Introduction

Akebia quinata (Thunb.) Decne., a dicotyledonous Lardizabalaceae Akebia Decne plant (Yang et al. 2022), is a traditional Chinese herbal medicine; the medicinal part is its dry stem, which is widely distributed in Shanxi, Anhui, Hubei, and Hunan (Liu and Wang 2011; Fu 2000; Li et al. 2010). The picking method has developed from wild picking to artificial cultivation (Zheng and Jin 2021). There are many medicinal active ingredients in Akebia quinata plants, such as calceolarioside B (Ou et al. 2015), These compounds have a variety of pharmacological effects, such as antibacterial, anti-inflammatory, anti-tumor, antioxidant, and immunomodulatory effects (Jiang et al. 2017; Julião Lde et al. 2009; Chu et al. 2009; Zhu et al. 2009; Ablise et al. 2008; Chen et al. 2009).

Phenylethanoid glycosides (PhGs) are glycosides composed of caffeic acid, phenylethanol aglycone, and glycosyl; hence, PhGs are also called phenylpropanoid glycosides (Wang et al. 2022). Calceolarioside B is a phenylpropanoid (Wang et al. 2014); moreover, there are other phenylpropanoids in Akebia quinata plants, such as caffeic acid and esculetin (Liu et al. 2020). Phenylpropanoids are naturally occurring compounds composed of benzene ring and three straight-chain carbon (C6-C3) groups; generally, they have a phenol structure and, thus are phenolic substances, including simple phenylpropanoid, coumarins, flavonoids, lignans, and lignins (Gu et al. 2012). The predominant parent sugar nucleus of phenylethanoid compounds is β-D-glucopyranose (Gao et al. 2015); during biosynthesis, most of these compounds are transformed from aromatic amino acids such as phenylalanine and tyrosine derived from shikimic acid, through a series of reactions such as deamination and hydroxylation. Phenylpropanoid compounds mainly exist in dicotyledonous plants such as Magnoliaceae and Rosaceae, and are found in traditional medicinal plants such as Rehmannia glutinosa (Wang et al. 2022), Scrophularia ningpoensis (Li et al. 2020), and Forsythia suspense (Peng et al. 2021). Phenylpropanoid metabolism is the most common pathway in plant secondary metabolism. Moreover, 4CL is a key enzyme in the biosynthesis of plant flavonoids and lignin, which is involved in the synthesis of a variety of secondary metabolites; it plays an important role in plant growth and development.

In recent years, with the development and application of high-throughput sequencing technology, RNA-Seq has been gradually applied in the study of a variety of medicinal plants. In the early stage, our research group completed transcriptome sequencing and database construction of Akebia trifoliate (Qian et al. 2022) and Langmei (Ou et al. 2020), mined the key enzyme genes of Akebia trifoliata triterpenoid saponins synthesis, and analyzed the expression of key enzyme genes of citrate synthesis in Langmei fruits. Presently, Akebia quinata research is mostly conducted in terms of quality evaluation and pharmacological action (Li 2018; Liu et al. 2012); nevertheless, the molecular understanding of Akebia quinata phenylpropanoid biosynthesis pathway has not been reported.

Based on the four tissue samples of Akebia quinata (flowers, leaves, roots, and stems), we applied high-throughput sequencing technology for transcriptome analysis and used qRT-PCR for data verification to mine the related genes and key enzymes of the Akebia quinata phenylpropanoid biosynthesis pathway, which laid a theoretical foundation for further exploration and development of phenylpropanoid biosynthesis and biology.

Materials and methods

Plant material and RNA extraction

In this study, the fresh Akebia quinata was selected as the experimental material and was harvested from Dashushan, Hefei City, Anhui Province on March 26, 2021. Four tissues (flowers, leaves, roots, and stems) were collected and identified as Akebia Decne Lardizabalaceae by Associate Professor Jinmei Ou of Anhui University of Chinese Medicine. Fresh plant samples were washed several times with ultra-pure water and then dried using a filter paper. Four tissue types were placed into 10 mL centrifuge tubes, instantly frozen in liquid nitrogen, and then stored in an ultra-low temperature refrigerator at −80 °C until their RNA extractions. The flowers, leaves, roots, and stems from three replicates were pooled together. The total RNA was extracted from each tissue using an RNA kit.

cDNA library construction and sequencing

The mRNA enrichment method was used to purify qualified total RNA with magnetic beads using Oligo (dT) and enrich the mRNA with a polyA tail. An appropriate amount of breaking reagent was added to the resulting mRNA sample and segmented under high temperature conditions. A strand of cDNA was synthesized using the broken mRNA as a template. Subsequently, the two-strand cDNA was synthesized by configuring the two-strand synthesis reaction system, and the two-strand cDNA was purified and recovered using the kit; the sticky end was repaired, base “A” was added to the 3′ end of the cDNA and connected the sequencing connector, and fragment size was selected. Finally, the appropriate fragment length for PCR amplification was selected to obtain the cDNA library. Constructed libraries were examined using Agilent 2100 Bioanalyzer and ABI StepOnePlus real-time PCR System (Zhao et al. 2020; Shan et al. 2020a; Shi et al. 2019). After passing the test, Illumina HiSeq sequencing platform (BGI, Shenzhen, China) was used to conduct high-throughput sequencing analysis on the complete cDNA libraries of the four tissues of the qualified Akebia quinata.

De novo transcriptome assembly and gene functional annotation

After completing the transcriptome sequencing, Trimmomatic 0.36 was used to filter out reads containing joints (joint contamination), unknown base N content greater than 5%, and low quality reads to obtain high-quality reads. Clean reads with high quality were assembled using Trinity 2.0.6 software (Kim et al. 2017), and unigenes were obtained using Tgicl software for clustering and redundancy removal (Pertea et al. 2003), which were used as reference sequences for subsequent analysis. Using NCBI Blast+ (Pirooznia et al. 2008) software, unigenes NCBI nucleic acid sequence database (NT, https://www.ncbi.nlm.nih.gov/blast/db), NCBI non-redundancy protein sequence databases (NR, https://www.ncbi.nlm.nih.gov/blast/db), protein orthologous database (KOG, http://www.ncbi.nlm. nih.gov/KOG), the Swiss-Prot protein sequence databases (SwissProt, http://ftp.ebi.ac.uk/pub/databases/swissprot), the Kyoto encyclopedia of genes and genomes (KEGG, http://www.genome.jp/kegg), multi-sequence alignment were matched with the annotation database of protein family domain constructed by hidden Markov model (PFAM, http://pfam.Xfam.org). According to the annotation information of the NR library, the similarity between the transcription sequences of species and related species, as well as the functional information of homologous sequences were checked, and the genetic ontology (GO, http://geneontology.org) information of related unigenes was matched with Blast2GO 2.5.0. E < 1 × 10–5 was set to obtain the corresponding functional annotations and classification information of unigenes (Shan et al. 2020b).

Analysis of the structural characteristics of 4CL

The 4CL gene was translated using Translate tool online software, and the complete opening reading frame (ORF) sequence was screened. The protein multiple sequence alignment was implemented using DNAMAN and Clustalx software; the 4CL secondary structure was predicted by SOPMA, and NCBI-CDD database was used to predict conservative domains of 4CL. The SWISS-MODEL (https://swissmodel.expasy.org/) online system and PyMOL software were used to simulate 4CL tertiary structure.

Differentially expressed gene analysis

High-quality readings for each sample were checked using Bowtie2 (version 2.2.5) software (Langmead and Salzberg 2012) against the genome sequences of Akebia quinata, and the gene and transcript expression levels of each organ of Akebia quinata were calculated using the RSEM (version 1.2.8) software and TPM formula (Li and Dewey 2011). The standard expression of each unigene (fragments per kilobase of exon model per million mapped fragments, FPKM) was obtained. Differentially expressed genes (DEGs) were analyzed and compared between the stem and other tissues (flowers, leaves, and roots), using Poisson distribution method. Functional annotation and KEGG pathway enrichment analysis were performed on DEGs. By default, DEGs were those with false discovery rate (FDR) ≤ 0.001 and multiple of difference > 2.

Analysis of transcription factors (TFs) encoding genes

Genes with the ability of encoding TFs were predicted, and the TF families of genes were classified. The ORF of unigenes was detected by Getorf (EMBOSS version 6.5.7.0) (Rice et al. 2000); the ORF was aligned to the TF protein domain using Hmmsearch (version 3.0); and the competent unigenes were then identified based on the TF family features described by PlantTFDB (Mistry et al. 2013).

Analysis of the key gene expression levels in phenylpropanoid biosynthesis using qRT-PCR

To validate the Akebia quinata transcriptome data sets, the expression of key enzyme genes for phenylpropanoid biosynthesis was determined by quantitative PCR. According to the principles of primer design, genes involved in phenylpropanoid biosynthesis and primers for reference genes (actin) were designed using Primer Premier (version 5.0). After reverse transcription PCR, each PCR tube was prepared in a 15 μL mixture volume containing diluted cDNA (2 μL), 2.5 μM gene primer (upstream + downstream) (1.5 μL), 2 × qPCR mixture (7.5 μL), and Water Nuclease-Free (4 μL), under the following conditions: pre-denaturation at 95 °C for 30 s, then 40 cycles of denaturation at 95 °C for 15 s, and annealing or extension at 60 °C for 30 s; three technical replicates were performed for each sample. The relative expression level of each selected unigene was data-normalized with the actin gene (Un 4699), and the relative expression levels of each unigene were statistically analyzed using the 2−ΔΔCt calculation method (Livak and Schmittgen 2001).

Results

Transcriptome sequencing and data assembly

Through transcriptome sequencing of Akebia quinata, 56.49 Gb of data were obtained. The clean reads Q20 and Q30 values for each sample were greater than 98.38% and 94.54%, respectively, indicating that the transcriptome data were highly reliable. A total of 123,576 unigenes were obtained after assembly and de-redundancy using Trinity software, with an average sequence length of 1328 bp; the length of N50 was 2083 bp, and the content of GC was 40.65% (Table 1). The unigenes obtained from the assembled Akebia quinata sequence were mainly distributed in 200–500 bp (Fig. 1), accounting for 69.23% of the total this indicated that the integrity of unigenes was good, and thus the unigenes could be functionally annotated.

Table 1 Quality evaluation of unigenes from the transcriptome data of Akebia quinata
Full size table
Fig. 1
figure 1

Unigenes length distribution of Akebia quinata assembly

Full size image

Functional classification and expressed overview of unigenes

Among the 123,576 unigenes, 86,789, 72,122, 67,481, 69,642, 69,196, 64,795, and 69,624 unigenes were finally annotated by the seven functional databases NR, NT, Swiss-Prot, KEGG, KOG, PFAM, and GO, respectively. Moreover, 38,486 unigenes were annotated by all databases, accounting for 31.14%; 90,273 unigenes were annotated by any database, accounting for 73.05% (Table 2).

Table 2 Unigenes functional annotation results
Full size table

Data related to Akebia quinata were compared with those of the KOG database, and their functions were classified and annotated. The results showed that 69,196 unigenes were annotated to the KOG database. Furthermore, 13,666 unigenes were annotated in “general function prediction only”, followed by 8304, 6027, 4407, and, 3428 unigenes annotated in “signal transduction mechanisms”, “posttranslational modification, protein turnover, and chaperones”, “transcription”, and “carbohydrate transport and metabolism”, respectively (Fig. 2). In addition, based on the 86,789 unigenes annotated in the NR database, the proportion of different species related to the annotated information was calculated; the species distribution map is shown in Fig. 3. The results showed that 28,939 (33.34%) unigenes were mapped to Macleaya cordata (Lin et al. 2018), which had the highest degree of similar sequence matching; moreover, 17,447 (20.10%), 9,614 (11.08%), 3,506 (4.04%), 2,186 (2.52%), and 25,097 (28.92%) unigenes were mapped to Nelumbo nucifera (Chen et al. 2019), Aquilegia coerulea (Singh et al. 2020), Vitis vinifera (Nassiri-Asl and Hosseinzadeh 2016), Nyssa sinensis (Yang and Ji 2017) and others, respectively. These species provide a reference for the sequence annotation of Akebia quinata.

Fig. 2
figure 2

KOG functional classification of unigenes of Akebia quinata

Full size image
Fig. 3
figure 3

Species comparison of homologous sequences of Akebia quinata

Full size image

The unigene results of all comparisons from the NR database were further annotated in the GO database, and statistically annotated to the classification map of three aspects of GO: biological process, cellular component, and molecular function (Fig. 4). According to the information in the GO database, 69,624 unigenes were annotated: in the biological process, 40,916 unigenes are mainly concentrated in “cellular process” and 32,079 unigenes are concentrated in “metabolic process”; in the cellular component, 56,303 unigenes focused on “cellular anatomical entity” and 28,792 unigenes focused on “intracellular”; in the molecular function, 46,671 unigenes are mainly clustered in “binding” and 44,856 unigenes clustered in “catalytic activity”.

Fig. 4
figure 4

Classification of GO function of unigenes of Akebia quinata

Full size image

In the transcriptome data of flowers, leaves, roots, and stems, there were 85,866, 90,701, 91,838, and 91,624 unigenes, respectively, of which 47,630, 45,680, 42,297 and 43,776 unigenes had FPKM ≤ 1 with low expression; 29,349, 34,707, 39,225 and 36,137 unigenes had FPKM = 1–10 with medium expression; and 8887, 10,314, 10,316 and 11,711 unigenes had FPKM ≥ 10 with high expression (Fig. 5A). According to the analysis of Graphpad Prism 8.0.2 and FPKM gene expression level, there were significant differences in gene expression levels between stems and flowers of Akebia quinata; there were no significant differences between stems and leaves and between stems and roots. The overall expression level of unigenes was highest in stems, followed by roots, leaves, and flowers (Fig. 5B).

Fig. 5
figure 5

Gene expression in four tissues of Akebia quinata. A The expression distribution of unigenes in four tissues (flowers, leaves, roots and stems). The X-axis represents the sample name, and the Y-axis represents the number of genes. Different gene expression levels are expressed by color shade: FPKM ≤ 1 is low expression, FPKM 1–10 is medium expression, FPKM ≥ 10 is high expression. B The degree of dispersion of the data distribution can be observed by showing the unigenes expressed in the four organizations through the box chart. The X-axis are subtissue types of Akebia quinata, the Y-axis is log10 (FPKM + 1)

Full size image

Identification of genes related to phenylpropanoid biosynthesis by KEGG enrichment analysis

The gene sequence of Akebia quinata involved in the metabolic pathway was analyzed using the KEGG database. The data showed that 69,642 unigenes were annotated in the Akebia quinata transcriptome, including 3036 unigenes in cellular processes, mainly concentrated in transport and catabolism; 4255 unigenes in environmental information processing, mainly concentrated in signal transduction; 14,273 unigenes in genetic information processing, mainly in translation; 39,377 unigenes in metabolism, mainly concentrated in global and overview maps; and 2942 unigenes in organismal systems, mainly concentrated in environmental adaptation (Fig. 6). In the transcriptome data of Akebia quinata, 1805 unigenes were located in other pathways related to secondary metabolic biosynthesis. KEGG pathway enrichment results showed that 1174 unigenes were located in phenylpropanoid biosynthesis, the highest proportion of unigenes (Fig. 7).

Fig. 6
figure 6

KEGG functional classification of unigenes of Akebia quinata

Full size image
Fig. 7
figure 7

Classification of other secondary metabolic biosynthetic pathways in Akebia quinata

Full size image

By analyzing the transcriptome data of Akebia quinata, we found 13 enzymes involved in phenylpropanoid biosynthesis, including PAL, C4H, 4CL, HCT, C3’H, CCOAOMT, CCR, F5H, COMT, CAD, CHS, CHI, and F3H; of these, 45 and 27 unigenes encoded CCR and CAD, respectively (Table 3). The complete protein sequences of 13 key enzyme genes are shown in the appendix, as detailed in the supplementary material. Based on the screening conditions (FPKM of unigenes > 1 and distance from the complete coding frame < 50), 10, 5, 12, 8, 1, 5 unigenes were found to encode PAL, C4H, 4CL, HCT, C3’H, CCOAOMT enzymes, respectively in the simple phenylpropanoid pathway; furthermore, 29, 2, 8, 13, 8, 7, 3 unigenes encoded CCR, F5H, COMT, CAD, CHS, CHI, F3H enzymes, respectively. The relative expression levels of these key enzyme genes in various tissues were shown in the form of heat map; in the phenylpropanoid biosynthesis pathway of Akebia quinata, C4H, CCR, and F3H were mainly highly expressed in flowers; PAL and CHI were mainly highly expressed in leaves; CHS was mainly highly expressed in roots; and 4CL, CCOAOMT, F5H, COMT, and C3’H were mainly highly expressed in stems (Fig. 8).

Table 3 Distribution of unigenes in four tissues during phenylpropanoid biosynthesis of Akebia quinata
Full size table
Fig. 8
figure 8

Biosynthetic route of phenylpropanoid in Akebia quinata. F, flowers; L, leaves; R, roots; S, stems. The expression level of key enzyme genes, green means low expression of genes, on the contrary, high expression of red. Cluster of transcripts and Unigene are abbreviated as “CL” and “Un”, respectively

Full size image

Expression analysis of structural characteristics of 4CL, a key enzyme

Based on the analysis of Akebia quinata transcriptome data, 12 unigenes encoding 4CL enzyme were found, and their nucleotide sequences were translated into protein sequences and compared with those of the NCBI Blast database; 10 unigenes were confirmed to have complete ORF. The self-sequences of the 10 unigenes were compared using DNAMAN, and the results showed a sequence identity of 72.97%, which indicated a high sequence homology of 4CL; in addition, the spatial structure similarity of unigenes constructed using SWISS MODEL was high. Among them, the CL 12789-2 unigene with the highest expression level of complete reading frame (length of ORF = 1629 bp) encoded 542 amino acids. Our team selected the 4CL (Un 64258, 76.27%) enzyme with the highest sequence homology to the template protein to describe its secondary and spatial structure. Secondary structure prediction showed that the 4CL enzyme consisted of 29.78%, 20.59%, 6.99%, and 42.65% alpha helix, extension chain, beta turn, and irregular crimp, respectively (Fig. 9). Based on the prediction of the spatial structure of the 4CL enzyme, two conserved peptide motifs were found in its amino acid sequence, including the conserved AMP binding domain (P187-V199, PYSSGTTGLPKGV) and the conserved GEICIRG domain (386–392 amino acid sequence) (Fig. 10). The peptide motif BoxII, the conservative region of GEICIRG, is reported to be absolutely conserved in the 4CL enzyme, and its central C residue is considered to be directly involved in the catalytic process (Stuible and Kombrink 2001).

Fig. 9
figure 9

Prediction of the secondary structure of 4CL protein. Blue: alpha helix; red: extension main chain; green: beta turn; purple: irregular crimp

Full size image
Fig. 10
figure 10

4CL enzyme spatial structure model of Akebia quinata. A The spatial structure model of 4CL (Un 64258; template: 3a9u.1.A; sequence identity: 76.27%); B peptide motif Box I is conserved AMP binding domain, expressed by green sphere; peptide motif Box II is GEICIRG conservative domain, expressed by blue sphere

Full size image

Analysis of DEGs

The results showed that the four tissues of Akebia quinata co-expressed 76,419 unigenes, and the specific expressions of flowers, leaves, roots, and stems were 10,048, 4350, 12,802, and 4451 unigenes, respectively (Fig. 11A). Using flowers as control, 36,494 DEGs were found in stems, with 18,102 and 18,392 up- and down-regulated genes, respectively. Using leaves as control, 11,285 DEGs were found in stems, including 5264 and 6021 up- and down-regulated genes, respectively. Moreover, 19,771 DEGs were found in stems, when roots were used as control, with 8225 and 11,546 up- and down-regulated genes, respectively (Fig. 11B).

Fig. 11
figure 11

Gene expression and number of DEGs in four tissues of Akebia quinata. A unigenes Venn diagram of four tissues; B number of DEGs: compared with flowers, leaves and roots, the number of DEGs with high or low expression levels in stems was defined as “up-regulated” and “down-regulated”, respectively. a: Starch and sucrose metabolism; b: pyruvate metabolism; c: porphyrin and chlorophyll metabolism; d: photosynthesis—antenna proteins; e: phenylpropanoid biosynthesis; f: pentose and glucuronate interconversions; g: glycolysis/gluconeogenesis; h: glycine, serine and threonine metabolism; i: glycerophospholipid metabolism; j: galactose metabolism; k: flavonoid biosynthesis; l: fatty acid metabolism; m: cysteine and methionine metabolism; n: cyanoamino acid metabolism; o: citrate cycle (TCA cycle); p: carbon metabolism; q: carbon fixation in photosynthetic organisms; r: biosynthesis of amino acids; s: amino sugar and nucleotide sugar metabolism; t: 2-oxocarboxylic acid metabolism

Full size image

The results of KEGG enrichment analysis of DEGs in Akebia quinata showed that 8213 DEGs in flowers versus stems were annotated to 134 metabolic pathways; 3075 DEGs in leaves versus stems were annotated to 127 metabolic pathways; 4971 DEGs in roots versus stems were annotated to 128 metabolic pathways; and the first three metabolic pathways with more enriched genes were the same, including carbon metabolism, biosynthesis of amino acids, and starch and sucrose metabolism (Fig. 12A–C).

Fig. 12
figure 12

KEGG pathway enrichment of differential genes. The ordinate is the first 20 metabolic pathways of enrichment of differential genes, and the abscissa represents the enrichment factors; A–C represent the significant enrichment pathways of the number of differential genes in stem compared with flowers, leaves and roots, respectively

Full size image

In Akebia quinata, 506 DEGs in stems versus flowers were annotated to the phenylpropanoid biosynthesis pathway (ko00940), of which 250 DEGs were specifically up-regulated in stems; 247 DEGs in stems versus leaves were annotated to the phenylpropanoid biosynthesis pathway, of which 109 DEGs were specifically up-regulated in stems; and 436 DEGs in stems versus roots were annotated to the phenylpropanoid biosynthesis pathway, of which 185 DEGs were specifically up-regulated in stems.

Analysis of TFs related to phenylpropanoid biosynthesis

TFs have been shown to regulate various secondary metabolic processes. In the Akebia quinata transcriptome database, 30 unigenes involved in the phenylpropanoid biosynthesis pathway were identified, including 24, 2, 2, 1, and 1 unigenes belonging to the MYB, Tify, NAC, FHA, and GRAS families, respectively.

Validation of RNA-Seq data with qRT-PCR

In order to verify the differential gene expression and transcriptome data in different tissues, six unigenes involved in phenylpropanoid biosynthesis were selected for qRT-PCR analysis. The expression level of CL 12789-2 (4CL) and CL 14341-24 (CCR) were highest in flowers; similarly, the expression level of Un 17571 (COMT) was highest in leaves, whereas those of CL 4383-2 (COMT), Un 14954 (CAD), and CL 233-17 (PAL) were highest in stems. The qRT-PCR results were consistent with those from the transcriptome data (Fig. 13).

Fig. 13
figure 13

Expression analysis of six unigenes encoding enzymes involved in phenylpropanoid biosynthesis. Relative expression of A CL 4383-2 (COMT), B Un 14954 (CAD), C CL 12789-2 (4CL), D CL 14341-24 (CCR), E Un 17571 (COMT), and F CL 233-17 (PAL) was analyzed by qRT-PCR using the actin gene (Un 4699) as the reference gene for normalization. Blue bars represent qRT-PCR data, red lines represent FPKM values. The data represents the average ± standard error of three repetitions. The left Y-axis is the relative expression level of genes detected by qRT-PCR, and the right Y-axis is the FPKM values of RNA-Seq data

Full size image

Discussion

Phenylpropanoids, one of the most important bioactive components of Akebia quinata, have good pharmacological properties. In this study, we identified genes involved in Akebia quinata phenylpropanoid biosynthesis, thereby facilitating the understanding of phenylpropanoid biosynthesis and regulatory mechanism, which are important in promoting the comprehensive development and utilization of Akebia quinata plants.

Furthermore, we performed transcriptome sequencing on four tissues of Akebia quinata using the BGISEQ platform, and 56.49 Gb data were obtained. After reassembly and de-redundancy, 123,576 unigenes were derived. The total length, average length, N50, and GC content were 164,180,390 bp, 1328 bp, 2083 bp, and 40.65%, respectively. Subsequently, the obtained unigene was compared to those of seven functional databases NR, NT, SwissProt, KOG, KEGG, GO, and PFAM; finally, 86,789 (70.23%), 72,122 (58.36%), 67,481 (54.61%), 69,196 (55.99%), 69,642 (56.36%), 69,624 (56.34%), and 64,795 (52.43%) unigenes obtained functional annotations in the abovementioned databases, respectively. The obtained RNA-Seq data provide high-quality annotation assembly for the following analysis results, and the clean reads Q30 of each sample was greater than 94.54%, indicating that the Akebia quinata transcriptome data were reliable.

The annotation results of the NR database showed that Akebia quinata without reference genome had the highest genetic relationship with Macleaya cordata of Papaveraceae, and a large number (69,624) of unigenes were assigned and annotated to the GO database, which indicated that there was significant diversity in the transcriptome of different Akebia quinata tissues. A total of 69,642 unigenes were annotated in the KEGG database, and 2267 unigenes were involved in other secondary metabolic biosynthesis pathways, of which 1174 unigenes were enriched in the phenylpropanoid biosynthesis pathway. Our analysis showed that 13 key enzymes were involved in the phenylpropanoid biosynthesis pathway, and 5 key enzymes were mainly highly expressed in stems. Based on the expression levels of genes encoding key enzymes, we found that 4CL (CL 12789-2) and CCR (CL 14341-24) had the highest expression levels in flower tissue; COMT (Un 17571) had the highest expression level in leaf tissue; and COMT (CL 4383-2), CAD (Un 14954), and PAL (CL 233-17) had the highest expression level in stem tissue. Six genes—4CL (CL 12789-2), CCR (CL 14341-24), COMT (Un 17571), COMT (CL 4383-2), CAD (Un 14954), and PAL (CL 233-17)—were detected using the qRT-PCR method, and the authenticity and reliability of our transcriptional data were verified.

The 4CL gene exists in many plants, such as Lycium chinense (Zhang et al.) and Arabidopsis thaliana (Zhai et al. 2017). The ORF length of the 4CL gene of Akebia quinata is 1629 bp, which encodes 542 amino acids. Through BLAST comparative analysis, the amino acid encoded by this gene is closely related to that in Thalictrum thalictroides of Ranunculaceae. Based on the prediction and analysis of the secondary structure of Akebia quinata protein, it was found that the secondary structure of Akebia quinata 4CL protein was composed of 29.78%, 20.59%, 6.99%, and 42.65% alpha helix, extension chain, beta turn, and irregular crimp, respectively. The polypeptide chain of 4CL protein is further coiled or folded on the basis of various secondary structures to form a regular three-dimensional spatial structure. The 4CL amino acid sequence of Akebia quinata, akin to that of other plants, contains the AMP binding domain “PYSSGTTGLPKGV” and the conserved sequence “GEICIRG” (Gao et al. 2020; Tian et al. 2017). The expression of the 4CL gene, which has a certain specificity in different tissues of plants, is highest in stems, followed by flowers, roots, and leaves; the relative expression of 4CL in roots, leaves, and flowers of Osmanthus sp. showed that flowers were larger than roots and leaves (Mu et al. 2016).

Gene expression level analysis showed a significant difference in gene expression between stems and flowers, and the overall expression level of unigene was highest in stems. Compared with leaves and roots, stems showed the highest number of up-regulated genes compared with flowers; hence, stems demonstrated higher gene expression levels than flowers. There were 250 up-regulated genes enriched in phenylpropanoid biosynthesis in DEGs of stems versus flowers, which was more than those in stems versus leaves and stems versus roots. These up-regulated genes may be more related to the higher value of traditional Chinese medicine in Akebia quinata stems.

Conclusion

We analyzed the transcriptome for four Akebia quinata tissues (flower, leaf, root, and stem), and identified many genes related to phenylpropanoid biosynthesis using RNA-Seq; moreover, we excavated the key enzymes involved in phenylpropanoid biosynthesis and analyzed the DEGs. Some key enzyme genes were verified using qRT-PCR, and the results were consistent with the RNA-Seq data. The study findings have deepened the understanding of the metabolic pathways related to Akebia quinata phenylpropanoid biosynthesis, and thus are helpful to further analyze the molecular mechanism of regulating Akebia quinata phenylpropanoid biosynthesis

Author contribution statement

CQ, YY, and RW contributed to the sample collection and sequencing. CQ performed the transcriptome data analysis and wrote the manuscript. BH, DW and JO revised the manuscript. JO provided financial support. All authors read and approved the manuscript.