Genome-Wide Identification and Expression Analysis of the CAD Gene Family in Walnut (Juglans regia L.)

Lignin deficiency in the endocarp of walnuts causes kernel bare, leads to inconvenient processing and transportation of walnuts, and easily produces insect damage and mildew, thereby affecting the quality of walnuts. Cinnamyl alcohol dehydrogenase (CAD) is one of the key rate-limiting enzymes in lignin synthesis and plays an important role in the synthesis of lignin in the endocarp of walnut. However, knowledge about CAD gene family members and their evolutionary and functional characteristics in walnuts is limited. In this study, all 18 JrCADs were identified, and phylogenetic relationships, gene structure, protein motifs, collinearity analysis, and expression patterns of the JrCADs were also analyzed. All JrCADs could be divided into three groups based on the phylogenetic tree, gene structure, and motif analysis also support this grouping. Transcriptome data demonstrated that JrCADs have different expression patterns in walnut endocarps at different developmental stages. Combined with qRT–PCR data, we finally identified several candidate JrCADs involved in the process of endocarp sclerosis. This study showed that the JrCAD family members are highly conservative in evolutionary characteristics and they might participate in a variety of hormone responses. JrCAD17 and JrCAD18 are highly expressed in all periods of walnut endocarp harding, they are closely related to lignin accumulation.


Introduction
Walnut (Juglans regia L.) is an important nut in China and widely grown in Xinjiang province. (Feng et al. 2018). There are abundant walnut germplasm resources in Xinjiang province, and the walnut varieties can be divided into three types according to the thickness of the endocarp: bare shell, thin shell, and thick shell. Most of the walnut varieties in Xinjiang province are thin-shelled walnuts, such as 'Wen 185ʹ and 'Xinxin 2', with shell thicknesses of 0.9 to 1.4 mm. The 'Xinlu' walnut variety is a bare shell, and the uneven distribution of lignin in the endocarp causes the walnut shell to be bare. The walnut consists of the husk, shell (endocarp) and kernel, and both the hull and shell protect the kernel (Jahanban-Esfahlan et al. 2019). The endocarp of walnut plays an important role in fruit development, transportation, and storage processes (Khir et al. 2011); moreover, there is a positive correlation between the structure of the endocarp and the quality of walnuts , and studying the structure of walnuts endocarp is very important. According to previous research, the development of the walnut endocarp is associated with lignin , and lignin can account for 30-60% of a walnut by weight (Fu et al. 2018). The lignin deposition process correlates with the thickness of walnut shells, and it also provides strength and stiffness to plant cells against biological attack and stress response from the external environment (Zhao et al. 2016;Vanholme et al. 2019;Wu et al. 2021).
The process of lignification in the walnut endocarp is the oxidation of phenylalanine to the three major lignin monomers catalyzed by a series of enzymes by the phenylpropane pathway and the monolignol pathway, which are then deposited in the secondary cell wall . The lignin monolignol pathway involves a variety of enzymes, and coumaryl alcohol, sinapyl alcohol, and coniferyl alcohol are generated in this pathway. Then, three types of lignin polymers are formed by peroxidase and laccase (Fig. 1). To clarify the mechanism of action of lignin, numerous gene families regulating lignin biosynthesis were identified and characterized in different species, including the laccase (LAC) gene family (Cheng et al. 2019;Xu et al. 2019a, b;Simões et al. 2020), the NAC (NAM, ATAF, CUC) transcription factor gene family (Shan et al. 2019;Yao et al. 2020), the peroxidase (PRX) gene family (Meng et al. 2021), the COMT gene family (Liu et al. 2021b), the CCR gene family (Barakat et al. 2011), and the CAD gene family in the phenol propane metabolic pathway. In addition, lignin monomer synthases such as COMT or CAD also regulate lignin biosynthesis, inactivation of the synthases, and reduced lignin content and fractions in maize or pine (Provan et al. 1997;Lapierre et al. 2000). When the activity of CAD was inhibited, the composition structure of lignin was changed (Baucher et al. 1996).
CAD is the final step in the biosynthesis of lignin monomers, and converts cinnamaldehyde to cinnamyl alcohol in the presence of NADPH, which in turn forms lignin polymers hydroxyphenyl (H), guaiacyl (G), and syringyl (S) (Cheng et al. 2017). The CAD gene family has been studied in Arabidopsis (Kim et al. 2004), Oryza (Park et al. 2018), and Populus (Chao et al. 2014). Nine CAD genes were identified in Arabidopsis, of which two CAD genes (AtCAD4 and AtCAD5) encode highly active proteins that play a role in lignin synthesis in Arabidopsis, and double mutation of the two genes resulted in a dramatic reduction in lignin content in flower stems (Sibout et al. 2003(Sibout et al. , 2005. Twelve CAD genes were identified in Oryza, and OsCAD2 was strongly expressed in lignified tissues, as indicated by GUS analysis and RNA blotting (Hirano et al. 2011). Downregulation of CAD genes in tobacco, poplar, and pine did not show significant morphological changes, however, the proportion of lignin monomers composition of the cell wall is changed (MacKay et al. 1997;Ralph et al. 2001;Kim et al. 2002;Anderson et al. 2015). Breeding a walnut variety with suitable shell thickness is one of the main objectives, and the kernel yield of walnuts is related to shell thickness. The endocarp of walnuts is rich in lignin, and the formation of walnut shells is closely related to the process of lignin. Analysis of the key genes in the lignin metabolic pathway will provide further insight into the linkage between lignin and the walnut shell. Generally, most studies of CAD are mainly focused on the roots and stems of herbaceous crops or model plants (Kim et al. 2010;Gabotti et al. 2015;Liu et al. 2021a) and few studies have been performed on fruits, especially nuts. The endocarp of walnuts, which is rich in lignin, plays an important role in protecting and transmitting information to the developing kernel (Xiao et al. 2020). Studying the enzymatic mechanism of endocarp lignin in walnuts can fill a gap in the related field.
Due to functional diversity during evolution, not all CAD family members are involved in lignin synthesis. Screening and identification of CAD gene families can provide relevant theories for subsequent studies on the regulation of lignin metabolism. In this study, we used Xinjiang province walnut variety 'Xinlu' as a material. During the ripening process of 'Xinlu', the endocarp becomes underdeveloped, Fig. 1 The main processes of plant lignification. PAL phenylalanine ammonia lyase; C4H cinnamate 4-hydroxylase; 4CL 4-coumaric acid: coenzyme a ligase; HCT: hydroxy cinnamoyl transferase; C3H coumarate 3-hydroxylase; CCoAOMT caffeoyl-CoA O-methyl transferase; F5H ferulate 5-hydroxylase; COMT atechol-O-methyltransferase; CCR cinnamoyl CoA reductase; CAD cinnamyl alcohol dehydrogenase; LAC laccase; POD peroxidase resulting in kernel exposure, and it is an ideal material for studying the thickness of walnut shells. Here, 18 JrCAD genes were obtained by screening the walnut genome database. The sequence characteristics, phylogenetic tree, gene structure, conserved motifs, protein interaction prediction and collinearity analysis, and RNA-seq expression profiles of the JrCAD family were analyzed using bioinformatics, and the gene expression information of the JrCAD genes in the lignified parts and nonlignified parts of walnut endocarp at different developmental times was analyzed using qRT-PCR.

Plant Materials
Xinjiang province walnut 'Xinlu' was used as the experimental material, it has a bared shell, the bared phenotype is relatively stable and it was almost not affected by climate condition. The endocarp status of 'Xinlu' walnut before 57 DAF (days after full blooming) was recorded as Inchoative-hue, the unhardened endocarp after 57 DAF was recorded as Pulpy-hue, and the hardened endocarp was recorded as Scleritic-hue . In this experiment, walnut samples were harvested from June 7 (approximately 50 DAF) to July 26 (approximately 100 DAF), covering the entire hardening period of the endocarp. The Scleritic-hue (nonexposed) and Pulpyhue (exposed) parts by peeling the endocarp of the 'Xinlu' walnut variety were used for transcriptome sequencing.

Identification of CAD Genes in Juglans Regia
The latest version of the walnut genome protein sequence file (GCF_001411555.2_ Walnut_2.0_protein.faa) was downloaded from the NCBI website (https:// www. ncbi. nlm. nih. gov/ genome/ 17683) (Marrano et al. 2020). The hmmer models of CAD gene structural domains (ADH_ZinC_N(PF00107), ADH_N(PF08240)) (Youn et al. 2006) were downloaded from the Pfam website (Finn et al. 2016), protein sequences containing this structural domain were retrieved using the hmmer search program in hmmer 3.2 software (Zhang and Wood 2003), and the identification criteria were E-value < 0.01. A localized BLAST P search was also performed on the protein data of walnuts (Marchin et al. 2005), using the 9 CAD proteins of Arabidopsis as the probe for alignment. The protein sequences obtained by the above two methods were combined, and the protein sequences with duplicates, no complete coding frame, and incomplete sequencing were removed. Candidate CAD members were uploaded to the NCBI-CDD (https:// www. ncbi. nlm. nih. gov/ cdd) databases for further confirmation of the inclusion of the correct structural domain. The isoelectric points and molecular weights of CADs were obtained from the website ExPASy (https:// web. expasy. org/ protp aram/).

Phylogenetic Analysis
To explore the phylogenetic relationships and functions among JrCADs, CAD protein sequences of Arabidopsis thaliana (Kim et al. 2004), Oryza sativa (Park et al. 2018), Juglans microcarpa (analyzed in this experiment), and Populus tomentosa (Chao et al. 2014) were obtained as references, and the phylogenetic tree was constructed using the neighbor-joining method in MEGA 7.0 with a bootstrap value set to 1,000 replicates (Thompson et al. 1994).

Analysis of Gene Structure, Protein Motifs, and Conserved Structural Domains
Gene structure was mapped using the online website GSDS 2.0 (Hu et al. 2015) (http:// gsds. gao-lab. org/ index. php). Protein motifs of JrCADs were obtained by MEME (http:// meme-suite. org/ meme/ tools/ meme) (Bailey et al. 2009). The conserved type of the member structural domains was analyzed using a Pfam search (http:// pfam. xfam. org/ search), and the obtained motif and domain results were further mapped using TBtools .

Protein Interaction Prediction, Collinearity Analysis, and Promoter Element Analysis
The protein interaction network of JrCADs was analyzed using the online website STRING (https:// string-db. org/), using A. thaliana as a reference. The MCScanX tool (Wang et al. 2012) was used to perform collinearity analysis of JrCADs and A. thaliana with J. microcarpa. To conduct cis-element analysis, 2000 bp upstream of the CAD genes of walnut were analyzed using PlantCARE (plant cis-acting regulatory element) with default parameters (Lescot et al. 2002).

Gene Expression Analysis
Nine tissue samples from five time points during endocarp hardening of walnuts were selected (P50d, P64d, S64d, P71d, S71d, P78d, S78d, P92d, S92d). Total RNA was isolated by an RNA-prep Pure Plant Kit (Tiangen, Beijing, China), and three biological replicates were selected for each sample. Extraction of RNA followed by transcriptome sequencing. The expression level FPKM values of JrCAD genes were obtained from the transcriptome database sequenced by the subject group (unuploaded), which initially revealed the expression pattern of walnut CAD genes in lignin.

Real-Time Fluorescence Quantitative PCR Analysis
Ten genes (JrCAD1,JrCAD4,JrCAD6,JrCAD7,JrCAD8,JrCAD14,JrCAD15,JrCAD16,JrCAD17,JrCAD18) were selected from the JrCADs identified as potentially involved in walnut lignin biosynthesis. For internal control of the gene, 18S rRNA was used. Details of the primer information are listed in Supplementary Materials Table S1. Before the experiment, primer specificities were verified. Each experiment was conducted in triplicate. Gene expression was calculated using the 2 −△△CT method (Livak and Schmittgen 2001), and the mean of three biological replicates indicated their relative expression levels.

Identification and Characterization of J regia CAD Genes
A total of 18 candidate JrCADs were identified (JrCAD1-18, Table 1) by blasting and/or Hmmer searching from the J. regia genome. The JrCADs possess amino acids (aa) 355-388, with an isoelectric point (pI) of 5.65-7.16. The 18 JrCADs were unevenly distributed on chromosomes Chr1, Chr2, Chr8, Chr10, Chr11, Chr13, and Chr16, and some genes, such as JrCAD9, JrCAD10, and JrCAD11, were located very close to each other on one chromosome. Supplementary Materials Table S2 lists CAD gene family information in Juglans microcarpa.

Analysis of the Phylogenetic Tree, Gene Structures, Motif Composition, and Conserved Domains of CAD Family Members in J. regia
A phylogenetic tree was constructed with the NJ method by using 60 CAD sequences of JrCADs together with AtCADs, OsCADs, JmCADs, and PtoCADs (Fig. 2). Our results showed that the 60 CADs could be classified into three groups: Class I (including 10 JrCADs), Class II (2), and Class III (6). Notably, Class II contains two Arabidopsis proteins, AtCAD4 and AtCAD5, and a rice OsCAD2, which were proposed to be associated with lignin biosynthesis, as they were preferentially expressed in actively lignifying tissues (Sibout et al. 2005;Hirano et al. 2011), implying that JrCAD17 and 18 located in Class II might also be involved in lignification. Table S3 List of plant genes used in CAD gene phylogenetic analyses.
The gene structures of JrCADs were analyzed with the composition and location of gene exons and introns method according to a previous study on Populus (Barakat et al. 2009), and the results indicated that the 18 JrCADs were classified into three groups (Fig. 3). Both groups I and II have 5 exons, among which the third and fourth exons exhibited a large difference in sequence length between the two groups, whereas group III has 6 exons. All the genes have UTRs except for JrCAD2 and JrCAD7, which have no upstream or downstream UTRs.
Conserved motif analysis showed that the 18 JrCADs were classified into group I, II, and III, and this grouping is similar to the gene structure grouping. Group I contains motifs 1-10, and groups II and III are similar to group I but lack motif 9. When the motif difference is chosen to be 15, group I can be divided into four branches, which is consistent with the phylogenetic tree branching case. Group II specifically possesses motif 14, and group III specifically possesses motif 12 (Fig. 4a).
The medium-chain dehydrogenase/reductase (MDR) superfamily contains highly conserved zinc-binding domains (ADH-N and ADH-zinc-N). And in this study, the conserved structural domain analysis showed that the JrCADs contain the two domains ADH-N and ADH-zinc-N (Fig. 4b).
High sequence similarity of JrCADs amino acid sequences, the zinc-binding sites and NADPH-binding sites of JrCADs are shown in Fig. 5. The 18 JrCADs contain three conserved structures, including the Zn 1 binding motif GHE(X) 2 G(X) 5 G(X) 2 V,  the Zn 2 binding motif GD(X) 9,10 C(X) 2 C(X) 2 C(X) 7 C, and the NADP(H) binding site GXG(X) 2 G. The supplementary materials Figure S2 shows the multiple sequence alignment of the complete JrCAD protein sequences.

Promoter Region Analysis of JrCAD
To investigate the possible regulatory mechanisms of the JrCAD genes, we predicted cis-acting elements in the upstream 2 k bp sequence of the gene start codon and discovered 14 cis-acting elements related to growth, adversity, and resistance (Fig. 6).
The results showed that the 18 JrCAD promoters exhibited large differences in the number, order, and type of cis-acting elements, although some of them were located on the same evolutionary branch. Two cis-acting elements were related to growth, with the circadian element involved in circadian and the CAT-box in regulation of hyphal tissue. Seven hormone-related cis-acting components were identified, including ABRE (related to abscisic acid), CGTCA-motif (related to methyl jasmonate), GARE-motif (gibberellin-responsive element), TCA-element and W-box (related to salicylic acid), TGA-element (part of an auxin-responsive element), and ERE (related to ethylene). Five relevant cis-acting elements of stress response were found, such as ARE (hypoxic stress element) and WUN-motif (wound-responsive element), LTR (low-temperature responsiveness), TGACG-motif (MeJA-responsiveness), MBS (MYB binding site involved in drought-inducibility). The JrCAD1 promoter contains the largest number of cis-acting elements with 23, while the JrCAD13 promoter contains the fewest with only 7, and JrCAD13 has only cis-acting elements associated with resistance in addition to ERE. Combined with the phylogenetic tree, although the JrCADs are on the same branch, there are still some differences in the distribution of cis-acting elements in the promoter region. For instance, JrCAD18 has elements involved in the response to LTR. Unlike JrCAD18, JrCAD17 has motifs that bind to Myb and TCA or are involved in the response to auxin (TGA).

Interaction Network Prediction and Collinearity Analysis of JrCADs
We constructed an interaction network of the JrCAD proteins based on their relationship with homologs in Arabidopsis (Fig. 7). The results showed that each AtCAD protein corresponds to at least one JrCAD protein, the relationship network contains 14 nodes and 39 groups, and the CAD genes are associated with the pathways of phenylpropanoid metabolic process, which are coregulated with the C4H, F5H, CCR, and HCT. To further investigate the evolutionary mechanisms of JrCADs, collinearity analysis was performed on Juglans regia and Juglans microcarpa, Arabidopsis thaliana (Fig. 8). The collinearity relationships with J. regia showed that between J.

Fig. 6
Predicted cis-elements in JrCAD promoters. The scales below provide the length of promoters microcarpa and A. thaliana, we identified pairs of homologs: 14 between J. regia and J. microcarpa and 6 between J. regia and A. thaliana, indicating that CAD genes were highly conserved among J. regia. Multiple colinear gene pairs were found in which the collinearity relationships between JrCAD16 and JrCAD8 with AtCAD6 and between JrCAD17 and JrCAD18 with AtCAD5 were more homologous to each other.

Expression Patterns of JrCADs Genes and qRT-PCR Validation
In our previous studies, we found that 'Xinlu' walnut completes kernel expansion and development at 50 DAF, and then the endocarp starts hardening . The process of lignin deposition was observed by phloroglucinol staining, and we found that the external lignin-producing parts of the walnut endocarp were stained red, the stained part was called scleritic-hue, and the unstained part was pulpy-hue. Internal staining of walnut fruits revealed that the central vascular bundle was stained at 50 DAF, while the endocarp area was not stained, indicating that the endocarp hardening process had not yet started. The base of the walnut showed staining at 57 DAF when the lignification process began. At 78 DAF, a large amount of staining began to appear. pulpy-hue parts stop developing and cannot form the endocarp, therefore leading to the phenomenon of walnut exposure (Fig. 9).
To investigate the expression pattern of JrCADs, endocarp transcriptome data from five developmental stages of the walnut fruit hardening process were extracted Fig. 7 Interaction network of JrCAD proteins. The interaction network of the JrCAD proteins was constructed using homologous CAD proteins in Arabidopsis, and the colors of the lines indicate different data sources. Proteins are represented by network nodes. Edges represent associations of proteins, blue balls represent cinnamyl alcohol dehydrogenase, red balls represent proteins required for the lignin synthesis process, and green balls represent phenylpropanoid metabolic process. Black lines represent coexpression  Lignin staining of 'Xinlu' walnut at different stages of hardening. Staining of kernel sections of walnuts at four developmental stages by phloroglucinol. Lignin was stained red by phloroglucinol under acidic conditions, and the kernels were placed in a mixture of phloroglucinol-alcohol-hydrochloric acid staining solution for 5 to 6 min and observed (Yu et al. 2019). Sch means scleritic-hue, Puh means pulpyhue, Cvb means carpel vascular bundle, Sc means seed coat, Rlu means region of lignin undeposition, Rlb means region of lignin deposition in this study. The results showed that the 18 JrCAD genes could be divided into 3 groups based on their expression patterns (Fig. 10). Five genes in group A were only expressed in one tissue at one time during endocarp development. During scleritic-hue, high expression of JrCAD3 and JrCAD5 was found at 71 DAF, and during pulpy-hue, high expression of JrCAD2 and JrCAD13 was found at 78 DAF. The 7 JrCAD genes in group B were almost always expressed. Group B was further divided into 2 branches, among which JrCAD11 and JrCAD15 showed higher expression in the scleritic-hue than in pulpy-hue at 64-78 DAF; however, the two genes showed lower expression in scleritic-hue than pulpy-hue after 78 DAF. For the other 5 JrCADs of group B, the expression was consistently higher in the scleritic-hue than the pulpy-hue. Group C genes exhibited tissue and time specificity; JrCAD8 and JrCAD16 at 71 DAF and JrCAD10 at 78 DAF showed higher expression in the pulpy-hue than in the scleritic-hue, JrCAD7 and JrCAD16 showed lower expression in the scleritic-hue than the pulpy-hue before 78 DAF, but they showed higher expression in the scleritic-hue than the pulpy-hue after 78 DAF.
The relative expression of ten JrCAD genes (JrCAD1, JrCAD4, JrCAD6-8, JrCAD14-18) with differential expression in the transcriptome data was further validated by quantitative real-time PCR (Fig. 11). The qRT-PCR results of the ten genes were generally consistent with the transcriptome data, although some difference was found between the two methods. In the construction of the expression pattern clustering of JrCAD based on transcriptome data, JrCAD1, JrCAD4, JrCAD14, JrCAD17, JrCAD18 were clustered in the same class; likewise, the qRT-PCR results showed that the expression of these genes was consistently higher in the scleritic-hue parts than in the pulpy-hue parts during the hardening process of the walnut endocarp. Among them, JrCAD18 (12-fold) was highly expressed in the Fig. 10 Heatmap of the JrCADs expressed differently in five developmental periods of the endocarp P and S, respectively, pulpy-hue and scleritic-hue scleritic-hue parts at 92 DAF compared to the pulpy-hue parts at 50 DAF. JrCAD genes have different expression patterns at different times. JrCAD8 possessed high expression at 64 DAF in the scleritic-hue state; subsequently, its expression in the pulpy-hue state was higher than that in the scleritic-hue state. JrCAD7 and JrCAD16 were more highly expressed in pulpy-hue at 71-78 DAF. These genes are differentially expressed in the endocarp and could be studied as a priority.

Discussion
A complete endocarp is very important for the development of walnut; however, some thin-shelled walnut varieties ('Wen185', 'Xinxin 2') show the phenomenon of bare shells during the growth process due to environmental factors and artificial management (Ning et al. 2018). There is a change in shell thickness in Iran due to environmental changes (Arzani et al. 2008). And the 'Xinlu' walnut variety was almost not affected by climate condition, the bared phenotype is relatively stable . We believe that the hardening process of the endocarp is the Fig. 11 qRT-PCR analysis of JrCAD genes in different endocarp development periods. The horizontal coordinate is the day after full blooming, and the vertical coordinate is the relative expression transformation of the pulpy-hue to the scleritic-hue, and the completion of this process results in the formation of a complete shell, while the lack of lignin affects the transformation process of the pulpy-hue, resulting in the formation of a bare shell. We cloned and analyzed the genes of complete walnut varieties and incomplete walnut varieties in Xinjiang province during lignin synthesis and found that there were differences in the expression of PAL, 4CL, and other genes, which affected lignin synthesis (Guo et al. 2018. Therefore, the study of genes involved in lignin synthesis in the walnut endocarp can further clarify the mechanism of walnut shell development. CAD converts cinnamaldehyde to cinnamyl alcohol during lignin synthesis, forming three lignin precursors (Cheng et al. 2017). Downregulation of genes involved in the synthesis of lignin precursors often results in a decrease in lignin synthesis or a change in the structure of lignin (Anderson et al. 2015) CAD genes are differentially expressed in the same tissue or in different parts of the same tissue (Shin et al. 2007), thus, the study of CAD function is believed to contribute to understanding plant growth and development. In recent years, CAD genes have been identified in the genomes of many plant species (Kim et al. 2004;Guillaumie et al. 2007;Ma 2010;Chao et al. 2014;Park et al. 2018;Wang et al. 2021). In this study, 18 CAD genes were identified from the walnut genome, the number of which was close to those of flax (16) (Preisner et al. 2018) and strawberry (14) , suggesting that CAD genes generally exist as a gene family in plants. The phylogenetic tree, gene structure and conserved motif analysis revealed that the JrCADs could be classified into three groups, which is similar to those in rice (Park et al. 2018) and Populus (Barakat et al. 2009), indicating that they may have a close evolutionary relationship ). Using the interaction network, we identified some JrCADs that are homologous to AtCAD proteins, these CAD proteins, which are associated with Arabidopsis, would have been biased by the functional roles produced by these Arabidopsis during evolution. The collinearity analysis results revealed that 14 and 6 genes of J regia, respectively, have collinear gene pairs with corresponding genes from J. microcarpa and A. thaliana, indicating that these genes may belong to the same ancestor (Kong et al. 2017). Based on the above results, further analysis of these genes can be performed.
To clarify the expression pattern of JrCADs, we analyzed the cis-acting elements of the JrCAD promoter region. The results revealed the presence of regulatory elements associated with phytohormones or resistance in all CAD members. In previous reports, the CAD activity and lignin content increased under the action of ethylene during the storage process of beans (Xie et al. 2020). With increasing MeJA treatment time, the expression of genes related to lignin synthesis, such as CAD, increased (Concha et al. 2013). Interestingly, the promoter sequences of JrCAD17 and JrCAD18 do not have ethylene response elements, whereas they have cis-acting elements related to salicylic acid and auxin and response components to low temperature. Salicylic acid was reported to promote lignin synthesis in plants to increase resistance to pathogenic bacteria (Napoleão et al. 2017), and it could also affect plant xylem development by responding to growth hormone signals (Gallego-Giraldo et al. 2011;Qu et al. 2021). In addition, lignin deposition is differentially regulated at low temperatures in pericarp and kernel tissues (dos Santos et al. 2015). And heat treatment could retard the low temperature-induced increase in cinnamyl alcohol dehydrogenase (CAD) activity, which in turn affects lignification (Meng et al. 2021). With a low light conditions, CAD gene expression is regulated and positively correlated with increasing lignin concentration . This finding indicates that JrCAD genes are regulated by different environmental and internal factors and play different functional roles in the lignin biosynthesis.
It has been shown that CAD genes are involved in lignin synthesis in different tissues of plants or in different conditions of the same tissue. For instance, TaCAD is highly expressed in the stem, while it showed low expression in the root of wheat (Ma 2010). In the same tissue, the expression of SbCAD is higher in the basal part of the stem than in the upper part of the stem (Tsuruta et al. 2007). The diversity and tissue specificity of expression profiles can provide further insight into the functional differences that exist in the CAD gene family. Analysis of the expression pattern of JrCAD in the endocarp by transcriptome data and qRT-PCR revealed that most of the JrCADs were differentially expressed between the scleritic-hue and pulpy-hue, indicating that JrCADs are directly or indirectly involved in the endocarp developmental process. The role of lignin in plant development has been confirmed in model plants, and analysis of homology relationships and gene expression profiles based on phylogenetic trees could provide important information for gene function studies. In the phylogenetic tree, JrCAD17 and JrCAD18 are close to AtCAD4 and AtCAD5 in group II, among which AtCAD5 plays a major role in lignin synthesis and plant defense processes in Arabidopsis (Sibout et al. 2005). qRT-PCR analysis showed that the JrCAD17 and JrCAD18 genes were highly expressed in the scleritic-hue, indicating that the two genes may play important roles in the process of endocarp hardening. JrCAD8, JrCAD16, AtCAD7, and AtCAD8 belong to group I, and the expression patterns of AtCAD7 and AtCAD8 are similar to those of AtCAD4 and AtCAD5 (Tavares et al. 2000;Kim et al. 2007). In our study, JrCAD8 and JrCAD16 were more highly expressed in the pulpy-hue than in the scleritic-hue after 71 DAF, and it is hypothesized that these two genes act in the same expression pattern on the endocarp as AtCAD7 and AtCAD8. The expression of the CAD gene is related to tissue aging. In our study, JrCAD4 had a continuous upward trend of expression during endocarp sclerosis. EjCAD1 in loquat has high expression during ripening (Shan et al. 2008), and CmCAD2 and CmCAD5 in melon were still highly expressed until harvest (Jin et al. 2014).
In this study, we identified 18 gene members of the JrCADs and obtained their expression data. The gene characteristics, evolutionary tree family characteristics, gene structure, protein interaction network prediction and collinearity analysis of CADs were analyzed, and their expression during endocarp sclerosis in walnut was analyzed by combining transcriptome sequencing and qRT-PCR. The CAD genes are generally involved in the developmental process of walnut lignin synthesis and have some synergistic regulatory mechanism. The study of the walnut CAD gene family provides candidate genes for lignin synthesis during endocarp development in walnut and provide a theoretical basis for functional validation such as transgene overexpression and gene silencing.