Analysis of the transcriptome of Panax notoginseng root uncovers putative triterpene saponin-biosynthetic genes and genetic markers

Abstract

Background: Panax notoginseng (Burk) F.H. Chen is important medicinal plant of the Araliacease family. Triterpene saponins are the bioactive constituents in P. notoginseng. However, available genomic information regarding this plant is limited. Moreover, details of triterpene saponin biosynthesis in the Panax species are largely unknown.

Results: Using the 454 pyrosequencing technology, a one-quarter GS FLX titanium run resulted in 188,185 reads with an average length of 410 bases for P. notoginseng root. These reads were processed and assembled by 454 GS De Novo Assembler software into 30,852 unique sequences. A total of 70.2% of unique sequences were annotated by Basic Local Alignment Search Tool (BLAST) similarity searches against public sequence databases. The Kyoto Encyclopedia of Genes and Genomes (KEGG) assignment discovered 41 unique sequences representing 11 genes involved in triterpene saponin backbone biosynthesis in the 454-EST dataset. In particular, the transcript encoding dammarenediol synthase (DS), which is the first committed enzyme in the biosynthetic pathway of major triterpene saponins, is highly expressed in the root of four-year-old P. notoginseng. It is worth emphasizing that the candidate cytochrome P450 (Pn02132 and Pn00158) and UDP-glycosyltransferase (Pn00082) gene most likely to be involved in hydroxylation or glycosylation of aglycones for triterpene saponin biosynthesis were discovered from 174 cytochrome P450s and 242 glycosyltransferases by phylogenetic analysis, respectively. Putative transcription factors were detected in 906 unique sequences, including Myb, homeobox, WRKY, basic helix-loop-helix (bHLH), and other family proteins. Additionally, a total of 2,772 simple sequence repeat (SSR) were identified from 2,361 unique sequences, of which, di-nucleotide motifs were the most abundant motif.

Conclusion: This study is the first to present a large-scale EST dataset for P. notoginseng root acquired by next-generation sequencing (NGS) technology. The candidate genes involved in triterpene saponin biosynthesis, including the putative CYP450s and UGTs, were obtained in this study. Additionally, the identification of SSRs provided plenty of genetic makers for molecular breeding and genetics applications in this species. These data will provide information on gene discovery, transcriptional regulation and marker-assisted selection for P. notoginseng. The dataset establishes an important foundation for the study with the purpose of ensuring adequate drug resources for this species.

Figure 1

Putative pathway for triterpene saponin biosynthesis in P. notoginseng. Enzymes found in this study are boxed. Abbreviations: AACT, acetyl-CoA acetyltransferase; β-AS, β-amyrin synthase; DMAPP, dimethylallyl diphosphate;DS, dammarenediol-II synthase; FPP, farnesyl diphosphate; FPPS, farnesyl diphosphate synthase; Glc, glucose; GPP, geranyl pyrophosphate; GGPP, geranylgeranyl diphosphate; GGPPS, geranylgeranyl pyrophosphate synthase; GT, glycosyltransferase; HMG-CoA, 3-hydroxy-3-methylglutaryl coenzyme A; HMGR, HMG-CoA reductase; HMGS, HMG-CoA synthase; IPP, isopentenyl diphosphate; IPPI, IPP isomerase; MVD, mevalonate diphosphate decarboxylase; MVK, mevalonate kinase; P450, cytochrome P450; PMK, phosphomevalonate kinase; SE, squalene epoxidase; SS, squalene synthase.

Figure 2

Length distributions of reads and contigs of P. notoginseng. A. Size distribution of 454 sequencing reads of P. notoginseng after removal of adaptor sequences. B. Length distribution of contigs (assembled sequences) in P. notoginseng 454-EST dataset.

Figure 3

GO analysis of P. notoginseng unique sequences based on cellular component (Figure 3A), molecular function (Figure 3B), and biological process (Figure 3C).

Figure 4

Alignment of DS amino acid residues from P. ginseng, P. quinquefolium, and P. notoginseng. P. ginseng DS and P. quinquefolius DS showing 99.5% identity, P. quinquefolius DS and P. notoginseng DS showing 99.0% identity, P. ginseng DS and P. notoginseng DS showing 98.7% identity.

Figure 5

Phylogenetic analysis of full-length CYP450s from P. notoginseng and characterized CYP450s from other plants. Amino acid sequences were aligned using the CLUSTALW program, and evolutionary distances were computed using MEGA4 with the Poisson correction method. Bootstrap values obtained after 1000 replications are indicated on the branches. Values less than 50% are not shown. The GenBank/EMBL/DDBJ accession numbers of the sequences are: U48434 (Solanum chacoense), O81971 (Glycine max), AJ238612 (Catharanthus roseus), AF124816 (Mentha × piperita), Z17369 (Helianthus tuberosus), Q42797 (Glycine max), D14588 (Petunia × hybrida), AB035868 (Arabidopsis thaliana), U54770 (Solanum lycopersicum), AB433175 (Medicago truncatula), AB433176 (Medicago truncatula), AB433177 (Lotus japonicus), AB433178 (Lotus japonicus), AB433179 (Glycyrrhiza uralensis), AF537321 (Pisum sativum), AF537322 (Pisum sativum), AF318500 (Arabidopsis thaliana), AF318501 (Arabidopsis thaliana), AF326277 (Hordeum vulgare), AF135485 (Glycine max), AB231332 (Glycine max), AB023636 (Glycyrrhiza echinata), AB437320 (Glycyrrhiza uralensis), D83968 (Glycine max), X71657 (Solanum melongena), X71658 (Solanum melongena), X71656 (Solanum melongena), X71655 (Solanum melongena), L23209 (Zea mays), NM_001112599 (Zea mays), U29333 (Pisum sativum), U69134 (Arabidopsis thaliana), U18929 (Arabidopsis thaliana), AF150881 (Solanum lycopersicum × Solanum peruvianum), AF214008 (Brassica napus), U38416 (Arabidopsis thaliana), AF029856 (Sorghum bicolor), AF029858 (Sorghum bicolor), AJ583531 (Triticum aestivum).

Figure 6

Phylogenetic analysis between full-length UGTs of P. notoginseng and characterized UGTs from other plants. Amino acid sequences were aligned using the program CLUSTALW and evolutionary distances were computed using the Poisson correction method with MEGA4. Bootstrap values obtained after 1000 replications are indicated on the branches. Values less than 50% are not shown. The GenBank/EMBL/DDBJ accession numbers of the sequences are AY747627 (Medicago truncatula), AY747626 (Medicago truncatula), DQ915168 (Vaccaria hispanica), AB013598 (Glandularia × hybrida), AB027455 (Petunia × hybrida), AB013597 (Perilla frutescens var. crispa), AB013596 (Perilla frutescens var. crispa), AY262037 (Crocus sativus), AY345982 (Stevia rebaudiana), Q41819 (Zea mays), DQ158907 (Arabidopsis thaliana), Q9MB73 (Citrus unshiu), AB166766 (Eucalyptus perriniana), AB166765 (Eucalyptus perriniana), AJ889012 (Solanum lycopersicum), AB072919 (Nicotiana tabacum), DQ832169 (Vitis labrusca), AF190634 (Nicotiana tabacum), AF304430 (Brassica napus), AB078723 (Aspergillus oryzae), Z83833 (Arabidopsis thaliana), AB071961 (Panax ginseng), AB031274 (Scutellaria baicalensis), Z25802 (Petunia × hybrida), L34847 (Zea mays).