Introduction

Aconitum episcopale Leveille is an important medicinal plant of the genus Aconitum L. of Ranunculaceae family (Fig. 1A–C), commonly known as “Hei wu tou”, “Zi wu tou”, “Dula”, “Xiao hei niu”, and so on1. This species is distributed mainly in the northwest Yunnan and southwest Sichuan provinces of China, where it grows in 2400–3200 m mountains and shrubs at the altitudes2. There are about 350 species of Aconitum worldwide and 170 species in China. Among these, 76 are used in folk medicine3, and it has been reported that most of the Aconitum species are highly toxic. According to the Illustrated Guide of Medicinal Plants of the Bai Nationality, A. episcopale has been found to effectively cure rheumatic bone pain, bruise injury, traumatic hematoma, joint sprain, and other diseases. Furthermore, records of Flora of China and Medicine and Pharmacy of Yi Nationality have mentioned that the A. episcopale can detoxify, has anti-alcoholic property, and can also detoxify opium. According to data available and ethnic application proof, A. episcopale Leveille is the only one species of Aconitum plant that can relieve the poisoning caused by the Aconitum plants, and the chemical component responsible for its detoxifying potential was found to be episcopalisine4. More than 20 compounds have been isolated from the ethanolic extract of the roots of the A. episcopale, including episcopalisine, deacetylheterophylloidine, heterophylloiding, songorine, talatisamine, 14-acetyltalaisamine, talatisamine, 14-acetyltalatisamine, vilmorrianine C, 8-deacetylyunaconitine, indaconitine, sczukininc etc. by various research groups5,6,7,8,9.

Figure 1
figure 1

(A) The wild habitat of A. episcopale. (B) Plant of A. episcopale. (C) Crude drug from A. episcopale.

Full size image

The wild population of A. episcopale has reduced significantly due to its ruthless use and habitat destruction. Many adulterates of A. episcopale are available in the market, which could seriously endanger people's life and health. Because the accurate distinction is difficult in the morphological appearance of these species, a molecular method is urgently needed to distinguish A. episcopale from the other adulterated species. Chloroplasts are independent organelles in the plant cells, have a complete set of the genome, which is relatively conserved in the genetic composition, structure and contains more abundant mutation sites. These structural features, which allow chloroplast genomes to occupy a vital position in plant species' discrimination and evolutionary study, have been widely used as super barcodes for the species identification and phylogenetic studies10,11,12. As of February 2021, more than 5000 chloroplast genomes have been recorded in the GenBank database (https://www.ncbi.nlm.nih.gov/genbank/) of the National Center for Biotechnology Information (NCBI).

For the current study, we assembled and analyzed the cp genome sequence of A. episcopale using Illumina PE sequencing. At the same time, the genome was compared with other published cp genomes of the same family and genus. The results determined the evolutionary position of A. episcopale in the Ranunculaceae family, which provided an essential theoretical basis for molecular identification and phylogeny of A. episcopale.

Results

Genome sequencing and assembly

In our study, the DNA of A. episcopale was paired-end sequenced using Illumina Novaseq sequencing. A total of 45,118,066 raw reads and 6,767,709,900 raw bases were gained. The proportion of Q20 (nucleotides with quality values larger than 20 in reads) was 96.54%. As a result, we obtained a scaffold with its high quality of assembled 155,827 bp in length. Finally, the entire cp genome sequence of A. episcopale was deposited to the GeneBank (Accession Number: MZ189733.1).

General characteristics of chloroplast genomes from A. episcopale

The results showed that the cp genome of A. episcopale was 155,827 bp in length and presented a complete circular structure, including a pair of IR regions (52,436 bp), which divided the genome into an LSC region of 86,452 bp, and an SSC region of 16,939 (Fig. 2). The coding region (90,201 bp) accounted for 57.9% of the genome, and the intergenic region (65,626 bp) accounted for the remaining 42.1%. The total GC content of the cp genome was 38.09%, and the GC contents of the LSC, IR, and SSC regions were 36.2%, 43.0%, and 32.6%, respectively. Among the codons in the cp genome, the frequencies of adenine (A), thymine (T), guanine (G), and cytosine (C) were 47,849 (30.7%), 48,628 (31.2%), 29,146 (18.7%), and 30,204 (19.4%), respectively (Table 1). The A. episcopale chloroplast genome comprised of 132 genes, including 85 PCGs (seven duplicated genes), 37 tRNAs (seven duplicated genes), eight rRNAs (four duplicated genes), and two pseudogenes (rps19 and ycf1) (Table 2). Of the 132 identified genes, 20 genes contained introns, of which 14 genes (trnK-UUU, trnG-GCC, atpF, rpoC1, trnL-UAA, trnV-UAC, petB, petD, rpl16, rpl2, trnI-GAU, trnA-UGC, ndhA, ndhB) contained one intron and other two genes (ycf3 and clpP) possessed two introns. Eleven genes were situated in the LSC region, eight in the IRs region, and one in the SSC region (Table S1). Twenty genes contained Exon I, Exon II, Intron I, and their lengths were 6–775 bp, 35–1625 bp, and 493–2520 bp, respectively. The ycf3 and clpP genes located in the LSC region contained Exon I, Exon II, Intron I, IntronII, and Exon III (Table S1). The protein-coding sequence of the A. episcopale cp genome contained 64 codons, and the number of codons encoding corresponding amino acids varies from 22 to 1068. Of these, 61 codons encoded 20 amino acids, and the remaining three were found to be stop codons. Methionine (Met) and tryptophan (Try) are encoded by only one codon, whereas 2–6 codons encoded the remaining amino acids (Fig. 3). Of the 64 codons encoded by the protein-coding sequence of the A. episcopale chloroplast genome, 30 had the RSCU values > 1, which were high-frequency codons. Among them, 16 codons were ending in U, 13 codons ending in A, and one codon ending in G, indicating that the third base of the high-frequency codon of A. episcopale chloroplast genome prefers to use A or U (T) (Table S2).

Figure 2
figure 2

Chloroplast genome map of A. episcopale. Genes shown inside the circle are transcribed clockwise, and genes located outside the circle are transcribed counter-clockwise. Genes belonging to different functional groups have been colour-coded. The dashed area in the inner circle indicates the GC content of the chloroplast genome.

Full size image
Table 1 Summary of chloroplast genome characteristics of A. episcopale.
Full size table
Table 2 List of protein-coding genes present in the A. episcopale chloroplast genome.
Full size table
Figure 3
figure 3

Relative synonymous codon usage (RSCU) of 20 amino acids and stop codons in all protein-coding genes of the chloroplast genome of A. episcopale.

Full size image

Base composition of genome analysis

The base composition of LSC, SSC, IRS regions, and codons at different positions in the chloroplast genome of A. episcopale was analyzed (Table 3). In the LSC, SSC, and IRs region, AT content of SSC was the highest, followed by the LSC region and IRs region. The first, second, and third codon positions were 26,113 bp in length, and they had thymine (T) contents of 23.5–37.7%, cytosine (C) of 14.2–18.9%, adenine (A) of 28.3–31.6%, and guanine (G) of 16.5–27.0% (Table 3). The results indicated that the AT content of the chloroplast genome was higher than the GC content and also demonstrated the codon preference for using bases A and T(U).

Table 3 Analysis of the base composition of the genome of A. episcopale.
Full size table

Simple sequence repeat (SSR) analyses

SSRs are highly variable molecular markers within the same species and are mainly used in population genetics and polymorphism studies. These are important for gene expression, transcriptional regulation, and chromosome construction13,14. We investigated the type, distribution, and frequency of the repeat types of SSRs in the cp genome of A. episcopale. A total of 64 SSRs were identified (Fig. 4A). Among them, 52 (81%) SSRs were identified in the LSC region, including 29 mononucleotides, 12 dinucleotides, three trinucleotides, six tetranucleotides, and two pentanucleotides. The SSC region identified eight SSRs (13%), including two mono-, two di-, three tri-, and one tetra-. Four SSRs (6%) were identified in the IR regions, including two mono- and two di- (Figs. 4B, 5A).

Figure 4
figure 4

Repeat type and number analysis of SSRs in the chloroplast genome of A. episcopale. (A) Simple repeat sequence distribution in the complete cp genome. (B) Distribution of SSR in LSC, SSC, and IR regions.

Full size image
Figure 5
figure 5

Distribution, type, and frequency of the repeat types of SSRs in the chloroplast genome of A. episcopale. (A) Presence of SSRs in the LSC, SSC, and IR regions. (B–D) Presence of polymers in the LSC, SSC, and IR regions.

Full size image

Moreover, the different repeat motifs corresponding to the different types of SSRs were the most distributed in the LSC region, which included repeat motifs of 3, 4, 5, 6, 7, 9, 10, 11, 12, 14, and 16, while the IR regions were the least, with only two repeat motifs of 7 and 10 nucleotides. Among the five types of SSRs (mono-, di-, tri-, tetra-, penta-), the repeat motifs corresponding to mononucleotides were the most, while pentanucleotides were the least (Fig. 5B–D, Table S3).

Long repeat analysis

Long repeats have been found to effect the evolution, heredity, variation of life, and their indispensable role in gene expression, transcription regulation, chromosome construction, and physiological metabolism has also been established15. According to the structure, function, and position of the repetition, long repeats could be divided into forward repeat, reverse repeat, palindrome repeat, and complement repeat15,16. A total of 44 long repeats were detected in the chloroplast genome of A. episcopale, including 12 forward repeats (F), 20 reverse repeats (R), 12 palindromic repeats (P), and complementary repeats (C) were unfound. The length of most repeats was between 30 and 39 bp (34, 72.27%), followed by 40–49 bp (4, 9.10%), 50–59 bp (2, 4.55%), >=70 bp (3, 6.82%), and no repeats were between 60 and 69 bp. There were 11 long replicates distributed in the intergenic spacer region (non-coding region), and the remaining 33 long replicates distributed in the protein-coding region, mainly concentrated in the rpl20, ycf15, ycf2, and psaB gene regions (Fig. 6, Table S4).

Figure 6
figure 6

Distribution types and number of long repeats in chloroplast genomes of A. episcopale. REPuter was used to identify repeat sequences with length ≥ 30 bp and sequence identified ≥ 90%. F, P, R, and C indicate the repeat types F (forward), P (palindrome), R (reverse), and C (complement), respectively. Different colors indicate repeats with different lengths.

Full size image

Tandem repeat analysis

Tandem repeat sequences play a significant role in the genomic rearrangement and phylogenetic analysis17. In this study, a total of 26 tandem repeats were detected in the A. episcopale cp genome, 13 tandem repeats were distributed in the gene spacers and introns (12 in the gene spacer region, one in the clpP intron), and the rest tandem repeats were distributed in the protein-coding region (Table S5).

Comparison with cp genomes of other Aconitum species

The length of the A. episcopale chloroplast genome was similar to that of the A. vilmorinianum (MG678799.1), A. brachypodum (MH886505.1), A. pendulum (MN7191352.1), A. hemsleyanum (MG678800.1), and A. carmichaelii (KX347251.1) chloroplast genomes (Table 4). However, the A. episcopale cp genome had the most extended LSC region (86,452 bp) and A. brachypodum cp genome had the shortest LSC region (86,292 bp). A. episcopale, A. vilmorinianum, and A. brachypodum contain 132 genes, A. pendulum, A. hemsleyanum had 131 genes, and A. carmichaelii had only 130 genes. A. pendulum had 86 PCGs, 85 PCGs were present in A. episcopale, A. vilmorinianum, A. brachypodum, and A. hemsleyanum, and 84 PCGs were in A. carmichaelii. Notably, all the six Aconitum species chloroplast genomes possessed 37 tRNA genes.

Table 4 Comparison of the general features of the genus Aconitum chloroplast genomes.
Full size table

Comparative analysis of the A. episcopale chloroplast genome

Five published sequences representing Aconitum (A. hemsleyanum and A. carmichaelii), Consolida (C. ajacis), Delphinium (D. grandiflorum and D. anthriscifolium) of the Ranunculaceae family were selected for the comparison with the sequence of A. episcopale to evaluate the sequence divergence of different regions of these genomes. Pairwise alignment between the chloroplast genomes of A. episcopale and five other species were performed using mVISTA, with the annotated A. episcopale chloroplast genome as a reference. We observed that the cp genome structure was conserved in the Trib. Delphineae (Fig. 7). Through pairwise comparisons of whole genomes of six species, as with the vast majority of chloroplast genomes, the IR regions were clearly more conserved than the LSC and SSC regions, with the intergenic spacer region showing the greatest variation. The most variable gene regions of the six cp genomes identified through global alignment were rpl20, aacD, ndhF, ycf1, ycf2, ccsA, and matK. In addition, the intergenic spacer was most variable with matK-rps16, rpoB-trnC-GCA, petN-psbM, psbM-trnD-GUC, trnP-UGG-psaJ, rps18-rpl20, ycf3-trnS-GGA, ycf4-cemA, psbL-petL, and rpl16-rps3, and the four rRNA genes (rrn4.5, rrn5, rrn16, and rrn23) were found to be most conserved.

Figure 7
figure 7

Comparison of the chloroplast genome sequences of A. episcopale, A. hemsleyanum, A. carmichaelii, C. ajacis, D. grandiflorum, and D. anthriscifolium generated with mVISTA. Grey arrows indicate the position and direction of each gene. Red and blue areas indicate the intergenic and genic regions, respectively. The vertical scale indicates the percentage of identity, ranging from 50 to 100%.

Full size image

In chloroplast genomes, IR is the most conserved region. Contraction and expansion of the IR boundary are prevalent factors affecting the chloroplast genome size17. The JLB boundaries of A. episcopale, A. carmichaelii, and D. grandiflorum were found to be located inside the rps19 gene so that the rps19 gene extended 3–20 bp beyond the IRb region. However, the rps19 genes of C. ajacis, D. anthraciscifolium, A. hemsleyanum were at some distance from the JLB boundary. In addition, the rps19 gene located in the JLB boundary also had a range in the length of the LSC region and IRb region, that is, between 259 and 276 bp for LSC and 3–20 bp IRb. Moreover, the JSB boundary (IRb-SSC) was found within the genes ycf1 and ndhF, the region containing ycf1 in A. episcopale, A. hemsleyanum, and D. grandiflorum showed some contraction at the IRb boundary, whereas the IRb region of D. anthraciscifolium had no ycf1 gene. The ndhF genes of A. hemsleyanum, C. ajacis, and D. anthriscifolium showed different lengths in the SSC region, while there was no ndhF gene in SSC region of A. episcopale, A. carmichaelii, and D. grandiflorum. The JSA boundary (SSC-IRA) was located inside the ycf1 gene in all six Trib. Delphineae species. Within the ycf1 gene, the two parts separated by the JSA boundary (a part in the SSC region and a part in the IRa region) showed variation in length, ranging from 3639 to 4316 bp in the SSC region and 1060 bp–1677 bp in the IRa. JLA boundaries (IRa-LSC) in six species were located in the intergenic region trnH-psbA. The trnH and pasA also differed in length from the JLA boundary, with the trnH gene having the shortest length of 17 bp from the JLA boundary (Fig. 8).

Figure 8
figure 8

Comparison of the border regions among LSC, IR and SSC in the chloroplast genomes of A. episcopale, A. hemsleyanum, A. carmichaelii, C. ajacis, D. grandiflorum, and D. anthriscifolium. JLB: junction line between LSC and IRb; JSB: junction line between IRb and SSC; JSA: junction line between SSC and IRa; JLA: junction line between IRa and LSC.

Full size image

Variation zone analysis of the A. episcopale chloroplast genome and related species

Highly variable fragments of the chloroplast genomes can be used for phylogenetic studies and related species identification at the species level. It could also provide important information at the population level to detect the differences between the species and understand the changes in the population structure18,19. The cp genome sequence of A. episcopale was similar to that of A. hemsleyanum, A. vilmorinianum, and A. delavayi, which showed that the chloroplast genome of Aconitum was relatively conserved. Compared with the SSC region and LSC region, IR regions showed higher variability. In addition, these genomes were tested for the variation at the intervals, resulting in four highly variable regions, including two intergene regions ndhC-trnV-UAC (pi = 0.0042) and ccsA-ndhD (pi = 0.0067), two coding protein regions rpl12 (pi = 0.0033) and ycf1 (pi = 0.0042). These regions could further be used as highly variable fragments to identify the Aconitum species (Fig. 9).

Figure 9
figure 9

Sliding window analysis of the four Aconitum species chloroplast genomes.

Full size image

Phylogenetic analysis

In the current study, the complete cp genome sequences of 30 species of Ranunculaceae family and that of Berberis bealei and Epimedium xichangense of Berberidaceae from the GenBank database in NCBI. In addition, the genomic sequence of A. episcopale was obtained by sequencing in this study. Phylogenetic trees were constructed by the maximum likelihood (ML) method to determine the relationships and phylogenetic positions of A. episcopale and its related species (Fig. 10). The results showed that the 33 species were divided into three large groups. The 26 species of Aconitum were divided into six clades. A close relationship among the A. episcopale, A. delavayi, A. vilmorinianum, and A. hemsleyanum was also uncovered.

Figure 10
figure 10

Phylogenetic tree of A. episcopale and 33 species of Ranunculaceae family using maximum likelihood based on the complete chloroplast genomes. Bootstrap values based on 1000 replicates have been shown on each node.

Full size image

Discussion

In 2003, DNA barcoding was proposed by Herbert et al. as a means for species identification, which employs a short, standard stretch of DNA to enable the species identification by resolving intra- and interspecies genetic differences. Thereafter, this method has been used widely for species identification18. In the plant species identification, the commonly used DNA fragment contained nuclear genes ITS and ITS2, and the cp genomic sequence rbcL, matK, psbA-trnH19,20. Due to the strong identification ability of the ITS2 sequences, it has been proposed as a common DNA barcode for taxonomic identification of the medicinal plants21. However, common DNA barcoding does not possess sufficient variable loci for particular medicinal plant groups to identify the species such as Fritillaria22. Therefore, complete chloroplast genomes has been successfully used to identify certain medicinal plants with species closely related to it23. In this study, the chloroplast genome of ethnodrug Aconitum episcopale was sequenced in the next-generation sequencing. The chloroplast genome sequences of Aconitum species were found to be relatively conserved through comparative analysis, with less variation occurring between species. Through comparative analysis, we found that the differences between the chloroplast genome length, gene list, and GC contents between A. episcopale and other Aconitum species were small. In addition, we also found that the IR regions were more conserved among chloroplast genomes of Aconitum species than the LSC and SSC regions. This was in consistent with the results of previous comparative analytical studies of Aconitum species24. Four regions were identified as high variation regions including two intergene regions ndhC-trnV-UAC (pi = 0.0042) and ccsA-ndhD (pi = 0.0067) and two coding protein regions rpl12 (pi = 0.0033) and ycf1 (pi = 0.0042). These regions served as high variation segments for the species identification as described above, containing the most SNP sites, and these may serve as potential DNA barcodes for future species identification studies.

SSRs, also known as microsatellites, are highly variable molecular markers of the same species and are mainly used in the population genetics. They play an essential role in gene expression, transcription regulation, and chromosome construction. A total of 64 SSRs were detected in the cp genome of A. episcopale, and the results were similar to the SSR numbers detected by Kong et al. from A. sinomontanum var. angustius and A. finetianum25. Most mono- and dinucleotides consisted of multiple copies of A/T and AT/AT repeats, respectively. These SSRs could serve as effective biomarkers for the population genetic diversity studies of the chloroplast genome of A. episcopale, contributing to effective conservation measures for this medicinal plant.

Conclusions

In the present study, we used Illumina Novaseq sequencing technology to obtain the complete chloroplast genome sequence of A. episcopale. Its cp genome was 155,827 bp in size and encoded 132 genes, 20 of which contained introns. The cp genome of A. episcopale contained 64 codons and encoded 20 amino acids, with the number of codons encoding corresponding amino acids varying from 22 to 1068. We identified a total of 64 SSRs. Four high variation regions (ndhC-trnV-UAC, ccsA-ndhD, rpl12, and ycf1) were identified by the nucleotide polymorphism analysis of the cp genome. In addition, we also performed a phylogenetic analysis of 33 whole cp genomes and colinearity analysis between A. episcopale and five related species. The above analytical results provided an essential theoretical basis for the molecular identification and phylogeny of the medicinal plant A. episcopale and valuable reference information for its effective conservation strategies.

Materials and methods

Sample preparation

Fresh leaves without disease spots of A. episcopale were collected from the Raboluo Village, Weixi County, Yunnan Province, in September 2020 (27°31′42″ N, 99°34′90″ E; elevation 2843.19 m). After that, the leaves were dried and stored with discolored silica gel. The specimen was stored in the Plant and Medicinal Herbology, College of Pharmacy, Dali University. The specimen number was recorded as WTWX001.

Extraction of genomic DNA and sequencing

Fresh leaves of A. episcopale were collected, and total DNA was extracted using the E.Z.N. A® Plant DNA kit (OMEGA, Beijing). DNA quality was assessed using agarose gel electrophoresis. This DNA was further paired-end sequenced using the Illumina Novaseq 6000 (Shanghai Biozeron Biotech Co., Ltd., China) platform. Since there could have been some data with lower quality from Illumina's raw sequencing data, to make subsequent assemblies more accurate, raw data was quality clipped using Trimmomatic software (http://www.usadellab.org/cms/index.php?page=trimmomatic)26. The specific steps were as follows: remove the adaptor sequence in reads, remove bases containing non AGCT at the 5’ end before shearing, trim the ends of reads with sequencing quality value less than Q20, remove reads with N ratio up to 10%, discard adaptor and small fragment with length less than 75 bp after quality trimming.

Chloroplast genome assembly, annotation, and submission

After the raw data was filtered by Trimmomatic, the NOVOPlasty (https://github.com/ndierckx/NOVOPlasty). 27was used to perform the chloroplast genome assembly. First, ~ 100 million reads were randomly selected and aligned to the cp genome sequence of Aconitum delavayi (NC_038097.1) using BWA (mem, default parameters)28. A perfect matched read to the psbA gene was selected as the seed input for NOVOPlasty. Two optional sequences, which were both single circular cp genomes, were produced, and the one with the same SSC direction to Aconitum delavayi was selected. Clean reads were aligned back to this sequence using BWA and inspected in IGV to exclude any assembly error. Finally, a custom-made script, which took the sequence and the corresponding bam file as input, was employed to correct ambiguous bases. The sequence was self aligned using blastn to determine the two IR regions and the start position of the LSC region. It was manually reorganized to final cp genome sequence, with the typical quadripartite structure as “LSC-IRb-SSC-IRa”. The GeSeq (https://chlorobox.mpimp-golm.mpg.de/geseq.html)29 was used for the gene prediction and annotation on the final assembled genome. The chloroplast genome sequences of reference used for the annotation were from Aconitum delavayi (NC_038097.1), Aconitum vilmorinianum (NC_038094.1), Aconitum hemsleyanum (NC_038095.1), and Aconitum contortum (NC_038098.1). Thereafter, the annotated sequences were submitted to GenBank. The genome was displayed using the software OrganellarGenomeDRA (http://ogdraw.mpimp-golm.mpg.de/cgi-bin/ogdraw.pl)30.

Codon preference analysis

Codon usage frequency and relative synonymous codon usage (RSCU) in cpDNA from A. episcopale were determined using the Codon W software31.

Simple repeat analysis

Simple sequence repeat, also known as microsatellites (SSRs), are widespread in the genome and generally consist of repeats of 1–6 bp to a lower extent, mainly in units of repeats of 2–3 nucleotides, such as (GA)n, (AC)n, and (GAA)n, among others. Microsatellite loci analysis was performed on the assembled A. episcopale chloroplast sequence genome using MISA tools32. The parameters were set to the definition (unit_size, min_repeats): 1–10, 2–5, 3–4, 4–3, 5–3, 6–3, and the minimum distance between two SSRs was set to 0 bp. Custom-made script is shown in "Custom-made script of MISA.txt" of supplementary file.

Long repeat and tandem repeats analysis

Long repeat analysis was conducted using the software REPuter (http://bibiserv.techfak.uni-bielefeld.de/Repeat/)33 with parameters set to the minimum repeat size of 30 bp and hamming distance of three. Tandem repeats were detected using the Tandem repeats finder (https://tandem.bu.edu/trf/trf.html) online program with parameter settings chosen as default34.

Comparative analysis of A. episcopale cp genome

The complete cp genome sequences of Aconitum hemsleyanum (NC_038095.1), Aconitum carmichaelii (KY407560.1), Consolida ajacis (NC_041534.1), Delphinium grandiflorum (NC_049872.1), and Delphinium anthriscifolium (MK253461.1) were from the GenBank database in NCBI (https://www.ncbi.nlm.nih.gov/genbank/). Thereafter, the comparative analysis of whole cp genome sequences of A. episcopale obtained by new sequencing in this study and the five published species mentioned above were carried out using the mVISTA online tool under the Shuffle-LAGAN model35. Genome sequence of A. episcopale was used as the reference sequence. The boundaries of the chloroplast genomes of the above six species were mapped using IRscope (https://irscope.shinyapps.io/irapp/)36. After aligning the whole chloroplast genome sequences of A. hemsleyanum, A. vilmorinianum, A. delavayi, and A. episcopale, sliding window analysis was performed by DnaSP with the parameters set to step size of 200 bp and window length of 600 bp37.

Phylogenetic analysis

We obtained the cp genome sequences of 32 species from the GenBank database in NCBI (https://www.ncbi.nlm.nih.gov/genbank/) and the genome sequences of A. episcopale were obtained from the current research. Of these, Berberis bealei and Epimedium xichangense, two species of the Berberidaceae family, were selected as outgroups. Complete cp genome sequences of the above species and A. episcopale were aligned by MAFFT online tool38. Next, phylogenetic relationships among species were constructed using the maximum likelihood method with a bootstrap value of 1000 in IQtree software39. Further, images of the ML tree were processed using the software MEGA X40.

All the experiments were carried in accordance with national and international guidelines. The plant materials procured and used in the study comply with China’s guidelines and legislation.

Specimen collection statement

The collection of fresh leaves obtained the permission of the owner.