Genome-Wide Identiﬁcation and Functional Analysis of the Basic Helix-Loop-Helix (bHLH) Transcription Family Reveals Candidate PtFBH Genes Involved in the Flowering Process of Populus trichocarpa

: As one of the largest TF families+ in plants, the basic helix-loop-helix (bHLH) family plays an important part in the growth and development of many plants. FLOWERING BHLH (FBH) encodes a bHLH-type transcriptional factor related to the ﬂowering process. Poplar is a model woody plant as well as an important economic tree species with a small genome. However, the characteristics of the bHLHs and FBHs gene family in the newest version of Populus trichocarpa genome have not been analyzed yet. We identiﬁed 233 PtbHLHs and 10 PtFBHs in the newest version genome, and PtbHLHs were classiﬁed into 21 groups with FBH subfamily occupying one, supported by phylogenetic analysis, exon–intron patterns, and conserved protein motifs. These PtHLHs were distributed on 19 chromosomes unevenly and expressed in nucleus mainly. Gene duplication and synteny analysis have indicated that the PtbHLHs gene family has undergone strong puriﬁcation selection during the evolution process. The cis -elements analysis has suggested that PtbHLHs may be related to the growth and development. Conserved residues of FBHs among Arabidopsis and poplar were also identiﬁed. Expression of 227 PtHLH genes (6 unmatched, 13 no expressed) showed diverse patterns in different tissues, implying their multiple functions. Protein–protein interaction network prediction and expression patterns in three states of the ﬂowering process (Flowers-Dormant, Flowers-Expanding and Flowers-Expanded) suggested that some members of PtbHLH and PtFBH family may be involved in the ﬂowering process. Our comprehensive and systematic analysis can provide some valuable clues and basic reference toward further investigations on physiological and molecular functions of PtbHLHs .


Introduction
There are four types of transcription factors (TFs) in higher plants categorized by the sequence of arginine and lysine residues in the DNA binding region: Zinc finger, HTH, bZIP, and HLH [1][2][3][4]. As one of the largest TF families in plants [5], the basic helix-loop-helix (bHLH) family plays an important part in the growth and development of many plants. Each member of bHLH family contains a bHLH domain, which consists of around 50-60 conserved amino acid sequences. The bHLH domain contains two functional regions: a basic amino acid region and an HLH region. The basic region is composed of approximately 15 amino acids at the N-terminal and responsible for binding to the cis-element E-box (CANNTG) or G-box (CACGTG) while the HLH region of a length of approximately 50 amino acids in a structure of two alpha helixes separated by a loop accessed on 1 September 2021) using the 153 bHLH gene locus gained from PlantTFDB database (http://planttfdb.cbi.pku.edu.cn/, accessed on 1 September 2021).
MEGA-X was used for the evolutionary and phylogenetic analyses (https://www. megasoftware.net/, accessed on 1 September 2021) [25]. A total of 233 PtbHLH amino acid sequences, were used to generate an unrooted phylogenetic tree using the neighbor-joining method with 1000 bootstrap replications, JTT substitution model, and 50% partial detection. Multiple sequence alignments were performed by MUSCLE [26] and then those sequences were clipped to 71 bp each using TBtools by Multiple Alignment Trimming before building trees [27].

Chromosome Location and Gene Duplication Analysis
Each PtbHLH gene was matched with the chromosomes of poplar using TBtools program based on the genome annotations of poplar [28]. The MCScanX software was used to identify duplicated genes with E value set to e −5 [29]. We used BioEdit Sequence Alignment Editor [30] to align the CDS sequence of the replicated gene pair, then DnaSP5 (https://dnasp.software.informer.com/, accessed on 1 September 2021) to calculate the Ks and Ka of the replicated gene pair, and the evolution time (T) calculation was based on the Ks value: T = Ks/2λ, λ = 9.1 × 10 −9 [31]. Members of bHLHs in Arabidopsis and rice were obtained from PlantTFDB database (http://planttfdb.gao-lab.org/family.php?fam=bHLH, accessed on 1 September 2021).

Protein Properties and Structure Analysis
Physical and chemical features of each PtbHLH protein were predicted by use of the ExPASy website (http://web.expasy.org/protparam/, accessed on 1 September 2021) [32]. The MEME tool (http://meme-suite.org/tools/meme, accessed on 1 September 2021) was used to identify 15 conserved motifs of bHLHs in P. trichocarpa with default parameters [33]. The gene structure was visualized by using TBtools program based on the genome annotations of poplar.

Cis-Acting Element Analysis and Multiple Sequence Alignment of the PtbHLHs
For each PtbHLH gene, the gene upstream 2000 bp DNA sequence of the start codon was obtained from the Phytozome v12.1 database. We then used the PlantCRAE to extract cis-acting elements that will be visualized by use of the TBtools software [34]. The ciselements of 233 identified PtbHLH genes were classified as Light responsive, Hormone responsive, Stress responsive, Hormone responsive, and Light responsive.

Subcellular Location and Protein-Protein Interaction Network Prediction
TargetP Server online software (http://www.cbs.dtu.dk/services/TargetP/, accessed on 1 September 2021) was used to predict the subcellular location of bHLH proteins. The 16 full-length FBH protein sequences (six Arabidopsis FBHs and 10 poplar FBHs) were selected for multiple sequence alignment using the MAFFT (https://www.ebi.ac.uk/ Tools/msa/mafft/, accessed on 1 September 2021) online tool [35]. The alignment results were displayed by JalView software (http://www.jalview.org/download, accessed on 1 September 2021) [36]. Proteins of 10 PtFBHs and 18 PtbHLHs possibly involved in flowering development were used for predicting protein-protein interactions by the STRING website (https://string-db.org/, accessed on 1 September 2021). We chose the informative figures gained to display. From these figures obtained from STRING database we can predict several relatively reliable interactions and functions of PtbHLHs.

Expression Patterns of PtbHLH Genes in Different Tissues
We gained the gene expression database available on PopGenIE website (ftp:// plantgenie.org/Data/PopGenIE/, accessed on 1 September 2021) to explore the PtbHLH gene expression patterns in various tissues and different growth stages [37]. In the database, six PtbHLH genes could not be matched and were omitted. Flowers-Expanded, Flowers-Dormant, Flowers-Expanding, Roots-Control, Seeds-Mature, Leaves-Control, and Cambium-Phloem-Dormant were sampled to investigate the expression patterns of PtbHLH genes. Subsequently, log2FC values were used for evaluating the gene expression and the heatmap was generated using TBtools. The RT-qPCRs were performed to confirm the DEGs analysis. We used the Plant RNA kit (Aidlab-Biotech, Beijing, China) to extract the total mRNA. The RT-qPCR reactions were performed in a real-time PCR system using SYBR Green (Applied Biosystems, Waltham, MA, USA) according to the manufacturer's instructions. The Actin2/7 gene was used as a reference, the RNA relative expression of each gene was calculated using the the 2 −∆∆CT method. The RT-qPCR reactions were repeated three times. The specific primers are shown in Table S8. The experimental data were subjected to analysis of variance by Statistical Product and Service Solutions 17.0 (SPSS). For statistical analyses, Student's t-test was used to generate every p value.
We then analyzed physicochemical properties and predicted subcellular location of proteins encoded by these 233 PtbHLH genes (Tables S2 and S3). PtbHLH proteins are predicted to locate in various organelles such as nucleus, inner membrane system, mitochondria, and chloroplasts. According to statistics, 208 PtbHLH proteins are mainly expressed in plant cell nuclei, accounting for 89% of the total number. Therefore, it is speculated that PtbHLHs mainly play a role in the nucleus. Their number of amino acids varies from 84 (PtbHLH107) to 970 (PtbHLH11). Molecular weights fall into the range of 9359.95 (PtbHLH107) to 104,899.77 kDa (PtbHLH11). The protein is basic when its value of theoretical isoelectric point (pI) is over 7, otherwise it is acidic. We predict 86 basic proteins and 147 acidic proteins from 233 PtbHLHs, whose pl values range from 4.44 to 11.74. Interestingly, we found that only one of the proteins have the grand average of hydropathicity index greater than 0 (PtbHLH107), which means it is a hydrophobic protein, and the other indices are less than 0, which means they are hydrophilic proteins. The values of aliphatic index run the gamut from 50.64 to 130.00. The values of protein instability index are between 23.05 and 89.92. Nine PtbHLH are stable proteins (PtbHLH101, PtbHLH210, PtbHLH107, PtbHLH198, PtbHLH128, PtbHLH11, PtbHLH232, PtbHLH185, and PtbHLH189), whose values of protein instability index are less than 40, and 224 proteins unstable. Regarding the formulas of PtbHLH proteins, they are all listed in the form (Table S2).

Chromosomal Distribution and Synteny Analysis of PtbHLH Genes
According to the annotation of the poplar genome database, we found that 233 PtbHLH genes are widely distributed on 19 chromosomes of poplar and gained the gene density on every chromosome. Then a chromosomal distribution of the PtbHLH genes was mapped to locate the position of each member ( Figure 1). The map shows that chromosome 2 harbors the largest number of PtbHLH genes (27) while chromosome 16 and chromosome 17 both contain the least (7). In summary, PtbHLHs were found on all 19 chromosomes and on most chromosomes the number of the genes is between 7 and 12. Our study also analyzed the gene duplication events in PtbHLH genes. The MCScanX software was used to perform a collinearity analysis. The results showed that 223 PtbHLH genes came from whole genome duplication (WGD) or segmental duplication and the other 10 PtbHLH genes came from dispersed duplication (Table S4). There were six groups of tandem duplicated PtbHLH genes (PtbHLH91/PtbHLH92, PtbHLH97/PtbHLH96, PtbHLH222/PtbHLH223, PtbHLH229/PtbHLH231, PtbHLH229/PtbHLH230, PtbHLH230/PtbHLH231) and 110 pairs of segmentally duplicated PtbHLH genes ( Figure 2, Table S4). Gene duplication greatly promoted the expansion of the bHLH gene family in poplar genome, and WGD/segmental duplication events played a major driving role. In addition, we performed homology analysis to identify the conserved chromosomal segments within the poplar genome in order to explore the evolutionary process of the PtbHLHs. The distribution of homologous PtHLH gene pairs on chromosomal distribution is random. The evolution date of WGD/fragment replication events can be estimated by the Ks value (frequency of synonymous mutations) [43]. As previously studied, the P. trichocarpa genome has undergone at least three genome-wide replication events [16]. Therefore, we used the Ks value to estimate the evolution date of gene duplication events in the PtbHLH gene family (Table S5). The Ks value means that most of the PtbHLH genes were duplicated around the recent WGD event while some other genes originated from ancient WGD. The intensity and direction of selection can be represented by the ratio of Ka/Ks [44]. The Ka/Ks ratio of all homologous PtbHLH genes is less than 1, indicating that the PtbHLH gene family has undergone strong purification selection during the evolution process. According to the PtbHLH gene collinearity relationship between poplar and Arabidopsis or rice, we can easily know that there are a large number of homologous bHLH genes among them ( Figure S1). Our study also analyzed the gene duplication events in PtbHLH genes. The MCScanX software was used to perform a collinearity analysis. The results showed that 223 PtbHLH genes came from whole genome duplication (WGD) or segmental duplication and the other 10 PtbHLH genes came from dispersed duplication (Table S4). There were six groups of tandem duplicated PtbHLH genes (PtbHLH91/PtbHLH92, PtbHLH97/PtbHLH96, PtbHLH222/PtbHLH223, PtbHLH229/PtbHLH231, PtbHLH229/PtbHLH230, PtbHLH230/PtbHLH231) and 110 pairs of segmentally duplicated PtbHLH genes ( Figure 2, Table S4). Gene duplication greatly promoted the expansion of the bHLH gene family in poplar genome, and WGD/segmental duplication events played a major driving role. In addition, we performed homology analysis to identify the conserved chromosomal segments within the poplar genome in order to explore the evolutionary process of the PtbHLHs. The distribution of homologous PtHLH gene pairs on chromosomal distribution is random. The evolution date of WGD/fragment replication events can be estimated by the Ks value (frequency of synonymous mutations) [43]. As previously studied, the P. trichocarpa genome has undergone at least three genome-wide replication events [16]. Therefore, we used the Ks value to estimate the evolution date of gene duplication events in the PtbHLH gene family (Table S5). The Ks value means that most of the PtbHLH genes were duplicated around the recent WGD event while some other genes originated from ancient WGD. The intensity and direction of selection can be represented by the ratio of Ka/Ks [44]. The Ka/Ks ratio of all homologous PtbHLH genes is less than 1, indicating that the PtbHLH gene family has undergone strong purification selection during the evolution process. According to the PtbHLH gene collinearity relationship between poplar and Arabidopsis or rice, we can easily know that there are a large number of homologous bHLH genes among them ( Figure S1).

Phylogenetic Analyses of the PtbHLH in Poplar
A reasonable phylogenetic tree can reveal and help us better understand the evolutionary relationship of gene families. We firstly used 233 bHLH protein sequences of poplar to construct an original phylogenetic tree, whose branches were constructed according to its gene structure and motif distribution (Figures 3 and 4).

Phylogenetic Analyses of the PtbHLH in Poplar
A reasonable phylogenetic tree can reveal and help us better understand the evolutionary relationship of gene families. We firstly used 233 bHLH protein sequences of poplar to construct an original phylogenetic tree, whose branches were constructed according to its gene structure and motif distribution (Figures 3 and 4).  Phylogenetic tree based on protein sequences of bHLH gene family in poplar. The phylogenetic tree was constructed using the neighbor-joining (NJ) method, with 1000 bootstrap replicates, using Mega-X. Figure 3. Phylogenetic tree based on protein sequences of bHLH gene family in poplar. The phylogenetic tree was constructed using the neighbor-joining (NJ) method, with 1000 bootstrap replicates, using Mega-X.  From the phylogenetic tree (Figure 3), PtbHLHs can be well divided into 21 groups, named from group I to group XXI. It appears that group I has the maximal number of genes (26) followed by group XII (19). Group XIX, group XX, and group XXI have the minimal number of genes (4). As for groups III-XVIII, there are 17,17,16,15,15,12,11,11,10,10,8,7,7,7,7, and 6 PtbHLH members, respectively. Fortunately, one subgroup FBH subfamily was classified together all the time, which is exactly what we focused on. As a consequence, we drew the conclusion that there are 10 members in PtFBH subfamily.
Finally, we constructed a phylogenetic tree using 16 FBH proteins ( Figure 5), of which six were the previously reported in Arabidopsis [12]. As is shown in Figure 5, PtbHLH94 and PtbHLH206 are closest to FBH4 (AT2G42280) followed by PtbHLH12 and PtbHLH131, PtbHLH1, and PtbHLH57 are closest to FBH3 (AT1G51140), and PtbHLH173 and PtbHLH227 are closest to FBH2 (AT4G09180) and FBH1 (AT1G35460). From the phylogenetic tree (Figure 3), PtbHLHs can be well divided into 21 groups, named from group I to group XXI. It appears that group I has the maximal number of genes (26) followed by group XII (19). Group XIX, group XX, and group XXI have the minimal number of genes (4). As for groups III-XVIII, there are 17,17,16,15,15,12,11,11,10,10,8,7,7,7,7, and 6 PtbHLH members, respectively. Fortunately, one subgroup FBH subfamily was classified together all the time, which is exactly what we focused on. As a consequence, we drew the conclusion that there are 10 members in PtFBH subfamily.

Conserved Motif and Gene Structure Analysis of PtbHLHs
In general, genes in the same group share similar gene and protein structures. We found that the number of exons per PtbHLH gene varies from 1 to 11 (Figure 4). PtbHLH genes in the same subgroup usually have similar exons and introns: genes in group VII and group XI have 1 exon with no intron, group XXI 2 exons, group VIII 3 exons, group XX 4 exons, group I and group XVII 6-8 exons, group XVI 10-11 exons, and so on. Then we analyzed the conserved motifs of the PtbHLH proteins using the MEME website. In total, 15 conserved motifs were identified and designated as motifs 1-15 ( Figure 4, Table  S6). The motif 1 and motif 2 should be the specific motifs of the PtbHLH family because

Conserved Motif and Gene Structure Analysis of PtbHLHs
In general, genes in the same group share similar gene and protein structures. We found that the number of exons per PtbHLH gene varies from 1 to 11 (Figure 4). PtbHLH genes in the same subgroup usually have similar exons and introns: genes in group VII and group XI have 1 exon with no intron, group XXI 2 exons, group VIII 3 exons, group XX 4 exons, group I and group XVII 6-8 exons, group XVI 10-11 exons, and so on. Then we analyzed the conserved motifs of the PtbHLH proteins using the MEME website. In total, 15 conserved motifs were identified and designated as motifs 1-15 ( Figure 4, Table S6). The motif 1 and motif 2 should be the specific motifs of the PtbHLH family because the majority of the protein subgroups contain both while group XIX, group XV, group IX, and group XX do not. Similar to gene structure, PtbHLH proteins in the same subgroup share group-specific conserved motifs. As is in Figure 3, motif 3 is only present in all members of group I, group XVII, group XII, and group XVIII and one of the members in group IV; motif 5 is only present in all members of group XVII, group II, and one of the members in group XIII; motif 6 is only observed in group XVI, group X, group XIX, and motif 8 in group XVI, group X, group XI; motif 11 is unique to group I, motif 12 in group IX, motif 13 in group XV, and motif 14 in group IX.

Cis-Elements Analysis in Promoters of the PtbHLH Genes
From the visualized cis-elements in Table S7 and Figure S2, all members of the PtbHLH gene family contain a large number of light responsive elements, almost every member contains hormone responsive elements, and most members also contain stress responsive elements. The cis-elements in the promoter region may reflect the potential regulation and function of genes. These results suggest that the PtbHLH gene family may play an important role in regulation of growth and development.

Multiple Alignment Analysis of the FBH Subfamily
Sequence alignment of 16 proteins from the FBH subfamily in Arabidopsis and poplar is shown in the Figure 5 and Figure S3, where only amino acid sites with greater than 30% identity are displayed in color. At the same time, we obtained the LOGO of the conserved HLH domain of poplar and Arabidopsis FBH ( Figure 5), where we found the bHLH domain here containing 54 conserved amino acid sequences. The structure of the bHLH conservative domain is distributed as follows: 17 bp basic amino acid region at the N-terminal, 15 and 19 bp alpha helixes and separated by a 3 bp loop at the C-terminal. Generally, the loop region is not conservative and its length is variable. We proved the loop in FBHs is a conservative region with 3 bp amino acids when not in bHLHs normally. We also can get the consensus of the HLH domain: . We concluded that HLH domain sequences were highly conserved between poplar and Arabidopsis FBHs.

Expression Profile Analysis of PtbHLH Genes in Different Tissues
To better characterize tissue-specific gene expression patterns of the PtbHLH genes, we compared their transcriptional levels in various tissues by using publicly available RNA-Seq data [37]. Values in a total of five sample tissues were selected from Flowers-Expanded, Roots-Control, Seeds-Mature, Leaves-Control, and Cambium-Phloem-Dormant (6 unmatched, 13 no expressed). Based on the Log2fold change (FC) of each PtbHLH gene, we generated a hierarchical clustering analysis and heatmap to characterize the PtbHLH family expression profiles ( Figure S4).
Thirteen genes were not significantly upregulated or downregulated in all analyzed tissues, which may be due to differences in spatial and temporal expression patterns. From Figure S4, we know the number of differentially expressed genes (DEGs) in tissues: 43 genes were found significantly upregulated and 16 downregulated in Flowers-Expanded tissue; 63 genes upregulated and 16 downregulated in Roots-Control; 48 genes upregulated and 18 downregulated in Seeds-Mature; 30 genes upregulated and 18 downregulated in Leaves-Control; 47 genes upregulated and 66 downregulated in Cambium-Phloem-Dormant. A total of 52 genes were significantly upregulated in just one tissue, suggesting their function in a specific tissue. In addition, 14 genes exhibiting a significant upregulation in three or more of all the analyzed tissues may play key roles in this tissue development and growth of poplar. Besides, four of our PtbHLHs are in the list of the most tissue-representative genes generated by previous analysis [37]: both PtbHLH2 and PtbHLH130 in wood, PtbHLH134 in buds, and PtbHLH16 in seeds. Exploring the expression pattern of PtbHLH gene in different tissues provides the basis for the identification of functional genes in poplar.

The PtbHLH Protein-Protein Interaction Network Prediction and Some May Be Involved in the Flowering Process
Two heatmaps were then generated for analyzing their expression patterns in three states of the flowering process: Flowers-Dormant, Flowers-Expanding, and Flowers-Expanded ( Figure 6). One heatmap is based on the log2FC values of 43 significantly upregulated genes in Flowers-Expanded together with PtCO1 (Potri.017G107500), PtCO2 (Potri.004G108300), and PtFT (Potri.008G077700) ( Figure 6A). The other heatmap is based on 10 PtFBHs together with PtCO1, PtCO2, and PtFT ( Figure 6B). The qRT-PCRs were performed to confirm the results showing similar trends in the DEGs analysis ( Figure 6C). functional genes in poplar.

The PtbHLH Protein-Protein Interaction Network Prediction and Some May Be Involved in the Flowering Process
Two heatmaps were then generated for analyzing their expression patterns in three states of the flowering process: Flowers-Dormant, Flowers-Expanding, and Flowers-Expanded ( Figure 6). One heatmap is based on the log2FC values of 43 significantly upregulated genes in Flowers-Expanded together with PtCO1 (Potri.017G107500), PtCO2 (Potri.004G108300), and PtFT (Potri.008G077700) ( Figure 6A). The other heatmap is based on 10 PtFBHs together with PtCO1, PtCO2, and PtFT ( Figure 6B). The qRT-PCRs were performed to confirm the results showing similar trends in the DEGs analysis ( Figure 6C).  Statistical evidence apparently indicated that 18 members (PtbHLH73, PtbHLH187, PtbHLH216, PtbHLH93, PtbHLH142, PtbHLH101, PtbHLH18, PtbHLH104, PtbHLH100, PtbHLH161, PtbHLH137, PtbHLH116, PtbHLH165, PtbHLH181, PtbHLH13, PtbHLH114, PtbHLH164, and PtbHLH180) may play a role in the whole process of poplar flowering due to their upregulation in all three states just as PtFT ( Figure 6). Those members were suspected to be involved in flower development, so we performed the protein-protein interaction network prediction on them. The PtbHLH proteins were predicted to interact with each other, which is in line with previous reports that bHLH proteins activate or repress target genes by forming homodimers or heterodimers [45]. Useful information is as follows: PtbHLH165 is related to PtbHLH127 and PtbHLH150; PtbHLH181 is related to PtbHLH14; PtbHLH61 is related to PtbHLH160 and PtbHLH28 when PtbHLH160 and PtbHLH28 co-express and interact with each other; PtbHLH73 interacts with PtbHLH36 experimentally ( Figure S5).
PtbHLH131, PtbHLH57, and PtbHLH173 were significantly downregulated in the three states of flowering. With information obtained from the STRING database considered together, PtbHLH57 in co-expression with BTB/POZ domain-containing family protein was thought to be involved with flowering development probably ( Figure 6C). The PtCO1 and PtCO2 did not show any increase but even decreased. Maybe we can attribute this to a certain genetic diurnal fluctuation expression pattern or different spatial and temporal expression. In conditions of low nitrogen, early flowering phenotype happened because of the phosphorylation state of FLOWERING BHLH 4 in Arabidopsis [46], then we speculated that some of the PtFBHs exert flowering-related effects in certain conditions.

Discussion
We obtained 233 bHLH genes in the newest version v4.1 of genomic sequences in P. trichocarpa based on the analysis of the conserved domain. Widely distributed on 19 chromosomes of poplar (Figure 1), they were named PtbHLH1-PtbHLH233 according to their accession number on the poplar genome (Table S1). The majority of PtbHLH proteins (89%) were predicted to function in the nucleus (Table S3). Gene duplication and synteny analysis indicated that PtbHLH gene family has undergone strong purification selection during the evolution process ( Figure 2 and Figure S4). Based on the gene structure and protein motifs of 233 PtbHLHs, we classified them into 21 groups (Figure 3). Genes in the same group share similar gene and protein structure (Figure 4). Cis-elements analysis of the PtbHLH genes suggest that the PtbHLH gene family may play an important role in regulation growth and development (Table S7 and Figure S2).
Based on publicly available RNA-Seq data [37], we obtained differentially expressed genes (DEGs) in tissues selected from Flowers-Expanded, Roots-Control, Seeds-Mature, Leaves-Control, and Cambium-Phloem-Dormant ( Figure S4). Heterologous expression of poplar PtFBH1 (PtbHLH173) could lead to early flowering in Arabidopsis under both short and long day conditions [12], and histone demethylase JMJ28, a CO activator removes H3K9me2 from the CO locus through interacting with FBH in Arabidopsis [15]. Therefore, we identified 10 members of the PtFBH subfamily in PtbHLHs additionally and the results were supported by both phylogenetic trees ( Figure 5). We then explored their expression patterns in three states of the flowering process (Flowers-Dormant, Flowers-Expanding, and Flowers-Expanded) together with PtCO1, PtCO2, and PtFT ( Figure 6). Quite a few bHLH genes were found to be involved in flower development in Chinese jujube [47]. In total, 18 members (PtbHLH73, PtbHLH187, PtbHLH216, PtbHLH93, PtbHLH142, PtbHLH101, PtbHLH18, PtbHLH104, PtbHLH100, PtbHLH161, PtbHLH137, PtbHLH116, PtbHLH165, PtbHLH181, PtbHLH13, PtbHLH114, PtbHLH164, and PtbHLH180) may play a role in the whole process of poplar flowering due to their upregulation in all three states just as PtFT.
We then explored other possible characteristics of PtFBHs through protein-protein interaction analysis by STRING database. In tomato, TMF protein interacts with BTB/POZ domain-containing SlBOPs, homologs of BOP transcriptional cofactors in Arabidopsis, to promote inflorescence complexity. Elimination of SlBOP function renders tomato inflorescences into single flowers [48]. The predictable co-expression between PtbHLH57 and BTB/POZ domain-containing family protein has indicated that PtbHLH57 may be involved in inflorescence complexity regulation. PtbHLH57 is homologous to AtFBH3 (AT1G51140) ( Figure 5). Overexpression of FBH3 in Arabidopsis has caused early flowering by upregulating CO expression no matter which photoperiod [12]. Since we proved FBHs to be conserved between Arabidopsis and poplar here, they are likely to function similarly in flower development.
Many PtbHLHs work with other members within the family, such as PtbHLH165 with PtbHLH127 and PtbHLH150, PtbHLH181 with PtbHLH14, and PtbHLH61 with PtbHLH160 and PtbHLH28 when PtbHLH160 and PtbHLH28 were in co-expression and interacting ( Figure S5). Maybe PtbHLH61, PtbHLH160, and PtbHLH28 function together for a certain biological pathway. The same speculation is on PtbHLH73 and PtbHLH36 for they were proved to interact experimentally. Based on our study, further research is needed to explain the detailed function of PtbHLHs and PtFBHs in P. trichocarpa.

Conclusions
In our study, 233 bHLH transcription factors of P. trichocarpa were subjected to identification, classification, phylogenetic analysis, characterization of conserved motif and gene structure, subcellular localization analysis, gene duplication and synteny analysis, and exploration on cis-elements. Then, we also analyzed the expression patterns of 233 PtbHLH genes in different tissues and protein interaction network. This analysis of bHLHs in poplar also reveals that some PtbHLH and PtFBH members may be involved in the flowering process. All in all, our comprehensive and systematic analysis can provide some valuable clues and basic reference toward further investigations on physiological and molecular functions of PtbHLHs.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/f12111439/s1. Table S1. The information of bHLH gene family in poplar; Table S2. Physicochemical properties of bHLH proteins in poplar; Table S3. Subcellular localization analysis of bHLH proteins in poplar; Table S4. Duplication type of the PtbHLH gene; Table S5. PtbHLH evolutionary selection pressure prediction; Table S6. Sequence information of 15 identified motifs in MEME analysis; Table S7. Cis-elements in members of the PtbHLH gene; Table S8. The primers used for qRT-PCR; Figure S1. PtbHLH gene collinearity relationship between poplar and Arabidopsis or rice; Figure S2. The cis-acting elements analysis of putative promoter of the bHLH gene family in poplar; Figure S3. Multi-sequence alignment and domain analysis of the FBHs in poplar; Figure S4. The expression pattern analysis of PtbHLHs genes in poplar; Figure S5. Protein interaction network of PtFBHs proteins in poplar.
Author Contributions: Y.Y., H.X., X.G. and J.M. performed the experiments. L.L. and Y.Y. analyzed the data, prepared figures and tables, wrote the paper and reviewed drafts of the paper. All authors have read and agreed to the published version of the manuscript.