Introduction

Plants often suffer from a variety of natural stress, such as salt, drought, cold, pathogens, and insect pests (Shao et al. 2015; Khalil et al. 2022). To cope with the impact of these different environmental pressures on plant growth and development, plants have evolved a series of defense mechanisms to resile from these stresses (Aldhanhani et al. 2022). NAC transcription factor gene family is a large plant specific gene family, as a molecular switch that can activate or inactivate gene expression, it plays an important role in plant response to these stresses (Shao et al. 2015; Yu et al. 2016).

The NAC gene family is designated, after the initials of three genes, NAM (NO APICAL MERISTEM, petunia) (Souer et al. 1996), ATAF1/2 (ARABIDOPSIS TRANSCRIPTION ACTIVATION FACTOR, Arabidopsis) and CUC2 (CUP-SHAPE COTYLEDON, Arabidopsis) (Aida et al. 1997). The five N-terminal motifs (A, B, C, D, E) of NAC transcription factor protein form a conservative domain. The length is about 160 amino acids, and the flank contains α- Helix and β- Fold structure. The motif A related to the formation of dimers and the motifs C and D that have nuclear localization signals and participate in DNA binding are highly conserved. The substructure variation of motif B and E is large, which may be closely related to the functional diversity of NAC genes (Jia et al. 2019). On the contrary, C-terminal domains are very different, with transcriptional activation or inhibition activities, and participate in the regulation of multiple networks (Olsen et al. 2005).

At present, NAC transcription factor family has been widely studied in plants species. Under the induction of different plant development stages and various environmental factors, NAC can activate specific target proteins to regulate plant growth and development and plant secondary growth (Nakashima et al. 2012). For example, in Arabidopsis, 117 NAC genes were identified (Kim et al. 2006). Overexpression of the NAC family gene VNI2, which could inhibit the normal development of xylem vessels in roots and aboveground parts by inhibiting the expression of xylem vessel development genes (Yamaguchi et al. 2010). In rice (Oryza sative), 151 NAC genes were identified (Nuruzzaman et al. 2010). Among them, after OsNAC6 overexpression in the plants, the transgenic plants showed a semi dwarfing phenotype (Sakamoto et al. 2016). In Populus trichocarpa, 163 NAC genes were identified (Hu et al. 2010). It was reported that overexpression of PopNAC122 gene will reduce plant height, cell size and cell number (Grant et al. 2010).

In addition, NAC genes has also been proved to play an important role when plants were subjected to biotic and abiotic stresses (Lee et al. 2012). For example, many NAC genes have the function of enhancing plant drought resistance. AtNAC016 could improve plant drought resistance by inhibiting the expression of AREB1, a negative regulator of ABA signaling pathway in Arabidopsis (Sakuraba et al. 2015). In rice, after OsNAC10 gene overexpression, the drought resistance of the plant was enhanced, the results of field experiments showed that the rice yield under drought and control conditions was significantly higher than that of the wild type (WT) (Jeong et al. 2010). It was reported that after overexpression of TaNAC2, the resistance of plants to drought, salt and cold stress was enhanced in Triticum aestivum (Mao et al. 2012). To sum up, NAC gene family has been identified in most plants, and the functions of some genes have been verified. Nevertheless, little is known about this gene family in Dendrobium catenatum (D. candidum).

D. candidum is the third largest genus of Orchidaceae and contains approximately 1,450 species, with high ornamental and medicinal value in China (Zhang et al. 2016). NAC gene plays an important role in plant response to drought stress, which is crucial for plant growth and development (Hu et al. 2006; Zheng et al. 2009; Chen et al. 2018; Negi et al. 2018; Wu et al. 2023). However, few studies have focused on identifying NAC genes in D. candidum. We also examined that DcNAC043-1, DcNAC043-2 and DcMYB46 (the gene encoding DcNAC043 interacting protein) expression under drought stress. Simultaneously, the lignin content in the plant was detected under the condition of up regulation of DcNAC043s expression, to reveal the relationship) between DcNAC043s and drought stress responses. Our research provides insight into the evolution of NAC genes in plants and lays the foundation for elucidating the roles of NACs in regulating drought stress in D. candidum.

Materials and methods

Plant materials and treatments

The materials used in this study were collected by Ming-jin Huang, associate professor of Agricultural College/Dendrobium Institute of Guizhou University, in Yandang Mountain, Zhejiang Province, China, and then planted in Dendrobium Institute of Guizhou University (26°26′44″N/106°39′32"E). It was identified as Dendrobium candidum by the Department of Agriculture and Rural Affairs of Guizhou Province, named as Jinhu No. 1, and identified as “Qianren 20210012” (Non endangered species). To investigate the expression patterns of DcNAC043 and DcMYB46 genes under drought stress, the Dendrobium sown in nutritious soil in a growth chamber, with a 16 h light (25 °C)/8 h dark (20 °C) photoperiod, 60% humidity, and an 800 μmol m−2 s−1 light intensity. The photosynthetic photon flux density (PPFD) was measured using a quantum radiometer/photometer (LI 185 B; LI-COR Biosciences, Lincoln, NE, United States). One-years-old plants were transferred into new Nutrient soil for treatment. Dendrobium was treated with natural drought, three biological replicates of leaves were harvested after 0, 3, 6 and 9 day (d) of treatment (Fig. 1Aa–d). All samples were immediately frozen in liquid nitrogen and stored at – 80 ℃. All the samples were prepared for RNA extraction and quantitative real-time PCR (qRT-PCR) detection (Miao et al. 2021). The remaining dendrobium materials used in the experiment continue to grow in the Dendrobium Institute.

Fig. 1
figure 1

Phylogenetic relationships, gene structures of NAC proteins in A. thaliana and D.candidum, the amino acid sequences of AtNACs and DcNACs are aligned using ClustalW2. A Phylogenetic analysis of AtNAC and DcNAC proteins in Dendrobium. The phylogenetic tree is constructed by MEGA 7.0 using the neighbor-joining method with1000 bootstrap replicates and displayed using FigTree v1.4.0. At: A. thaliana; Dc: D. candidum. B Gene structures generated using the Gene Structure Display Server. Exons (CDS) and introns are shown as orange wedges and black lines, respectively

Full size image

Identification of NAC family genes in D. catenatum

D. candidum Genomic data (BioProject ID PRJNA453230) were obtained from NCBI (http://www.ncbi.nlm.nih.gov/) publicly availabl (Miao et al. 2021). Pfam website (http://pfam.xfam.org/) was used to download the hidden Markov model (HMM) profile of the NAC domain (PF02365) (Mistry et al. 2021). To identify NAC proteins in these species, 117 NAC proteins sequences from Arabidopsis were used as queries in a reciprocal Basic Local Alignment Search Tool Protein (BLASTP) analysis (Camacho et al. 2009; Liu et al. 2018). Multiple sequence alignment of the NAC proteins was performed in ClustalW2 (http://www.ebi.ac.uk/Tools/msa/clustalw2) using the default parameters (E value < 1e-5) (Larkin et al. 2007). According to the aligned sequences, a new HMM in HMMER (HMMER 3.1, http://hmmer.org/) was constructedand and used as the query (E value < 0.01) to search against the D. candidum genome sequence data (Potter et al. 2018). NAC gene candidate was based on the gene encoding NAC domain protein (Shen et al. 2019).

Phylogenetic and protein property analysis of the NAC family in D. catenatum

Multiple sequence alignments of NAC coding sequences between D. candidum and A. thaliana sequences mentioned above were conducted using ClustalW2 with default parameters (http://www.genome.jp/tools bin/clustalw) (Liu et al. 2018). After screening the sequences that were closely related to Arabidopsis, based on these sequences (NAC031, NAC033, NAC043, NAC009, NAC018, NAC002 and NAC029), the related sequences of species that were closely related to D. candidum were obtained from NCBI (http://www.ncbi.nlm.nih.gov/) (Table S1), and then ClustalW2 Performs was usd multiple sequence alignments, with default parameters. The phylogenetic trees were generated using the Molecular Evolutionary Genetics Analysis (MEGA) 7 program (Tokyo Metropolitan University, Tokyo, Japan) (Kumar et al. 2016), with the NJ method, the p-distance + G substitution model, 1000 bootstrap replications, and conserved sequences with a coverage of 80% (Liu et al. 2018). Using FigTree v1.4. (http://tree.bio.ed.ac.uk/software/figtree/) to realize the phylogenetic trees visualization (Chen et al. 2015).

The physicochemical properties of each DcNAC protein were predicted using the ExPASy online program (http://web.expasy.org/translate/) (Duvaud et al. 2021). The transmembrane transport peptides were predicted using Tmpred (http://www.ch.embnet.org/software/TMPRED_form.html), and signal peptides were predicted using SignalP4.1 (http://www.cbs.dtu. dk/services/SignalP/) (Cruz et al. 2017). The subcellular localization of proteins were predicted using Plant-mPLoc online program (http://www.csbio.sjtu.edu.cn/bioinf/plant-multi/) (Li et al. 2020). The coding sequence alignments were imported into KaKs_calculator3.0 (Chinese Academy of Sciences, Beijing, China) to calculate the synonymous mutation rate (Ks) and non-synonymous mutation rate (Ka) using the NG method (Zhang et al. 2021). STRING datebase (https://string-db.org/) was used to display the protein–protein networks of NAC proteins (Szklarczyk et al. 2021). Gene Structure Display Server (GSDS 2.0, http://gsds.cbi.pku.edu.cn) was used to display the exon/intron structures of DcNAC genes (He et al. 2019).

Motif identification and cis-acting elements analysis

Conserved motifs in the proteins were identified using the Expectation Maximization for Motif Elucidation program (MEME v4.12.0, http://meme-suite.org) with the following parameter settings: The maximum number of motifs was 15, and the optimum width was set from 6 to 200. Only motifs with an e-value of 1e-10 were retained for further analysis (Nystrom et al., 2021). The results were visualized using TBtools (Chen et al. 2020). PlantCARE software (http://bioinformatics.psb.ugent.be/webtools/plantcare) was used to predict the cis-acting elements within 2000 bp upstream of all DcNAC genes (Sun et al. 2022a, b).

Expression analysis of the DcNAC043 and DcMYB46 by qRT-PCR

Total RNA was extracted from the Dendrobium leaf using an RNA Extraction Kit (Tiangen, Beijing, China), and cDNA was synthesized from 1 µg of total RNA using M-MLV transcriptase (TaKaRa Biotechnology, Dalian, China) according to the manufacturer’s instructions. And cDNA was synthesized using the PrimeScript™ 1st Strand cDNA Synthesis Kit (Takara, Japan) according to the manufacturer’s instructions (Ge et al. 2018). These reactions were executed under the following conditions: 37 °C for 15 min, 85 °C for 5 s, and finally ended at 4 °C. The reaction system was 2 × SYBR Green Pro Taq HS Premix 10 μL, primer F 0.4 μL, primer R 0.4 μL, cDNA 2 μL, ddH2O 7.2 μL (Pan et al. 2023).

Gene-specific primer sequences for the DcNAC043 and DcMYB46 were designed using Geneious Prime (Version. 2022.2) (Zhang et al. 2021) (Table S2). Each reaction was carried out in biological triplicate in a reaction volume of 20 µL containing 1.6 µL of gene-specific primers (1.0 µM), 2.0 µL of cDNA, 10 µL of SYBR green, 0.2 µL of ROX Reference Dye II, and 6 µL of sterile distilled water. The PCR program was as follows: 95 ℃ for 3 min; 45 cycles of 10 s at 95 ℃ and 30 s at 60 ℃; and then melt curve 65–95 ℃, increment 0.5 ℃ for 5 s (Liu et al. 2018). Melting curves were generated to estimate the specificity of these reactions. Relative expression levels were calculated using the 2−∆∆Ct method, with ACT2 used as internal controls (Wu et al. 2015).

Observation on lignin in cell wall of stem in D. catenatum

Stems of Dendrobium after 0, 3, 6 and 9 d of natural drought stress were sliced into 20 μM slice sections. Thin sections were stained for lignin with a HCl-phloroglucinol solution [2% (w/v) phloroglucinol (dissolved in 95% (v/v) ethanol): 2 M HCl = 1:1, fresh preparation] and observed under a Zeiss Axioscope A1 microscope (Leica) with a × 0.5 optical adapter (Patten et al. 2005).

Statistical analysis

All experiments were assessed using at least three independent repeats, and all data are represented as the mean ± SD. The statistical analysis was performed using a Student’s t test, and the significance was indicated using asterisks: *(p value < 0.05) or **(p value < 0.01).

Results

Identification of NAC proteins in D. candidum and other plants

Using the 117 AtNAC protein sequences in Arabidopsis as queries, 33 NAC proteins were identified from D. candidum. All DcNAC proteins share a conserved domain: The NAM domain. Relationships based on the phylogenetic tree; the nomenclature used for DcNAC protein was based on the corresponding AtNAC orthologs (Table 1). Based on the phylogenetic tree, eight relatively conservative NAC members were selected for subsequent analysis (Fig. 1A). Using the nine DcNAC protein sequences in D. candidum as queries, other 12 NAC043, 1 NAC033, 13 NAC031, 13 NAC029, 5 NAC018, 10 NAC009 and 9 NAC002 proteins were identified from 37 crops (P. equestris. and others) (Table 1).

Table 1 List of NAC genes identified in D. candidum and other plants
Full size table

The number of amino acid residues in the DcNAC proteins ranged from 220 (TuNAC018) to 449 (AsNAC043), with an average of 336. The relative molecular weights (MWs) ranged from 24.48 (TuNAC018) to 49.93 kDa (PtNAC031). The isoelectric points (pIs) were predicted to range from 5.32 (DcaNAC002) to 9.06 (PeNAC029), and 28 members had pI values > 7 and others had pI values < 7. No transmembrane helix or signal peptide was found in any NAC protein (Table 1). All the NAC proteins were predicted to localize in the nucleus, which were consistent with the general nuclear localization of TFs.

Phylogenetic analysis and classification of the NAC proteins in D. candidum and other plants

To clarify the evolutionary relationships between NAC proteins from Arabidopsis and D. candidum, we aligned the sequences of 33 NAC proteins using AtNACs. Based on affinities presented by the evolutionary tree, and we removed groups with only AtNAC or DcNAC, NAC proteins were divided into four groups (Fig. 2A). Although different copies of the same species were still more closely related in evolutionary relationship than the same copies of different species, we can still find the closely related copies of Arabidopsis and D. candidum in these groups. Including the NAC31 clade (class II), the NAC033 and NAC043 clade (class III), and the NAC009, NAC029, NAC018 and NAC002 clade (class IV).

Fig. 2
figure 2

Phylogenetic relationships and conserved motifs between DcNAC and other plant NAC proteins. A The phylogenetic tree is constructed using MEGA 7.0 with the neighbor-joining method (1000 bootstrap replicates) and displayed using FigTree v1.4.0. NAC proteins in the phylogenetic tree are clustered into four distinct groups. At: A. thaliana, Dc: D. candidum (marked in red), Cl: Carex littledalei, Jr: Juglans regia, Pt: Populus trichocarpa, Zm: Zea mays, Dca: Dioscorea cayenensis, Pe: Phalaenopsis equestris, Dn: Dendrobium nobile, Ma: Musa acuminata; Ac: Ananas comosus, Eg: Elaeis guineensis, Cn: Cocos nucifera, Pd: Phoenix dactylifera, Qs: Quercus suber, Pa: Prunus avium, Me: Manihot esculenta, Tc: Theobroma cacao, Vv: Vitis vinifera, Os: Oryza sativa, As: Apostasia shenzhenica, Ql: Quercus lobata, Zj: Ziziphus jujuba, Hb: Hevea brasiliensis, Ao: Asparagus officinalis, Tu: Triticum urartu, Mn: Morus notabilis, Dk: Diospyros kaki, Nt: Nicotiana tomentosiformis, Bn: Brassica napus, Ss: Spatholobus suberectus, Rc: Ricinus communis, Csa: Cannabis sativa, Tg: Tulipa gesneriana, Pj: Phtheirospermum japonicum, Sp: Solanum pennellii, Cc: Citrus clementina, Cs: Citrus sinensis. B Conservative motifs of DcNAC proteins identified by MEME. The motifs were indicated by colored boxes and their numbers are shown in the scale below the diagram. C Sequences of the 15 conserved motifs in DcNACs identified in this study

Full size image

To further investigate the evolutionary relationships of NAC proteins in plants, we constructed a larger neighbor-joining (NJ) tree based on 8 NAC protein sequences identified from D. candidum, 63 NAC proteins from other 37 species (Phalaenopsis equestris et al.) were found and constructed a phylogenetic tree together. According to kinship, they could be divided into 4 classes (Fig. 2A). Class I and III were the two most conservative groups, class I contains only NAC031 clade and class III contains only NAC009 clade. Class II contains NAC033 and NAC043 two clade, probably because NAC033 had only two copies of AtNAC033 and DcNAC033. Class IV as the largest group, including NAC002, NAC018 and NAC029, with a total of 30 members (Fig. 2A). Interestingly, both DcNAC043, DcNAC031 and DcNAC009 appeared to be more closely related to Phalaenopsis equestris, probably because they were both orchids.

Gene structure and conserved Motif analyses

To compare the exon/intron structures of AtNACs and DcNACs genes, the coding sequences with their corresponding genomic sequences were aligned. 74 genes (43 AtNACs and 31 DcNACs) were mapped using the GSDS 2.0 software package. The results showed that all AtNACs used for analysis contained no introns, while almost all DcNACs contained at least one intron (except DcNAC002, DcNAC033, DcNAC070 and DcNAC074) (Fig. 1B). Among DcNACs, there were 6 copies of NAC100 and 3 copies of NAC098 in class I, except for NAC100-1 and NAC098-1, the rest of the copies contain an intron and 2 CDS regions. Similarly, two copies of DcNAC43 in class III, two copies of DcNAC009 in class IV, and three copies of DcNAC068 in class IV have similar gene structures (2 introns, 3 CDS regions; 1 intron, 2 CDS regions; 2 introns, 2 CDS regions, respectively) (Fig. 1B). It suggested that the gene structures of different copies of these genes were relatively conserved in the evolutionary process, and their functions may also be similar. In contrast, there seemed to be no regularity in the gene structure between different copies of other DcNAC genes, maybe there was a big difference in functionality.

To further understand the structural diversity of DcNAC proteins, we identified 15 putative motifs (motif 1–15) in the proteins using the MEME/MAST program (Fig. 2B, C). As shown in Fig. 2, all NAC protein sequences involved in the analysis contained motif 1, motif 2, motif 3, motif 4 and motif 5. However, each subgroup had its own unique motif, for example, in class I, the specific motif 14 of individual members, the specific motif 9 and motif 12 of class II, the specific motif 7, 11, 13 of class III, and the specific motif 6 of group IV. The finding of similar gene structures and conserved motifs within the same subfamily further supports the accuracy of the phylogenetic tree. The other side of the shield, the motif differences between different subfamilies also indicate functional diversity of the NAC gene family in D. candidum.

Responsive elements in DcNAC promoters

To further investigate the potential regulatory mechanisms of DcNAC cduring the abiotic stress response, a sequence 2 kb upstream from the translation initiation site of the DcNAC gene was submitted to PlantCARE to detect cis-elements. All genes involved in the analysis of promoter elements (DcNAC008-5, DcNAC009-1, DcNAC009-2, DcNAC018, DcNAC029, DcNAC031, DcNAC033, DcNAC043-1, DcNAC043-2, DcNAC068-1, DnNAC068-2, DcNAC068-3, DcNAC070, DcNAC074, DcNAC078, DcNAC094, DcNAC098-1, DcNAC098-3, DcNAC100-2, DcNAC100-3, DcNAC100-4, DcNAC100-5, DcNAC100-6) contain CAAT-box and TATA-box elements, which proved that all the genes can be transcribed normally and thus participate in plant growth and development.

Abiotic stress response elements and phytohormone response elements were analyzed. DcNAC009-1 and DcNAC043-2 contained 20 and 18 response elements (Hormones and Abiotic Stresses), respectively, which were the two genes with the highest element content (Fig. 3). The promoters of the two genes, DcNAC009-2 and DcNAC078, had the least stress response elements, and both had only five. Among them, MYB and MYC elements (response to drought and ABA induction) were the most numerous (49 and 66, respectively), and TCA-element (response to salicylic acid induction) was the least numerous (only 4). Among the promoters of 23 genes, drought stress response elements appeared most frequently, accounting for 48%.

Fig. 3
figure 3

The cis-elements in the promoter of DcNAC genes. The promoter sequences 2 kb of 33 DcNAC genes were analyzed. The names of the DcNAC genes are provided at the bottom of the figure. The number of cis-elements about hormone-stress (marked in orange), abiotic-stress (marked in blue) are presented

Full size image

Ka/Ks ratio calculation

The genomic information obtained from National Center for Biotechnology Information (NCBI) does not include the chromosomal location of the genes. Therefore, chromosomal location and collinearity analysis were not performed in this study. To explore the selective pressure on DcNAC genes, the non-synonymous/synonymous mutation ratio (Ka/Ks) was calculated for the genes involved in the evolutionary analysis, and values of Ka/Ks > 1, = 1 and < 1 indicate positive selection, neutral selection and purifying selection, respectively (Nekrutenko et al. 2002).

After removing gene pairs with no numerical results, the remaining genes were counted. As shown in Table 2, the Ka/Ks ratio for all DcNAC genes were < 1, ranging from 0.0807 (AtNAC087-2/DcNAC100-6) to 0.4293 (AtNAC003/DnNAC100-4), indicating the genes were negative selection during evolution and functionally conserved, which reduced the rate of change in aa profile. A comparation of the Ka/Ks ratio of DcNAC genes among the class I, II, III, IV showed that the average Ka/Ks ratio was higher in the class I (0.2040) than in the class III, II and IV (0.1534, 0.1376, and 0.1327, respectively), suggesting that DcNAC genes in the class I experienced higher selection pressure during the evolutionary history of D. candidum. However, the overall results showed that most DcNAC genes were slowly evolving in D. candidum.

Table 2 Non-synonymous and synonymous nucleotide substitution rates between AtNACs and the corresponding orthologs in D. candidum
Full size table

Protein–protein networks analysis of DcNAC043s

The previous analysis found that NAC043 gene was relatively conservative in the evolutionary process. Therefore, based on D candidum, with Arabidopsis and Populus protein database were refered, the protein interacting with DcNAC043s were predicted by STRING software. As the Fig. 4 showed that the two copies of DcNAC043 were consistent with Arabidopsis, and both predicted to interact with DcMYB46 (Zhong et al. 2007a, b). Except Cellulose synthase A catalytic subunit 8 (AtCESA8), Homeobox protein knotted-1-like 3 (DcHOS66) and DcNAC031, the proteins appearing in the interaction network belonged to R2R3-MYB family. These proteins have been reported to participate in the growth and development of plant secondary cell wall (Fang et al. 2020; Im et al. 2021; Ye et al. 2021). Interestingly, although the proteins interacting with PtNAC043 did not belong to the R2R3-MYB family, they have also been reported to directly or indirectly regulate the growth and development of plant secondary cell wall (Xing et al. 2008; Min et al. 2020).

Fig. 4
figure 4

NAC043 protein interaction network of A. thaliana, D. candidum (2 copies) and P. trichocarpa

Full size image

DcNAC043 responds to drought stress and regulates lignin synthesis

Previous studies showed that NAC043 responded to drought stress in ryegrass (Cheng et al. 2022). Based on the previous analysis, NAC043 had proved that its function was conservative. Therefore, we treated seedling of D. candidum under drought stress, and detected the expression of DcNAC043s during drought stress by qRT-PCR. The results showed that the change of DcNAC043s expression in leaves was related to the drought degree suffered by plants (Fig. 5B). With the increase of drought stress time, the expression of DcNAC043-1 and DcNAC043-1 in leaves also increased significantly (PDcNAC043-1 = 0.024, 0.017, 0.009 and PDcNAC043-2 = 0.03, 0.021, 0.017) (Fig. 5B). Consistent with the change trend of DcNAC043s, the expression of DcMYB46 also increased with the increase of drought time, which again proved that DcMYB46 was involved in regulating drought stress.

Fig. 5
figure 5

Morphology and staining of Dendrobium and expression of genes related to drought stress response. A a, b, c and d-I Morphology of dendrobium plants under drought stress for 0, 3, 6 and 9 days respectively. a-II, III and IV, lignin staining results of dendrobium plants under drought stress for 0 d, b-II, III and IV, lignin staining results of dendrobium plants under drought stress for 3 d, c-II, III and IV, lignin staining results of dendrobium plants under drought stress for 6 d, d-II, III and IV, lignin staining results of dendrobium plants under drought stress for 9 d. B Expression analysis of DcNAC043 and DcMYB46 under drought stress at 0,3,6 and 9 d. CK: Drought stress 0 day, as control, Ds_3d: Drought stress 3 days, Ds_6d: Drought stress 6 days, Ds_9d: Drought stress 9 days. ** denotes a significant difference at P < 0.01, as determined using Student’s t test, Bars represent SD of the average, mean ± SD

Full size image

In Citrus grandis, CgNAC043 was found to regulate lignin synthesis, thereby affecting the development of secondary cell wall (Li et al. 2022). Therefore, we used the lignin specific stay HCl—phloroglucinol to dye it for analysis. We observed that the longer the drought stress time was, the more serious the lignification of the plant was, which indicates that the more lignin was accumulated (Fig. 5A). It was indirectly proved that DcNAC043s also participates in the synthesis of lignin in Dendrobium.

Discussion

NAC-TF has been found and classified in many plants, but there is no detailed evidence about the NAC-TFs family of Dendrobium. In this study, although the published full genome information of Dendrobium showed that there were 75 annotated NAC genes, which was similar to the number of NAC genes identified in grapes (74) and sugarcane (88) (Wang et al. 2013; Manimekalai et al. 2017). However, by comparing BLASTP with 117 AtNAC proteins and based on conservation analysis, this study ultimately obtained 33 DcNAC proteins. It was far less than potato (110) (Singh et al. 2013), soybean (101) (Shang et al. 2016), cacao tree (102) (Shen et al. 2019), rice (151) (Nuruzzaman et al. 2010) and poplar (163) (Hu et al. 2010). The difference in the number of NAC members among these species may be caused by genome wide polyploidy events during the evolution and differentiation of each species.

Interestingly, analysis of phylogenetic tree, gene structure, conservative motif and Ka/Ks value showed that NAC was unevenly distributed in Arabidopsis and Dendrobium, indicating that NAC gene family existed before the differentiation of the two species. It was different from the ancient and inactive evolutionary nature of NAC gene family in other species (Hu et al. 2010; Nuruzzaman et al. 2010; Wang et al. 2013; Singh et al. 2013, 2016; Shang et al. 2016; Manimekalai et al. 2017; Shen et al. 2019). It may be that Dendrobium and Arabidopsis were exposed to different environments during evolution, so gene differentiation occurred, and the number of NAC genes in their subfamilies was also different.

However, among all DcNAC genes, DcNAC009s and DcNAC043s are different from other DcNAC genes and are relatively conserved in terms of gene structure and physicochemical properties. Therefore, to further prove the evolutionary conservation of NAC009 and NAC043, we analyzed the gene structures of NAC009 in 10 different species and NAC043 in 15 different species, respectively. The results showed that NAC009 and NAC043 in all species (except Arabidopsis) have similar gene structures; what is more, closely related species have very similar gene lengths (Fig. 6). Meanwhile, motif analysis of phylogenetic tree and gene structure based on multiple species shows that NAC043 genes were relatively conservative, which is similar to the existing research results on the gene structure of NAC043 (Sun et al. 2023). All the NAC043 proteins contained not only motif 1, motif 2, motif 3, motif 4 and motif 5 that all the NAC proteins had, but also had their own unique motif 9 (SSITADTKTQMFHSSEGALDQILQYMGRSCKZESEAISNS) and motif 12 (MSISVNGQSQV) (Fig. 2B, C). Therefore, we infered that genes in this subfamily shared a common parent and may have similar functions.

Fig. 6
figure 6

Phylogenetic relationship and gene structure of DcNAC009 and DcNAC043 in different species. The amino acid sequences of AtNACs and DcNACs are aligned using ClustalW2. The phylogenetic tree is constructed using MEGA 7.0 with the neighbor-joining method (1000 bootstrap replicates) and displayed using FigTree v1.4.0. Exons (CDS) and introns are shown as orange wedges and black lines, respectively. A Gene structure differences of NAC009 among different species, B Gene structure differences of NAC043 among different species

Full size image

In previous studies, NAC043 was shown to Ryegrass can respond to drought stress (Cheng et al. 2022). Based on this, we selected mature Dendrobium plants for drought stress treatment, and extracted plant leaves for qRT-PCR detection. The results showed that the expression of DcNAC043-1 and DcNAC043-2 in plants were up-regulated after drought stress induction, and the up-regulated trend was related to the duration of drought stress. The staining results of HCl-phloglucinol not only proved that DcNAC043s in Dendrobium responded to drought stress by regulating lignin content, but also further proved that DcNAC043s, like CgNAC043 gene in other species, had the function of regulating plant secondary cell wall growth (Li et al. 2022; Wu et al. 2023). Moreover, they were similar to the protein interaction network of AtNAC043 and CgNAC043. In the network, DcNAC043-1 and DcNAC043-2 also act as the master switch that activates the expression of many lignin biosynthetic genes (Zhong et al. 2006; Zhong et al. 2007a, b; McCarthy et al. 2009; Zhou et al. 2009; Hussey et al. 2011; Li et al. 2022) (Fig. 4). It was proved that DcNAC043s were involved in the synthesis of lignin again. However, whether DcNAC043-1 and DcNAC043-2 have functional redundancy needs further experimental verification.

Conclusion

Our research shows that there were significant differences in the evolution of NAC genes between Dendrobium and the model plant Arabidopsis. Therefore, the functions of many members of the DcNAC family need to refer to other species for further research. But DcNAC043-1 and DcNAC043-2, which had relatively conservative function, has proved its contribution to the drought stress of Dendrobium through our verification. And we believe that DcNAC043 plays a very important role in regulating the lignin synthesis of Dendrobium. However, further research is needed on their upstream and downstream regulatory relationships and interaction networks in lignin synthesis pathways, as well as their molecular mechanisms in response to drought stress.