Genome-Wide Identiﬁcation and Expression Analysis of Dendrocalamus farinosus CCoAOMT Gene Family and the Role of DfCCoAOMT14 Involved in Lignin Synthesis

: As the main component of plant cell walls, lignin can not only provide mechanical strength and physical defense for plants, but can also be an important indicator affecting the properties and quality of wood and bamboo. Dendrocalamus farinosus is an important economic bamboo species for both shoots and timber in southwest China, with the advantages of fast growth, high yield and slender ﬁber. Caffeoyl-coenzyme A-O-methyltransferase (CCoAOMT) is a key rate-limiting enzyme in the lignin biosynthesis pathway, but little is known about it in D. farinosus. Here, a total of 17 DfCCoAOMT genes were identiﬁed based on the D. farinosus whole genome. DfCCoAOMT1/14/15/16 were homologs of AtCCoAOMT1 . DfCCoAOMT6/9/14/15/16 were highly expressed in stems of D. farinosus ; this is consistent with the trend of lignin accumulation during bamboo shoot elongation, especially DfCCoAOMT14 . The analysis of promoter cis-acting elements suggested that DfCCoAOMTs might be important for photosynthesis, ABA/MeJA responses, drought stress and lignin synthesis. We then conﬁrmed that the expression levels of DfCCoAOMT2/5/6/8/9/14/15 were regulated by ABA/MeJA signaling. In addition, overexpression of DfCCoAOMT14 in transgenic plants signiﬁcantly increased the lignin content, xylem thickness and drought resistance of plants. Our ﬁndings revealed that DfCCoAOMT14 can be a candidate gene that is involved in the drought response and lignin synthesis pathway in plants, which could contribute to the genetic improvement of many important traits in D. farinosus and other species.


Introduction
Bamboo forests, with about 1500 species worldwide, cover approximately 31.5 million hectares in tropical and subtropical regions of the world [1]. They help address the issue of global warming by fixing carbon dioxide in the atmosphere [2,3]. Moreover, bamboo resembles wood in appearance, weight, hardness, texture and strength, making bamboo an ideal substitute for many traditional wood products. D. farinosus is an important sympodial bamboo species for both bamboo shoots and materials; it is mainly distributed in southwest China, with the advantages of high fiber content, a large biomass, cold tolerance, rapid growth and high yield, which are also important for paper making and ecological greening [4].
Lignin is a complex polyphasic aromatic polymer that is not only a key product of the phenylpropane pathway, but also an important component of plant cell walls, playing an important role in enhancing plant mechanical strength and contributing to the defense against pests and diseases [5]. However, lignin is not conducive to paper

Genome-Wide Identification of CCoAOMT Genes in D. farinosus
Based on the reported protein sequence of the Arabidopsis CCoAOMT gene family and the methyltransferase structural domain from Pfam (PF01596), a total of 17 CCoAOMT genes were identified based on the D. farinosus genome and named as DfCCoAOMT1 to DfCCoAOMT17 (Table 1). Detailed information about these DfCCOAOMT gene sequences is presented in Supplemental Table S1 and Supplementary Document S1. These lengths of DfCCoAOMT proteins ranged from 220 aa (DfCCoAOMT13) to 311 aa (DfC-CoAOMT17), the molecular weight ranged from 24.30 KDa (DfCCoAOMT13) to 34.00 KDa (DfCCoAOMT17) and the isoelectric point (pI) ranged from 4.88 (DfCCoAOM6) to 9.44 (DfC-CoAOMT3). A subcellular localization prediction indicated that all CCoAOMT proteins were localized in the cytoplasm and chloroplasts.

Phylogenetic Relationship, Gene Structure and Conserved Motif Analysis of the CCoAOMT Gene Family in D. farinosus
To clarify the evolutionary relationships and classification of the CCoAOMT gene family in D. farinosus, a phylogenetic tree was constructed using the full-length protein sequences of CCoAOMT from D. farinosus, Arabidopsis, rice and Populus with IQ-TREE v1.6.12 and the Model Finder algorithm ( Figure 1A). The results showed that the DfC-CoAOMT genes were divided into four groups (Classes I~IV); Classes I, II and IV contained 4, 10 and 3 DfCCoAOMT genes, respectively, and Class III had no DfCCoAOMT genes. However, in Arabidopsis, AtCCoAOMT genes were mainly clustered in Class III, only two genes (AtCCoAOMT3 and AtCCoAOMT4) were clustered in Class IV and one (AtCCoAOMT1) was clustered in Class I. DfCCoAOMT1, DfCCoAOMT14, DfCCoAOMT15 and DfCCoAOMT16 were homologs of AtCCoAOMT1, OsCCoAOMT6, PtCCoAOMT1 and PtCCoAOMT4, respectively. DfCCoAOMT6 and DfCCoAOMT9 were homologs of OsCCoAOMT1 and OsC-CoAOMT3, respectively, suggesting that they might have a similar function.
Next, the gene structure and conserved motif composition of DfCCoAOMT were analyzed ( Figure 1B). The MEME online website was used to investigate the conservation and diversity of motifs in different DfCCoAOMT proteins. The results showed that all the proteins were found to be related to O-methyltransferase; motifs 2, 3, and 6 were common motifs in all proteins. Motif 1 was present in 16 DfCCoAOMT proteins except DfCCoAOMT13, and motif 9 appeared in 15 DfCCoAOMT proteins apart from DfCCoAOMT4/8. Motif 9 was only observed in Class IV, which might be associated with the functional diversity of DfCCoAOMT genes. Then, the exon-intron gene structure of DfCCoAOMTs were analyzed. The exon number of DfCCoAOMTs ranged from 3 to 9 and the intron number ranged from 2 to 8. Six genes (DfCCoAOMT6/7/8/10/11/13) contained four exons and three introns, and DfCCoAOMT1/5/9/12/14/15 contained three exons and two introns; this difference in gene structure may be an evolutionary genetic difference. Meanwhile, the members of the same group had similar exon-intron structures and motif compositions. Next, the gene structure and conserved motif composition of DfCCoAOMT were analyzed ( Figure 1B). The MEME online website was used to investigate the conservation and diversity of motifs in different DfCCoAOMT proteins. The results showed that all the proteins were found to be related to O-methyltransferase; motifs 2, 3, and 6 were common

Chromosome Distribution and Collinearity Analysis of DfCCoAOMT Genes
Gene shrinkage and expansion are important factors for plant gene functional diversity and environmental adaptation. Tandem repeats and fragment repeats are the main causes of gene expansion in plants. Therefore, the chromosome distribution and collinearity of the DfCCoAOMT gene family were analyzed (Figure 2 and Supplemental Figure S1). In D. farinosus, these DfCCoAOMT genes were located on chromosome groups 3, 6, 7 Int. J. Mol. Sci. 2023, 24, 8965 5 of 18 and 11, respectively; the DfCCoAOMT genes were most abundant on chromosome group 7 (10/17). Furthermore, there was an obvious tandem duplication in the seventh chromosome group, DfCCoAOMT5/6/7/8 formed a tandem repeat gene cluster on chromosome A07 and DfCCoAOMT9/10/11/12 formed a tandem repeat gene cluster on chromosome B07. Four pairs of putative paralogous DfCCoAOMT genes were identified and located on different chromosomes; these results suggested that there were fragment duplication events in the DfCCoAOMT gene family and led to expansion of the family members.

Tissue Expression Profile of DfCCoAOMTs and Analysis of Expression at Different Developmental Periods in D. farinosus
To better understand the function of the DfCCoAOMT genes in D. farinosus, we performed a transcriptome sequencing analysis of the expression levels of various tissues and different developmental periods in D. farinosus (Supplemental Table S3). As shown in Fig

Tissue Expression Profile of DfCCoAOMTs and Analysis of Expression at Different Developmental Periods in D. farinosus
To better understand the function of the DfCCoAOMT genes in D. farinosus, we performed a transcriptome sequencing analysis of the expression levels of various tissues and different developmental periods in D. farinosus (Supplemental Table S3). As shown in Figure 3A, DfCCoAOMT gene expressions were different among the Young leaf (Yleaf), Mature leaf (MLeaf), Lateral bud (Labud), Sheath of bamboo shoot (ShB), Node, Rhizome bud (Rhbud), Rhizome neck (Rhne), Root, Branch, Internode (Inod) and Culm sheath (CSh). Among these 17 genes, DfCCoAOMT10/11/12 were specifically highly expressed in the root, implying that they mainly functioned during root development. Eight genes (DfC-CoAOMT1/2/3/4/7/8/13/17) showed a low expression in all tissues. DfCCoAOMT6/9/14/15/16 showed a high expression level in various tissues, especially in the Inod, Rhne, root and node. These results suggested that DfCCoAOMT genes function in organ development and a variety of biological processes in D. farinosus. Since lignin deposition and secondary wall thickening usually accompany rapid growth of bamboo, we then analyzed the transcript levels of bamboo shoots during different growth periods ( Figure 3B). Only DfCCoAOMT6/9/14/15/16 showed a high expression in fast-growing culm, indicating that these five genes might be involved in lignin synthesis during the rapid growth period of bamboo culms.

Prediction of Cis-Component Analysis of DfCCoAOMTs
To explore the mechanism of DfCCoAOMT genes in stress response and development, the cis-regulatory elements in the 2 kb upstream sequence of DfCCoAOMT genes were predicted by PlantCARE and PLACE ( Figure 4A,B). In the DfCCoAOMT genes' promoter, a variety of stress-, growth-and development-related cis-acting elements were identified, such as the low temperature response element (LTR), gibberellin (GA) response element, drought response element (MBS), light response element (G-box, Box4, MRE, Ibox, ACE and LAMP element), MeJA response element (CGTCA-motif and TGACG-motif), salicylic acid (SA) response element and abscisic acid (ABA) response element (ABRE). Light-and growth-related cis-acting elements were present in almost all DfCCoAOMT promoter sequences. MeJA and ABA response elements were identified in the promoter of CCoAOMT2/5/6/8/9/14/15. Various lignin synthesis-related elements, such as MYB elements, were identified in the promoters of CCoAOMT1/2/4/5/9/10/11/14/15/16/17, especially CCoAOMT14. Since lignin deposition and secondary wall thickening usually accompany rapid growth of bamboo, we then analyzed the transcript levels of bamboo shoots during different growth periods ( Figure 3B). Only DfCCoAOMT6/9/14/15/16 showed a high expression in fast-growing culm, indicating that these five genes might be involved in lignin synthesis during the rapid growth period of bamboo culms.

Prediction of Cis-Component Analysis of DfCCoAOMTs
To explore the mechanism of DfCCoAOMT genes in stress response and development, the cis-regulatory elements in the 2 kb upstream sequence of DfCCoAOMT genes were predicted by PlantCARE and PLACE ( Figure 4A,B). In the DfCCoAOMT genes' promoter, a variety of stress-, growth-and development-related cis-acting elements were identified, such as the low temperature response element (LTR), gibberellin (GA) response element, drought response element (MBS), light response element (G-box, Box4, MRE, I-box, ACE and LAMP element), MeJA response element (CGTCA-motif and TGACG-motif), salicylic acid (SA) response element and abscisic acid (ABA) response element (ABRE). Light-and growth-related cis-acting elements were present in almost all DfCCoAOMT promoter sequences. MeJA and ABA response elements were identified in the promoter of CCoAOMT2/5/6/8/9/14/15. Various lignin synthesis-related elements, such as MYB ele- We then examined the expression pattern of DfCCoAOMT2/5/6/8/9/14/15 after ABA and MeJA treatments using qRT-PCR ( Figure 4C). The results showed that the expression levels of DfCCoAOMT6/8/9/14 decreased rapidly after ABA and MeJA treatments, reaching the lowest levels at 3 h and 6 h, respectively, and then gradually bound. In contrast, the expression levels of DfCCoAOMT2/5 were significantly increased in response to ABA after 3 h, reaching the highest levels at 6 h. These results suggested that these DfCCoAOMT genes might be involved not only in lignin synthesis, but also in plant biotic and abiotic stress responses.
In addition, due to the differences in lignification between different internodes of bamboo shoots, the microscopy of individual stem sections of 2 m tall bamboo shoots was performed. Phloroglucinol-Cl staining showed a gradual decrease in internode lignin deposition from base to tip in bamboo shoots and measurements of lignin content in different internodes showed consistent results ( Figure 5A,B). Furthermore, the expression levels of DfCCoAOMT genes in different internodes were detected and showed that We then examined the expression pattern of DfCCoAOMT2/5/6/8/9/14/15 after ABA and MeJA treatments using qRT-PCR ( Figure 4C). The results showed that the expression levels of DfCCoAOMT6/8/9/14 decreased rapidly after ABA and MeJA treatments, reaching the lowest levels at 3 h and 6 h, respectively, and then gradually bound. In contrast, the expression levels of DfCCoAOMT2/5 were significantly increased in response to ABA after 3 h, reaching the highest levels at 6 h. These results suggested that these DfCCoAOMT genes might be involved not only in lignin synthesis, but also in plant biotic and abiotic stress responses.
In addition, due to the differences in lignification between different internodes of bamboo shoots, the microscopy of individual stem sections of 2 m tall bamboo shoots was performed. Phloroglucinol-Cl staining showed a gradual decrease in internode lignin deposition from base to tip in bamboo shoots and measurements of lignin content in different internodes showed consistent results ( Figure 5A,B). Furthermore, the expression levels of DfCCoAOMT genes in different internodes were detected and showed that DfC-CoAOMT5/6/9/14 had high expression levels in the first to third internode, consistent with the trend of lignin deposition. In contrast, DfCCoAOMT15/16 were highly expressed in all internodes, suggesting that they are also involved in other biological functions of bamboo shoots ( Figure 5C). DfCCoAOMT5/6/9/14 had high expression levels in the first to third internode, consistent with the trend of lignin deposition. In contrast, DfCCoAOMT15/16 were highly expressed in all internodes, suggesting that they are also involved in other biological functions of bamboo shoots ( Figure 5C).

Overexpression of DfCCoAOMT14 Improves Lignin Biosynthesis in Transgenic Tobacco
To determine the role of DfCCoAOMTs in lignin biosynthesis, DfCCoAOMT14 was cloned and CaMV 35S-driven plant expression vectors were constructed and then transformed into tobacco. The expression of DfCCOAOMT14 in transgenic plants and wild type was detected using qRT-PCR and we found that the expression levels of line 1, 2 and 3 were significantly higher than in wild type ( Figure 6C). The phenotype of transgenic plants showed growth retardation compared with wild type ( Figure 6A). In addition, the third internode of the 40-day-old transgenic plants was used to observe the anatomical cross-section; the results showed that overexpression of DfCCOAOMT14 significantly increased xylem lignification in transgenic plants compared with wild type; the xylem widths in lines 1, 2 and 3 increased by approximately 46.5%, 9.2% and 12.3% compared with wild type, respectively ( Figure 6B,D). Determination of lignin content in stems from transgenic plants also showed a significant increase ( Figure 6E). Then, the expression levels of several lignin biosynthesis-related genes were detected using qRT-PCR and it was found that the expression levels of four genes (PAL, C4H, CAD and COMT) in transgenic lines were significantly higher than those of wild type ( Figure 6F). These results suggested The values represent the mean ± SD, and different letters indicate significant differences according to Duncan's multiple range test and least significance difference (LSD) (p < 0.05).

Overexpression of DfCCoAOMT14 Improves Lignin Biosynthesis in Transgenic Tobacco
To determine the role of DfCCoAOMTs in lignin biosynthesis, DfCCoAOMT14 was cloned and CaMV 35S-driven plant expression vectors were constructed and then transformed into tobacco. The expression of DfCCOAOMT14 in transgenic plants and wild type was detected using qRT-PCR and we found that the expression levels of line 1, 2 and 3 were significantly higher than in wild type ( Figure 6C). The phenotype of transgenic plants showed growth retardation compared with wild type ( Figure 6A). In addition, the third internode of the 40-day-old transgenic plants was used to observe the anatomical cross-section; the results showed that overexpression of DfCCOAOMT14 significantly increased xylem lignification in transgenic plants compared with wild type; the xylem widths in lines 1, 2 and 3 increased by approximately 46.5%, 9.2% and 12.3% compared with wild type, respectively ( Figure 6B,D). Determination of lignin content in stems from transgenic plants also showed a significant increase ( Figure 6E). Then, the expression levels of several lignin biosynthesis-related genes were detected using qRT-PCR and it was found that the expression levels of four genes (PAL, C4H, CAD and COMT) in transgenic lines were significantly higher than those of wild type ( Figure 6F). These results suggested that the elevated expression levels of DfCCoAOMT14 could enhance lignin biosynthesis in plants.
that the elevated expression levels of DfCCoAOMT14 could enhance lignin biosynthesis in plants.

Overexpression of DfCCoOMT14 Improved Drought Resistance of Transgenic Plants
Previous studies have shown that lignin synthesis significantly affects the plant responses to drought [16]. We performed a drought treatment on transgenic plants for 37 days and found that the height and growth of transgenic plants performed better than wild type ( Figure 7A). Moreover, we measured the drought-related physiological indicators in transgenic plants and wild type after the drought treatment; the levels of malondialdehyde (MDA), superoxide dismutase (SOD), peroxidase (POD), proline (Pro) and relative water content (RWC) in transgenic plants were significantly higher than in wild type ( Figure 7C-G). qRT-PCR showed that the expression levels of the drought-related genes DREB and RD29A were also significantly higher in transgenic plants

Overexpression of DfCCoOMT14 Improved Drought Resistance of Transgenic Plants
Previous studies have shown that lignin synthesis significantly affects the plant responses to drought [16]. We performed a drought treatment on transgenic plants for 37 days and found that the height and growth of transgenic plants performed better than wild type ( Figure 7A). Moreover, we measured the drought-related physiological indicators in transgenic plants and wild type after the drought treatment; the levels of malondialdehyde (MDA), superoxide dismutase (SOD), peroxidase (POD), proline (Pro) and relative water content (RWC) in transgenic plants were significantly higher than in wild type ( Figure 7C-G). qRT-PCR showed that the expression levels of the drought-related genes DREB and RD29A were also significantly higher in transgenic plants compared with wild type ( Figure 7B). These results suggested that DfCCoAOMT14 improved the drought resistance of plants.
compared with wild type ( Figure 7B). These results suggested that DfCCoAOMT14 improved the drought resistance of plants. The asterisks indicate a significant difference compared to vector-transformed plants using the one-way analysis of variance (* p < 0.05, ** p < 0.01).

Discussion
As a cross-linked phenolic polymer, lignin is not only a major component of plant cell walls, determining the degree of lignification, but is also involved in the plant's ability to resist injury in response to abiotic and biotic stresses [21]. Lignin biosynthesis is a complex and delicate process; lignin monomers are synthesized from phenylalanine in the cytoplasm through a series of enzymatic reactions, and then transported across the membrane to participate in the synthesis of cell walls. As an important key enzyme for lignin synthesis, CCoAOMT catalyzes the methylation of caffeoyl-coenzyme A (caffeoyl-CoA) to feruloyl-coenzyme A (feruloyl-CoA), which mediates the biosynthesis of G-type lignin monomers to S-type lignin monomers, affecting the ratio of S/G lignin components [13]. We explored the number and function of CCoAOMT genes involved in lignin synthesis, The asterisks indicate a significant difference compared to vector-transformed plants using the one-way analysis of variance (* p < 0.05, ** p < 0.01).

Discussion
As a cross-linked phenolic polymer, lignin is not only a major component of plant cell walls, determining the degree of lignification, but is also involved in the plant's ability to resist injury in response to abiotic and biotic stresses [21]. Lignin biosynthesis is a complex and delicate process; lignin monomers are synthesized from phenylalanine in the cytoplasm through a series of enzymatic reactions, and then transported across the membrane to participate in the synthesis of cell walls. As an important key enzyme for lignin synthesis, CCoAOMT catalyzes the methylation of caffeoyl-coenzyme A (caffeoyl-CoA) to feruloylcoenzyme A (feruloyl-CoA), which mediates the biosynthesis of G-type lignin monomers to S-type lignin monomers, affecting the ratio of S/G lignin components [13]. We explored the number and function of CCoAOMT genes involved in lignin synthesis, which play important roles in improving fiber quality and lignocellulosic biomass utilization. The CCoAOMT gene family involved in the lignin biosynthesis pathway has been identified and analyzed in several species, including Arabidopsis, rice, tea tree, tobacco and poplar [22][23][24]. In this study, 17 DfCCoAOMT genes were identified based on the D. farinosus whole genome data, which is much more than the number of genes in Arabidopsis (7), poplar (6) and rice (6) (Table 1 and Figure 1A). According to the phylogenetic analysis of these CCoAOMT protein sequences from D. farinosus and other species, these CCoAOMTs could be clustered into four groups and no gene was present in Class III; DfCCoAOMT1/14/15/16 clustered with Arabidopsis CCoAOMT1 in Class I, 10 genes clustered in Class II and 3 genes clustered in Class IV. Proteins clustered in the same group usually exhibit similar functions; AtCCoAOMT1 has been shown to be the true CCoAOMT1 gene and is involved in lignin biosynthesis, suggesting that DfCCoAOMT1/14/15/16 in Class I may also be primarily involved in this biological function [24].
In addition, the gene structure determines protein conformation and the protein structure determines function; genes with similar protein structures usually have similar functions ( Figure 1B). According to our analysis, all DfCCoAOMTs have structural motif 2/3/6, which might be the key structure to maintain the functional conservation of this gene family. While motif 1 of the N-terminal region was also present in all the remaining DfCCoAOMTs except DfCCoAOMT13, since the protein sequences of the Nterminal and C-terminal regions are essential for enzyme activity and substrate binding, the deletion of motif 1 might lead to the specificity of the function of DfCCoAOMT13. DfCCoAOMT1/14/15/16 had the same protein structure, and the genes' exon-intron distribution was also consistent except for DfCCoAOMT16. DfCCoAOMT3/4/17 had a specific structural motif 17, which might lead to a functional diversification of DfCCoAOMT genes. Gene duplication was considered to be a major factor leading to gene family expansion and gene functional diversity. Based on the chromosomal localization analysis of the D. farinosus 17 CCoAOMT genes, 9 DfCCoAOMT genes were found to be densely distributed on chromosome 7, showing local tandem and proximal gene duplication; this is similar to that in Arabidopsis [25] and Populus [26], which may be due to gene evolution. The other remaining DfCCoAOMT genes were scattered on chromosomes 3, 6 and 11 ( Figure 2). In poplar, CCoAOMTs were mainly evenly distributed on chromosomes 1, 8, 9 and 10 [17]. Compared with Arabidopsis, rice and poplar, the DfCCoAOMT gene family was significantly expanded, which might lead to more secondary metabolites and enhance the adaptability of D. farinosus to the environment. Of course, it requires more research to prove.
Cis-regulatory elements in gene promoter regions are critical for regulation of gene expression patterns and transcriptional expression [27]. Different cis-acting elements usually represent that genes are involved in different biological processes. Previous studies have found that AC elements are commonly found in the promoter regions of lignin biosynthetic enzyme genes, such as CCoAOMT, PAL, C4H, COMT and CAD, and are important for the transcriptional regulation of the lignin biosynthetic pathway [28]. H-box elements were involved in the regulation of lignin biosynthesis [29]. The Mybplant motif in the promoter of the phenylpropanoid biosynthesis genes regulates lignin biosynthesis by binding P-box elements [30]. In our study (Figure 4), all DfCCoAOMT gene promoters, except DfCCoAOMT3/6/7/12/13, contained AC elements or Mybplant elements, especially DfC-CoAOMT2/14/16, suggesting that these genes might be involved in lignin biosynthesis. Furthermore, other cis-acting elements were identified in the DfCoAOMT gene promoters, which were associated with ABA, auxin, SA, MeJA, GA, LTR, drought and light. It has been well documented that lignin synthesis is closely related to environmental stress and could help plants resist mechanical damage and pathogenic bacterial attack. In switchgrass, drought and injury treatments rapidly induced an elevated expression level of PvCCoAOMT [31]. In Arabidopsis, ABA regulated lignin synthesis by phosphorylating NST1 [32]. A MeJA treatment also rapidly increased the expression of lignin synthase genes (including CCoAOMT) to promote lignin synthesis [18,33]; the JA core transcription factor MYC2 could regulate secondary wall synthesis by regulating the expression of NST1 [34,35]. Therefore, we sprayed ABA and MeJA on 1-year-old plants and then detected the expression of DfCCoAOMT2/5/6/8/9/14/15; we found that the expression levels of DfC-CoAOMT6/8/9/14 were significantly inhibited by ABA and MeJA and reached the lowest level at 3 and 6 h, respectively, while the expression levels of DfCCoAOMT2/5 were rapidly induced in response to the ABA treatment and reached the highest level at 6 h ( Figure 4C).
These results suggested that ABA and MeJA signaling could regulate lignin synthesis by affecting the expression of DfCCoAOMT genes, and different DfCCoAOMT genes played different roles in response to ABA and MeJA, but how they coordinate environmental signals and lignin synthesis needs further investigation.
According to the analysis of transcriptome data from different growth periods/tissues of D. farinosus (Figure 3), the DfCCoAOMT genes had differential expression profiles in tissues. DfCCoAOMT1/2/3/4/7/8/13/17 showed low expression levels in all tissues and different growth periods, suggesting that they were not involved in lignin synthesis during the normal growth and development of D. farinosus. DfCCoAOMT10/11/12 were specifically highly expressed in roots, indicating that they were mainly involved in root development. DfCCoAOMT6/9/14/15/16 showed a high expression in several tissues, especially in stems and roots, suggesting that they were involved in lignin synthesis in several tissues during plant growth and development. The elongation process of bamboo shoots is usually accompanied by secondary wall thickening and lignin accumulation. To further screen the key DfCCoAOMT gene in the lignin synthesis pathway [36,37], we analyzed sections of 2 m tall bamboo shoots from each internode and found that the lignin content gradually decreased from the base to the top of the bamboo shoots ( Figure 5). qRT-PCR showed that the expression trends of DfCCoAOMT5/6/9/14 were consistent with lignin accumulation, combined with the transcriptome data analysis, suggesting that these three genes might be the main enzyme genes involved in lignin synthesis during the development of D. farinosus shoots. According to our results, there were still a large number of DfCCoAOMT genes that might not be directly involved in lignin accumulation in normal development; they would be preferred to the synthesis of other secondary metabolites or participate in the stress response. For instance, VvCCoAOMT4, along with many CCoAOMT genes from other plant species [38][39][40], was shown to be involved in the synthesis of flavonoids, especially anthocyanins. Of course, whether the same function exists in D. farinosus requires further verification using transgenic technology.
Compared to Arabidopsis and Populus, the CCoAOMT gene family in D. farinosus has more members and more complex functions. DfCCoAOMT14 might be a key gene that is involved in lignin synthesis in plants. To further validate our analysis, DfCCoAOMT14 was overexpressed in tobacco ( Figure 6) and we found that the transgenic plants were slightly smaller than wild type, but the lignin content in stems was significantly increased and the xylem width was thickened. qRT-PCR confirmed that the expression levels of genes related to lignin synthesis were significantly up-regulated, indicating the function of DfC-CoAOMT14 involved in lignin synthesis. Drought stress experiments confirmed that overexpression of DfCCoAOMT14 significantly improved the drought resistance of plants ( Figure 7); this result is consistent with previous studies of Pinus sylvestris CCoAOMT [16]. Moreover, the content of lignin is crucial to the properties and quality of bamboo, which is a major limiting factor for bamboo pulp papermaking. Mutation of the CCoAOMT gene in Arabidopsis significantly down-regulated lignin content, especially G-type lignin [10]. Reverse repression of CCoAOMT in alfalfa significantly reduced lignin content but increased the ratio of S/G lignin components, which facilitated lignin degradation and separation [41]. So far, changing the expression levels of the key enzyme genes in the lignin biosynthesis pathway is still the main means to regulate lignin content. Therefore, this study can provide a theoretical basis for improving the wood properties of bamboo and breeding superior bamboo species with a high biomass or low lignin.

Identification of CCoAOMT Genes of D. farinosus
For the identification of the CCoAOMT family in D. farinosus, the methyltransferase domains (PF01596, CCoAOMT) were downloaded from Pfam (http://pfam.xfam.org, accessed on 1 May 2023) and they were used as a query in Hidden Markov Model (HMM) searches for candidate CCoAOMT genes in D. farinosus, with the cutoff at 0.01 [42].The Conserved Domain Search (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi, accessed on 1 May 2023) online software in NCBl was used to remove the sequences without the methylation domain. The online tool ExPASY (http://web.expasy.org/protparam/, accessed on 1 May 2023) was used in calculating the molecular weight (MW) and theoretical isoelectric point (pI) of DfCCoAOMT proteins. The subcellular localization of DfCCoAOMT proteins was predicted using Cell-PLoc 2.0 (Cell-PLoc 2.0 package (sjtu.edu.cn, accessed on 1 May 2023)). A total of 17 CCoAOMT family members were retained in D. farinosus for further analyses (Table 1). The candidate CCoAOMT gene sequences of D. farinosus were identified from the whole-genome data of D. farinosus, and all DfCCoAOMT gene sequences are attached in Supplementary File 1.

Phylogenetic Analysis and Sequence Alignment of the CCoAOMT Family of D. farinosus
The CCoAOMT family protein sequences of Arabidopsis thaliana, rice and Populus trichocarpa were downloaded from Phytozome (https://phytozome.jgi.doe.gov/, accessed on 1 May 2023) (Supplemental Table S1). Phylogenetic trees were constructed with IQ-TREE v1.6.12 using the Model Finder algorithm. We performed multiple sequence alignment analyses of CCoAOMT protein between D. farinosus and other species [43]. We constructed a phylogenetic tree between the CCoAOMT genes of D. farinosus by using the Clustal W program in MEGA 7.0 software and the neighbor-joining (NJ) method with 1000 bootstrap replications [44].

Chromosomal Localization and Collinearity Analysis of DfCCoAOMT Genes
The internal collinearity relationship was analyzed by using the Multiple Collinearity Scan toolkit (MCScanX) and we obtained collinearity gene pairs in the D. farinosus CCoAOMT family genes [45]. The collinearity analysis maps were then constructed using the Dual Systeny Plotter software (https://github.com/CJ-Chen/TBtools, accessed on 1 May 2023) [44]. Then, the Ka, Ks and MYA values were calculated using WGD v1.1.1 (Supplemental Tables S4 and S5).

Conserved Motifs and Gene Structure Analysis of CCoAOMT Family of D. farinosus
The conserved motifs of DfCCoAOMTs were determined using the MEME online tool (version 5.3.0, http://meme-suite.org/tools/meme, accessed on 1 May 2023) [46]; the number of motifs output was 10. The motif analysis and gene structure were visualized using the TBtools software. The Gene Structure Display Server 2.0 online tool (GSDS, http://gsds.gao-lab.org/index.php, accessed on 1 May 2023) [47] was used to illustrate the structure and exon/intron organization of DfCCoAOMT genes and the genomic length. The 2000 bp upstream promoter sequences of DfCCoAOMT genes were submitted to the PlantCARE database [48] (http://bioinformatics.psb.ugent.be/webtools/plantcare/ html/) and PLACE (https://www.dna.affrc.go.jp/PLACE/) to predict cis-elements and to subsequently screen cis-elements manually.

Plant Materials and Hormone Treatment
All materials used in this study were from the Institute of Bamboo Research, Southwest University of Science and Technology. One-year-old young leaves of D. farinosus were treated with 100 mM ABA and MeJA, respectively, and sampled in liquid nitrogen before treatment and after 3 h, 6 h and 9 h of treatment.

RNA Extraction and qRT-PCR
Total RNA of D. farinosus leaves was extracted using the TRIzol reagent (Takara, Dalian, China). The RNA quality was monitored using 1% denaturing agarose gel and a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Beijing, China). Then, cDNA was synthesized from 1 to 2 µg total RNA using the FastKing cDNA First Strand Synthesis Kit (TIANGEN, Beijing, China). The specific primer used for qRT-PCR was designed by using Primer Premier 5.0 (Supplemental Table S2). Tubulin was used as an internal reference gene. The transcript levels of DfCCoAOMT were obtained using the SYBR qPCR Master MIX kit (Vazyme, Nanjing, China) on the CFX96TM Real-Time System thermal cycler (BIO-RAD, CA, USA). The relative expression levels of DfCCoAOMT genes were calculated with the 2-∆∆Ct method [49]. Each sample had three independent biological replicates-each with three technical replicates.

RNA Sequencing
RNA sequencing was performed on 1-year-old D. farinosus materials in a growing chamber. The samples of D. farinosus included Young leaf (Yleaf), Mature leaf (MLeaf), Culm sheath (CSh), Sheath of bamboo shoot (ShB), Lateral bud (Labud), Node, Root, Branch, Internode (Inod), Rhizome bud (Rhbud), Rhizome neck (Rhne) and shoots of different height. The analysis included three biological duplications. Total RNA was extracted by using the TRIzol reagent (Takara, Dalian, China); the quantity and quality of total RNA were detected by a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Beijing, China). Seventeen samples were sequenced by the NovaSeq 6000 platform (Illumina, Beijing, China). The clean reads of each sample were compared with the reference genome of Dendrocalamus farinosus, and the alignment rate ranged from 74.47% to 91.57%. The raw reads were further processed with a bioinformatic pipeline tool, the BMKCloud (www.biocloud.net, accessed on 1 May 2023) online platform. All of these RNA-seq data were mapped to the D. farinosus reference genome using HISAT2 and the fragments per kilobase per million (FPKM) were calculated using StringTie. The heatmap was generated by TBtools based on the transformed data of log 2 (FPKM + 1) values. The details of the sample information and the FPKM values of CCoAOMT genes are shown in Supplemental Table S3, respectively.

Overexpression of DfCCoAOMT14 in Nicotiana Tabacum
Overexpressed tobacco lines were obtained by introducing the pCAMBIA1302-DfCCo-AOMT14 with the CaMV 35S promoter recombinant vector into the Agrobacterium tumefaciens strain EHA105 using the leaf disc method [50]. Infected tobacco leaf discs were inoculated on a MS medium containing 0.1 mg/L NAA, 0.5 mg/L 6-BA, 400 mg/L cephalexin and 9 mg/L Hyg in the dark at 27 ± 1 • C for 2 d. The regenerated buds were transferred to a MS medium containing 0.1 mg/L NAA, 400 mg/L cephalexin and 9 mg/L Hyg for the formation of complete plants. Genomic DNA was extracted from the leaves of transgenic plants and wild type and then amplified under the following conditions: preheating at 95 • C for 5 min, followed by 30 cycles of denaturation at 95 • C for 30 s, annealing at 60 • C for 30 s, extension at 72 • C for 1 min and finally extension at 72 • C for 10 min (Supplemental Figure S2).The PCR product was checked using 1% agarose gel electrophoresis. The expression of DfCCoAOMT14 in transgenic plants was detected by qRT-PCR and N. tabacum NtActin was used as the internal references for normalization [51].

Lignin Analysis
For plant cell wall preparation, the second to eighth internodes of 2 m bamboo shoots were ground to powder in liquid nitrogen. The determination method of lignin content was based on the method of Li et al. [52] and was appropriately improved as follows: weighed 0.5 g fresh sample, added liquid nitrogen to grind to fine powder, added 5 mL 0.1 M phosphate buffer (pH 7.2), incubated at 37 • C for 30 min and centrifuged for 5 min. Then, added 5 mL 80% ethanol to incubate at 80 • C for 1 h, centrifuged for 5 min and added 10 mL acetone to extract once, centrifuged for 5 min and dried at 60 • C. For the thioglycolic acid lignin analysis, approximately 20 mg of dried plant cell wall was incubated at 80 • C in 750 µL of water, 250 µL of concentrated HCl and 100 µL of thioglycolic acid for 3 h. The mixture was centrifuged and the pellet was washed with 1 mL of water and resuspended in 1 mL of 1 M NaOH on a rocking plate at room temperature overnight. The mixture was centrifuged again, and the supernatant was collected and mixed with 200 µL of concentrated HCl. After being vortexed and incubated at 4 • C for 4 h, the mixture was centrifuged and the pellet was dissolved in 1 mL of 1 M NaOH. The absorbance of a 50-fold dilution of the supernatant in 1 M NaOH was measured at 280 nm. The lignin content of Nicotiana tabacum overexpressing DfCCoAOMT14 was also determined by the same method.

Histochemical Staining
The complete and thin sections were stained in 2% phloroglucinol solution for 5 min, then were stained with 30% hydrochloric acid for 1 min, washed with water, observed and photographed under a Leica microscope.

Drought Treatment in DfCCoAOMT Transgenic Tobacco
In total, 45-day-old transgenic tobacco plants and wild-type plants were treated in drought stress for 37 days. The treated plants were then subjected to the measurement of relevant physiological indices, including malondialdehyde (MDA), superoxide dismutase (SOD), proline (Pro), peroxidase (POD) and leaf relative water content (RWC). The fresh weight (FW) of tobacco leaves was weighed. The fresh tobacco leaves soaked in distilled water for 24 h in the condition of 16 h light/8 h dark at room temperature and the turgid weight (TW) was weighed. The leaves were then dried for 24 h at 80 • C to obtain the total dry weight (DW) [53]. Leaf relative water content (RWC) was estimated according to the method of Turner (1981) [54]: RWC (%) = (fresh weight − dry weight)/(turgid weight − dry weight) × 100. MDA content, Pro content, SOD activity and POD activity were all measured using the related kit (Solarbio, Beijing, China). Subsequently, qRT-PCR was used to detect the expression levels of drought-related genes including NtERD10C, NtDREB and NtRD29A in drought-treated plants [55]; tobacco NtActin served as an internal reference.

Statistical Analysis
The data in this study were statistically analyzed by using SPSS 25.0 and Origin 2018. Data are presented as means ± SD of three independent replicates. The one-way analysis of variance (ANOVA) was used for the significant difference analysis and the significant difference of p < 0.05 was statistically significant.