Chloroplast (cp) genomes of Lilium amabile, L. callosum, L. lancifolium, and L. philadelphicum were fully sequenced. Using these four novel cp genome sequences and five other previously sequenced cp genomes, features of the cp genomes were characterized in detail among species in the genus Lilium and other related genera in the order Liliales. The lengths and nucleotide composition showed little variation. No structural variation was found among the cp genomes in Liliales. Gene contents were conserved among four newly sequenced cp genome in Lilium species, the only differences being in two pseudogenes. We identified 112 genes in 13 functional categories, 18 of which carried introns that were conserved among the species in Liliales. There were 16–21 SSR loci (>12 bp, >3 repeats) in the cp genomes in Lilium and the genomic locations of these loci were highly variable among the species. Average mutations were 15 SNPs per 1kb and 5 indels per 1kb, respectively, in the cp genomes of the newly sequenced four Lilium species. Phylogenetic classifications revealed some discrepancies between trees based on the cp genomes and previous classifications based on the morphology and geographic distributions.
Citation: Kim J-H, Lee S-I, Kim B-R, Choi I-Y, Ryser P, Kim N-S (2017) Chloroplast genomes of Lilium lancifolium, L. amabile, L. callosum, and L. philadelphicum: Molecular characterization and their use in phylogenetic analysis in the genus Lilium and other allied genera in the order Liliales. PLoS ONE 12(10): e0186788. https://doi.org/10.1371/journal.pone.0186788
Editor: Lorenzo Peruzzi, Università di Pisa, ITALY
Received: June 8, 2017; Accepted: October 6, 2017; Published: October 24, 2017
Copyright: © 2017 Kim et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This work was carried out with a grant from National Research Foundation of Korea (https://www.nrf.re.kr/index) (Project No. C1013144-01-01) to NSK, and a Golden Seed Project grant from the Rural Development Administration of Korea (http://www.rda.go.kr; http://www.rda.go.kr/main/mainPage.do) to JHK (project No. 213007-05-1-CGM00). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Lilies, the plants in the genus Lilium, are perennial herbaceous flowering plants with over 110 species distributed widely in temperate and boreal zones in the Northern Hemisphere . All lilies grow from large bulbs, plant height ranging from 50 cm to 200 cm. Because lilies bear large and showy flowers in diverse colors, which are often fragrant, many commercial cultivars have been produced by interspecific hybridization . Currently lilies are the number three flowering crop after roses (Rosa) and mums (Chrysanthermum) worldwide .
Taxonomical classification of the genus Lilium has been disputed and repeatedly modified since its first botanical classification into five sections based on the morphological characters by Endlicher in 1836 . In 1949, Comber divided the genus into seven sections based on 13 different morphological characteristics and germination types . Although the seven-section system has been slightly modified by subsequent cytogenetic and interspecific hybridization analyses [6–7], it is basically solid with only a few species being re-assigned to different sections. Recently, Pelkonen and Pirttilä  reviewed the lily classifications based on the morphology, cytogenetic and molecular analyses, proposing a classification into seven sections as follows; Martagon, Pseudolirium (American group), Archelirion (Oriental group), Lilium (Candidium group), Sinomartagon (Asiatic group), Leucolirion (Trumpet group), and Daurolirion (L. bulbiferum and Dauricum group).
Chloroplasts are cellular organelles in photosynthetic plants and algae. The chloroplast genomes (cp genome) vary typically between 120 and 170 kb in, and are comprised of a quadripartite structure that includes two copies of invert repeat (IR) regions separated by a large-single copy (LSC) and a small-single copy (SSC) region [9–10]. The number of genes encoded in cp genome varies from 100–120 genes that are often arranged in an operon-like manner and transcribed as polycistronic precursor mRNAs which are processed into mature mRNAs by splicing and nucleolytic cleavage [10–12]. The inheritance of the cp genome is predominantly by maternal inheritance except in a few species of eudicots in the families of Geraniaceae, Campanuclaceae and Fabaceae which have biparental cp genome inheritance . Because the uniparental inheritance does not allow sequence shuffling by recombination, the cp genome sequences have been the primary choice for delineating maternal lineages in plant systematic studies [13–15]. In Lilium and allied genera, Hayashi and Kawano  analyzed the phylogenetic relationships using two cp genes, rbcL and matK, sequences according to which the species in the genus Lilium can be grouped into three different major groups. The authors argued that the molecular-systematic results were not congruent with the classifications based on morphology. In the phylogenetic analysis of Lilium species endemic in Qinghai-Tibet Plateau (Q-T Plateau) using matK sequences, Gao et al.  grouped these lilies into 9 lineages in which the species in different sections of Comber  and Pelkonen and Pirttilä  were mixed. Moreover, the phylogenetic grouping using the matK gene sequences were different from grouping based on the nuclear ITS sequence .
The advent of the next-generation sequencing technology and various bioinformatics tools have allowed easier gaining of more cp genome sequences in diverse plant species [18–20]. In lilies, the whole cp genome sequences have been reported for L. taliense , L. tsingtauense , L. hansonii , L. fargesii , L. cernuum , L. distichum , L. longiflorum , and L. superbum (KP462883). In the present work we are adding four more Lilium species with a sequenced whole cp genome; L. amabile, L. callosum, L. lancifolium, and L. philadeliphicum. The four species were chosen to add the chloroplast genomes in the Korean endemic Lilium species in the section Sinomartagon and compare them with the cp genome of L. philadelphicum that is a native North American species in the section Pseudolilium . The current report contains the comprehensive genomic and phylogenomic analyses of the cp genomes in the genus Lilium.
Materials and methods
DNA preparations, sequencing, and assembly
Chloroplast genomes of four Lilium species were sequenced: L. lancifolium, L. amabile, L. callosum and L. philadelphicum. L. lancifolium (Accession No GWL0702), L. amabile (Accession No GWL15789), and L. callosum (Accession No GWL3662) were accessions that have been maintained at the Lilium germplasm nursery in Kangwon National University, Korea. L. philadelphicum was an accession collected from its natural habitat (46° 2' 5.63"N; 81° 46' 23.172" W) close to Sudbury, Ontario, Canada, in June 2016. L. philadelphicum is not on the list of the endangered or protected species, and no permissions were required for collections of leaves for this specimen from its natural habitats.
Fresh leaves (~100 mg) were sampled from young plants. Cellular DNA was extracted using the DNAeasy Plant Maxi Kit (QIAGEN, Valencia, CA, USA). DNA (5 ug) samples were then sheared to an average size about 300 bp by nebulization with compressed N2 gas. Quality of the sheared DNA was assessed using a Bioanalyzer 2200 (Agilent Technologies, Santa Clara, CA, USA), and a paired-end library was constructed using the Illumina Paired-End Library Kit (Illumina, San Diego, CA, USA). Genomic DNA sequencing was then carried out on a single lane of a HighSeq 2000 flow cell by Phyzen Inc. (Seoul, Korea). The sequence was filtered and assembled using de novo assembly package software, CLC Assembly Cell v.4.2.1 (https://www.qiagenbioinformatics.com/products/clc-assembly-cell/, Quigen Co., Ltd. Hilden, Germany) for a complete chloroplast genome assembly using the dnaLCW method (de novo assembly of low coverage whole-genome shotgun sequencing method) as suggested protocol of Kim et al. . The ambiguous sequences including structural borders and mono-polymer were manually edited. The complete chloroplast genome map was produced using reported chloroplast genomes from other Lilium species as references (KM103364 in L. hasonii, KC968977 in L. longiflorum, KX592156 in L. fargesii, KP462883 in L. superbum) [20–26]. The circular chloroplast genome map was then drawn using the OrganellarGenomeDRAW tool (ORDRAW) .
Gene and simple sequence repeat (SSR) annotation
Gene annotation of the newly sequenced cp genomes was performed using the Dual Organellar GenoMe Annotator (DOGMA) , and all initiation and stop codons were manually confirmed in the DOGMA-annotated data. Predicted introns were further checked by comparison with other cp genome sequences, and all annotated transfer RNA (tRNA) genes were verified using ARAGORN . SSR sequences were detected with the UGENE program (http://ugene.net/) by a command “Find tandems” with a default set a minimum size 12 bp and repeat count 3.
The nine cp genome sequences were aligned using MAFFT version 7 program (http://mafft.cbrc.jp/alignment/software/). The VCF (variant call format) was built using Msa2vcf (http://lindenb.github.io/jvarkit/MsaToVcf.html). Then, the SNPs and indels were identified manually.
Sequence identity and phylogenetic analysis among the cp genomes in Liliales
Cp genomes of 13 species in the order Liliales (nine Lilium species, two Fritillaria species, one of each Smilax and Alstroemeria species) were used for sequence identity and phylogenetic analyses. The cp genome of Allium cepa (order Asparagales) was used as an out-group in the analyses. Except for the four newly sequenced cp genomes, the cp genomes were downloaded from GenBank. A multiple sequence alignment was then generated in ClustalW, and gaps were edited using the MEGA5 program . For sequence identity comparison and sequence divergence along the cp genomes, sequences were compared and plotted using the mVISTA program (http://genome.lbl.gov/vista/mvista/submit.shtml). For phylogenetic analyses, two data sets were used; one with the whole cp genome sequences and another with protein coding sequences. After maximum parsimony analysis was performed with PAUP v4b10 , maximum likelihood (ML) analyses were performed with 1000 bootstrap replicates using RAxML-HPC BlackBox v.8.1.24 at Cipres Science Gateway site (http://www.phylo.org/tools/obsolete/raxmlhpc2.html#) .
Cp genome length and AT contents among the Lilium species
The complete cp genomes of four Lilium species were successfully assembled using high-quality Illumina sequence data filtered by CLC Assembly Cell software. The cp genomes were assembled with average coverage depth 177x in L. amabile, 92x in L. callosum, 58x in L. lancifolium, 116x L. philadelphicum, respectively, using at least 13 Gbp genome sequence data generated by Illumina sequencer platform (S1 Table). Table 1 summarizes the length of cp genomes and GC contents in Lilium species. Total lengths of the cp genomes range from 152,175 in L. philadelphicum to 153,235 in L. fargesii. The lengths of LSC range from 81,580 in L. philadelphicum to 82,230 in L. longiflorum, and those of SSC from 17,038 in L. fargesii to 17,620 in L. hansonii, respectively. The lengths of IRs varies from 26,491 in L. callosum to 26,990 in L. fargesii. The nucleotide compositions of cp genomes had a high AT content in the range of 62.93% in L. philadelphicum to 63.01% in L. fargesii. The IR regions showed lower AT ratio than the LSC and SSC regions in all Lilium species. Thus, the length and nucleotide variations were low among the cp genomes in the Lilium species. The four newly sequenced cp genomes in the current study did not show any structural and gene order variations (Fig 1). The cp genomes were deposited to GenBank with accession numbers KY940844 for L. lancifolium, KY940845 for L. amabile, KY940846 for L. callosum, and KY940847 for L. philadelphicum, respectively.
The gene orders of the cp genomes of L. amabile, L. callosum, and L. lancifolium were identical but different from the cp genome of L. philadelphicum by two pseudogenes (red arrows). The former three cp genomes have a pseudogene ndhG in SSC region, but this pseudo gene was absent in L. philadelphicum. The pseudogene cemA in LSC was present in the cp genome in L. philadelphicum, but absent in the former three cp genomes.
Genes encoded in the cp genomes in lilies
In each cp genome of the four newly sequenced Lilium species, we annotated a total of 156 genes, of which 102 are protein-coding genes, 46 are tRNA genes, and 8 are ribosomal RNA (rRNA) genes (S2 Table). Because some genes are duplicated or triplicated, the 156 genes are classified into 112 different genes. Table 2 shows the 112 genes that are classified into 13 functional categories, with no differences among the four newly sequenced cp genomes. The LSC and SSC regions contain 96 and 12 genes, respectively, and each IR region has 24 genes that are inversely oriented to one another. There are two pseudogenes, ndhG in L. philadeliphicum and cemA in L. amabile, L. callosum and L. lancifolium, which carried premature stop codons (Table 2).
Eighteen genes contain introns; ten protein-coding genes (rps16, atpF, rpoC1, petB, petD, rpl16, rpl2, ndhB, rps12, ndhA) and six tRNA genes (trnK-UUU, trnG-UCC, trnL-UAA, trnV-UAC, trnI-GAU, trnA-UGC) have single introns, whereas two protein-coding genes (clpP and ycf3) have two introns each. One intron-containing gene (rps12) is trans-splicing, having the first exon in the LSC and the second and third exons in IR regions (Table 2).
Of the 18 intron-containing genes, introns in 17 genes were conserved among the species in the genera Lilium, Fritillaria, and Smilax in the order Liliales (Table 3). The intron in trnG-UCC was not present in the L. fargesii and two Fritillaria species. Six genes including the trnG-UCC showed intron absence in Allium cepa in the order Asparagales.
SSR sequences in the cp genomes in Lilium species
We identified 96 SSR loci with a threshold of over 10 bp and 3 repeats and the 96 SSR loci consisted of 14 di-nucleotide repeats, 74 tri-nucleotide repeats, and 8 tetra-nucleotide repeats in L. lancifolium cp genome (data not shown). When the stringency was increased to a threshold over 12 bp and 3 repeat count, the number of SSR loci was narrowed to 42 SSR loci which consisted of eight di-nucleotide repeats, 12 tri-nucleotide repeats, 17 tetra-nucleotide repeats, and five penta-nucleotide repeats (Table 4). The SSR loci were mostly present in the LSC regions except of the three loci in SSC. No SSR locus was present in the invert repeat regions (IRs). Twelve, three, and 27 SSR loci were present in intronic regions, exons and intergenic spacers, respectively. The number of SSR loci varied from 16 in L. lancifolium to 21 in L. fargesii and the presence/absence polymorphisms were highly variable among the species. Of the 42 SSR loci, only four loci were present in all the Lilium cp genomes. L. amabile and L. callosum shared exact SSR loci and repeat numbers. The SSR loci in L. lancifolium were all present in L. amabile and L. callosum, but one locus (trnL-UAA) at LSC was different in the number of repeats as (AT)8 in L. lancifolium and (AT)10 in L. amabile and L. callosum
SNPs and Indels among cp genomes in Lilium species
We identified 3,018 mutations which consisted of 2,271 SNPs and 747 indels among the 4 newly sequenced cp genomes (Table 5, S3 Table). The average variations were 15 SNPs per 1 kb and 5 indels per 1kb, respectively. The most variable region was in the introns with 67.7 mutations per 1 kb, followed by the intergenic region with 36 mutations per 1 kb. Of the 112 genes, 80 genes showed variations (Fig 2, S4 Table). Of the 80 genes with SNPs, only 27 had indels. The number of SNPs in a gene was not related with the number of indels, 19 genes having more SNPs than indels, while 7 genes had more indels than SNPs (S5 Table). Gene length was highly correlated with the number of SNPs, but the the number of indels was not related with the gene length. Four of the 46 tRNA genes showed variations.
Sequence divergence along the cp genomes among species in Liliales
We identified no major structural variations such as inversions or large deletions in cp genomes of the 9 Lilium species. Sequence divergence hotspot regions along the cp genomes were analyzed among nine Lilium species. Five other species (two Fritillaria species, Smilax china, Alsroemeria aurea, and Allium cepa) were included in the cp genome variation survey (Fig 3). Among the Lilium species, most sequence variations were found in the noncoding intergenic regions in the LSC and SSC regions. Two hypervariable regions were identified in the gene-sparse intergenic regions in LSC, and are designated by bars at the top of Fig 3. The sequence variations in the IR regions were comparably lower than the LSC and SSC regions. In comparisons beyond the Liliales, sequence variations were also present in intergenic regions throughout the cp genomes. As expected, sequence divergence among the species in Liliaceae (the genera Lilium and Fritillaria) was lower along the whole cp genomes, compared to the divergence among all the species.
Phylogenetic analysis among species in Liliales
Phylogenetic trees based on the whole-cp genomes and those based on the 71 protein coding genes were not different from each other, and showed clustering which agreed with the taxonomical hierarchical order (Fig 4). Allium cepa in the order Asparagales was out-clustered from the species in Liliales. Among the species in Liliales, Alstroemeria aurea in the family Alstroemeriacea and Smilax china in the family Smilaceae were out-grouped from the Liliaceae species. The two Fritillaria species showed distinct clustering from the species in the genus Lilium. The nine Lilium species were clustered in two groups; one group with three Sinomartagon lilies (L. lancifolium, L. callosum, and L. amabile), one Martagon lily (L. hansonii), and one Leucolirion lily (L. longiflorum), and another group with two Pseudolirium lilies (L. superbum and L. philadelphicum), one Sinomartagon lily (L. fargesii), and one Martagon lily (L. distichum).
Phylogenetic trees based on the whole cp genome sequences (A) and functional genes (B) among 13 species in the order Liliales and Allium cepa. The trees were made using maximum likelihood algorithm and the numbers on the nods designate the bootstrap values.
This report contains novel cp genome sequences of four Lilium species and other previously sequenced cp genomes in Liliales for the purpose of genomics and phylogenomics analyses, based on the whole cp genome sequences. The cp genomes in nine Lilium species, including five previously sequenced cp genomes showed remarkably low variations in length, base compositions, gene contents, intron retentions, and genome structure. Cp genomes of certain lineages of land-plants have undergone gene losses and significant structural rearrangements . A good example is the genus Astragalus in the family Fabaceae, in which inversions and gene losses resulted in the variations in cp genome structure and gene contents between species . In the analysis of 81 genes from 64 plastid genomes, Jansen et al.  reported 62 independent gene and intron losses that are limited to more derived monocot and eudicot clades. Kim and Kim  surveyed gene losses among cp genomes in monocots and noted that gene losses were frequent events in some monocot families. Among three families, Liliaceae, Smilaceae, and Alstromeriaceae in the order Liliales, they found that gene content and order were conserved except of the infA loss in Smilax and Altroemeria. Introns in cp genes were known to be generally conserved in land-plant cp genomes. We observed an intron loss polymorphism in trnG-UCC gene among the Lilium speices and two Frillaria species. This intron, however, was present in Smilax china and Alstromemeria aurea in Liliales, but absent in Allium cepa in Asparagales. The presence/absence polymorphism of this gene was also reported both among monocot and eudicot species . Thus, the intron loss of this gene must have happened independently, rather than in a lineage specific manner.
Simple sequence repeats (SSR) occur in both nuclear and cp genomes in all plants. Cp SSRs have been demonstrated as robust marker systems in population genetics and ecology [35–37], but has some drawbacks due to low variation compared to the high polymorphism in nuclear SSRs . Prior to this report, several cp genomes in Lilium species have been reported [20–26], but no data on the cp SSRs are available. SSRPs (simple sequence repeat polymorphisms) are derived from two mechanisms such as unequal crossing-over and DNA replication slippage . However, there is no unequal crossing-over in the cp genome SSRs, resulting in the low intra-specific polymorphisms as noted by Wheeler et al. . Because once the SSR sequences occur de novo in the cp genome, they may stay in the position in the lineages. Thus, the presence/absence polymorphisms of the SSR locus between species may be useful indicators in the analysis of genetic relatedness. In practice, L. amabile and L. callosum, shared the exact loci, these two species also showed a very close phylogenetic relatedness.
Cp genome structural changes have been noted in several unrelated lineages in flowering plants such as Geraniaceae , Onagraceae , Campanulaceae , and Fabaceae . Inversions and heteroplasmic variations have been reported within the genus Astragalus in the family Fabaceae . However, no structural variations were observed among the cp genomes in the genus Lilium in the current study. Conservation of the cp genome structure in Liliales has also been reported by Kim and Kim , supporting our finding of constrained structural variation in the cp genomes in the genus Lilium. In a comparison between two cp genomes of tropical trees in the genus Machilus in the family Lauraceae, Song et al.  counted 297 mutation events including a micro-inversion, 65 indels, and 231 substitutions. In the coding regions, they counted 95 SNPs between the two species. The number mutations in the cp genomes in Lilium species observed in the current study was comparatively higher. The discrepancy between the two studies may derive from the difference in the number cp genomes: four cp genomes in our study compared to two cp genomes in the study by Song et al .
We identified two hypervariable regions in the LSC regions. Zhang et al.  surveyed the mutations in cp genome wide variations in five Epimedium species in the family Berberidaceae, in which overall variation patterns along the cp genomes are congruent with our results, but they did not observe such prominent hypervariable. In our analysis, the two hypervariable regions were also found in the Fritillaria species in Liliaceae. Shaw et al. [14, 45–46] surveyed noncoding cp DNA sequences among angiosperm species to choose the regions for phylogenetic and phylogeographic studies, in which they showed that most variations are in the noncoding intergenic regions in LSC and SSC regions. Moreover, they reported two variable regions within the LSC and one within the SSC. The two hypervariable regions in our study were the same regions as in their report in LSC. However, Smilax and Alstoemeria species in the order Liliales do not have the conspicuous hypervariable regions which show variations along the LSC and SSC regions. Thus, the two hypervariable regions might be limited to the Liliaceae or to the tribe Lilieae.
Cp genome sequences have been employed for phylogenetic analysis in the genus Lilium by several investigators [20–24, 26]. We are adding four novel cp genomes to have more comprehensive analyses on interspecific relationships. Our analyses basically confirms the phylogenetic trees based on the whole cp genome sequences and protein coding genes. The nine Lilium species were clustered into two groups in the phylogenetic trees (Fig 3), which was consistent with the sequence divergence patterns generated by the mVISTA program (Fig 2). Our results are congruent with the results of Bi et al. . In their study, seven Lilium cp genomes were grouped into two groups in which the L. superbum (section Pseudolirion) and L. fargesii (section Sinomartagon) were grouped into one cluster and L. longiflorum (section Leucolirion) and L. hansonii (section Martagon) into another. However, the cp genome-based phylogenetic trees are incongruent with recent classification of the morphological features and geographic origin . This was also reported by Hayashi and Kawano  in their study of phylogenetic relationships based on two cp genes, rbcL and matK, among Lilium species and related genera. Gao et al.  also noted that the phylogenetic groupings were dissimilar among the Lilium species collected from Q-T plateau in China based on the nuclear ITS and cp matK sequence variations. The phylogenetic relationship inferred from retrotransposon based markers showed the L. lancifolium in Sinomartagon was not grouped with L. callosum and L. amabile in Sinomartagon section . The two Martagon lilies, L. hansonii and L. distichum were clustered in the same group in their report, but these two species were separated into different groups in our study. The high bootstrap values indicate the robustness in the current analysis. Thus, the discrepancies might be derived from the phylogenetic inferences from maternal inheritance of cp genomes and biparental inheritance of nuclear genomes.
The comparative genomic and phylogenomic analyses of the cp genomes in the genus Lilium and other related genera in the order Liliales revealed high conservation in length, AT ratios, gene contents and genome structures. There were 18 intron-containing genes. One intron loss was observed in species- relationship independent manner. We observed 16–21 SSR loci and high variations of presence/absence polymorphisms among the cp genomes among the species in the genus Lilium. Compared to the limited length and structure variations, there were significant numbers of sequence variations of SNPs, indels and SSR loci in the cp genomes of the genus Lilium. The two hyper-variable regions in the LSC may need to be compared with cp genomes of other distantly related genera for a better understanding of selection constraints along the cp genomes. Discrepancies in the positions of some species in the phylogenetic trees should be further analyzed. The presence/absence polymorphisms in SSR loci in the cp genomes may be expanded to more species to trace for the maternal lineages, as the SSRs stay in the current loci after de novo occurrence.
S1 Table. The data summary of genome sequence reads generated by Illumina sequencer platform and mapped sequence reads to cp genome de novo assembly.
S2 Table. The list of genes encoded in the cp genomes in L. amabile, L. callosum, L. lancifolium and L. philadelphicum.
S3 Table. Number of SNPs and indels in 80 genes in L. amabile, L. callosum, L. lancifolium and L. philadelphicum.
S4 Table. The SNPs and their locations in the cp genomes in nine Lilium species.
- 1. Peruzzi L (2016) A new infra-familial taxonomic setting for Liliaceae, with a key to genera and tribes. Plant Biosyst 150(6): 1341–1347
- 2. Van Tuyl JM, Arens P (2011) Lilium: Breeding history of the modern cultivar assortment. Acta Hort. 900:223–230
- 3. Benschop M, Kamenetsky R, Le Nard M, Okubo H (2010) The global flower bulb industry: Production, utilization, research. Hortic Rev 36: 1–115.
- 4. Endlicher SL (1836) Genera Plantarum. F. Beck, Vienna.
- 5. Comber HF (1949) A new classification of the genus Lilium. Lily Year Book RHS 13: 86–105
- 6. Asano Y (1986) A numerical taxonomic study of the genus Lilium in Japan. J Fac Agricult, Hokkaido Univ 62: 333–341
- 7. Noda S (1987) ‘Lily road’ toward the Japanese Islands–a cytological view point. In: The Lilies in Japan (ed. M Shimizu) pp. 98–110. Seibundo-shinkouya, Tokyo. Japan
- 8. Pelkonen VP, Pirttilä AM (2012) Taxonomy and phylogeny of the genus Lilium. Floricul Ornament Biotechnol 6: 1–8
- 9. Shinozaki K, Phme M, Tanaka M, Wakasugi T, Hayashida N, Matsubayashi T, et al. (1986) The complete nucleotide sequence of the tobacco chloroplast genome: its gene organization and expression. EMBO J 5(9): 2043–2049 pmid:16453699
- 10. Wicke S, Schneeweiss GM, dePamphilis CW, Müller KF, Quandt D (2011) The evolution of the plastid chromosome in land plants: gene content, gene order, gene function. Plant Mol Biol 76: 273–297 pmid:21424877
- 11. Stoebe B, Kowallik KV (1999) Gene-cluster in chloroplast genomics. Trends Genet 15: 344–3437 pmid:10461201
- 12. Stern DB, Goldschmidt-Clermont M, Hanson MR (2010) Chloroplast RNA metabolism. Annu Rev Plant Biol 61: 125–155 pmid:20192740
- 13. Olmstead RG, Palmer JD (1994) Chloroplast DNA systematics: a review of methods and data analysis. Am J Bot 81: 1205–1224
- 14. Shaw J, Lickey EB, Beck JT, Farmer SB, Liu W, Miller J, et al. (2005) The tortoise and the hare II: Relative utility of 21 noncoding chloroplast DNA sequences for phylogenetic analysis. Am J Bot 92(1) 142–166 pmid:21652394
- 15. Nikiforova SV, Cavalieri D, Velasco R, Goremykin V (2013) Phylogenetic analysis of 47 chloroplast genomes clarifies the contribution of wild species to the domesticated apple maternal lines. Mol Biol Evol 30 (8):1751–1760. pmid:23676769
- 16. Hayashi K, Kawano S (2000) Molecular systematics of Lilium allied genera (Liliaceae): Phylogenetic relationships among Lilium and related genera based on the rbcL and matK gene sequence data. Plant Species Biol 15: 73–93
- 17. Gao YD, Harris AJ, Zhou SD, He XJ (2013) Evolutionary events in Lilium (including Nomocharis, Liliaceae) are temporally correlated with organelles of the Q-T plateau and the Hengduan Mountains. Mol Phylogenet Evol 68: 443–460. pmid:23665039
- 18. Brozynska M, Furado A, Henry RJ (2014) Direct chloroplast sequencing: Comparison of sequencing platforms and analysis tools for whole chloroplast barcoding. PLOS One 9(10): e110387. pmid:25329378
- 19. Lei W, Ni D, Wang Y, Shao J, Wang X, Yang D, et al. (2016) Intraspecific and heteroplasmic variations, gene losses and inversions in the chloroplast genome of Astragalus membranaceus. Sci Rep 6:21669. pmid:26899134
- 20. Zhang Q, Bi Y, Zhang M, Chen X, Yang F, Xue J, et al. (2016) The complete chloroplast genome of Lilium taliense, an endangered species endemic to China. Conservation Genet Resour 9 (2): 201–203.
- 21. Song JH, Yoon CY, Do HDK, Lee WB, Kim JH (2016) The complete chloroplast genome sequence of Lilium tsingtauense Gilg (sect. Martagon, Liliaceae). Mitochon DNA Part II: 1(1) 318–320.
- 22. Kim KH, Hwang YJ, Lee SC, Yang TJ, Lim KB (2016) The complete chloroplast genome sequence of Lilium hansonii Leitlin ex D.D.T. Moore. Mitochon DNA Part A: 27(5) 3678–3679. pmid:26404645
- 23. Bi Y, Du Y, Chen X, Yang F, Xue J, Zhang X, et al. (2016) The complete chloroplast genome sequence of Lilium fargesii (Lilium, Liliaceae). Conservation Genet Resour 8: 419–422.
- 24. Du Y, Bi Y, Chen X, Yang F, Xue J, Zhang X (2016) The complete chloroplast genome of Lilium cernuum: genome structure and evolution. Conservation Genet Resour 8: 375–378.
- 25. Hwang YJ, Lee SC, Kim KH, Choi BS, Park JY, Yang TJ, et al. (2016) The complete chloroplast genome of Lilium distichum Nakai (Liliaceae). MitochonDNA Part A 27: 4633–4634.
- 26. Kim JS, Kim JH (2013) Comparative genome analysis and phylogenetic relationship of Order Liliales insight from the complete plastid genome sequence of two lilies (Lilium longiflorum and Alstroemeria aurea). PLOS One 8(6) e68180. pmid:23950788
- 27. Kim K, Lee SC, Lee J, Yu Y, Yang Y, Choi BS, et al. (2015) Complete chloroplast ad ribosomal sequences for 30 accessions elucidate evolution of Oryza AA genome species. Sci Rep 5:15655. pmid:26506948
- 28. Loshe M, Drechsel O, Bock R (2007) OrganellarGenomeDRAW (ORDRAW): a tool for the easy generation of high quality custom graphical maps of plastid and mitochondrial genomes. Curr Genet 52: 267–274. pmid:17957369
- 29. Wyman SK, Jansen RK, Boore JL (2004) Automatic annotation of organellar genome with DOGMA. Bioinformatics 20: 3252–3255. pmid:15180927
- 30. Lanslett D, Canback B (2004) ARAGORN, a program to detect tRNA genes and trnRNA genes in nucleotide sequences. Nucleic Acids Res 32: 11–16. pmid:14704338
- 31. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: Molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28(10): 2731–2739. pmid:21546353
- 32. Sworfford DL (2003) PAUP*. Phylogenetic Analysis Using Parsimony (* and other Methods). Version 4b10. Sunderland, Massachusetts, Sinauer, USA
- 33. Stamatakis A (2006) RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed model. Bioinformatics 22: 2688–2690. pmid:16928733
- 34. Jansen RK, Cai Z, Raubeson LA, Daniell H, dePamphillis CW, Leebens-Mack J, et al. (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 (49): 19369–19374. pmid:18048330
- 35. Powell W, Morgante M, McDevitt R, Vendramin GG, Rafalski JA (1995) Polymorphic simple sequence repeat genomes: applications to the population genetics in plants. Proc Natl Acad Sci USA 92(17): 7759–7763 pmid:7644491
- 36. Clark CM, Wentworth TR, O’Malley DM (2000) Genetic discontinuity revealed by chloroplast microsatellite in eastern North American Abies (Pinaceae). Am J Bot 87(6) 774–782 pmid:10860908
- 37. Huang J, Yang X, Zhang C, Yin X, Liu S, Li X (2015) Development of chloroplast microsatellite markers and analyses of chloroplast diversity in Chinese Jujube (Ziziphus jujube Mill.) and wild Jujube (Ziziphus acidojujuba Mill.). PLOS One 10(9) e0134519. pmid:26406601
- 38. Wheeler GL, Dorman HE, Buchanan A, Challagundla L, Wallace LE (2014) A review of the prevalence, utility, and caveats of using chloroplast simple sequence repeats for studies of plant biology. App Plant Sci 2(12) 1400059.
- 39. Park Y-J, Lee JK, Kim N-S (2009) Simple sequence repeat polymorphisms (SSRPs) for evaluation of molecular diversity and germplasm classification of minor crops. Molecules 13: 4546–4569.
- 40. Palmer JD, Nugent JM, Herbon LA (1987a) Unusual structure of geranium chloroplast DNA: a triple-sozed inverted repeat, extensive gene duplications, multiple inversions, and two repeat familes. Proc Natl Acad Sci USA 84: 769–773 pmid:16593810
- 41. Greiner S, Wang X, Rauwolf U, Silber MV, Mayer K, Meuer J, et al. (2008) The complete nucleotide sequences of the five genetically distinct plastid genomes of Oenothera, subsection Oenothera: I. Sequence evolution and plastome evolution: Nucl Acids Res 36: 2366–2378. pmid:18299283
- 42. Haberle RC, Fourcade HM, Boore JL, Jansen RK (2008) Extensive rearrangement in the chloroplast genome of Trachelum aceruleum are associated with repeat and tRNA genes. J Mol Evol 66: 350–361, pmid:18330485
- 43. Palmer JD, Osorio B, Aldrich J, Thompson WF (1987b) Chloroplast DNA evolution among legumes: loss of a large inverted repeat occurred prior to other sequence rearrangements. Curr Genet 11: 275–286.
- 44. Song Y, Dong W, Liu B, Xu C, Yao X, Gao J, et al. (2015) Comparative analysis of complete chloroplast genome sequences of two tropical trees Machilus yunnanensis and Machilus balansae in the family Lauraceae. Front Plant Sci 6: 662. pmid:26379689
- 45. Shaw J, Lickey EB, Schilling EE, Small RL (2007) Comparison of whole chloroplast genome sequences to choose non-coding regions for phylogenetic studies in Angiosperms: The tortoise and the hare III. Am J Bot 94(3): 275–288 pmid:21636401
- 46. Shaw J, Shafer HL, Leonard OR, Kovach MJ, Schorr M, Morris AB (2014) Chloroplast DNA sequence utility for the lowest phylogenetic and phylogeographic inferences in Angiosperms: The tortoise and the hare IV. Am J Bot 101(11): 1987–2004. pmid:25366863
- 47. Lee SI, Kim JH, Park KC, Kim NS (2015) LTR-retrotransposons and inter-retrotransposon amplified polymorphism (IRAP) analysis in Lilium species. Genetica 143: 343–352. pmid:25787319