter region of DlbHLHs, different homeopathic elements involved in light response and hormone response co-exist, indicating that DlbHLHs are related to the regulation of the flower development of D. latiflorus.

1. Introduction

Bamboo species exhibit synchronized flowering and die-off cycles, controlled by complex genetic mechanisms [1]. The collective death of bamboo has resulted in massive losses for forest institutions and private cultivators and has created a serious ecological crisis [2,3,4,5,6]. The molecular regulation of bamboo flowering is a critical research challenge.

Molecular studies have identified numerous flowering-related genes involved in pathways like photoperiod, vernalization, circadian rhythms, temperature, aging, hormones, and sugar production [7,8]. The bHLH domain comprises approximately 60 amino acids, characterized by two amphipathic α-helices that are separated by a variable-length loop region [9]. Members of the bHLH family collaborate with various other proteins to facilitate the regulation of flower development in plants [10]. For example, MdbHLH4 in apples promotes flowering by promoting GA accumulation and FLOWERING LOCUS T (FT) gene expression [11]. A new bHLH transcription factor in chrysanthemum, CmbHLH110, can speed up flowering [12]. Blue light receptors called cryptochromes (CRY1/2) prevent hypocotyl elongation and regulate the onset of flowers [13], CRYPTOCHROME-INTERACTING BASIC HELIX–LOOP–HELIX 1 (CIB1) is a bHLH protein that interacts with CRY2 to promote FT transcription. In response to blue light, A protein complex that can be formed by CRY2, CIB1, and CONSTANS (CO) can then encourage floral initiation [14,15]. In brief, the transcription factor gene family bHLH is involved in controlling the growth and development of plants, abiotic stress, and metabolic network responses [16]. However, the regulatory interactions between bHLH transcription factors (TFs) and hormones in bamboo remain unclear.

D. latiflorus is a clumpy, woody bamboo that blooms sporadically [17]. It has great economic value and is commonly grown in tropical and subtropical regions [18,19]. D. latiflorus often blooms sporadically, and the seed-setting rate of the plant flowering in its natural state is very low [20], which seriously restricts the genetic improvement process of D. latiflorus [21,22]. To understand the significance of the bHLH gene family in regulating the flower development of D. latiflorus, the bHLH gene family was identified and analyzed in this study. Subsequently, the potential molecular function of bHLHs was inferred through differentially expressed genes at four different stages of flower development. This study provides a foundation for understanding the molecular mechanisms of flower development in bamboo.

2. Results

2.1. bHLH TFs in D. latiflorus

We discovered 309 members of the bHLH gene family in D. latiflorus (Table S1), designated as DlbHLH1-DlbHLH309 under their respective chromosomal locations. Moreover, a comprehensive physicochemical characterization was performed on the 309 DlbHLH proteins. The encoded proteins by the aforementioned DlbHLH proteins exhibited a size range of 88 (DlbHLH72) to 1628 (DlbHLH195) amino acids, with molecular weights spanning from 9820.08 Da to 183,630.63 Da and isoelectric points (pI) ranging from 4.61 to 11.05. All proteins were unstable in vitro, according to the instability index, except for DlbHLH27, DlbHLH36, DlbHLH57, DlbHLH129, and DlbHLH139. Subcellular localization predictions showed that 260 DlbHLHs were nuclear, while smaller numbers were found in the chloroplasts (24), cytosol (9), endoplasmic reticulum (5), peroxisome (4), mitochondria (4), plasma membrane (2), and vacuole membrane (1). The prediction results of the transmembrane structure (TM) showed that none of the other family members had transmembrane structures except DlbHLH88 and DlbHLH189, which had one and two transmembrane structures, respectively. None of the DlbHLH members had signal peptide sequences according to signal peptide prediction analytics. The diversity of subcellular localization indicates that members of the DlbHLH gene family conduct a wide range of biological functions.

2.2. Phylogenetic Analysis of the DlbHLHs

To elucidate the phylogenetic relationship among the DlbHLHs, bHLHs from D. latiflorus (309), Arabidopsis thaliana (158) [23], Triticum aestivum (324) [24], and Oryza sativa (190) [25] were used to construct an ML tree (Figure 1). The phylogenetic tree containing 981 bHLH genes was divided into 23 subfamilies based on the taxonomy and topology suggested in earlier works [26,27] (Table S1). These subfamilies ranged in size from 2 to 37 members, with subfamily 3 having the greatest number of DlbHLHs (37), and subfamilies 16, 17, and 22 having just 1 DlbHLH. In multiple clusters during evolution, The bHLH gene family demonstrated substantial species bias, indicating that particular gene duplication happened after species divergence [28].

2.3. Gene Structure and Conserved Motifs of DlbHLHs

The gene structure and conserved sequences of the 309 DlbHLH genes based on their evolutionary relationships were analyzed and provided (Figure 2). Using the MEME, a conserved pattern search on 309 DlbHLH genes identified 20 conserved motifs (motifs 1–20; Figure 2B). DlbHLH genes from the same subfamily shared similar types, amounts, and distributions of conserved motifs. Motifs 1, 2, and 10 are found in the majority of DlbHLH genes, while others are found only in a few. For example, (motifs 1, 2, 3, 10) and (motifs 1, 2, 4, 9, 10) are present in every member of the families of subfamilies 8 and 11, respectively. Afterward, we evaluated the DlbHLH protein domains and exhibited the locations of the bHLH-conserved domains that were matched to the D. latiflorus gene structure (Figure 2C). The findings showed there were 16 distinct bHLH domain types with considerable diversity, and that, with few exceptions, gene members belonging to the same subfamily had similar bHLH domain types.

We also examined the architecture of the DlbHLH genes, which ranged in length from 472 bp (DlbHLH157) to 42,590 bp (DlbHLH88) (Figure 2). The number of exons in the 309 DlbHLHs ranged from 1 to 17 (Figure 2D). The amount of DlbHLHs with three CDSs (51) is the highest, followed by DlbHLHs with seven CDSs. DlbHLH191 has the greatest number of exons and introns among all members, with 17 and 16, respectively. DlbHLH178 had as many as five UTRs, indicating that its alternatively spliced version may be the most complicated [29].

2.4. Chromosomal Location, Collinearity, and Evolution Analysis of DlbHLHs

Gene duplication is an important first step in the process of extending the functioning of a multigene family [30]. In each subgenome, we found that the DlbHLHs are evenly distributed on eleven or twelve chromosomes. Nevertheless, each chromosome has an unequal distribution of DbHLH numbers. chr2.1 and chr25.1, which belong to the B and C subgenomes, respectively, had the largest number of DlbHLHs, with sixteen each. The second most abundant are chr1.1 and 7.1, each with fifteen DlbHLHs. ch35.1 has the lowest number of DlbHLHs, only two.

To investigate the DlbHLH family gene duplication events, homology analysis identified 379 fragment repeat pairs and 4 tandem repeat pairs (Figure 3; Table S2). Subfamily 4 has the largest number of duplicate gene pairs (48 pairs, all of which are fragment copies). The results indicate that DlbHLH fragment replication events are the primary factors influencing the evolution of DlbHLHs. Through a comparison of mutation rates between Ka (nonsynonymous) and Ks (synonymous), the selection pressure of gene duplications was assessed [31], so we computed the Ka/Ks ratio for all 383 gene pairs (Table S3). All gene pairs had Ka/Ks ratios below 1, with most under 0.5, suggesting that DlbHLH gene pairs underwent purifying selection during evolution.

We chose four sample species for collinearity comparative study to investigate the gene replication timing of DlbHLHs and determine its phylogenetic relationship: Arabidopsis thaliana, Triticum aestivum, Oryza sativa, and Phyloglostachys edulis (Figure 4). Collinear analysis of different species is a way to study their evolution and affinities [32]. A total of 877 PebHLHs were collinearly correlated with the bamboo genome, followed by wheat (832), rice (399), and Arabidopsis (32) (Table S4). Among them, There are 23 DlbHLHs (DlbHLH4, DlbHLH7, DlbHLH24, DlbHLH26, DlbHLH47, DlbHLH54, DlbHLH59, DlbHLH70, DlbHLH116, DlbHLH137, DlbHLH151, DlbHLH169, DlbHLH199, DlbHLH208, DlbHLH218, DlbHLH220, DlbHLH226, DlbHLH237, DlbHLH245, DlbHLH247, DlbHLH284, DlbHLH285, and DlbHLH289) (Table S4-2) in all four representative species. The presence of collinear associations between species suggests that genes in these bHLH gene families are essential in evolution.

2.5. Cis-Acting Elements in DlbHLH Promoters

Cis-regulatory elements (CREs) contribute to development, and one common reason for evolutionary change is their divergence [33]. A total of 8488 CREs were predicted in the promoter regions of DlbHLHs (Table S5), with 19 representative CREs shown in Figure S1. Among these, DlbHLH238 had the most CREs, a total of 62 elements. The overall number of CREs was categorized into three major groups (Figure S1), and the number of CREs differed within subfamilies (Figure 5). Growth and development (3909) comprised the following categories: root-specific (9, 0.2%), circadian control (55, 1.4%), seed-specific regulation element (87, 2.22%), endosperm expression element (65, 1.66%), light-responsive element (3375, 86.33%), meristem expression element (282, 7.21%), and AT-rich DNA binding protein (36, 0.9%). According to Figure 5, within the promoter area of every DlbHLH, CREs linked to light responsiveness (such as the GATA-motif, GATA-motif, G-box, ACE, and Box I, etc.) are extensively expressed. Apart from components that react to light, there are additional regulatory components necessary for growth and development. They might be missing in many subfamilies such as AT-rich for AT-rich DNA binding protein, GCN4_motif for endosperm, RY-element for seed, and circadian for circadian control.

The second class was related to hormone responsiveness (3166), including abscisic acid-responsive (1175, 37.11%), auxin-responsive (222, 7%), gibberellin-responsive (244, 7.70%), MeJA-responsive (1206, 38.01%), salicylic acid-responsive (151, 4.77%), and zein metabolism regulation (168, 5.30%). MeJA-responsive elements (CGTCA-motif and TGACG-motif) and abscisic acid-responsive elements (ABREs) are widely present in the promoter regions of almost all DlbHLH subfamilies. CREs associated with salicylic acid response are present in most subfamilies except subfamilies 20 and 22. CREs associated with the auxin reaction (AuxRE, AuxRR-core, TGA-element, and TGA-box) are found in most subfamilies except subfamily 23 in their promoter region. Gibberellin-responsive elements (GARE-motif, TATC-box, and P-box) are present in the promoter region of 163 of the 309 DLbHLHs except for the subfamily 22. In addition, many DlbHLHs, such as DlbHLH8, DlbHLH12, DlbHLH15, DlbHLH42, DlbHLH54, DlbHLH65, DlbHLH78, DlbHLH86, DlbHLH143, DlbHLH216, DlbHLH233, and DlbHLH289, in the promoter regions contain multiple identical hormone responses, which suggests that the specific hormones may have a more rapid and stronger reaction. At the same time, some DlbHLH response elements, such as DlbHLH12, DlbHLH102, DlbHLH109, DlbHLH143, DlbHLH152, DlbHLH297, DlbHLH219, and DlbHLH238, in the promoter regions contain a variety of hormones, implying that they might be involved in some hormone regulatory networks.

The third class was stress-responsive CREs (1413), containing anaerobic induction (775, 54.85%), defense and stress-responsive elements (98, 6.94%), low-temperature responsive elements (246, 17.40%), MYB drought inducibility (277, 19.60%), MYB flavonoid biosynthetic elements (11, 0.78%), and wound-responsive elements (6, 0.42%). Among these, CREs for anaerobic induction and MYB drought inducibility are present and abundant in all subfamilies. CREs associated with defense and stress-responsive elements are present in most subfamilies except subfamily 17. CREs associated with defense and low-temperature responsive elements are present in most subfamilies except subfamilies 19 and 22. Furthermore, several subfamilies are shown to have higher amounts of stress-related CREs.

Previous studies have shown that JASMONATE ASSOCIATED MYC2-LIKE1 (JAM2) negatively regulates JA response by antagonizing MYC2 [34]. We identified some homologous genes of JAM2, namely DlbHLH1, DlbHLH14, DlbHLH19, DlbHLH38, DlbHLH132 DlbHLH148, DlbHLH183, and DlbHLH285, and these gene promoter regions have Me-JA response elements. The findings showed that there was substantial variation in the number and makeup of cis-regulatory elements in the promoter regions of various DlbHLHs, both within and between subfamilies. This discovery implies that a wide range of CREs associated with hormones, plant growth, developmental processes, and stress response control the expression of bHLH genes in D. latiflorus.

2.6. Transcription Factor Binding Sites (TFBSs) in DlbHLH Promoters

The binding sites for 42 TF families were identified in the promoters of the DlbHLH family members (Figure S2). The greatest number of C2H2 binding sites was observed to be 289. This was followed by MYB and NAC, with 235 and 228, respectively, while the lowest number was recorded for YABBY, with a mere 3. In addition, TFBSs in each DlbHLH promoter differ in type, number, and location. The promoter region of DlbHLH98 has 24 binding sites, while DlbHLH244 has only 4 binding sites, namely C2H2, MYB, B3, and ERF.

2.7. Protein–Protein Interaction Network of the DlbHLH Members

Based on the homologous alignment of DlbHLH in Arabidopsis, we constructed a protein–protein interaction network between DlbHLHs (Figure S3). The results showed that there was an interaction between the DlbHLH proteins. AT4G0050 (PIF8) and AT2G14760 (bHLH94) exhibit a direct interaction. In addition, there are some polygenic interactions. Examples include AT3G47640 (bHLH47), AT2G22759 (bHLH18), AT5G5960 (bHLH41), and AT1G68810 (bHLH30). Notably, we identified several genes involved in hormone-regulated flower development pathways. For example, AT1G32640 (MYC2) has been reported to be involved in jasmonic acid-mediated Arabidopsis flowering inhibition [35]. AT5G08130 (BIM1) is a Brassinosteroid signal transduction element that synergistically regulates pollen fertility with SQUAMOSA PROMOTER BINDING PROTEIN-LIKE 8 (SPL8) [36]. AT1G01260 (JAM2) regulates petal senescence by JA [37]. These genes are directly or indirectly related to each other.

2.8. Expression Patterns of DlbHLHs in Four Flower Developmental Stages

The expression profiles of 309 DlbHLHs in four phases of flower development were described using RNA-seq data to investigate the potential functions of the DlbHLHs (Figure 6 and Figure 7). After excluding genes with low levels of expression, 156 DlbHLHs remained. Using hierarchical clustering, the DlbHLHs could be divided into four blocks (block A–C); genes from the same subfamilies may have diverse expression patterns. The DlbHLHs in different blocks exhibited disparate patterns of expression in space and time: (A) DlbHLHs in block A are preferentially expressed in bud stage, including genes from subfamily 2, subfamily 3, subfamily 4, subfamily 5, subfamily 7, subfamily 9, subfamily 10, subfamily 11, subfamily 12, subfamily 18, subfamily 19, and subfamily 21. (B) Most of the DlbHLHs in region B were expressed in four flower developmental stages, among which DlbHLH135 was highly expressed in bud stage and DlbHLH143 was highly expressed in middle-flower development, exhibiting a trend in initially growing and then reducing during bloom development (C). The majority of DlbHLHs in the C region show high expression levels at all four phases of floral development. This finding suggests that these genes may play a function in floral formation. More importantly, CRY2-INTERACTING BHLH 4 (CIB4) has been shown to participate in FT transcription and thus to promote flowering [38]. Many DlbHLHs have been identified as CIB4 homologs, such as DlbHLH21, DlbHLH39, DlbHLH97, DlbHLH118, and so on. Among them, DlbHLH97 and DlbHLH118 are only significantly expressed in the bud stage, showing a developmental stage-specific expression pattern. We identified BIM1 as a pollen sterility gene among homologous genes [36], respectively, DlbHLH51, DlbHLH71, DlbHLH223, DlbHLH228, DlbHLH228, DlbHLH238, and DlbHLH259. In addition to DlbHLH238 in block C, which is highly expressed in all four stages of flower development, other homologous genes are located in block B. FBH1 is a bHLH-related protein and a transcriptional activator of the CO gene. It can favorably control CO-mediated blooming time [39]. DlbHLH214 of block C was highly expressed at all four stages of flower development, indicating similar functions in D. latiflorus. MYC2 regulates the flowering time of Arabidopsis by inhibiting FT transcription in the JA pathway [35,40]. In Figure 6, four homologous DlbHLHs of MYC2 (DlbHLH233, DlbHLH246, DlbHLH264, and DlbHLH271) are expressed in different stages of flower development priority. Moreover, DlbHLH246, DlbHLH264, and DlbHLH271 promoter regions exist in the Me-JA responsive elements. These results show a similar function in the regulation of D. latiflorus flower development.

2.9. Validation of DlbHLH Expression

To verify the reliability of the transcriptome data, qRT-PCR was used to analyze the expression profiles of eight representative DlbHLHs at four flower developmental stages (Figure 8; Table S7). Consistent with the transcriptome data, DlbHLH95, DlbHLH123, and DlbHLH264 were highly expressed in late flower development, and DlbHLH97, DlbHLH118, and DlbHLH165 were significantly highly expressed in the F1 stage. And DlbHLH135 and DlbHLH150 were highly expressed in all four flower developmental stages; both of these genes showed the trend in first increasing and then decreasing, suggesting that they are involved in the four flower developmental stages.

3. Discussion

An increasing body of information indicates that bHLH TFs participate in the diverse developmental processes of eukaryotes [41,42,43,44]. D. latiflorus is the most widely cultivated clumping bamboo species in southern China and is of significant economic value for both human consumption and wood production [45]. The flowering period of D. latiflorus is long and irregular [17,22]. The establishment of its genome has paved the way for the resolution of the previously intractable issue of flowering patterns [21,46]. Although numerous bHLH genes in other plants have been examined, not much is known about the bHLH genes in D. latiflorus. We identified 309 bHLH genes in the D. latiflorus genome, categorizing them into 23 subfamilies according to phylogenetic relationships. The results indicated significant variation in the structure and function of bHLH TFs in different plants by comparing what is similar and different to the categories of other plant species [16]. The conserved motifs and gene structures of genes belonging to the same subfamily were shown to be similar (Figure 2). Motifs 1, 2, and 10 were conserved in almost all bHLH proteins and were critical for the operational specificity of these transcription factors [47]. Overall, the DlbHLH gene family is diverse and complex, with subfamily-specific structural traits supporting their phylogenetic categorization. In contrast, structural variability across distinct DlbHLHs may reflect the functional variety in the bHLH gene family.

Tandem or segmental duplication is the most commonly evaluated mechanism for gene family expansion [48]. DlbHLH genes were found to be extensively distributed across all subgenomes according to chromosome location. A total of 384 gene pairs (379 segmental repeats and 4 tandem repeats) with gene repetition time were discovered in D. latiflorus, showing that the growth of the DlbHLH gene family may be predominantly through fragment repeats. The Ka/Ks ratio for all gene pairings was less than one, indicating that purifying selection had a larger role than positive selection in the evolution of the D. latiflorus bHLH gene family. Among the plants used in the collinearity study, Phyllostachys edulis and D. latiflorus have the closest evolutionary relationship. The bulk of DlbHLH members are found in close clades with the bHLHs of rice and wheat, although their genetic lineage from Arabidopsis is comparatively distant. Structural research may also provide useful details on evolutionary duplication moments. Structural analysis can yield significant insights into evolutionary replication events. Certain genes that repeat have comparable patterns of expression and belong to the same subfamily. This observation indicates potential functional redundancy among them [49]. DlbHLH21 and DlbHLH39, for example, share a similar motif composition, bHLH domain type, and gene structure, as well as similar expression patterns. As mentioned in the previous section, DlbHLH21 and DlbHLH39 are both homologous genes of CIB4, so we speculated that they may have functional redundancy in participating in FT transcriptional activation to promote flowering. However, many duplicate gene pairs show distinct expression patterns. For example, DlbHLH124 and DlbHLH63 have distinct exon–intron architectures and expression patterns. We postulate that structural alterations cause variances in the expression of these duplicated genes, leading to functional diversity.

D. latiflorus is significant economically, but it also possesses the unique characteristics of rapid growth and a lengthy flowering interval, pollen sterility, and so on [50,51,52]. Identifying the mechanism that controls floral growth is therefore extremely important. The bHLH transcription factor is essential for plant growth and development. It is involved in a variety of physiological processes and has a wide range of functions in plant hormone synthesis and metabolism [10,53]. Interestingly, we identified several BIM1 homologs, a bHLH family gene that regulates pollen fertility. In the future, in situ hybridization experiments can be conducted to explore their temporal and spatial expression in anthers. In addition, we also found some homologous genes involved in hormone-regulated flowering pathways, especially jasmonic acid. Specifically, the mechanism of how the hormone interacts with the DlbHLH gene to regulate flowering remains to be further verified.

There are many factors regulating plant flowering, and the process is affected by light, temperature, and endogenous signal transduction. We found that light-responsive elements are present in almost all DlbHLH promoter regions that regulate photocontrolled transcriptional activity. A large number of DlbHLHs are highly expressed in some stages of flower development, which may act as a photoperiodic regulator of flowering in D. latiflorus. According to previous research, hormones play a crucial role in regulating bamboo flowering [54]. We found 3166 hormone-associated cis-acting elements, covering all DlbHLHs. This implies a potential link between DlbHLHs and hormonal response. In addition, abscisic acid has been observed to regulate the photoperiodic pathway in response to flowering, particularly under conditions of abiotic stress. In conditions of drought, the ABA hormone negatively regulates flowering through a pathway that requires the SOC1 protein [55], and jasmonic acid has been shown to cause pollen sterility [56]. Many DlbHLH promoter regions contain either abscisic acid-responsive elements (287 of 309) or jasmonic acid-responsive elements (245 of 309). This could explain why bamboo has more flowers but lower pollen and seed-setting rates. It would be beneficial for future studies to place a greater emphasis on the investigation of the regulatory mechanisms of endogenous hormones in the development of bamboo flowers. The unique flowering characteristics of bamboo may be influenced by chromosome polyploidy, and some genes related to flowering pathways have been positively selected over a long evolutionary time scale. To sum up, the DlbHLH genes may play a key role in the development of D. latiflorus flowers. However, the exact physiological mechanisms underlying their function require further experimental studies for elucidation.

4. Materials and Methods

4.1. Plant Materials

In this study, the D. latiflorus flowers from three flowering clumps in four development stages (F1, young bud stage; F2, flowering mid-term; F3, full-bloom stage, and F4, fading stage) were collected from a Baisha state-owned forest farm in Fuzhou, Fujian province of China (26°20′ N, 119°07′ E), in July 2023 (Figure 7); three biological replicates were collected for each sample. Before picking the flowers, alcohol was evenly sprayed on their surface to achieve thorough coverage. To ensure thorough disinfection, the alcohol was applied to the floral surfaces for 30 s to 1 min. The alcohol-sprayed flowers were then gently immersed in deionized water for approximately 2 min to remove any remaining alcohol and any dead microbes from the surface. The harvested flowers were immediately combined and flash-frozen in liquid nitrogen. They were stored at −80 °C.

4.2. RNA-Seq and Transcriptome Data Analysis

The RNA prep Pure Plant Kit (Tiangen, Beijing, China) was utilized to extract the total RNA from the flowers. Afterward, 2% agarose gel electrophoresis was used to evaluate the quality of the total RNA. A NanoPhotometer^® spectrophotometer (IMPLEN, Westlake Village, CA, USA) and a Qubit^® RNA assay kit with a Qubit^® 2.0 fluorometer (Life Technologies, Carlsbad, CA, USA) were used to measure the concentration of RNA. An Agilent^® Bio-analyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA) was used to measure RNA integrity using an RNA Nano 6000 assay kit (Agilent Technologies, Santa Clara, CA, USA). A benchmark RNA integrity number of 7 was established as the benchmark for quality evaluation. For Illumina^®, the NEB-Next^® UltraTM RNA library prep kit (NEB, Ipswich, MA, USA) was utilized during the library creation process. Using the Illumina 6000 platform, all samples underwent sequencing to provide 150 bp paired-end reads. We processed raw RNA-Seq reads using fastp v0.23.2 [50], eliminating sequences with adapters, bases with Qphred ≤ 20, and N-base percentages larger than 15%. The HiSAT2 software v2.0.1 [57] was used to map the clean RNA-seq reads to the D. latiflorus genome with default parameters. FeatureCounts [58] was used to quantify the mapped reads, and DESeq2 [59] (|log2 (Fold Change)| ≥ 1 and Padj ≤ 0.05) was used to screen for differentially expressed genes. The TPM value of the DlbHLHs was taken from all transcriptome data and normalized to log2 (TPM+1), and heat maps were created using TBtools [60].

4.3. qRT-PCR Validation

The same sample used for RNA-seq was utilized for qRT-PCR to confirm the expression of DlbHLHs in other tissues. RNA was extracted from four flower stages using the Trelief Hi-Pure Pure Plant RNA Kit (Tiangen, Beijing, China) and reverse transcribed into cDNA with the PrimeScript^™ RT reagent kit (Perfect Real Time, Takara, Japan). Quantitative real-time PCR analysis of 8 selected DlbHLHs was performed using SYBR Green premix Ex Taq Kit (TaKaRa, Dalian, China) on an Applied Biosystems 7500 Real-Time System (Applied Biosystems, Foster City, CA, USA). The qRT-PCR amplification sequence was as follows: 95 °C pre-denaturation for 5 s, 95 °C denaturation for 30 s, 60 °C annealing for 30 s, 72 °C extension for 30 s, and 40 cycles. The GAPDH gene was used as an internal reference [61]. Three replicates of each gene were carried out, and the fold change was calculated using the 2^−ΔΔCT method. The primers were designed through the NCBI primer design online (https://www.ncbi.nlm.nih.gov/tools/primer-blast/ (accessed on 25 July 2024)). The primers used in this work are listed in Table S6.

4.4. Identification of D. latiflorus bHLH Transcription Factor Family

Arabidopsis bHLH sequences were obtained from the TAIR database (https://www.arabidopsis.org/ (accessed on 28 June 2024)), as queries used local alignment search tools (BLAST) [62] against D. latiflorus. The HMM profile of the bHLH transcription factor gene family (PF00010) was downloaded from Pfam (https://www.ebi.ac.uk/interpro/entry/pfam/ (accessed on 1 July 2024)) [63]. HMMER [64] search was used to search the bHLH transcription factor family in the D. latiflorus genome with default parameters. Consequently, the integrity of the presumed bHLH transcription factor family was further examined by InterPro [65].

4.5. Bioinformatic Analysis of D. latiflorus bHLH Transcription Factor Family

The physicochemical properties were analyzed using ExPASy (https://www.expasy.org/ (accessed on 12 July 2024)) [66]. Subcellular localization prediction of bHLH transcription factor family members was performed using the WoLF PSORT online website (https://wolfpsort.hgc.jp/ (accessed on 12 July 2024)) [67]. We utilized MEME (https://meme-suite.org/meme/) [68] to find the conserved motifs of DlbHLHs, with the following parameters: maximum e-value = 1 × 10⁻⁵; minimum motif length = 6; maximum motif length = 200; and number of motifs = 20. We extracted 2000 bp sequences from each DlbHLH gene and used PlantCARE (https://bioinformatics.psb.ugent.be/webtools/plantcare/html/ (accessed on 12 July 2024)) [69] and PlantTFDB (https://planttfdb.gao-lab.org/ (accessed on 12 July 2024)) [70] to predict cis-elements and TFBs. Afterward, we created conserved domain sequence logos and gene structures. The domain analysis result file was obtained by NCBI Batch CD (https://www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi/ (accessed on 12 July 2024)). Using the STRING database (http://string-db.org (accessed on 12 July 2024)), we conducted protein interaction analysis and created a protein interaction network based on the DlbHLH homologs in Arabidopsis.

4.6. Phylogeny, Gene Duplications, and Selection Pressure Analysis

To study the evolutionary relationship of D. latiflorus, we examined homologs of DlbHLH genes in the model plants Arabidopsis, O. sativa, and T. aestivum using BLAST software v2.10.0 [62] with a maximum e-value of 1 × 10⁻²⁰. These sequences were concatenated with those from D. latiflorus, and MUSCLE v5.0 [71] was used to construct the multiple protein sequence alignment, which was then trimmed using trimAL [72]. The IQ-TREE 2 program [73] was used to generate the phylogenetic tree of related proteins. Using the maximum likelihood (ML) method, the bootstrap was set to 1000 and the phylogenetic tree was visualized using iTOL (https://itol.embl.de/) [74].

To assess the degree of overlap between DlbHLHs and other species, the annotation file of the D. latiflorus genome was used to determine the subgenome assignment, chromosomal location, and exon count of DlbHLHs [46]. All D. latiflorus proteins were analyzed for similarity using Blast, and data on fragments and tandem repeats were obtained using MSCanX [75]. Moreover, the DlbHLH orthologs across species were analyzed in the same way, with the data visualized using NGenomeSyn 1.41 [76]. The genome accession numbers of D. latiflorus, Arabidopsis, wheat, and rice are JACBGG000000000, GCA_000001735, GCA_900519105, and GCA_001433935. The collinearity among DlbHLHs was visualized using the Advanced Circos module in TBtools. The values for non-synonymous substitution (Ka) and synonymous substitution (Ks) in duplicated DlbHLH gene pairs were then calculated.

5. Conclusions

In this study, 309 members of the bHLH gene family were identified in D. latiflorus. Based on phylogenetic relationships with the bHLH genes in Arabidopsis, wheat, and rice, these genes were divided into 23 subfamilies. Structural analysis showed that DlbHLH proteins are relatively conserved in specific subfamilies. Collinearity analysis indicated that the amplification of the DlbHLH gene family may be mainly carried out by fragment duplication, some duplicate gene pairs (DlbHLH21 and DlbHLH39) may have functional redundancy, and many duplicate gene pairs (DlbHLH63 and DlbHLH124) show different expression profiles, indicating functional diversification. Many DlbHLHs have been identified to be involved in the regulation of flower development. Different homeopathic regulatory elements of hormone response are present in its promoter region, suggesting that DlbHLH may be involved in the regulation of flower development. These results laid a foundation for further research on the regulatory mechanism of bHLH protein in the development of bamboo flowers.