search for


Comparative SNP Analysis of Chloroplast Genomes and 45S nrDNAs Reveals Genetic Diversity of Perilla Species
Plant Breeding and Biotechnology 2018;6:125-139
Published online June 1, 2018
© 2018 Korean Society of Breeding Science.

Kyeong-Seong Cheon1, In-Seon Jeong1, Kyung-Hee Kim2, Myoung-Hee Lee3, Tae-Ho Lee1, Jeong-Hee Lee4, Ung-Han Yoon1, Romika Chandra5, Ye-Ji Lee1, and Tae-Ho Kim1,*

1Genomics Division, National Institute of Agricultural Sciences, Rural Development Administration, Jeonju 54874, Korea, 2Department of Plant Science, Plant Genomics and Breeding Institute and Research Institute of Agriculture and Life Sciences, College of Agriculture and Life Sciences, Seoul National University, Seoul 08826, Korea, 3Legume and Oil Crop Research Division, National Institute of Crop Science, Rural Development Administration, Milyang 50424, Korea, 4SEEDERS, Daeduk Industry-Academic Cooperation Building, Daejeon 34015, Korea, 5Department of Biology, Kunsan National University, Gunsan 54150, Korea
Correspondence to: *Corresponding author: Tae-Ho Kim,, Tel: +82-63-238-4563, Fax: +82-63-238-4554
Received March 20, 2018; Revised April 10, 2018; Accepted April 10, 2018.
This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Perilla species belong to the Lamiaceae family of flowering plants and are widely grown in East Asia, for use in a traditional herbal medicine or functional food. To identify single nucleotide polymorphisms (SNPs) in Perilla species and conduct a phylogenomic analysis, we determined the complete sequences of the chloroplast (cp) genome and 45S nuclear ribosomal DNA (45S nrDNA) of six cultivated and three wild Perilla species. The complete cp genome ranged in size from 152,588 bp to 152,656 bp and the length variation in cp genomes was 68 bp. The length of the 45S nrDNA ranged from 6,235 bp to 8,303 bp and the main variation of length differences was in the intergenic spacer (IGS) region. Comparative analysis of the cp genome sequences of nine Perilla species showed low genetic diversity at the intra- and inter-species level. Using SNP analysis, we detected 42 synonymous SNPs (sySNPs) from 27 genes and 37 non-synonymous SNPs (nsSNPs) from 15 genes. A comparison of the 45S nrDNA sequences revealed two SNPs in the 18S rRNA, five SNPs in the 26S rRNA, three SNPs and two InDels in the internal transcribed spacer (ITS) 1 region, and six SNPs in the ITS 2 region. Our phylogenomic analysis suggests that the tetraploidization of Perilla cultivars may have arisen from the P. citriodora genome. The genotyping data from this study may be used to develop molecular markers associated with useful traits for use in Perilla breeding.

Keywords : Genetic diversity, SNP analysis, Perilla, Chloroplast genomes, 45S nrDNAs

Perilla is a herb belonging to the mint family, Lamiaceae, that is widely cultivated in East Asian countries, including Korea, China, and Japan. Perilla is used in traditional herbal medicine for treating various conditions, such as depression, anxiety, tumors, cough, bacterial and fungal infections, allergies, intoxication, and intestinal disorders (Markakis 1982; Wang 1997; Clifford 2000). The raw leaves are used as condiments and flavoring agents and the seeds are a source of oil that is rich in omega-3 fatty acids and alpha-linolenic acid (ALA). The cultivated species, Perilla frutescens Britt., is an allotetraploid plant with chromosome numbers of 2n = 4× = 40, whereas the three wild species, P. citriodora, P. hirtella, and P. setoyensis, have 2n = 2× = 20 chromosomes (Honda et al. 1994; Ito et al. 1996). The cultivated and wild species can be distinguished based on the properties of their trichomes. The trichomes of the cultivated species are sparsely distributed, straight, and long, whereas those of the wild species are dense, crooked, and short.

The chloroplast (cp) genome, which originated from cyanobacteria through endosymbiosis, is widely used in evolutionary and phylogenetic studies, DNA barcoding, and population-level genetic diversity analyses (Powell et al. 1995; Jansen et al. 2007; Hollingsworth et al. 2011). The cp genome is highly conserved within plant species due to its association with the evolutionarily conserved process of photosynthesis; thus, intra-species polymorphisms are rare, but inter-species polymorphisms are common in the genic or intergenic regions. The coding regions of 45S nuclear ribosomal DNA (45S nrDNA) are highly conserved and show low substitution or mutation rates within a species (Nazar et al. 1976; Ma et al. 1998; Sumida et al. 2004), which limits its use in elucidating phylogenic relationships between members of the same species. By contrast, the internal transcribed spacer (ITS) regions have relatively high levels of inter-species variation, resulting from insertions, deletions, and point mutations. Therefore, with few exceptions, ITS sequence analysis is suitable for reconstructing phylogenetic relationships, even among members within the same species. Therefore, both the cp genome and 45S nrDNA sequences are important genetic materials for clarifying genetic relationships and surveying genetic diversities among intra- and inter-specific specimens. The NCBI database is the primary resource for complete cp genome and 45S nrDNA sequences derived from various organisms. However, as many of the sequences in this database are a single representative sequence from a species, it is difficult to conduct a comparative analysis using complete cp or nr sequences from multiple accessions of a species. Only a few genetic diversity studies have been conducted at the intra-species level using complete cp, 45S nrDNA, or both in plants such as Allium cepa (onion), Malus domestica (apple), and Panax ginseng (ginseng) (Cho et al. 2006; Nikiforova et al. 2013; Kim et al. 2015). In the diversity study of Perilla genus, restriction fragment length polymorphism (RFLP) and sequence characterized amplified region (SCAR) markers have been used to distinguish between Perilla species (Kim et al. 2010; Lee et al. 2011). Therefore, despite the potential usefulness for genetic studies or as a source of medicine and food in Perilla, knowledge of intra-species sequence variation within this genus is limited.

In this study, we generated the complete cp genome and 45S nrDNA sequences of six cultivated and three wild Perilla species using next generation sequencing (NGS) technology. We conducted a comparative sequence analysis of the complete cp genome and 45S nrDNA sequences. These sequences can be used to develop molecular markers for phylogenomic analyses of Perilla species.


Plant materials

Six cultivars (P. frutescens ‘Deulkkae’, ‘Chajogi’, ‘Apureunchajogi’, ‘PureunJureumchajogi’, ‘Jureumchajogi’, and ‘Pureunchajogi’) and three wild species (P. citriodora, P. setoyensis, and P. hirtella) of Perilla were used for genomic DNA preparation and sequencing (Table 1, Supplementary Fig. S1). Individual plants (4–8) of all nine Perilla species were used for PCR analysis to validate polymorphic sites. Leaves of young Perilla species above were sampled from the National Institute of Crop Science, Rural Development Administration (RDA) and stored at −80°C until use.

DNA preparation and extraction

Perilla seeds were planted in 120-hole trays in the greenhouse at the Department of Southern Area Crop Science, National Institute of Crop Science, RDA, Republic of Korea. Total genomic DNA was isolated from 20-day-old Perilla seedlings using a modified cetyltrimethylammonium bromide (CTAB) method (Murray et al. 1980). Briefly, fresh leaf tissues (200 mg) were frozen with liquid nitrogen, powdered using a mechanical mill, and incubated in 800 μL of extraction buffer (Intron Biotechnology, Inc, Republic of Korea; 2% CTAB, 100 mM Tris-HCl (pH 8.0), 20 mM EDTA, 1.4 M NaCl, and 1% 2-mercaptoethanol) in 2 mL microcentrifuge tubes at 65°C for 1 hour in a water bath. After incubation, 800 μL of chloroform/octanol, 24:1 (vol/vol) was added and samples were mixed by inversion for 2 minutes to form an emulsion. Samples were then centrifuged at 4,200 rpm for 10 minutes at 4°C. The upper aqueous solution (450 μL) was transferred into a new 1.5 mL microcentrifuge tube and an equal volume of isopropanol was added. Samples were mixed by gentle inversion for 2 minutes. The precipitated genomic DNA was centrifuged for 10 seconds and the liquid was gently poured out. The precipitated genomic DNA was rinsed twice with 1 mL of 70% ethanol and dried overnight at room temperature. The genomic DNA was dissolved in 600 μL of 1× TE buffer and stored at −20°C for further studies.

De novo assembly of the cp genome and 45S nrDNA

Approximately 5× coverage of the whole genome sequence was attained using an Illumina NextSeq instrument for all nine Perilla samples. About 650 Mb of the estimated P. citriodora genome size was used to calculate the sequence coverage of each sample in this study. Low quality bases (Phred score of ≤ 30) were removed and the reads containing high quality bases were used to assemble the cp genome and 45S nrDNA sequence using a CLC Genome Assembler (ver. 4.6 beta, CLC Ins., Aarhus, Denmark) with parameters of 200–600 bp of autonomously controlled overlapping size. The cp contigs were selected from the CLC de novo assembly and ordered using a BLASTN analysis against the NCBI organelle genome resource database ( Then a complete cp sequence was constructed by determining and joining overlapping terminal sequences. The complete cp assembly was validated with the previously reported cp genomes using BlastZ with an E-value of 1e-6 (Schwartz et al. 2003) and one merged gap was experimentally confirmed by PCR amplification followed by Sanger sequencing. PCR was conducted in a 25 μL volume with 20 ng of genomic DNA template, 0.2 μM of each primer (forward, 5′-CATTTG CATAAGATCATAAG-3′ and reverse, 5′-TTAGAAAG GTGGGCTTTTAT-3′), 0.2 mM of each dNTP, 1× PCR buffer, and 0.4 U of Taq DNA polymerase (Vivagen, Seongnam, Republic of Korea). PCR amplification was performed with denaturation at 95°C for 5 minutes, followed by 35 cycles of 95°C for 40 seconds, 40°C for 40 seconds, and 72°C for 40 seconds, and a final elongation at 72°C for 10 minutes.

The 45S nrDNA sequences were assembled into a single contig with the same assembly method described above. Assembly errors caused by tandem repeats (TRs) were manually corrected and the complete 45S nrDNA sequence was validated by PCR amplification. The complete cp genome and 45S nrDNA sequences were further validated by ABI Sanger sequencing when ambiguous sequences were not able to be resolved manually.

Gene annotation

Gene annotation of the cp genome was performed using the DOGMA program ( (Wyman et al. 2004) with a similarity parameter of 60% for protein-coding and 80% for RNA genes. The exact coordinates of cp genes were manually confirmed and corrected using BLASTN searches against the NCBI plastid database ( The 45S nrDNA genes (18S rDNA, 5.8S rDNA, and 26S rDNA) were predicted using RNAmmer ( (Lagesen et al. 2007), and then intergenic sequences were annotated as ITS1 (between 18S rDNA and 5.8S rDNA) and ITS2 (between 5.8S rDNA and 16S rDNA). The complete structure of 45S nrDNA was confirmed using 45S nrDNA of P. ginseng cv. Chunpoong (GenBank Acc. KM036295) as a reference. The circular cp genome maps were drawn using OGDRAW ( (Lohse et al. 2007).

Comparative analysis of the sequences and variation among the Perilla species

Comparative sequence analysis was performed to identify sequence variations among the nine Perilla cp genomes. First, multiple sequence alignment was carried out with MAFFT ( (Katoh et al. 2002) and mVISTA with the LAGAN alignment program ( (Frazer et al. 2004), and the aligned multiple sequences were manually confirmed using BioEdit program (Hall 1999). TRs in the cp genome were characterized using Tandem Repeat Finder software v2.0 ( (Bensen 1999) with the following alignment parameters: 2 for matches, 7 for mismatches, and 7 for InDels. Final TRs were selected with the following criteria of 30 for minimum score, 500 bp for maximum period size, and 100% sequence similarity.

To validate polymorphisms among the nine Perilla plants, specific primers were designed based on polymorphic sites found among the cp genome and 45S nrDNA sequences of nine Perilla plants. Primers for SNP and InDel sites (Supplementary Table S1) were designed using Primer 3 ( (Koressaar et al. 2007; Untergrasser et al. 2012). PCR amplifications for SNP and InDel regions were performed in 20 μL reaction mixtures containing 20 ng genomic DNA template, 1× Ex-Taq buffer, 200 μM dNTPs, 0.5 U Ex-Taq DNA polymerase (TaKaRa Bio, Shiga, Japan), and 0.4 μM of each primer. Reaction conditions were as follows: denaturation at 94°C for 3 minutes; 32 cycles of denaturation at 94°C for 30 seconds, annealing at 60°C for 30 seconds, and extension at 72°C for 30 seconds; and a final extension at 72°C for 5 minutes. The PCR products (10 μL each) were confirmed by electrophoresis on a 2.5% agarose gel using 0.5× TAE buffer. Then, the amplified products (5 μL each) were digested with restriction enzymes.

For phylogenomic analyses of the nine cp genomes, the best rotation for the genomes was first determined using the CSA tool (Fernandes et al. 2009) in the presence of the Olea europaea cp genome as an outgroup. Then, multiple alignment was performed using MAFFT (Katoh et al. 2013) with the reorder and auto options. The phylogenomic tree was constructed using the maximum likelihood method implemented in the Dnaml program in the PHYLIP package (Felsenstein 2005) with O. europaea as outgroup. Bootstrap analysis (with 1000 bootstrap replicates) was conducted using phangorn (Schliep et al. 2011) in the R environment.


Complete cp genome and 45S nrDNA sequences

Low-coverage whole genome sequencing was performed for the nine Perilla species, including six cultivated species (i.e., ‘Deulkkae’, ‘Chajogi’, ‘Apureunchajogi’, ‘PureunJureumchajogi’, ‘Jureumchajogi’, and ‘Pureunchajogi’) and three wild species (i.e., P. citriodora, P. setoyensis, and P. hirtella). The sequencing results yielded 118,243 × 106 bp (corresponding to 1,165 × 106 reads) or were equivalent to ~5× coverage of the diploid P. citriodora (Mo et al. 2017) genome. High quality sequences with a Phred quality value of > Q20 were used to assemble the cp genome and 45S nrDNA sequence for each species. About 5% of the total reads were aligned to the assembled cp genome; thus, the estimated cp coverage was, on average, 2,790× (Supplementary Table S2).

The complete cp genomes for the nine Perilla species ranged from 152,588 bp in length for ‘Chajogi’ to 152,656 bp for P. hirtella (Fig. 1, Table 2). The cp genomes contained a pair of inverted repeat regions (IRa and IRb) that were each 25,676 bp, with the exception of 25,675 bp, in the case of P. setoyensis (Table 2). The two IR regions divide the cp genome into a large single copy (LSC) region and a small single copy (SSC) region. The LSC regions were 83,699 bp to 83,757 bp in length and the SSC regions were 17,537 bp to 17,547 bp (Table 2). The gene content and order were identical among the cp genomes of the nine Perilla species, with each having a total of 114 genes, including 80 protein-coding genes, 30 transfer RNA (tRNA) genes, and four ribosomal RNA (rRNA) genes. A total of 18 genes containing introns were annotated, nine of which are protein-coding genes, six of which are tRNA genes that contain one intron, and three of which (rps12, clpP, and ycf3) contain two introns (Table 3). Similar gene contents were observed in the cp genomes of Salvia miltiorrhiza (Qian et al. 2013), Origanum vulgare L. (Lukas et al. 2013), and Panax ginseng (Kim et al. 2004; Kim et al. 2015). Each of the 45S nrDNA sequences was assembled into a single contig by de novo assembly using low-coverage whole genome sequences (dnaLCW). The 45S nrDNA sequences, including the 18S rRNA, ITS 1, 5.8S rRNA, ITS 2, 26S rRNA, and an IGS were highly homogeneous, consisting of single units of between 6,235 bp and 8,303 bp in length (Fig. 2, Supplementary Table S2).

Homology among the Perilla cp genome sequences

To estimate the genetic diversity of the Perilla species at the inter- and intra-species level, we analyzed nucleotide substitutions in the cp genome sequences. The overall cp genome sequence identity of the nine Perilla species was plotted with mVISTA, using the annotation of P. citriodora as a reference. The cp genomes of the cultivated species ‘Apureunchajogi’, ‘PureunJureumchajogi’, and ‘Jureumchajogi’ were identical in terms of sequence and length, while the cp genomes of the remaining six species had over 99.7% sequence similarity. Among the nine Perilla species, the wild species P. setoyensis showed the highest level of variation (Supplementary Fig. S2).

Sequence variations in the cp genome and 45S nrDNA sequences among Perilla species

To obtain insight into the genetic diversity among the nine Perilla species, we analyzed their cp genome and 45S nrDNA sequences. We identified 233 SNPs within the cp genomes of the nine Perilla species, including 154 SNPs in the intergenic regions and 79 SNPs in the coding sequences (Supplementary Table S3). Relatively large sequence length differences were identified in the LSC region and most of the detected SNPs (~73.4%) were located in the LSC region. To identify variations among the coding sequences, we compared each of the individual annotated gene sequences (Supplementary Table S3). All tRNA and rRNA genes were conserved and no SNPs were found. More than five SNPs were detected in each of four genes (i.e., rpoC2, ycf1, ycf2, and ndhF). For instance, ycf1, which had the highest number of SNPs of all the coding sequences examined, contained eight non-synonymous SNPs (nsSNPs) and one synonymous SNP (sySNP), (Fig. 3). Interactive comparative analysis revealed 42 sySNPs from 27 genes and 37 nsSNPs from 15 genes. Most nsSNPs were found in P. setoyensis (28 nsSNPs) and P. hirtella (18 nsSNPs). In P. citriodora, two nsSNPs were detected (one in rpl22 and the other in rps8), while the other six cultivated species had no nsSNPs (Table 4). In total, 124 species-specific SNPs (ssSNPs), approximately 53.2% of the total number of SNPs, were identified in cp genomes. Most of the ssSNPs (83 and 36, respectively) were also present in the wild species P. setoyensis and P. hirtella. Of the ssSNPs, 76 were identified in intergenic regions and 48 in coding regions (Supplementary Table S3). Only one ssSNP from ‘Chajogi’ (C to A at position 87,317) and ‘Deulkkae’ (C to A at position 114,010) was identified among the six cultivated Perilla species (Supplementary Table S3). Five TRs were observed in a comparison of the cp genome sequences of the nine Perilla species. Among these, five variable numbers of TRs were derived from simple InDels ranging from 1–27 bp in length (Table 5). Of these, one variable number of TRs was found in the intron of petB, while the others occurred in the intergenic regions. P. hirtella had two species-specific variable numbers of TRs, allowing this species to be easily distinguished from the other Perilla species. One TR was located in the intron of petB; P. hirtella had three copies of a 6-bp TR (AAAGAA), while the other Perilla species had only two copies of this TR. The other TR was found in the intergenic region between petA and psbJ; P. hirtella had two copies of a 7-bp TR (ATTATAT), while the other species had only one copy. Also, P. setoyensis and P. hirtella had two copies of a 27-bp TR, while the other species had only one copy in the intergenic region between trnKUUU and matK (Table 5).

Moreover, we identified 18 variations (16 SNPs and 2 InDels) in the 45S nrDNA comparison, with P. citriodora containing the most variations, i.e., 14 (13 SNPs and one InDel) followed by P. hirtella (11 SNPs) (Table 6). In contrast to the cp genome variation results (Supplementary Table S3), P. setoyensis had the fewest variations (one SNP and two InDels) in our 45S nrDNA analysis. A comparison of the 45S nrDNA sequences revealed two SNPs in the 18S rRNA, five SNPs in the 26S rRNA, three SNPs and two InDels in ITS 1, and six SNPs in ITS 2. In this comparison, only one species-specific InDel was detected at position 2,006 of ITS 1 in P. setoyensis. Also, six ssSNPs were identified among the 45S nrDNAs of the nine Perilla species with the following substitutions: G-to-T at position 1,915 of ITS 1, C-to-A at position 1,974 of ITS 1 and 2,409 of ITS 2, C-to-T at positions 2,263 and 2,408 of ITS 2, and T-to-C at position 5,642 of the 26S rRNA.

To validate the polymorphisms identified in the cp genome and 45S nrDNA sequences in three wild species and six cultivated species of Perilla, we performed PCR analysis using several polymorphic sites and gDNAs that were extracted from individual Perilla plants (Fig. 4). We designed four cleaved amplified polymorphic sequence (CAPS) markers for the SNP position if any restriction enzyme sites were available for the clear validation of the SNP genotype (Supplementary Table S1). Each SNP in ndhF, ycf1, and ycf2 (at 110,180, 123,676, and 146,157 positions, respectively) was unique to P. setoyensis and/or P. hirtella (Fig. 4A, 4B, and 4C) and one SNP in the ITS 1 region (at 1,915 position) was unique to P. citriodora. These SNPs were detectable using CAPS markers (Fig. 4D). To validate the TRs, a region between trnKUUU and matK that was derived from a simple InDel of 27 bp and was the longest of the five TRs, was confirmed by PCR analysis (Fig. 4E).

Phylogenomic analysis based on the cp genomes

Phylogenomic analyses using entire cp genome alignments confirmed the placement of the Perilla species in the Lamiaceae family (Fig. 5). All Perilla genomes formed a cluster that was distinct from that of other outgroup species (S. miltiorrhiza and Lavandula angustifolia) in the Nepetoideae subfamily. Among the three wild Perilla species, we found that P. setoyensis and P. hirtella were more closely related to each other than to P. citriodora. However, P. citriodora was more closely related to the six cultivated species. As expected, all cultivars clustered with P. citriodora, reflecting the close relationship between the cultivated species and P. citriodora.


The complete cp genomes of plants are a valuable resource for developing molecular markers to study the evolution of both closely and distantly related organisms. Molecular markers are helpful when the taxonomic value of morphological characteristics is limited (Cui et al. 2006). Cp genomes and 45S nrDNA sequences are highly conserved within species; however, nucleotide substitutions preserved within these sequences can be used to elucidate the evolutionary relationships and genetic diversity among angiosperms (Wolfe et al. 1987; Leebens-Mack et al. 2005; Jansen et al. 2007). Therefore, cp genomes and 45S nrDNA sequences have been widely used to decipher complex evolutionary lineages of angiosperms and to develop DNA barcoding tools (Hollingsworth et al. 2011) for cultivar identification and species discovery.

We performed de novo assembly to obtain the complete cp genomes and 45S nrDNA sequences of 18 accessions from nine Perilla species using low-coverage NGS data. We successfully obtained the complete sequences of the cp genome of three wild and six cultivated accessions of Perilla (Fig. 1). We compared the cp genome sequences of P. citriodora with those of Sesamum indicum (Pedaliaceae) (Yi et al. 2012) and S. miltiorrhiza (Lamiaceae) (Qian et al. 2013) and found that the similarity between P. citriodora and S. indicum was 89.6% and between P. citriodora and S. miltiorrhiza was 93.9% (Supplementary Table S4). The cp genomes exhibited an identical gene order and content compared to most angiosperm cp genomes (Jansen et al. 2005; Yang et al. 2010; Wicke et al. 2011; Yi et al. 2012), indicating the highly conserved nature of these Perilla cp genomes.

Previously, ndhF, rps15, rps33, matK, ccsA, rpoA, rpl22, and ycf1 genes were reported to be variation hotspots (Downie et al. 1996; Drescher et al. 2000; Kim et al. 2004; Huotary and Korpelainen 2012; Yi et al. 2012; Qian et al. 2013). Of these, matK and rbcL in the cp genome are the main sites used to develop DNA barcodes in plants and we identified four SNPs in these two genes in the Perilla species (Fig. 3, Supplementary Table S3). Among these, three nsSNPs in matK can be used as a molecular marker to distinguish between P. hirtella and the other eight species, and one sySNP in rbcL can distinguish P. setoyensis and P. hirtella from the other seven Perilla species (Supplementary Table S3). Four variable numbers of TRs except for mono-polymorphism can be used to distinguish P. setoyensis and/or P. hirtella from the remaining Perilla species (Table 5). Furthermore, variations were detected in all regions of the 45S nrDNA (18 in total, including two InDels), except in the 5.8S rRNA (Table 6). Of the 18 variations in the 45S nrDNA, seven (six SNPs and one InDel) are unique polymorphic markers that can be used to discriminate between Perilla species. Gel-based assays also used in this study (Fig. 4) allowed us to develop species-specific markers that can be used to select specific genotypes from diverse populations, which is especially critical for Perilla breeding.

In our comparative sequence analysis, only two ssSNPs in the cp genome had very low levels of genetic diversity among the six cultivated Perillas. The 122 ssSNPs detected among the three wild species indicated that natural diversity is preserved in wild Perilla species (Supplementary Table S3). Thus, we revealed high levels of genetic variations among the three wild Perilla species and high levels of similarity among the six cultivated species in the cp genome of nine Perillas. Moreover, phylogenomic analysis showed that all cultivated Perilla species clustered with P. citriodora (Fig. 5). According to a previous study (Mo et al. 2017), a whole cp genome comparison between diploid P. citriodora and cultivated tetraploid P. frutescens showed that these species had over 99.0% sequence similarity, and thus P. citriodora was much more closely related to P. frutescens than to other wild Perilla species. Previously reported RFLP and randomly amplified polymorphic DNA (RAPD) analyses also indicated that P. citriodora was the closest relative of the cultivated species, P. frutescens, among the wild species (Ito et al. 1998). These results suggested that the six cultivated species might share a maternal ancestor. Our results also support the finding that the tetraploid genomes of P. frutescens are composed of two copies of the P. citriodora genome and two copies of an unknown Perilla species (Honda et al. 1994). Therefore, the P. citriodora genome may have contributed to the formation of the tetraploid genome of these cultivated species.

Collectively, we analyzed the complete cp genome and 45S nrDNA sequences using NGS technology and analyzed the SNPs and TRs for molecular marker that differed among the nine Perillas. Polymorphisms identified in this study can be used to analyze the origin and geographic distribution of economically important species of Perilla and can be applied to identify commercial cultivars and to determine their purity using DNA barcodes derived from the cp genome and 45S nrDNA sequences.


This work was financially supported by the National Agricultural Genome Program (PJ013355) of RDA, Republic of Korea.

Supplementary Information
Fig. 1. Chloroplast genome map of the nine Perilla species. Genes labeled inside the circle are transcribed clockwise and genes outside are transcribed counter-clockwise. Genes belonging to different functional groups are color coded. The dark gray ring corresponds to the GC content, while the light gray ring corresponds to the AT content.
Fig. 2. Assembly of the complete 45S nrDNA unit. The completed 45S nrDNA unit of P. citriodora, consisting of 18S rRNA, an internal transcribed spacer (ITS) 1, 5.8S rRNA, ITS 2, 26S rRNA, and the intergenic spacer region. The blue shading indicates the mapping depth of raw reads on the 45S nrDNA sequence of P. citriodora.
Fig. 3. Number of SNPs for different genes and the frequency of non-synonymous (ns) vs. synonymous (sy) SNPs. Interactive comparative analysis of the sequences of nine Perilla species showed that 33 genes contained 81 SNPs, including 44 sySNPs in 27 genes and 37 nsSNPs, which modify amino acid residues, in 15 genes. All genes of the transfer RNA and ribosomal RNA were conserved.
Fig. 4. Validation of SNP and InDel polymorphic sites using 4–8 individuals of each Perilla species. (A-C) SNP analysis of SNP positions in ndhF (A) using primer set CAPS-1 (SacII), in ycf1 (B) using primer set CAPS-2 (VspI), in ycf2 (C) using primer set CAPS-3 (AluI), and in ITS1 (D) using primer set CAPS-4 (BsmAI). (E) PCR analysis of InDel regions in trnKUUU-matK using primer set InDel-1. Abbreviated names on the gels indicate 100-bp marker (M), ‘Chajogi’ (CH), ‘Apureunchajogi’ (AC), ‘PureunJureumchajogi’ (PJC), ‘Jureumchajogi’ (JC), ‘Deulkkae’ (DK), ‘Pureunchajogi’ (PC), and negative control (N). Horizontal arrows represent 300 bp of the 100-bp marker.
Fig. 5. Phylogenomic analysis of chloroplast genomes of Perillas and representative species in the Lamiaceae family. The Olea europaea chloroplast genome was used as outgroup. GenBank accession numbers are in parentheses, and the number at an internal node represents the bootstrap value.

List of cultivated and wild Perilla species used in this study.

noSpeciesSample codeStatusGenBank Accession number

Chloroplast45S nrDNA
1Perilla citriodoraP. citriodoraWildKT220690KT220699
2Perilla frutescens Britt. var. acuta Kudo‘Chajogi’CultivatedKT220685KT220694
3Perilla frutescens Britt. var. japonica Hara. for. discolor Makino‘Apureunchajogi’CultivatedKT220686KT220695
4Perilla frutescens Britt. var. crispa Hand.-Mazz.‘PureunJureumchajogi’CultivatedKT220687KT220696
5Perilla frutescens Britt. Var. crispa Hand.-Mazz. frutescens atropurpurea‘Jureumchajogi’CultivatedKT220688KT220697
6Perilla frutescens Britt. var. japonica Hara‘Deulkkae’CultivatedKT220689KT220698
7Perilla frutescens Britt. var. viridis Makino‘Pureunchajogi’CultivatedKT220684KT220693
8Perilla setoyensisP. setoyensisWildKT220692KT220701
9Perilla frutescens var. hirtellaP. hirtellaWildKT220691KT220700

Chloroplast genome sizes of the nine Perilla species based on de novo assembly.

SampleTotal Length (bp)LSCz) (bp)SSCy) (bp)IRax) (bp)IRbw) (bp)
P. citriodora152,60283,70517,54525,67625,676
P. setoyensis152,60783,71017,54725,67525,675
P. hirtella152,65683,75717,54725,67625,676

z)Large single copy,

y)Small single copy,

x)Inverted repeat region a,

w)Inverted repeat region b.

Chloroplast gene products of the nine Perilla species.

Group of genesName of genes
Photosystem IpsaA, B, C, I, J, ycf3y), ycf4
Photosystem IIpsbA, B, C, D, E, F, H, I, J, K, L, M, N, T, Z
Cytochrome b6/fpetA, Bz), Dz), G, L, N
ATP synthaseatpA, B, E, Fz), H, I
NADH oxidoreductasendhAz), Bz,x), C, D, E, F, G, H, I, J, K
Large subunit ribosomal proteinsrpl2z,x), 14, 16z), 20, 22, 23x), 32, 33, 36
Small subunit ribosomal proteinsrps2, 3, 4, 7x), 8, 11, 12y,x,w), 14, 15, 16z), 18, 19
RNA polymeraserpoA, B, C1z), C2
Unknown function protein coding geneycf1x), 2x), 15x)
Other genesaccD, ccsA, cemA, clpPy)matK, InfA
Ribosomal RNAsrrn16x), 23x), 4.5x), 5x)
Transfer RNAstrnA-UGCz,x), trnC-GCA, trnD-GUC, trnE-UUC, trnF-GAA, trnG-UCCz), trnG-GCC, trnH-GUG, trnI-CAUx), trnI-GAUz,x), trnK-UUUz), trnL-UAAz), trnL-UAG, trnL-CAAx), trnfM-CAU, trnM-CAU, trnN-GUUx), trnP-UGG, trnQ-UUG, trnR-ACGx), trnR-UCU, trnS-GCU, trnS-GGA, trnS-UGA, trnT-GGU, trnT-UGU, trnV-UACz), trnV-GACx), trnW-CCA, trnY-GUA

z)Gene containing a single intron,

y)Gene containing two introns,

x)Two gene copies in IRs,

w)Trans-splicing gene.

Non-synonymous substitutions that modify amino acid residues among the nine Perilla species.


P. citriodoraCHy)ACy)PJCy)JCy)DKy)PCy)P. setoyensisP. hirtella
accD58337–58339Phe (TTT)x)Phe (TTT)Phe (TTT)Phe (TTT)Phe (TTT)Phe (TTT)Phe (TTT)Ile (ATT)Phe (TTT)
ccsA113262–113264Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Phe (TTT)
clpP70863–70865Val (CTC)Val (CTC)Val (CTC)Val (CTC)Val (CTC)Val (CTC)Val (CTC)Ile (TTC)Val (CTC)
matK3342–3344Ser (GAA)Ser (GAA)Ser (GAA)Ser (GAA)Ser (GAA)Ser (GAA)Ser (GAA)Ser (GAA)Leu (AAA)
2130–2132Phe (AAA)Phe (AAA)Phe (AAA)Phe (AAA)Phe (AAA)Phe (AAA)Phe (AAA)Phe (AAA)Ser (GAA)
2094–2096Glu (TCC)Glu (TCC)Glu (TCC)Glu (TCC)Glu (TCC)Glu (TCC)Glu (TCC)Gly (CCC)Gly (CCC)
ndhF110721–110723Met (CAT)Met (CAT)Met (CAT)Met (CAT)Met (CAT)Met (CAT)Met (CAT)Ile (TAT)Met (CAT)
110195–110197Arg (GAA)Arg (GAA)Arg (GAA)Arg (GAA)Arg (GAA)Arg (GAA)Arg (GAA)Arg (GAA)Cys (AAA)
110180–110182Ser (TGT)Ser (TGT)Ser (TGT)Ser (TGT)Ser (TGT)Ser (TGT)Ser (TGT)Arg (GGT)Arg (GGT)
110068–110070Tyr (TAA)Tyr (TAA)Tyr (TAA)Tyr (TAA)Tyr (TAA)Tyr (TAA)Tyr (TAA)Ser (GAA)Ser (GAA)
110021–110023Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Gly (CCT)Lys (TTT)
110020–110022Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Gly (CTT)Lys (TTT)
109895–109897Val (CAA)Val (CAA)Val (CAA)Val (CAA)Val (CAA)Val (CAA)Val (CAA)Ile (TAA)Val (CAA)
109888–109890Ser (GAT)Ser (GAT)Ser (GAT)Ser (GAT)Ser (GAT)Ser (GAT)Ser (GAT)Phe (AAT)Ser (GAT)
petB75774–75776Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Ile (ATT)Val (GTT)Val (GTT)
rpl2283142–83144Phe (GAA)Phe (GAA)Phe (GAA)Phe (GAA)Phe (GAA)Phe (GAA)Phe (GAA)Leu (TAA)Phe (GAA)
83135–83137Gln (GAG)Lys (TAG)Lys (TAG)Lys (TAG)Lys (TAG)Lys (TAG)Lys (TAG)Gln (GAG)Gln (GAG)
rpoA78559–78561Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Asp (GTC)Asp (GTC)
77863–77865Asp (ATC)Asp (ATC)Asp (ATC)Asp (ATC)Asp (ATC)Asp (ATC)Asp (ATC)Glu (CTC)Asp (ATC)
rpoC217958–17960Leu (GGA)Leu (GGA)Leu (GGA)Leu (GGA)Leu (GGA)Leu (GGA)Leu (GGA)Leu (GGA)Phe (AGA)
17565–17567Ile (TAA)Ile (TAA)Ile (TAA)Ile (TAA)Ile (TAA)Ile (TAA)Ile (TAA)Phe (AAA)Ile (TAA)
rps382916–82918Ser (CTT)Ser (CTT)Ser (CTT)Ser (CTT)Ser (CTT)Ser (CTT)Ser (CTT)Ser (CTT)Asn (TTT)
rps879996–79998Pro (GTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Pro (GTC)Pro (GTC)
79982–79984Leu (CAA)Leu (CAA)Leu (CAA)Leu (CAA)Leu (CAA)Leu (CAA)Leu (CAA)Phe (AAA)Leu (CAA)
rps164943–4945Val (CAG)Val (CAG)Val (CAG)Val (CAG)Val (CAG)Val (CAG)Val (CAG)Leu (GAG)Val (CAG)
rps1983719–83721Asn (ATT)Asn (ATT)Asn (ATT)Asn (ATT)Asn (ATT)Asn (ATT)Asn (ATT)Lys (TTT)Asn (ATT)
109217–109219Glu (GAA)Glu (GAA)Glu (GAA)Glu (GAA)Glu (GAA)Glu (GAA)Glu (GAA)Glu (GAA)Lys (AAA)
127091–127093Glu (CTT)Glu (CTT)Glu (CTT)Glu (CTT)Glu (CTT)Glu (CTT)Glu (CTT)Glu (CTT)Lys (TTT)
126527–126529Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Lys (TTT)Glu (CTT)Glu (CTT)
ycf1125843–125845Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Ala (CTC)Thr (TTC)Ala (CTC)
123676–123678Thr (GTA)Thr (GTA)Thr (GTA)Thr (GTA)Thr (GTA)Thr (GTA)Thr (GTA)Thr (GTA)Ile (ATA)
123615–123617Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Glu (CTC)Asp (ATC)Glu (CTC)
122632–122634Ile (ATG)Ile (ATG)Ile (ATG)Ile (ATG)Ile (ATG)Ile (ATG)Ile (ATG)Ser (CTG)Ser (CTG)
ycf288843–88845Met (TGA)Met (TGA)Met (TGA)Met (TGA)Met (TGA)Met (TGA)Met (TGA)Thr (CGA)Met (TGA)
147465–147467Met (ATC)Met (ATC)Met (ATC)Met (ATC)Met (ATC)Met (ATC)Met (ATC)Thr (GTC)Met (ATC)
90151–90153Val (TTG)Val (TTG)Val (TTG)Val (TTG)Val (TTG)Val (TTG)Val (TTG)Ala (CTG)Val (TTG)
146157–146159Val (ACT)Val (ACT)Val (ACT)Val (ACT)Val (ACT)Val (ACT)Val (ACT)Ala (GCT)Val (ACT)

z)The positions of SNPs were based on the P. citriodora sequence.

y)CH, ‘Chajogi’; AC, ‘Apureunchajogi’; PJC, ‘PureunJureumchajogi’; JC, ‘Jureumchajogi’; DK, ‘Deulkkae’; PC, ‘Pureunchajogi’.

x)Amino acid (nucleotide).

Variable number of tandem repeats present in the nine Perilla species.

RegionLocationSize (bp)Copy number variationConsensus

P. citriodora, ACz), PJCz), JCz), DKz)CHz) PCz)P. setoyensisP. hirtella
LSCatpH - atpIintergenic114141615T
LSCpetA - psbJintergenic71112ATTATAT
SSCx)rpl33 - trnLUAGintergenic82122CAAATTCC

z)CH, ‘Chajogi’; AC, ‘Apureunchajogi’; PJC, ‘PureunJureumchajogi’; JC, ‘Jureumchajogi’; DK, ‘Deulkkae’; PC, ‘Pureunchajogi’.

y)Large single copy.

x)Small single copy.

Summary of nucleotide polymorphisms in 45S nrDNA sequences of the nine Perilla species.

TypeLocusNucleotide Positionz)9 perilla species

P. citriodoraCHy)ACy)PJCy)JCy)DKy)PCy)P. setoyensisP. hirtella

z)The positions of SNPs were based on the P. citriodora sequence.

y)CH, ‘Chajogi’; AC, ‘Apureunchajogi’; PJC, ‘PureunJureumchajogi’; JC, ‘Jureumchajogi’; DK, ‘Deulkkae’; PC, ‘Pureunchajogi’.

x)No detection.

  1. Benson, G (1999). Tandem repeats finder: a program to analyze DNA sequences. Nucleic Acids Res. 27, 573-580.
  2. Cho, KS, Yang, TJ, Hong, SY, Kwon, YS, Woo, JG, and Park, HG (2006). Determination of cytoplasmic male sterile factors in onion plants (Allium cepa L.) using PCR-RFLP and SNP markers. Mol. Cells. 21, 411-417.
  3. Clifford, MN (2000). Anthocyanins nature, occurrence and dietary burden. J Sci Food Agric. 80, 1063-1072.
  4. Cui, L, Veeraraghavan, N, Richter, A, Wall, K, Jansen, RK, and Leebens-Mack, J (2006). ChloroplastDB: the Chloroplast Genome Database. Nucleic Acids Res. 34, D692-D696.
    KoreaMed CrossRef
  5. Downie, SR, Katz-Downie, DS, and Cho, KJ (1996). Phylogenetic analysis of Apiaceae subfaimily Apioideae using nucleotide sequences from the chloroplast rpoC1 intron. Mol Phylogenetic Evol. 6, 1-18.
  6. Drescher, A, Ruf, S, Calsa, T, Carrer, H, and Bock, R (2000). The two largest chloroplast genome-encoded open reading frames of higher plants are essential genes. Plant J. 22, 97-104.
    Pubmed CrossRef
  7. Felsenstein, J (2005). PHYLIP (Phylogeny Inference Package) version 3.6. Distributed by the author.
  8. Fernandes, F, Pereira, L, and Freitas, AT (2009). CSA: an efficient algorithm to improve circular DNA multiple alignment. BMC Bioinformatics. 10, 230.
    Pubmed KoreaMed CrossRef
  9. Frazer, KA, Pachter, L, Poliakov, A, Rubin, EM, and Dubchak, I (2004). VISTA: computational tools for comparative genomics. Nucleic Acids Res. 32, 273-279.
  10. Hall, TA (1999). BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series No. 41, 95-98.
  11. Hollingsworth, PM, Graham, SW, and Little, DP (2011). Choosing and using a plant DNA barcode. PLoS One. 6, e19254.
    Pubmed KoreaMed CrossRef
  12. Honda, G, Yuba, A, Kojima, T, and Tabata, M (1994). Chemotaxonomic and cytogenetic studies on Perilla frutescens var. citriodora (“Lemon Egoma”). Nat Med. 48, 185-190.
  13. Huotary, T, and Korpelainen, H (2012). Complete chloroplast genome sequence of Elodea canadensis and comparative analyses with other monocot plastid genomes. Gene. 508, 96-105.
  14. Ito, M, and Honda, G (1996). A taxonomic study of Japanese wild Perilla (Labiatae). J Phytogeogr Taxon. 44, 43-52.
  15. Ito, M, Kato, H, Oka, Y, and Honda, G (1998). Phylogenetic analysis of Japanese Perilla species by using DNA polymorphisms. Nat Med. 52, 248-252.
  16. Jansen, RK, Cai, Z, Raubeson, LA, Daniell, H, Depamphilis, CW, and Leebens-Mack, J (2007). Analysis of 81 genes from 64 plastid genomes resolves relationships in angiosperms and identifies genome-scale evolutionary patterns. Proc Natl Acad Sci USA. 104, 19369-19374.
    Pubmed KoreaMed CrossRef
  17. Jansen, RK, Raubeson, LA, Boore, JL, dePamphilis, CW, Chumley, TW, and Haberle, RC (2005). Methods for obtaining and analyzing whole chloroplast genome sequence. Methods Enzymol. 395, 348-384.
  18. Katoh, K, Misawa, K, Kuma, K, and Miyata, T (2002). MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30, 3059-3066.
    Pubmed KoreaMed CrossRef
  19. Katoh, K, and Standley, DM (2013). MAFFT multiple sequence alignment software version 7: Improvements in performance and usability. Mol Biol Evol. 30, 772-780.
    Pubmed KoreaMed CrossRef
  20. Kim, HK, Cho, YS, Yang, JW, Choi, YW, Kang, JS, and Lee, YJ (2010). Analysis of genetic variation of Perilla frutescens var. crispa germplasm using RAPD. Journal of Life Science. 20, 119-123.
  21. Kim, KH, Lee, SC, Lee, JK, Lee, HO, Joh, HJ, and Kim, NH (2015). Comprehensive survey of genetic diversity in chloroplast genome and 45S nrDNAs within Panax ginseng species. PLoS One. 10, e0117159.
  22. Kim, KJ, and Lee, HL (2004). Complete chloroplast genome sequences from Korean ginseng (Panax schingeng Nees) and comparative analysis of sequence evolution among 17 vascular plants. DNA Res. 11, 247-261.
    Pubmed CrossRef
  23. Koressaar, T, and Remm, M (2007). Enhancements and modifications of primer design program Primer3. Bioinformatics. 23, 1289-1291.
    Pubmed CrossRef
  24. Lagesen, K, Hallin, P, Rødland, EA, Stærfeldt, HH, Rognes, T, and Ussery, DW (2007). RNAmmer: consistent and rapid annotation of ribosomal RNA genes. Nucleic Acids Res. 35, 3100-3108.
    Pubmed KoreaMed CrossRef
  25. Lee, MH, Yang, KW, Ha, TJ, Jung, CS, Pae, SB, and Hwang, CD (2011). Development of SCAR marker for identification of the Perilla species. Kor J Breed Sci. 43, 186-193.
  26. Leebens-Mack, J, Raubeson, LA, Cui, L, Kuehl, JV, Fourcade, MH, and Chumley, TW (2005). Identifying the basal angiosperm node in chloroplast genome phylogenies: sampling one’s way out of the Felsenstein zone. Mol Biol Evol. 22, 1948-1963.
    Pubmed CrossRef
  27. Lohse, M, Drechsel, O, and Bock, R (2007). OrganellarGenomeDRAW (OGDRAW): a tool for the easy generation of high-quality custom graphical maps of plastid and mitochondrial genomes. Curr Genet. 52, 267-274.
    Pubmed CrossRef
  28. Lukas, B, and Novak, J (2013). The complete chloroplast genome of Origanum vulgare L. (Lamiaceae). Gene. 528, 162-169.
  29. Ma, YJ, Qu, FY, and Xu, JJ (1998). Sequence differences of rDNA-ITS2 and species-diagnostic PCR assay of Anopheles sinensis and Anopheles anthropophagus from China. J Med Coll PLA. 13, 123-128.
  30. Markakis, P (1982). Anthocyanins as food color. New York, U.S.A: Academic Press
  31. Mo, JS, Kim, K, Lee, MH, Lee, JH, Yoon, UH, and Kim, TH (2017). The complete chloroplast genome sequence of Perilla citriodora (Makino) Nakai. Mitochondrial DNA A DNA Mapp Seq Anal. 28, 131-132.
  32. Murray, MG, and Thompson, WF (1980). Rapid isolation of high molecular weight plant DNA. Nucleic Acids Res. 8, 4321-4326.
    Pubmed KoreaMed CrossRef
  33. Nazar, RN, Sits, TO, and Busch, H (1976). Sequence homologies in mammalian 5.8S ribosomal RNA. Biochemistry. 15, 505-508.
    Pubmed CrossRef
  34. Nikiforova, SV, Cavalieri, D, Velasco, R, and Gremykin, V (2013). Phylogenetic analysis of 47 chloroplast genome clarities the contribution of wild species to the domesticated apple maternal line. Mol Biol Evol. 30, 1751-1760.
    Pubmed CrossRef
  35. Powell, W, Morgante, M, McDevitt, R, Vendramin, GG, and Rafalski, JA (1995). Polymorphic simple sequence repeat regions in chloroplast genomes: applications to the population genetics of pines. Proc Natl Acad Sci USA. 92, 7759-7763.
    Pubmed KoreaMed CrossRef
  36. Qian, J, Song, J, Gao, H, Zhu, Y, Xu, J, and Pang, X (2013). The complete chloroplast genome sequence of the medicinal plant Salvia miltiorrhiza. PLoS One. 8, e57607.
    Pubmed KoreaMed CrossRef
  37. Schliep, KP (2011). Phangorn: phylogenetic analysis in R. Bioinformatics. 27, 592-593.
    KoreaMed CrossRef
  38. Schwartz, S, Kent, WJ, Smit, A, Zhang, Z, Baertsch, R, and Hardison, RC (2003). Human-mouse alignments with BLASTZ. Genome Res. 13, 103-107.
    Pubmed KoreaMed CrossRef
  39. Sumida, M, Kato, Y, and Kurabayashi, A (2004). Sequencing and analysis of the internal transcribed spacers (ITSs) and coding regions in the EcoR I fragment of the ribosomal DNA of the Japanese pond frog Rana nigromaculata. Genes Genet Syst. 79, 105-118.
    Pubmed CrossRef
  40. Untergrasser, A, Cutcutache, I, Koressaar, T, Ye, TJ, Faircloth, BC, and Remm, M (2012). Primer3 - new capabilities and interfaces. Nucleic Acids Res. 40, e115.
  41. Wang, H, Cao, G, and Prior, RL (1997). Oxygen radical absorbing capacity of anthocyanins. J Argic Food Chem. 45, 304-309.
  42. Wicke, S, Schneeweiss, GM, dePamphilis, CW, Müller, HF, and Quandt, D (2011). The evolution of the plastid chromosome in land plants: gene content, gene order, gene function. Plant Mol Biol. 76, 273-297.
    Pubmed KoreaMed CrossRef
  43. Wolfe, KH, Li, WH, and Sharp, PH (1987). Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs. Proc Natl Acad Sci USA. 84, 9054-9058.
    Pubmed KoreaMed CrossRef
  44. Wyman, SK, Jansen, RK, and Boore, JL (2004). Automatic annotation of organellar genomes with DOGMA. Bioinformatics. 22, 3252-3255.
  45. Yang, M, Zhang, X, Liu, G, Yin, Y, Chen, K, and Yun, Q (2010). The complete chloroplast genome sequence of data palm (Phoenix dactylifera L.). PLoS One. 5, e12762.
  46. Yi, DK, and Kim, KJ (2012). Complete chloroplast genome sequence of important oilseed crop Seasmum indicum L. PLoS One. 7, e35872.