Molecular Mechanisms of Phenylpropane-Synthesis-Related Genes Regulating the Shoot Blight Resistance of Bambusa pervariabilis × Dendrocalamopsis grandis

Bambusa pervariabilis × Dendrocalamopsis grandis shoot blight caused by Arthrinium phaeospermum is a fungal disease that has affected a large area in China in recent years. However, it is not clear which genes are responsible for the disease resistance of B. pervariabilis × D. grandis. Based on the analysis of transcriptome and proteome data, two genes, CCoAOMT2 and CAD5, which may be involved in disease resistance, were screened. Two gene expression-interfering varieties, COF RNAi and CAD RNAi were successfully obtained using RNAi technology. Quantitative real-time fluorescence (qRT-PCR) results showed that CCoAOMT2 gene, CAD5 gene and seven related genes expression was down-regulated in the transformed varieties. After inoculating pathogen spore suspension, the incidence and disease index of cof-RNAi and cad-RNAi transformed plants increased significantly. At the same time, it was found that the content of total lignin and flavonoids in the two transformed varieties were significantly lower than that of the wild-type. The subcellular localization results showed that both CCoAOMT2 and CAD5 were localized in the nucleus and cytoplasm. The above results confirm that the CCoAOMT2 and CAD5 genes are involved in the resistance of B. pervariabilis × D. grandis to shoot blight through regulating the synthesis of lignin and flavonoids.


Introduction
Bambusa pervariabilis × Dendrocalamopsis grandis is the product of the female parent B. pervariabilis and the male parent D. grandis. B. pervariabilis × D. grandis is widely used in papermaking and in returning farmland to forest, which also promotes the construction of ecological barriers in the middle and lower reaches of the Yangtze River. However, in recent years, it has been found that B. pervariabilis × D. grandis shoot blight caused by Arthrinium phaeospermum (Corda) M.B. Ellis has caused damage to more than 3000 hm 2 , leading to large economic losses [1] and greatly threatening the process of returning farmland to forest and the construction of ecological barriers in the Yangtze River Basin. At present, in addition to physiological research on the disease, molecular biology research has also made some progress. However, there is a lack of research on the molecular mechanism of the disease resistance of B. pervariabilis × D. grandis.
Disease resistance breeding is an effective method to improve the disease resistance of forest trees. It has the advantages of good economic benefits and environmental friendliness. The conventional methods of disease resistance breeding are systematic breeding and cross-breeding, but the breeding period is long. In recent years, with the development of molecular biology and tissue culture technology, bioengineering methods have been gradually applied to disease resistance breeding. Breeding resistant cultivars using resistance genes is currently the most effective and economical strategy for controlling crop diseases [2]. However, in bamboo plants, especially in B. pervariabilis × D. grandis, this kind of research is very few. Our previous studies have shown that many genes involved in cell wall composition synthesis, redox reactions, signal transduction and synthesis of secondary metabolites were significantly different in B. pervariabilis × D. grandis after pathogen infection. Twenty-one candidate genes for blight resistance, such as pme53, CCoAOMT2, cad5, pod, gdsl-ll and Myb4l, were found [3,4]. In higher plants, the intermediate products produced by the glycolysis pathway and the pentose phosphate pathway can be converted to phenylalanine through the shikimic acid pathway [5]. Phenylalanine, as the initial reaction substrate of the phenylpropanoid biosynthesis pathway in plants, can finally synthesize intermediate products such as ferulic acid, cinnamic acid and chlorogenic acid through the phenylpropanoid metabolism pathway, and then produce lignin, flavonoids, isoflavones and other substances [6]. Phenylpropane compounds are important components of plant defense in plants. Changes in many phenylpropane compounds and related enzymes involved in the synthesis of phenylpropane compounds are widely involved in plants' responses to biological and abiotic stresses in the process of plant growth and development, and help plants resist external physical and biological invasion [7][8][9]. Both CCoAOMT (caffeoyl coenzyme, a 3-O-methyltransferase) and CAD (Cinnamyl alcohol deaminase) are important enzymes in the phenylpropanoid biosynthesis pathway in plants ( Figure S1). Studies have shown that the CCoAOMT gene and the CAD gene are not only involved in lignin and flavonoid synthesis in the phenylpropanoid biosynthesis pathway but also related to plant resistance [10,11]. When plants are under the conditions of drought [12], low temperature [12,13], salt stress [14,15], hormone induction [15] and pathogen infection [16,17], CCoAOMT and CAD participate in the mechanisms of plant stress responses. However, research on the CCoAOMT gene and the CAD gene in bamboo generally focuses on gene cloning and expression analysis. There are few studies on the function of the CCoAOMT gene and the CAD gene in bamboo, although there is a report on the function of the CAD gene in Bambusa multiplex [18]. In particular, there is no report on the function of the CCoAOMT and the CAD genes in B. pervariabilis × D. grandis.
The methods of gene function identification are generally divided into transgenic technology, gene knockout technology and gene silencing technology. RNAi (RNA interference), also known as post-transcriptional gene silencing (PTGS), is characterized by specificity. An advantage of RNAi is that it can accurately identify and cut the mRNA of the corresponding gene, but has no interference with irrelevant genes, and has a significant interference effect on the target gene, which also confirms the signal amplification effect in the process of RNA interference [19][20][21][22][23][24][25]. In plants, using RNAi, dsRNA can be designed and synthesized and introduced into plants. Through the RNA interference mechanism in plants, the key pathogenic genes of pathogens can be silenced, so as to improve disease resistance, some examples of it are the resistance by RNAi against some viruses, fungi, pests, nematodes, parasitic weeds [26][27][28][29], etc.
In this study, the disease-resistant candidate genes of B. pervariabilis × D. grandis were screened by early transcriptome and proteome sequencing technology. On this basis, RNAi technology was used to interfere with the expression of disease-resistant candidate genes of B. pervariabilis × D. grandis, so as to verify the function of disease-resistant genes of B. pervariabilis × D. grandis, and finally to determine the main disease-resistant genes and analyze the internal causes of resistance to shoot blight of B. pervariabilis × D. grandis on the molecular level. This study provides a basis for in-depth research on the disease-resistance mechanism of B. pervariabilis × D. grandis and the molecular mechanism of pathogen-host interaction, as well as the development of new strategies for sustainable and effective control of forest branch and stem diseases.

Gene Sequence Analysis and Protein Structure Prediction
Using the NCBI online blast nucleic acid sequence analysis tool, the results from the database were compared. They showed that the CDS sequences of CCoAOMT2 and CAD5 genes are 100% similar to the genes annotated with the agreed function of Phyllostachys edulis (FP094556.1) and Bambusa multiplex (GU9855522.1). According to the comparison results, the annotation of different families, genera and model plants were selected as the phase protein gene sequence using MEGA 5.1. The phylogenetic tree was constructed using the neighbor-joining algorithm ( Figure 1). MEME was used to analyze the conserved sequences of the CCoAOMT2 and CAD5 genes. According to the principle of E-value < 0.05, 18 and 20 important domains of CCoAOMT2 and CAD5 genes were predicted, respectively ( Figure 2). The amino acid residues of the CCoAOMT2 and CAD5 genes were 248 and 368, respectively, the total average hydrophilicity was −0.238 and 0.015, respectively, the theoretical isoelectric points were 5.21 and 6.06, respectively, and the predicted molecular weights were 27.84 kd and 39.47 kd, respectively. The proteins translated by the CCoAOMT2 and CAD5 genes were modeled by SWISS-MODEL. The comparison results showed that the proteins that most matched CCoAOMT2 and CAD5 in the database were 5 kva.1.A, which was annotated as sorghum caffeyl-Coenzyme A methyltransferase, and 2cf5.1. B, which was annotated as Arabidopsis cinnamyl dehydrogenase. The consistency is higher than 70%, and its function can be preliminarily judged to be consistent ( Figure 3).

Gene Sequence Analysis and Protein Structure Prediction
Using the NCBI online blast nucleic acid sequence analysis tool, the results from the database were compared. They showed that the CDS sequences of CCoAOMT2 and CAD5 genes are 100% similar to the genes annotated with the agreed function of Phyllostachys edulis (FP094556.1) and Bambusa multiplex (GU9855522.1). According to the comparison results, the annotation of different families, genera and model plants were selected as the phase protein gene sequence using MEGA 5.1. The phylogenetic tree was constructed using the neighbor-joining algorithm ( Figure 1). MEME was used to analyze the conserved sequences of the CCoAOMT2 and CAD5 genes. According to the principle of E-value < 0.05, 18 and 20 important domains of CCoAOMT2 and CAD5 genes were predicted, respectively ( Figure 2). The amino acid residues of the CCoAOMT2 and CAD5 genes were 248 and 368, respectively, the total average hydrophilicity was −0.238 and 0.015, respectively, the theoretical isoelectric points were 5.21 and 6.06, respectively, and the predicted molecular weights were 27.84 kd and 39.47 kd, respectively. The proteins translated by the CCoAOMT2 and CAD5 genes were modeled by SWISS-MODEL. The comparison results showed that the proteins that most matched CCoAOMT2 and CAD5 in the database were 5 kva.1.A, which was annotated as sorghum caffeyl-Coenzyme A methyltransferase, and 2cf5.1. B, which was annotated as Arabidopsis cinnamyl dehydrogenase. The consistency is higher than 70%, and its function can be preliminarily judged to be consistent ( Figure 3).

B. pervariabilis × D. grands Total RNA Extraction and Reverse Transcription
RNA mass was detected using 1% agarose gel electrophoresis ( Figure S2). The two bands of 28S and 18S from top to bottom are clearly visible in the figure, and although the 5.8S was slightly dispersed, it did not affect the follow-up experiment. The RNA extraction was complete, and the RNA extraction product can be used for subsequent experiments. The reverse transcription kit of Trans Biotechnology Co., Ltd. (Beijing, China) was used to reverse the total RNA of B. pervariabilis × D. grands to cDNA.

Construction and Transformation of RNAi Expression Plasmids of CCoAOMT2 and CAD5 Genes
Using the pHANNIBAL vector as the intermediate vector, the forward and reverse target fragments were inserted into both ends of the intron sequence of the pHANNIBAL vector, respectively ( Figures S3-S7), and then the complete fragment was obtained by enzyme digestion, as shown in Figure 4a. The pCAMBIA1301-35SN expression vector was constructed to interfere with the expression of two candidate disease resistance genes (CCoAMT2, CAD5). The positive colonies obtained after E. coli transformation sequencing with Zf/r-COF and Zf/r-CAD primers are shown in Figure 4b. The correctly sequenced recombinant plasmid DNA was transferred into Agrobacterium tumefaciens, and transgenic B. pervariabilis × D. grandis was cultivated according to the methods of Yuan [30]. The tissue culture process is shown in Figure 5. The positive plants were screened by PCR with hygromycin primer hyg501 f/R (Table S1). The results are shown in Figure 4b.

B. pervariabilis × D. grands Total RNA Extraction and Reverse Transcription
RNA mass was detected using 1% agarose gel electrophoresis ( Figure S2). The two bands of 28S and 18S from top to bottom are clearly visible in the figure, and although the 5.8S was slightly dispersed, it did not affect the follow-up experiment. The RNA extraction was complete, and the RNA extraction product can be used for subsequent experiments. The reverse transcription kit of Trans Biotechnology Co., Ltd. (Beijing, China) was used to reverse the total RNA of B. pervariabilis × D. grands to cDNA.

Construction and Transformation of RNAi Expression Plasmids of CCoAOMT2 and CAD5 Genes
Using the pHANNIBAL vector as the intermediate vector, the forward and reverse target fragments were inserted into both ends of the intron sequence of the pHANNIBAL vector, respectively ( Figures S3-S7), and then the complete fragment was obtained by enzyme digestion, as shown in Figure 4a. The pCAMBIA1301-35SN expression vector was constructed to interfere with the expression of two candidate disease resistance genes (CCoAMT2, CAD5). The positive colonies obtained after E. coli transformation sequencing with Zf/r-COF and Zf/r-CAD primers are shown in Figure 4b. The correctly sequenced recombinant plasmid DNA was transferred into Agrobacterium tumefaciens, and transgenic B. pervariabilis × D. grandis was cultivated according to the methods of Yuan [30]. The tissue culture process is shown in Figure 5. The positive plants were screened by PCR with hygromycin primer hyg501 f/R (Table S1). The results are shown in Figure 4b.

Expression Analysis of CCoAOMT2 Gene and CAD5 Gene
The expression of the CCoAOMT2 gene in the cof-RNAi transformed strain after 8 weeks of transplanting was significantly lowered compared with that of wild-type B. pervariabilis × D. grandis ( Figure 6). This shows that the expression of the CCoAOMT2 gene in B. pervariabilis × D. grand was inhibited after the transfer of the CCoAOMT2 gene conserved region fragments of sense fragments and antisense fragments, and the expression of the CAD5 gene of the cad-RNAi transformed plant after 8 weeks of transplanting was also significantly lowered compared with wild-type B. pervariabilis × D. grandis, compared with wild-type B. pervariabilis × D. grandis, indicating that the expression of CAD5 genes in B. pervariabilis × D. grandis was significantly inhibited after the transgression of the gene conserved region of CAD5 gene fragments of sense fragments and antisense fragments.

Expression Analysis of CCoAOMT2 Gene and CAD5 Gene
The expression of the CCoAOMT2 gene in the cof-RNAi transformed strain after 8 weeks of transplanting was significantly lowered compared with that of wild-type B. pervariabilis × D. grandis ( Figure 6). This shows that the expression of the CCoAOMT2 gene in B. pervariabilis × D. grand was inhibited after the transfer of the CCoAOMT2 gene conserved region fragments of sense fragments and antisense fragments, and the expression of the CAD5 gene of the cad-RNAi transformed plant after 8 weeks of transplanting was also significantly lowered compared with wild-type B. pervariabilis × D. grandis, compared with wild-type B. pervariabilis × D. grandis, indicating that the expression of CAD5 genes in B. pervariabilis × D. grandis was significantly inhibited after the transgression of the gene conserved region of CAD5 gene fragments of sense fragments and antisense fragments.

Disease Resistance Level of Transgenic Plants
Taking the wild-type B. pervariabilis × D. grandis as the control, it was transplanted 3 months later with A. Phaeospermum spore suspension (10 6 cfu/mL) and treated with cof-RNAi, cad-RNAi, and the wild-type B. pervariabilis × D. grandis, respectively. The incidence rate was calculated after 25 days of inoculation and the disease index was calculated ( Table 1). The incidence of the wild-type B. pervariabilis × D. grandis was the lowest in the statistical control group, which was 16.7 ± 4.7%, and the lowest in the three groups of the

Disease Resistance Level of Transgenic Plants
Taking the wild-type B. pervariabilis × D. grandis as the control, it was transplanted 3 months later with A. Phaeospermum spore suspension (10 6 cfu/mL) and treated with cof-RNAi, cad-RNAi, and the wild-type B. pervariabilis × D. grandis, respectively. The incidence rate was calculated after 25 days of inoculation and the disease index was calculated ( Table 1). The incidence of the wild-type B. pervariabilis × D. grandis was the lowest in the statistical control group, which was 16.7 ± 4.7%, and the lowest in the three groups of the wild-type B. pervariabilis × D. grandis disease index, which was 9.2 ± 3.1. The CCoAOMT2 gene RNAi interferer strain cof-RNAi had the highest incidence of 86.7 ± 4.7%, which was significantly different from the incidence of the control group, and the disease index was higher, 42.5 ± 2.0, which was significantly different from the control group. The incidence of the CAD5 gene RNAi interferer strain cad-RNAi was the second-highest, at 83.3 ± 9.4%, which was significantly different from that of the control group, and the disease index was 55.8 ± 5.1, which was significantly higher than the disease index of the control group. There were no significant differences in the incidence and disease index between the CCoAOMT2 gene RNAi strain cof-RNAi and the CAD5 gene RNAi interferer cad-RNAi.
The data are the average of three sets of repetitions; data in the same column followed by different lowercase letters indicate significant differences among different varieties after inoculation by the LSD test (p < 0.05).
After 25 days of inoculation, on a small number of the leaves of wild-type B. pervariabilis × D. grandis there appeared a small number of disease spots, the color of the branches was a healthy green, the dead branch phenomenon was not observed, and there were new tillers growing. After cof-RNAi and cad-RNAi inoculation for 25 days, spots appeared on more than 75% of the leaves, and some of the leaf spots covered the entire leaf, resulting in yellow wilt on the leaves, and some leaves finally falling off; further, brown spots appeared on the branches, and the branches lost water and yellowed (Figure 7). At the same time, there were obvious differences in the plant height, tillering number, leaf size and other indicators between wild-type B. pervariabilis × D. grandis, the cof-RNAi transformation plant, and the cad-RNAi transformation plant. The wild-type B. pervariabilis × D. grandis plants were taller, had more tillers, and their leaf size was significantly larger than that of cof-RNAi and cad-RNAi.  Lignin content in the stem tissues of wild-type B. pervariabilis × D. grandis, cof-RNAi and cad-RNAi were determined by lignin spectrophotometry ( Table 2). The wild-type B.

Total Lignin Analysis and Total Flavonoids 2.6.1. Total Lignin Analysis
Lignin content in the stem tissues of wild-type B. pervariabilis × D. grandis, cof-RNAi and cad-RNAi were determined by lignin spectrophotometry ( Table 2). The wild-type B. pervariabilis × D. grandis lignin content was the highest, at 4.064 mg/g, while the cof-RNAi transform plant had the lowest lignin content of 2.727 mg/g, thus recording a decrease of 33%, and the lignin content of the cad-RNAi transform plant was slightly higher, at 3.152 mg/g, thus recording a decrease of 22%. Overall, compared with the lignin content of wild-type B. pervariabilis × D. grandis, the lignin content of the RNAi transforming plant and the cad-RNAi transforming plant was not significant. The data are the average of three sets of repetitions; data in the same column followed by different lowercase letters indicate significant differences among different varieties after inoculation by the LSD test (p < 0.05).

Total Flavonoids Analysis
Flavonoid content in the stem tissues of wild-type B. pervariabilis × D. grandis, cof-RNAi and cad-RNAi were determined by spectrophotometry ( Table 3). The wild-type B. pervariabilis × D. grandis flavonoids content was the highest, at 5.63 mg/g, while the cad-RNAi transform plant had the lowest flavonoids content of 3.01 mg/g, and the flavonoids content of the cof-RNAi transform plant was slightly higher, at 3.84 mg/g. Overall, compared with the flavonoid content of wild-type B. pervariabilis × D. grandis, the flavonoid content of the RNAi transforming plant and the cad-RNAi transforming plant was not significant. The data are the average of three sets of repetitions; data in the same column followed by different lowercase letters indicate significant differences among different varieties after inoculation by the LSD test (p < 0.05).

Correlation Analysis
Correlation analysis was performed on the relationships between the gene expression, lignin content and flavonoid content and the incidence, disease index, lignin content data and flavonoid content, respectively ( Table 4). The correlation between the expression of the gene of interest and the incidence, disease index, lignin content, and flavonoid content and the correlation between the lignin content, flavonoid content and the incidence and disease indexes were significant. It was noted that there was no significant correlation between lignin content and flavonoid content. The above results show that CCoAOMT2 and CAD5 are involved in the synthesis of lignin and flavonoids, which plays a role in the defense mechanism of B. pervariabilis × D. grandis against shoot blight.
Insights, Inc. (Cascade, CO, USA) (Table S1). In the cof-RNAi transform strain, the expression of seven related proteins showed a certain downregulation at the transcription level compared with that of wild-type B. pervariabilis × D. grandis (Figure 8), of which hydroxycinnamoyl transferase expression was the most significantly downregulated, followed by Glycosyltransferase, cinnamoyl CoA reductase, cytosolic aldehyde dehydrogenase RF2D and 4-coumarate CoA ligase, and the relative expression of the remaining two genes was slightly downregulated, of which peroxidase 4 had the lowest decrease in relative expression, which was not significantly different from the control group. In the cad-RNAi transform strain, the expression of seven related proteins in the cad-RNAi transforming plant showed a certain downregulation compared with the expression of wildtype B. pervariabilis × D. grandis, of which hydroxycinnamoyl transferase was the most significantly downregulated, followed by 4-coumarate CoA ligase, cinnamoyl CoA reductase and glycosyltransferase, and the remaining three genes were slightly and relatively downregulated. Among them, the expression of peroxidase 4 was consistent with the expression of wild-type B. pervariabilis × D. grandis in the control group: there was no upregulation or downregulation, and there was almost no difference.

Subcellular Localization of CCoAOMT2 Gene and CAD5 Gene
After the sequence verification of the complete coding region gene amplification fragments of CCoAOMT2 and CAD5 were correct, the gel of interest was recovered and the recovered fragments were electrophoretically detected on a 1% agarose gel ( Figure S8). The figure was marked as 1 where the lane was the complete CDS region of CCoAOMT2, To verify the expression of the seven interplay proteins at the transcription level, seven pairs of primers were designed using the software Oligo V. 7, Molecular Biology Insights, Inc. (Cascade, CO, USA) (Table S1). In the cof-RNAi transform strain, the expression of seven related proteins showed a certain downregulation at the transcription level compared with that of wild-type B. pervariabilis × D. grandis (Figure 8), of which hydroxycinnamoyl transferase expression was the most significantly downregulated, followed by Glycosyltransferase, cinnamoyl CoA reductase, cytosolic aldehyde dehydrogenase RF2D and 4-coumarate CoA ligase, and the relative expression of the remaining two genes was slightly downregulated, of which peroxidase 4 had the lowest decrease in relative expression, which was not significantly different from the control group. In the cad-RNAi transform strain, the expression of seven related proteins in the cad-RNAi transforming plant showed a certain downregulation compared with the expression of wild-type B. pervariabilis × D. grandis, of which hydroxycinnamoyl transferase was the most significantly downregulated, followed by 4-coumarate CoA ligase, cinnamoyl CoA reductase and glycosyltransferase, and the remaining three genes were slightly and relatively downregulated. Among them, the expression of peroxidase 4 was consistent with the expression of wild-type B. pervariabilis × D. grandis in the control group: there was no upregulation or downregulation, and there was almost no difference.

Subcellular Localization of CCoAOMT2 Gene and CAD5 Gene
After the sequence verification of the complete coding region gene amplification fragments of CCoAOMT2 and CAD5 were correct, the gel of interest was recovered and the recovered fragments were electrophoretically detected on a 1% agarose gel ( Figure S8). The figure was marked as 1 where the lane was the complete CDS region of CCoAOMT2, and the enzyme cutting site of 6 bp and the protective base of 2 bp were attached at each end of the fragment, leading to a total of 763 bp. Further, the 2 lane was marked as the complete CDS region of CAD5, and the enzymatic resination site and 2 bp of the protective base were attached at each end of the fragment, leading to a total of 1124 bp. From the figure it can be seen that the CCoAOMT2 recovered fragment was located between 750 bp and 1000 bp of the 2Kmarker, and the CAD5 recovered fragment was located between 1000 bp and 2000 bp of the 2Kmarker indicating that the amplification product was correct; this can be used for the next experiment. CCoAOMT2 and CAD5 gene amplification products and pCAMBIA1300-EGFP-MCS plasmids were transferred to Agrobacterium EHA105 after double-cutting recovery by KpnI. and XbaI., respectively, and the fusion plasmids were named cof-EGFP and cad-EGFP, respectively. After 2-3 d of incubation on YEP plates, positive colonies were selected to expand the culture, and these were injected into the back of Tobacco Ben's leaves for 1 month of incubation. After 48-72 h of incubation under suitable conditions, images were recorded by laser confocal microscopy imaging, and the results found that the no-loaded tobacco fluorescence signal of the positive control conversion pCAMBIA1300-EGFP-MCS appeared randomly in the whole plant cells, and at the same time transferred to the markers localized to the membrane and nucleus, respectively. Further, the tobacco localization infected with cof-EGFP and cad-EGFP fusion plasmid Agrobacterium was found to be similar (Figure 9). The fusion proteins cof-EGFP and cad-EGFP coincided with the nuclear localization marker fluorescence signal, but not with the membrane localization marker fluorescence signal, indicating that CCoAOMT2 and CAD5 are localized to the nucleus and cytoplasm.

Discussion
In this experiment, the clone sequences of the two genes of interest were highly similar to the genes annotated as CCoAOMT and CAD in the database, and most of these genes came from bamboo; in particular, the most similar sequences were from bamboo, which was consistent with the results of previous studies [31,32]. In a conservative regional analysis of the sequence, both gene sequences were found to be very conserved, and the identified conserved motif covered almost the entire sequence. In particular, an S-adenosyl methionine binding site, which is its functional domain [33], was identified in the CCoAOMT2 gene, which initially determined that the gene expression product could perform methyl transfer functions. This result is consistent with that found for Eucalvptus urophylla [34] and Dendrocalamus sinicus [31]. In the protein tertiary structure analysis, the model results matched by the comparison of the sequences showed that the two sequences

Discussion
In this experiment, the clone sequences of the two genes of interest were highly similar to the genes annotated as CCoAOMT and CAD in the database, and most of these genes came from bamboo; in particular, the most similar sequences were from bamboo, which was consistent with the results of previous studies [31,32]. In a conservative regional analysis of the sequence, both gene sequences were found to be very conserved, and the identified conserved motif covered almost the entire sequence. In particular, an S-adenosyl methionine binding site, which is its functional domain [33], was identified in the CCoAOMT2 gene, which initially determined that the gene expression product could perform methyl transfer functions. This result is consistent with that found for Eucalvptus urophylla [34] and Dendrocalamus sinicus [31]. In the protein tertiary structure analysis, the model results matched by the comparison of the sequences showed that the two sequences matched the highest degree of caffeyl-CoA methyltransferase and cinnamyl alcohol dehydrogenase, respectively, and the consistency reached more than 70%, meaning that the model can be considered to have fully represented the real results.
The phenomenon of RNAi is widely confirmed in plants [35][36][37], where there are two different interfering mechanisms at the molecular level; one is the siRNA-producing pathway, and the other is the miRNA-producing pathway [38]. In this experiment, by constructing a ds-RNA transgenic plasmid, the B. pervariabilis × D. grandis transformed strain was successfully obtained. The expression of the CCoAOMT2 and CAD5 genes of interest was analyzed by q-PCR technology, and the expression of both genes in the transformed strain was significantly inhibited. The results showed that there was an RNA interference mechanism through the siRNA pathway in B. pervariabilis × D. grandis, which was similar to that found in carnation [39].
In the study of anti-blight disease in melon, it was found that CmCAD2 and CmCAD5 played a certain anti-disease role. In this experiment, the incidence of CCoAOMT2 and CAD5 gene expression was significantly higher than that of the control group, reaching 86.7%, 83.3%, 70.0% and 66.6% higher than that of the control group, respectively. At the same time, the disease index was also significantly increased to 42.5 and 55.8, respectively, which were higher than the control group values of 33.3 and 46.6, respectively. These results were similar to the previous research results [16], which also showed that CCoAOMT2 and CAD5 genes were involved in B. pervariabilis × D. grandis's resistance to the shoot blight caused by the dark spore node rhombus. It was found that the AtCAD4 and AtCAD5 genes in the AtCAD gene family of Arabidopsis thaliana were significantly expressed after 6 h of Phytophthycosis infestation with Arabidopsis [40], while the other genes were not significantly different in the family, and the AtCAD4 and AtCAD5 genes were closely related to plant resistance in the CAD gene family. In the present experiment, the disease index of the CAD5 gene expression interferer was higher than that of the CCoAOMT2 gene expression interferer, which was consistent with the above-mentioned result. Due to the interference of CAD5 gene expression, which is more significantly related to disease resistance, plant disease resistance is reduced, showing a higher disease index. The CCoAOMT2 gene was more sensitive to the RNAi mechanism, so it recorded a higher incidence in B. pervariabilis × D. grandis populations.
Lignin biosynthesis is mainly regulated by three functional enzymes, namely PAL, C4H, and 4CL in the amphetamine pathway, COMT, CCoAOMT, and F5H associated with monomer synthases, and CAD and CCR [41], which are downstream regulatory enzymes for lignin synthesis. In this experiment, the lignin content of the plants inhibited by expression inhibition of the two genes was determined by interfering with the expression of the lignin biosynthetic enzymes CCoAOMT and CAD, respectively, and it was found that the lignin content of the transformed plant decreased to varying degrees [42][43][44][45][46][47], especially the CCoAOMT expression interference transform strain. This may have been due to the fact that the expression of the CCoAOMT2 gene is more sensitive to RNA interference. This result is consistent with the more significant decrease recorded in the expression of the CCoAOMT2 gene of interest. In general, after the interference in CCoAOMT2 and CAD5 gene expression, the lignin content decreased significantly compared with the control group, which proved that the CCoAOMT2 and CAD5 genes were involved in regulating the synthesis of lignin in B. pervariabilis × D. grandis, which was consistent with the previous research results. Flavonoids are important substances for plants to resist stress under stress and have a variety of bioactive functions. In this experiment, it was found that the content of flavonoids decreased significantly, and the incidence rate and disease index of B. pervariabilis × D. grandis also increased significantly. It shows that flavonoids are related to the resistance of B. pervariabilis × D. grandis to shoot blight.
Hydroxycinnamoyl transferase (HCT) belongs to the plant acyltransferase family (BAHD family), which is able to catalyze the conversion of acyl-coenzyme A into a variety of downstream products [48]. In Arabidopsis, it was found that lignin synthesis was inhibited, and plant dwarfing was evident after interference with the expression of quinic acid/shikimic acid-hydroxycinnamyltransferase (HCT), suggesting that hydroxycinnamyltransferase was associated with lignin synthesis in plants [49]. It was also found that HCT has the characteristics of the broad spectrum of the reaction substrate, which can catalyze the acylation of a variety of substrates, including cinnamoyl CoA, p-coumarincoa CoA, caffeanoyl CoA, feruloyl CoA, and mustardyl CoA [50]. In this experiment, the expression of the CCoAOMT2 and CAD5 genes related to lignin synthesis was interfered with, resulting in blocked caffeyl CoA methylation and oxidation resistance of hydroxycinnamaldehyde in lignin synthesis, which led to the decrease in lignin content, while hydroxycinnamoyltransferase was positively correlated with lignin synthesis. Therefore, the expression of hydroxycinnamoyltransferase, a key enzyme of lignin synthesis, was downregulated after the interference of the expression of CCoAOMT2 and CAD5 genes.
The subcellular localization of the CCoAOMT2 and CAD5 proteins was predicted using a database prior to the trial, showing that the CCoAOMT2 and CAD5 proteins were most likely to be localized to the cytoplasm, but it was also observed that the localization results of the two genes were different in different plants [51][52][53][54]. The CDS cloning of the two genes in this experiment was then connected to a plasmid with EGFP for fusion expression. The results showed that CCoAOMT2 and CAD5 were localized to the cell membrane and nucleus, which was generally consistent with the prediction results, and there was no nucleus localization in the prediction results of CAD5. Therefore, the second fusion expression was tried, and the experimental results were consistent with the first experimental results. The two experiments thus proved the credibility of the experimental results. Previous results found for Kuerle Pear [55], in which CCoAOMT and CAD localization in the cytoplasm were observed, are consistent with the results of this experiment. However, it was noted that CCoAOMT2 and CAD5 were also located in the nucleus in addition to the cytoplasm, which was different from the above results; further, according to Liu [56], who studied the subcellular localization of the two metabolic enzymes of pear lignin, CCoAOMT2 and CAD5 were found to have the function of the transcription factor and could bind to DNA in the nucleus to regulate gene expression, so they were also located in the nucleus.
In conclusion, CCoAOMT2 and CAD5 may be involved in the resistance of B. pervariabilis × D.grandis to shoot blight as disease resistance genes. However, the interaction between target genes and pathogens at the protein level is not clear. In the future, we can continue to deeply explore the disease-resistance mechanism of CCoAOMT2 and CAD5 at the protein level.

Materials
Bambusa pervariabilis × Dendrocalamopsis grandis seedlings were obtained from the forest protection laboratory of Sichuan Agricultural University, China. A. phaeospermum was isolated by tissue isolation method [57] from diseased B. pervariabilis × D. grandis, the accession number of A. phaeospermum ITS in NCBI database is OK626768. The isolate was stored in China Forestry Culture Collection Center, numbered cfcc 86860 (http://www. cfcc-caf.org.cn/, accessed on 6 April 2007).

Gene Sequence Analysis, Protein Structure Prediction
The candidate gene transcriptome sequences were aligned in the NCBI database, and the phylogenetic tree was constructed using MEGA5.1's neighbor-joining algorithm [58]. The conserved domains of proteins were predicted by MEME [59]. The isoelectric points of proteins were analyzed by EXPASY [60]. The tertiary structures of proteins were analyzed by SWISS-MODEL [61].

Total RNA Extraction and cDNA Synthesis of B. pervariabilis × D. grandis
Fresh shoots of B. pervariabilis × D. grandis were ground in liquid nitrogen, and total RNA was extracted from tissue cells using the PlantTrol RNA extraction kit (TransGen, TransGen Biotech Co., Beijing, China). The concentration and integrity of RNA were determined by Nanodrop microspectrophotometer (NANODROP, ThermoFisher Scientific-CN, Shanghai, China) and agarose gel electrophoresis (DYY-6D, LIUYI, Beijing, China), respectively. EasyScript ® All-in-One First-Strand cDNA Synthesis SuperMix (TransGen) of TransGen Biotech Co. China, was used for PCR, reverse transcription into cDNA was carried out, and the results were kept on standby at −20 • C.

Agrobacterium Transformation and B. pervariabilis × D. grandis Tissue Culture
Transformation of E. coli DH5α took place using ligation product α Competent cells. The recombinant plasmid was introduced into Agrobacterium EHA105 competent cells by freeze-thaw method, and the agrobacterium that was identified as positive by PCR was impregnated with B. pervariabilis × D. grandis embryogenic callus [29]. The basic medium and plant hormone conditions used in the process of tissue culture were as follows: MS medium was added with 4 mg/L 2,4-D for callus induction, MS basic medium was added with 3 mg/L 6-BA and 3 mg/L KT for callus differentiation, MS basic medium was added with 2 mg/L NAA for rooting. Primer hyg501-f/r was designed for the positive transformants screening by PCR amplification of hygromycin resistance genes. The transformed strains with positive CCoAOMT2 and CAD5 gene expression interference were named cof-RNAi and cad-RNAi, respectively.

Quantitative Detection of Gene Expression of CCoAOMT2 and CAD5 in RNAi
Positive Transformed Vaccine by q-PCR RNA from wild-type B. pervariabilis × D. grandis and cof-RNAi and cad-RNAi was extracted and reverse transcribed into cDNA. Using the cDNA as the template, primers were designed from the specific fragments of CCoAOMT2 and CAD5 genes in B. pervariabilis × D. grandis, and real-time quantitative PCR (q-PCR) specific primers qCof-F/R, qCAD5-F/R were designed (Table S1). The primers were synthesized by Tsingke Biotechnology Co., Ltd. (Beijing, China). GAPDH and Actin were used as internal reference genes. TransScript ® Green One-Step qRT-PCR SuperMix (TransGen) was used for qRT-PCR. The q-PCR system consisted of the following: 10 µL qPCR SuperMix, 0.4 µL RT/RI Enzyme Mix, 0.4 µL Passive Reference Dye, 7.4 µL ddH 2 O, 0.4 µL F/R, 1 µL RNA. The qPCR was performed: at 95 • C for 5 min, 94 • C for 30 s, 94 • C for 5 s, and 60 • C for 30 s. The third step and the fourth step were repeated 40 times. Each treatment group underwent qPCR three times, and the average value was calculated. Data were analyzed by the 2 −∆∆Ct method [62].

Pathogen Inoculation and Disease Resistance Detection
The spores of A. phaeospermum cultured for 10 days were washed off with sterile water to prepare 10 5 CFU/mL spore suspension. A total of 30 strains of cof-RNAi, 30 strains of 6 cad-RNAi and 30 wild-type seedlings were selected, and these were inoculated with spore suspension by acupuncture method [63]. Each plant was inoculated with 3 nodal forks, and each place was inoculated with 5 mL spore suspension, which was kept wet for 12 h, and sterile water was used as the control. The incidence rate and disease index were calculated after 25 days of inoculation.
where I, D and T are the incidence, the number of diseased plants and the number of total plants, respectively.
where DI is the disease index, DN is the disease level, L is the number of plants at each level, TN is the total number of diseased plants, and M is the maximum disease incidence.

The Content of Total Lignin and Total Flavonoids
The content of lignin and flavonoids were determined by using the branches of the cof-RNAi, cad-RNAi transformed plants and wild-type B. pervariabilis × D. grandis as plant materials The total amount of lignin was determined by the acetyl bromide method. The specific operation is to dry the sample at 80 • C to constant weight, crush it, sieve it through 40 mesh, weigh it about 5 mg (recorded as W) in 10 mL glass test tube (always use a glass test tube, not EP tube) and record it according to the instructions of Suzhou Keming Biotechnology Co., Ltd. (Suzhou, China), China lignin content kit (COMIN). The total amount of lignin can be calculated according to the standard curve and measurement data provided by the kit. y = 0.0694x + 0.0068, R2 = 0.9889 Lignin (mg/g) = (∆A − 0.0068) ÷ 0.0694 × V(mL) × 10 − 3 ÷ W(g) × T = 0.0294 × (∆A − 0.0068) ÷ W(g) × T where V is the total reaction volume of 2.04 mL, W is the sample mass, and T is the dilution multiple. The principle is that flavonoids can form a stable complex with aluminum salt and obtain a stable characteristic absorption peak in the visible spectrum. The content of total flavonoids was determined according to the method of Liu et al. [64]. 0.5 g of fresh tissue of the stem was placed in 70% ethanol for ultrasonic treatment at 35 • C for 45 min, and its concentration was measured by colorimetry; Add 1 mL of solution and 5 mL of deionized water to a 25 mL test tube, add 1 mL of 5% (mass volume fraction) NaNO 2 to the reaction mixture and keep it at room temperature for 6 min; Then add 1 mL of 10% (mass volume fraction) AlCl 3 · 6H 2 O and keep it at room temperature for 6 min; Then add 1 mL of 1 mol/L NaOH and fix the final volume to 25 mL with deionized water. The absorbance of the solution was immediately read at 512 nm.
The target gene expression, incidence rate, disease resistance index, lignin content and flavonoid content were measured according to the appeal method, and the software graphpad prism V. 8.0, GraphPad Software, LLC (San Diego, CA, USA) was used. Pearson correlation coefficient was used for correlation analysis.

Expression Analysis of Related Genes
The STRING 11.5 (STRING: STRING: functional protein association networks (stringdb.org)) protein interaction database was used to analyze CCoAOMT2 and CAD5, and then seven interacting proteins were screened from the annotation information data of B. pervariabilis × D. grandis transcriptome, which were 4-coumarate CoA ligase isoform 2 (4CL. 2), hydrocinnamoyltransferase, cinnamoyl CoA reductase, cytosolic aldehyde dehydrogenase RF2D, glycosyltransferase, cinnamoyl CoA reductase 2 and peroxidase. Taking the cDNA of wild-type B. pervariabilis × D. grandis as the control and GAPDH and Actin genes, two of the plant housekeeping genes, as the internal reference, the seven predicted interaction proteins were named 1-7 in the order of appeal, corresponding to primers 1-F/R to 7-F/R in Table S1, and the fluorescence quantitative identification and analysis were carried out according to the method in 4.2.5.

Subcellular Localization of CCoAOMT2 and CAD5 Gene
Primers CCoAOMT2-GFP-F/R and CAD5-GFP-F/R (Table S1) were designed according to the transcriptome sequencing sequences of CCoAOMT2 and CAD5 genes. The coding sequences of CCoAOMT2 and CAD5 genes were amplified by PCR with the B. pervariabilis × D. grandis cDNA as the template, and the PCR products were recovered. Kpn I and Xba I were selected as restriction endonuclease sites and connected by T4 DNA ligase. The recovered target fragments were connected with pCAMBIA1300-EGFP-MCS and then introduced into E. coli DH5α competent cells, which were screened for positive clones with primer yz-F/R (Table S1) and sequenced. The correctly inserted plasmid was extracted and transferred into Agrobacterium EHA105 to infect tobacco [65]. After 48-72 h, observations were made under laser confocal microscope (FV3000, Olympus, Tokyo, Japan).