https://orcid.org/0000-0003-4699-9799
ABSTRACT
Genome characterization of California poppy (Eschscholzia californica cv. “Hitoezaki”), which produces pharmaceutically important benzylisoquinoline alkaloids (BIAs), was carried out using the draft genome sequence. The numbers of tRNA and rRNA genes were close to those of the other plant species tested, whereas the frequency of repetitive sequences was distinct from those species. Comparison of the predicted genes with those of Amborella trichopoda, Nelumbo nucifera, Solanum lycopersicum, and Arabidopsis thaliana, and analyses of gene ontology and Kyoto Encyclopedia of Genes and Genomes pathway indicated that the enzyme genes involved in BIA biosynthesis were highly enriched in the California poppy genome. Further comparative analysis using the genome information of Papaver somniferum and Aquilegia coerulea, both BIA-producing plants, revealed that many genes encoding BIA biosynthetic enzymes, transcription factors, transporters, and candidate proteins, possibly related to BIA biosynthesis, were specifically distributed in these plant species.
Genome characterization of California poppy provides valuable information for identifying specific genes involved in plant-specialized metabolism.
Abbreviations
- BIA:
benzylisoquinoline alkaloid
- bp:
base pairs
- GO:
Gene Ontology
- KEGG:
Kyoto Encyclopedia of Genes and Genomes
- KOG:
euKaryotic clusters of Orthologous Group
- TE:
transposable element
California poppy (Eschscholzia californica), a member of the Papaveraceae family, is native to the United States of America and Mexico and mainly found on the west coast of North America. It is a basal eudicot with a beautiful ornamental orange flower and has been used as herbal medicine owing to the pharmacological effects of specialized compounds it contains. The California poppy is not only a model plant to study floral organ development in basal eudicot (Becker, Gleissberg and Smyth 2005; Carlson et al.2006), but also a model to study the biosynthesis of benzylisoquinoline alkaloids (BIAs), which are pharmaceutically important and chemically divergent specialized products found mainly in the Papaveraceae, Ranunculaceae, Berberidaceae, and Menispermaceae families (Kutchan 1995; Hagel and Facchini 2013; Sato 2013, 2020). Of the approximately 2500 compounds, morphine (analgesic), codeine (antitussive), magnoflorine (anti-HIV agent), and berberine (antimicrobial) are the representative BIAs. One of the major BIAs produced by the California poppy is sanguinarine, which shows antibiotic activity and has been used as a component of toothpaste. The biosynthetic pathway of sanguinarine and the transcriptional regulation mechanism by several transcription factors have been well elucidated at the molecular level (Dittrich and Kutchan 1991; Ikezawa, Iwasa and Sato 2007, 2009; Liscombe and Facchini 2007; Hagel et al.2012; Beaudoin and Facchini 2013; Takemura et al.2013; Yamada et al.2015; Hori et al.2018). However, many other specialized metabolite pathways in the California poppy, including some BIAs, such as macarpine and escholtzine biosynthesis pathways (Figure S1), and flower pigmentations have been scarcely elucidated, although there have been reports on the expressed sequence tag (EST) libraries and transcriptome data of E. californica (Carlson et al.2006; Xiao et al.2013).
To identify and characterize candidate genes involved in specialized metabolism directly from the genome sequence, we obtained the draft genome sequence (approximately 489 Mb) of the California poppy (ECA_r1.0), which covered >95% of the whole genome (503.8 Mb). Although a biosynthetic gene cluster similar to noscapine biosynthesis in Papaver somniferum (opium poppy) and steroidal glycoalkaloid biosynthesis in Solanum lycopersicum (tomato) and Solanum tuberosum (potato) (Itkin et al.2013; Winzer et al.2012) were not found in the California poppy genome scaffolds, we found several cytochrome P450 gene clusters and their novel function in the macarpine biosynthesis pathway (Hori et al.2018). Generally, the evolution of specialized metabolism is considered to be related to gene duplication and diversification of key enzyme genes such as cytochrome P450. Our draft genome sequence data indicated that certain P450 family genes, for example CYP80, CYP82, and CYP719, which encode important enzymes involved in the formation of the specific chemical structure of various BIAs, are enriched in the California poppy genome compared to Arabidopsis thaliana (Hori et al.2018).
On the other hand, further investigation of enzyme genes involved in the morphine biosynthesis pathway in P. somniferum, the same member of the Papaveraceae plant that produces morphine, showed no homologous genes for the production of morphine in the genome of California poppy (Hori et al.2018), which cannot produce morphinan alkaloids. The California poppy genome is valuable for studying the evolution of specialized metabolite biosynthesis associated with gene diversification as well as that of basal eudicot owing to its relatively small genome size (Soltis et al.2009).
In this report, we analyzed the structure and characteristics of the draft genome of the California poppy in comparison with the genomes of other plant species including the following: Nelumbo nucifera (sacred lotus) and tomato, which produce BIAs and steroidal glycoalkaloids, respectively (Friedman 2002; Nakamura et al.2013); well-researched model plant, A. thaliana; and the most primitive Angiosperm, Amborella trichopoda (Amborella). The comparisons were made using gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database. To focus on the exploration of BIA biosynthesis-related genes, we compared the California poppy genes, using orthology analysis, with those of other BIA-producing plants, specifically Aquilegia coerulea, a member of the Ranunculaceae plant family, and opium poppy, whose both genome sequences have been recently revealed (Filiault et al.2018; Guo et al.2018). Finally, we discussed the usefulness of the California poppy genome to discover candidate genes involved in BIA biosynthesis.
Materials and methods
Draft genome sequencing and gene prediction of the California poppy
E. californica plant and genomic DNA were prepared and sequenced as described previously (Hori et al.2018). The raw Illumina sequence reads were assembled using PLATANUS v1.2.11 (Kajitani et al.2014) and the training sets were constructed using BRAKER v1.3 and RNA-Seq data. The protein-coding genes were predicted by Augustus v3.0.3 using the training set. Transfer RNA genes were searched using tRNAscan-SE v1.3.1 and pseudogenes were filtered out (Lowe and Chan 2016). Ribosomal RNA genes were searched using BLAST with an E-value cutoff of 1E-10 using 5.8S rRNA and 25S rRNA (accession number: X52320.1) and 18S rRNA (accession number: X16077.1) as queries.
Detection of repetitive sequences
Repetitive sequences were detected using the following steps: first, repetitive sequences were detected using RepeatScout v1.05 (Price, Jones and Pevzner 2005), and subsequent repetitive sequences were searched against Repbase (Bao, Kojima and Kohany 2015) using RepeatMasker v4.0.6 (http://www.repeatmasker.org). The repetitive sequences found using RepeatMasker were classified into known repeats, while those found only using RepeatScout were classified into unique repeats in the California poppy.
Functional classification of genes
Functional domains located in the predicted amino acid sequences were searched against the InterPro database using InterProScan v4 (https://github.com/ebi-pf-team/interproscan). The predicted gene sequences were classified into the euKaryotic clusters of Orthologous Group (KOG) categories and the plant GO slim categories (Ashburner et al.2000; Tatusov et al.2003). Comparison of orthologous gene clusters of California poppy with those of A. coerulea, P. somniferum, and A. thaliana was performed using OrthoFinder v2.3.1 (Emms and Kelly 2015).
Comparison of metabolic pathways
Gene sequences obtained from genome databases (A. trichopoda: AmTr v1.0; N. nucifera: lotus marker; tomato: ITAG2.4; A. thaliana: Araport 11), as well as those of the predicted California poppy genes, were compared and mapped onto the KEGG reference pathways using BLAST searches against the GENES database in the KEGG. The criteria used were an E-value cutoff of 1E-10, length coverage ≥25%, and identity ≥50%, and the status of mapping was compared among the 5 plant species. Information on the programs and databases is listed in Table S1.
Results and discussion
Prediction of RNA-encoding genes
The draft genome sequence of the California poppy cv. “Hitoezaki” was determined using the Illumina HiSeq 2500 systems and assembled to 489 Mb, which is equivalent to 97% of the estimated genome size (503.8 Mb). The 41 612 protein-encoding genes predicted using Augustus showed significant sequence similarity to the registered genes, as reported in a previous manuscript (http://eschscholzia.kazusa.or.jp) (Hori et al.2018). Based on the prediction of protein-encoding genes, the genome comparison was carried out as described below.
When RNA-coding genes were searched (67 genes for rRNAs, made up of 3 genes for 5.8S RNA, 24 genes for 18S RNA, and 40 genes for 25S RNA), 1343 intact genes for tRNAs were identified in the assembled genomic sequences ECA_r1.0 (Table S2). Comparison of the number of genes for tRNAs in the genomes of other plant species indicated that the California poppy genome contains relatively more genes than other plant species such as Amborella (651), sacred lotus (960), A. thaliana (629), tomato (887), Dianthus caryophyllus L. (carnation) (1050), and Macleaya cordata (815). The opium poppy genome contains more tRNA genes (5467) than the California poppy genome possibly owing to its larger genome size (Table S3).
Repetitive sequences
When the repetitive sequences were searched, the assembled California poppy genomic sequences comprised 52 683 626 bp known repeats including 64 193 bp Class I short interspersed nuclear elements (SINEs), 2 426 826 bp long interspersed nuclear elements (LINEs), and 32 671 650 bp long terminal repeat (LTR) elements (containing 24 796 251 bp Copia and 5 293 043 bp Gypsy types). The sequences also comprised 3 480 271 bp Class II DNA elements, 7 493 549 bp simple repeats, 432 877 bp Helitrons, and 2 450 250 bp low-complexity sequences. In the known repeat sequences, LTR elements were observed most frequently. However, 183 872 768 bp novel repeats were defined by de novo repeat finding, the sum of which corresponds to 48.4% of the assembled genomic sequences (Table 1).
Repetitive sequences found in the California poppy draft genome
| Length occupied (bp) | Percent of whole genome (%) | |||||
|---|---|---|---|---|---|---|
| Known repeats | Interspersed repeats | Class I | SINE | Total | 64 193 | 0.01 |
| LINE | Total | 2 426 826 | 0.50 | |||
| LTR elements | Copia | 24 796 251 | 5.1 | |||
| Gypsy | 5 293 043 | 1.1 | ||||
| Total | 32 671 650 | 6.7 | ||||
| Class II | DNA elements | 3 480 271 | 0.71 | |||
| Unclassified | 69 | 0 | ||||
| Helitrons | 432 877 | 0.09 | ||||
| Low complexity | 2 450 250 | 0.50 | ||||
| Simple repeat | 7 493 549 | 1.5 | ||||
| Unknown | 26 147 | 0.01 | ||||
| Subtotal | 52 683 626 | 10.8 | ||||
| Unique repeats | Simple repeat | 92 522 | 0.02 | |||
| Unknown | 183 780 246 | 37.6 | ||||
| Subtotal | 183 872 768 | 37.6 | ||||
| Total | 236 5 56 394 | 48.4 |
| Length occupied (bp) | Percent of whole genome (%) | |||||
|---|---|---|---|---|---|---|
| Known repeats | Interspersed repeats | Class I | SINE | Total | 64 193 | 0.01 |
| LINE | Total | 2 426 826 | 0.50 | |||
| LTR elements | Copia | 24 796 251 | 5.1 | |||
| Gypsy | 5 293 043 | 1.1 | ||||
| Total | 32 671 650 | 6.7 | ||||
| Class II | DNA elements | 3 480 271 | 0.71 | |||
| Unclassified | 69 | 0 | ||||
| Helitrons | 432 877 | 0.09 | ||||
| Low complexity | 2 450 250 | 0.50 | ||||
| Simple repeat | 7 493 549 | 1.5 | ||||
| Unknown | 26 147 | 0.01 | ||||
| Subtotal | 52 683 626 | 10.8 | ||||
| Unique repeats | Simple repeat | 92 522 | 0.02 | |||
| Unknown | 183 780 246 | 37.6 | ||||
| Subtotal | 183 872 768 | 37.6 | ||||
| Total | 236 5 56 394 | 48.4 |
Repetitive sequences found in the California poppy draft genome
| Length occupied (bp) | Percent of whole genome (%) | |||||
|---|---|---|---|---|---|---|
| Known repeats | Interspersed repeats | Class I | SINE | Total | 64 193 | 0.01 |
| LINE | Total | 2 426 826 | 0.50 | |||
| LTR elements | Copia | 24 796 251 | 5.1 | |||
| Gypsy | 5 293 043 | 1.1 | ||||
| Total | 32 671 650 | 6.7 | ||||
| Class II | DNA elements | 3 480 271 | 0.71 | |||
| Unclassified | 69 | 0 | ||||
| Helitrons | 432 877 | 0.09 | ||||
| Low complexity | 2 450 250 | 0.50 | ||||
| Simple repeat | 7 493 549 | 1.5 | ||||
| Unknown | 26 147 | 0.01 | ||||
| Subtotal | 52 683 626 | 10.8 | ||||
| Unique repeats | Simple repeat | 92 522 | 0.02 | |||
| Unknown | 183 780 246 | 37.6 | ||||
| Subtotal | 183 872 768 | 37.6 | ||||
| Total | 236 5 56 394 | 48.4 |
| Length occupied (bp) | Percent of whole genome (%) | |||||
|---|---|---|---|---|---|---|
| Known repeats | Interspersed repeats | Class I | SINE | Total | 64 193 | 0.01 |
| LINE | Total | 2 426 826 | 0.50 | |||
| LTR elements | Copia | 24 796 251 | 5.1 | |||
| Gypsy | 5 293 043 | 1.1 | ||||
| Total | 32 671 650 | 6.7 | ||||
| Class II | DNA elements | 3 480 271 | 0.71 | |||
| Unclassified | 69 | 0 | ||||
| Helitrons | 432 877 | 0.09 | ||||
| Low complexity | 2 450 250 | 0.50 | ||||
| Simple repeat | 7 493 549 | 1.5 | ||||
| Unknown | 26 147 | 0.01 | ||||
| Subtotal | 52 683 626 | 10.8 | ||||
| Unique repeats | Simple repeat | 92 522 | 0.02 | |||
| Unknown | 183 780 246 | 37.6 | ||||
| Subtotal | 183 872 768 | 37.6 | ||||
| Total | 236 5 56 394 | 48.4 |
Comparison of the transposable element (TE) contents of the California poppy genome with those of other plant species, such as Amborella, sacred lotus, A. thaliana, tomato, carnation, opium poppy, and M. cordata (Table S4), showed that the relative content of known TEs in the California poppy genome was lower than that of unclassified repeats. In particular, the percentage of LTR elements in the whole California poppy genome was 6.7%, which is lower than that in Papaveraceae plants, opium poppy (45.6%), and M. cordata (26.6%) genomes. Simple repeats appeared to be relatively high, similar to that of carnations. The lower content of known TEs in the assembled genomic sequences in the California poppy may be attributed to the escape of TE sequences during the process of sequence assembly, since the California poppy assemblies in this study were highly fragmented at the positions of potential repetitive sequences, as mentioned in the carnation genome analysis (Yagi et al.2014).
Comparison of California poppy genes with those of other plant species
To explore the genomic characteristics of the California poppy, we compared the predicted genes of California poppy, including not only metabolic pathways but also other plant processes, to those of other angiosperm plant species, namely Amborella, sacred lotus, tomato, and A. thaliana. The complete gene sets of these groups (26 846, 26 685, 34 725, and 27 655 protein-encoding genes, respectively), were retrieved from the AmTr v1.0, lotus marker, ITAG2.4, and Araport11 databases, respectively (Table S1). The translated amino acid sequences of 41 612 coding sequences (CDSs), including splicing variants in the California poppy, were searched against the complete gene sets of Amborella, sacred lotus, tomato, and A. thaliana using BLAST with an E-value cutoff of 1E-10. To create a Venn diagram of homologous gene clusters, we used the CD-HIT program (Figure 1). Only 4 genes of N. nucifera and 17 genes of A. thaliana were excluded through the CD-HIT program; however, a total of 97 522 clusters were generated (Table S5). As shown in Figure 1, 19 819 clusters (containing 29 682 California poppy genes) were unique to E. californica, whereas 2272 gene clusters were shared with the other 4 plant species. The highest number (1459 clusters) of shared sequence clusters with E. californica was found in N. nucifera, which is the closest species and BIA-producing plant. The number of shared gene clusters with tomato (490), Amborella (350), and A. thaliana (257) is similar, which is relatively consistent with their phylogenetic relationship.
Classification of gene families in 5 plant species. Numbers in the Venn diagram show unique and shared gene families in each species.
To further compare the distribution of genes between the California poppy and the other 4 plant species, we classified the protein-encoding genes of all 5 species into KOG categories (Table S6). The number of genes classified into KOG (excluding genes in the “Poorly characterized” category) in California poppy, Amborella, sacred lotus, tomato and A. thaliana was 16 596 (42.1%), 9606 (47.6%), 12 497 (48.4%), 13 955 (46.9%), and 11 5 47 (41.7%), respectively. Compared to the other 4 species, the California poppy genes categorized in KOG D (cell-cycle control, cell division, and chromosome partitioning) and R (general function prediction only) were relatively high (Figure 2 and Figure S2). The ratio of the genes in the other KOG categories, including KOG Q (secondary metabolites biosynthesis, transport, and catabolism), was similar among the 5 plant species.
Gene assignment to KOG functional categories in California poppy.
To examine the classification of the genes based on GO slim, we assigned genes of the 5 plant species to GO terms. In the root category, the ratio of California poppy genes in the biological process (BP), cellular component (CC), and molecular function (MF) categories were 27.0%, 12.1%, and 43.2%, respectively, which were relatively close to that of Amborella genes. The distribution of gene ratio was quite similar among the 5 plant species (Figure 3a). The detailed results of GO classification showed that the genes of California poppy categorized in the BP, CC, and MF categories were also close to those of other plant species (Figure 3b–d and Table S7).
Classification of GO categories. The % of genes classified into root (a), biological process (b), cellular component (c), and molecular function (d) categories, respectively, is shown.
To investigate the classification of genes based on the enzymatic function of their proteins, we mapped the complete gene sets of the California poppy, Amborella, sacred lotus, tomato, and A. thaliana onto KEGG reference pathways. Results indicated that 1, 708, 838, 449, 614, and 633 genes of the California poppy were assigned into 5 major KEGG orthology (KO) categories, namely metabolism, genetic information processing, environmental information processing, cellular processes, and organismal systems (Figure S3a). The number of genes of the other 4 plants in each major category was close to that of the California poppy genes. In the metabolism category, the ratio of genes belonging to the biosynthesis of other secondary metabolites in the California poppy and sacred lotus was relatively high, whereas those belonging to energy metabolism in California poppy, Amborella, and sacred lotus was clearly low compared to the other plants (Figure S3b and Table S8). A more detailed analysis revealed that the percentage of California poppy genes mapped onto “Isoquinoline alkaloid biosynthesis” was apparently higher than that of the other 4 plant species (Figure 4). In fact, many genes mapped only in the California poppy were found in the pathway maps downstream of (S)-scoulerine, which is an important intermediate of BIAs produced in the California poppy (Figure S4). Sacred lotus also produces BIAs, but has only a small number of genes mapped onto “Isoquinoline alkaloid biosynthesis” compared to the California poppy, indicating that the diversification of BIAs found only in the California poppy related to the acquisition of such specific genes during evolution.
Classification of genes using KEGG reference pathways. The percentage of classified genes in each pathway in the “Biosynthesis of other secondary metabolites” category is shown.
Orthology analysis of gene families in BIA-producing plants
Recently, genome sequences of opium poppy and A. coerulea, which are BIA-producing plants, have been uncovered (Filiault et al.2018; Guo et al.2018). To further examine the orthologous gene families, focusing on the genes involved in the biosynthesis of BIA among BIA-producing plant species, we performed orthologous gene family analysis with the OrthoFinder program using complete gene sets of California poppy (41 612 protein-encoding genes), opium poppy (51 212), and A. coerulea (30 023), compared with A. thaliana (27 655) as a representative non-BIA-producing plant (Figure 5). The results showed that 16 901, 22 280, 9738, and 8815 gene cluster groups were unique to each species, respectively. These included 17 476, 22 871, 10 088, and 9394 genes, respectively, and 9980 gene cluster groups that were shared by all 4 plant species. The number of orthologous groups shared between the California poppy and either the opium poppy (743) or A. coerulea (349) was higher than that between the California poppy and A. thaliana (101). The orthologous groups shared by California poppy, opium poppy, and A. coerulea were the highest (824) compared to that of the other combinations including A. thaliana. A large number of shared genes suggests that not only 3 BIA-producing plants are closely related species, but that many genes in these groups might be diversified and involved in the biosynthesis of specific metabolites.
Classification of gene families in California poppy, opium poppy, A. coerulea, and A. thaliana. Numbers in the Venn diagram show unique and shared gene cluster groups, and those in parentheses show the number of genes in each species.
We extracted the orthologous groups containing genes annotated to BIA biosynthetic enzymes (see Figure S1) and listed these in Table S9. Many groups containing predicted biosynthetic enzyme genes, such as tyrosine/DOPA decarboxylase, (S)-norcoclaurine synthase, (S)-norcoclaurine 6-O-methyltransferase, (S)-N-methylcoclaurine 3′-hydroxylase, (S)-3′-hydroxy-N-methylcoclaurine 4′-O-methyltransferase, berberine bridge enzyme, (S)-stylopine synthase, (S)-canadine synthase, (S)-tetrahydroprotoberberine N-methyltransferase, and protopine 6-monooxygenase, were shared by the California poppy, opium poppy, and A. coerulea. However, the orthologous groups of genes encoding (S)-cheilanthifoline synthase, (S)-reticuline-7-O-methyltransferase, dihydrosanguinarine 10-hydroxylase, and dihydrobenzophenanthridine alkaloid 10-hydroxylase, that of the gene encoding pavine N-methyltransferase, and that of the gene encoding (S)-scoulerine 9-O-methyltransferase, were shared by the California poppy–opium poppy, California poppy–A. coerulea, and opium poppy–A. coerulea, respectively. The orthologous groups found in 2 plant species contain genes encoding specific proteins involved in the branch point or downstream in the biosynthetic pathway, which suggests that these enzymes might be implicated in the diversification of specific metabolites found in each plant. A recent phylogenomic approach has also suggested that the gene-encoding enzymes involved in a common pathway of BIA biosynthesis might be present in the common ancestor (Li et al.2020). Furthermore, the orthologous group containing a unique type of basic helix–loop–helix transcription factor gene to BIA biosynthesis in Coptis japonica, CjbHLH1, was also found in 3 BIA-producing plants (Yamada et al.2011). This result indicates the specific distribution of CjbHLH1 homologs in BIA-producing plants, as previously reported (Yamada, Koyama and Sato 2011).
To explore candidate genes related to BIA biosynthesis, we selected representative orthologous groups with functional annotation shared by 3 BIA-producing plant species. As shown in Table S10, the genes encoding not only BIA biosynthesis-related proteins (highlighted in yellow) but also uncharacterized enzymes, including dehydrogenases, cytochrome P450s, and methyltransferases, were found in the orthologous groups. Furthermore, the orthologous groups containing predicted transcription factor-encoding genes, for example bHLHs, MYBs, and APETALA2/ethylene responsive factors (AP2/ERFs), and predicted transporter-encoding genes, such as ABC transporters, multidrug and toxic compound extrusion (MATE) transporters, and purine permeases, were also found. It has been reported that AP2/ERF transcription factors belonging to the group IX subfamily regulate the biosynthesis of other alkaloids such as nicotine, indole alkaloids, and glycoalkaloids (van der Fits and Memelink 2000; Shoji, Kajikawa and Hashimoto 2010; Cárdenas et al.2016). The orthologous group, OG0011259, contains predicted group IX AP2/ERF genes, including 1 California poppy gene. In C. japonica, B-type ABC transporters, CjABCB1 and CjABCB2, and a MATE transporter, CjMATE1, are involved in the translocation and compartmentation of berberine (Shitan et al.2003; Shitan et al.2013; Takanashi et al.2017). The orthologous groups, OG0008001 and OG0012836, contain B-type ABC transporter genes that are homologous to CjABCB1, and the orthologous groups, OG0001473 and OG0007280, contain MATE-type transporter genes that are homologous to CjMATE1. Recently, novel BIA uptake purine permeases (BUPs) have been isolated from the opium poppy and their functions have been characterized (Dastmalchi et al.2019). Interestingly, BUP1 gene belongs to the orthologous group, OG0000544, which contains 1 California poppy gene. The other group, OG0000703, which contains 10 California poppy genes, also includes predicted purine permease genes. These orthologous genes might be involved in the biosynthesis and diversification of specific metabolites. The functional characterization of these uncharacterized enzymes and transcription factor genes is ongoing (Yamada et al.2020).
Conclusions
E. californica (California poppy), a member of the Papaveraceae plant species, is a useful model plant to study both floral organ development and evolution in basal eudicot, and the biosynthesis of BIAs including pharmaceutically important chemicals. The elucidation of draft genome sequences of the California poppy facilitates these studies. To characterize the features of the California poppy genome, we performed comparative analyses using several plant species. Our repetitive sequence analysis indicated that LTR elements were observed less frequently in the California poppy than in other plants. When we performed a functional classification of predicted protein-encoding genes with KOG and GO categories, there was no apparent difference among the California poppy, Amborella, sacred lotus, tomato, and A. thaliana genes. However, further classification of predicted genes using the KEGG reference pathway revealed that the ratio of California poppy genes categorized in “Biosynthesis of other secondary metabolites,” particularly “Isoquinoline alkaloid biosynthesis” was higher than that of other plant genes. Orthology analysis using recently revealed genome sequences of BIA-producing plants, P. somniferum and A. coerulea, showed that a large number of orthologous groups shared by 3 BIA-producing plants were found. Some of these contain not only BIA biosynthetic enzyme genes and a unique-type of bHLH transcription factor gene, but also various gene encoding enzymes, e.g. methyltransferases and P450 proteins, transcription factors (bHLH, MYB, and AP2/ERF proteins), and transporters (B-type ABC proteins), MATEs, and purine permeases; therefore, these genes might be involved in BIA biosynthesis. This study also suggests that the specific distribution of metabolism in limited plant species might be attributed to gene diversification during evolution. Our findings obtained from comparative genome analyses provide a useful resource for identifying and characterizing specific genes involved in specific metabolism in future studies.
Data availability
The genome and gene sequences of the California poppy are available from the Eschscholzia Genome DataBase (http://eschscholzia.kazusa.or.jp). In the database, BLAST searches against the genome and gene sequences and keyword searches against the annotation are also available.
Author contribution
Y.Y. and F.S. conceived and designed the research. H.H., Y.M, and A.T. conducted draft genome sequencing of the California poppy. H.H. performed comparative analysis using programs and acquired original datasets. Y.Y., H.H., and F.S. analyzed the data, and Y.Y. and F.S. wrote the manuscript. N.S. and F.S. discussed the results and contributed to the improvement of the manuscript. All authors have reviewed the manuscript.
Funding
This research was supported by the Ministry of Education, Culture, Sports, Science and Technology of Japan (MEXT) (Grant-in-Aid for Scientific Research (S) 26221201 to F.S.). The draft genome sequencing of the California poppy was supported by MEXT KAKENHI (No. 221S0002).
Disclosure statement
No potential conflict of interest was reported by the authors.
