Genome sequencing of herb Tulsi (Ocimum tenuiflorum) unravels key genes behind its strong medicinal properties

Abstract

Background: Krishna Tulsi, a member of Lamiaceae family, is a herb well known for its spiritual, religious and medicinal importance in India. The common name of this plant is 'Tulsi' (or 'Tulasi' or 'Thulasi') and is considered sacred by Hindus. We present the draft genome of Ocimum tenuiflurum L (subtype Krishna Tulsi) in this report. The paired-end and mate-pair sequence libraries were generated for the whole genome sequenced with the Illumina Hiseq 1000, resulting in an assembled genome of 374 Mb, with a genome coverage of 61 % (612 Mb estimated genome size). We have also studied transcriptomes (RNA-Seq) of two subtypes of O. tenuiflorum, Krishna and Rama Tulsi and report the relative expression of genes in both the varieties.

Results: The pathways leading to the production of medicinally-important specialized metabolites have been studied in detail, in relation to similar pathways in Arabidopsis thaliana and other plants. Expression levels of anthocyanin biosynthesis-related genes in leaf samples of Krishna Tulsi were observed to be relatively high, explaining the purple colouration of Krishna Tulsi leaves. The expression of six important genes identified from genome data were validated by performing q-RT-PCR in different tissues of five different species, which shows the high extent of urosolic acid-producing genes in young leaves of the Rama subtype. In addition, the presence of eugenol and ursolic acid, implied as potential drugs in the cure of many diseases including cancer was confirmed using mass spectrometry.

Conclusions: The availability of the whole genome of O.tenuiflorum and our sequence analysis suggests that small amino acid changes at the functional sites of genes involved in metabolite synthesis pathways confer special medicinal properties to this herb.

Fig. 1

Plant and leaf morphology of five Ocimum species prevalent in India viz., O. tenuiflorum subtype Krishna, O. tenuiflorum subtype Rama, O. gratissimum, O. sacharicum, O. kilmand. Leaf morphologies are quite different for the five species

Fig. 2

Distribution and clustering of orthologous genes of Tulsi genome to other related plant genomes. a. Distribution of gene families among five plant genomes. Ocimum tenuiflorum (Ote - green), Arabidopsis thaliana (Ath – black rectangle), Oryza sativa (Osa – red), Solanum lycopersicum (Sly – blue) and Mimulus guttatus (Mgu – black circle). The numbers in the Venn diagram represent shared and unique gene families across these 5 species obtained by OrthoMCL. b. Horizontal stacked bar plot of all the genes in 23 different genomes. This figure shows ortholog group distribution in all 23 plant species including Tulsi. Each row represents a plant species - Physcomitrella patens (Ppa), Selaginella moellendorffii (Smo), Oryza sativa (Osa), Setaria italic (Sit), Zea mays (Zma), Sorghum bicolor (Sbi), Aquilegia caerulea (Aca), Ocimum tenuiflorum (Ote), Mimulus guttatus (Mgu), Solanum lycopersicum (Sly), Solanum tuberosum (Stu), Vitis vinifera (Vvi), Eucalyptus grandis (Egr), Citrus sinensis (Csi), Theobroma cacao (Tca), Carica papaya (Cpa), Brassica rapa (Bra), Arabidopsis thaliana (Ath), Fragaria vesca (Fve), Prunus persica (Ppe), Glycine max (Gma), Medicago truncatula (Mtr), Populus trichocarpa (Ptr). The bar graph represents ortholog protein groups for that species subdivided into 22 categories depending on the degree of sharing with the other 22 plant species e.g., category 2 represents the number of orthologous groups that have representatives from the species of interest and from one more species out of the 23 species selected for the study

Fig. 3

Phylogenetic representation of five selected plant genomes viz., Solanum lycopercicum (72.45 %), Vitis vinifera (77.84 %), Medicago trucatula (63.47 %), and Arabidopsis thaliana (56.28 %). The numbers indicate percentage of association of these genomes with the metabolite genes of Ocimum genome. These percentages agree with the taxonomic phylogeny and hierarchy, suggesting that the evolution of genes involved in metabolic pathways is not a cause of recent expansions or sudden genome drifts. The inner circle represents chromosomes from respective homolog genome. Each scaffold is organized in the middle circle and is represented in chronological order as per position on chromosomes. The line represents location of each scaffold on the respective chromosome. Colors indicate  = < 2 genes, =2 genes, = > 2 genes, = Metabolite related genes. Height of orange columns in outermost circle represents amount of repeats in corresponding scaffolds

Fig. 4

Circular representation of O.tenuiflorum metabolite related genes mapped onto chromosomes of Solanum lycopersicum genome. Height of orange column in outer circle represents amount of repeats present in respective scaffold. The inner circle represents chromosomes from Tomato genome. Inner circle of rectangles represents scaffolds, each scaffold is organized in the middle circle and is represented in chronological order as per position on chromosomes. Color of each scaffold indicates following information: =2 genes, = > 2 genes, = Metabolite related genes. Connecting line between scaffolds and chromosome represents postion of the scaffold in genome. Red color of connecting line represents presence of metabolite related genes. Scaffold Numbers are mentioned in Additional file 24: Text A

Fig. 5

Transcript expression of Tulsi Krishna and Rama subtypes are expressed as RPKM values. Highly significant differentially abundant RNA scaffolds/transcripts were defined to have RPKM of atleast 5 in both and the fold-change difference between two subtypes should be atleast 8 times. Only the transcripts, for which the 95 % lower-confidence-bound of more abundant subtype and 95 % upper-confidence-bound of less abundant subtype, and had at least 8 times difference, were retained. Of these differentially abundant transcripts, top-50 in Krishna and Rama subtype were plotted in the form of heat-map. a. Differentially more abundant transcripts in Krishna. b. Differentially more abundant transcripts in Rama. (please look in Additional file 24: Text B and C for transcript IDs for a. and b)

Fig. 6

Expression quantification of selected genes by q-RT-PCR method. a. Fold changes of genes involved in color production, obtained through q-RT PCR. Blue colour horizontal bar is for chlorophyll a-b binding protein, red to denote Gamma-cadenine synthase and green to denote Anthocyanin. Mature leaf of Krishna subtype was used as control. It can be seen that, genes responsible for color production such as Chlorophyll a-b binding protein and gene in anthacyanin pathway are down-regulated as compared to mature Krishna leaf, which corresponds to phenotypic characteristics. b. Fold changes of genes involved in ursolic acid biosynthetic pathway, as obtained through qRT-PCR for 5 different Tulsi subtypes. Blue colour horizontal bar is for squalene epoxidase, red to denote alpha-amyrin synthase and green to denote Cytochrome P450 monooxygenase. Mature leaf of Krishna subtype was used as control. Mature leaf of Rama subtype has high expression of genes while expression in Ocimum kilmund is low. Expression of these genes are uniformly high in small, developing plants. Samples are as follows: 1) O. tenuiflorum (Rama) - Sampling Leaf. 2) O. tenuiflorum (Rama) - Sampling Root. 3) O. tenuiflorum (Rama) - Mature Leaf. 4) O. tenuiflorum (Krishna) - Sampling Leaf. 5) O. tenuiflorum (Krishna) - Sampling Root. 6) O. gratissimum - Sampling Leaf. 7) O. gratissimum - Sampling Root. 8) O. gratissimum - Mature Leaf. 9) O. sacharicum - Sampling Leaf. 10) O. sacharicum - Sampling Root. 11) O. sacharicum - Mature Leaf. 12) O. kilmund - Sampling Leaf. 13) O. kilmund - Sampling Root. 14) O. kilmund - Mature Leaf

Fig. 7

Number of genes involved in specialized metabolite synthesis in Tulsi genome. a. There are four classes of metabolites present in Ocimum genome viz., sesquiterpenes (52 %), flavonoids (19 %), terpenes (18 %) and phenylpropanoids (11 %). Number in the bracket is percentage of sepecialized metabolites present in the genome. 458 genes were identified as coding for enzymes involved in synthesis of specialized metabolites. b. Specialized metabolic pathways of disease relevance proposed in Ocimum tenuiflorum. Major classes of diseases investigated are indicated in different colors: anticancer , anticancer-antioxidant , antifungal , antiseptic , anti-infective , antioxidant , and anti-inflammatory . The enzymes have been labelled with 5–7 letters for convenience. The numbers after the’_’ in the enzyme label represent the number of putative hits found for the given enzyme in the genomic assembly of O. tenuiflorum. The metabolites involved in the disease relevance and the enzymes involved in the synthesis of these metabolites are as follows: APIGENIN (Flavone-synthaseI-FSYN1, Naringenin-NADPH-oxygen-oxidoreductase-NNOOX), LUTEOLIN (Flavone-synthaseI-FSYN1, Naringenin-NADPH-oxygen-oxidoreductase-NNOOX, Flavone-3-monooxygenase-F3MON), TAXOL (Taxadiene-synthase-TSYN, Taxadiene-5-alpha-hydroxylase-T5AHYD, Taxadien-5-alpha-ol-O-acetyltransferase-T5AOOA, Taxane-10-beta-hydroxylase-T10BHYD, Taxoid-14-beta-hydroxylase-T14BHYD, 2-alpha-hydroxytaxane-2-O-benzoyltransferase-2AH2OB, 10-deacetylbaccatin-III-10-O-acetyltransferase-10D10OA, 3-N-debenzoyl-2-deoxytaxol-N-benzoyltransferase-3ND2DNB, URSOLIC ACID (ursolic-aldehyde-28-monooxygenase-UA28M, Alpha-amyrin-synthase-AASYN), OLEANOLIC ACID (Beta-amyrin-synthase-BASYN, oleanolic-aldehyde-28-monooxygenase-OA28M), SITOSTEROL (24C-methyltransferase-24CMET), ROSMARINIC ACID I (4-coumaroyl-4-hydroxyphenyllactate-3-hydroxylase-4C4H3H, Tyrosine-transaminase-TTRAN), ROMARINIC ACID II (Hydroxyphenylpyruvate-reductase-HPPRE, Tyrosine-3-monooxygenase-TTRAN), METHYL CAHVICOL (Eugenol-o-methyltransferase-EOMET), EUGENOL (Alcohol-o-acetyltransferase-AOACE, Eugenol-synthase-ESYN, Isoeugenol-synthase-ISYN), LINALOOL (Farnesyl-pyrophosphate-synthase-FPSYN, R-linool-synthase-RLSYN, S-linool-synthase-SLSYN), CARYOPHYLENE (Alpha-humulene-synthase-AHSYN, Beta-caryophyllene-synthase-BCSYN), SELINENE (Alpha-selinene-synthase-ASSYN, Beta-selinene-synthase-BSSYN), CITRAL (Geraniol-synthase-GSYN, Geraniol-dehdrogenase-GDHYD)

Fig. 8

Phylogeny of terpene synthases of representative sequences of six classes from the plant kingdom along with putative Tulsi terpene synthases genes: The tree is color coded as tpsa:red, tbsb:blue, tpsc:yellow, tpsd: green, tpse: blue and tpsf:purple

Fig. 9

The synthesis of ursolic acid from squalene is a three-step process starting from squalene. A: Squalene epoxidase, B: α-amyrin synthase, C1: α-amyrin 28-monooxygenase [Multifunctional], C2: Uvaol dehydrogenase [Multifunctional] and C3: Ursolic aldehyde 28-monooxygenase. Squalene epoxidase and alpha amyrin synthase, along with alpha amyrin 28 mono-oxygenase, uvol dehydrogenase and ursolic aldehyde 28 mono-oxygenase, play important role in synthesis of ursolic acid. These three genes have been chosen for quantification of gene expression by q-RT PCR method in different tissues and species

Fig. 10

Phylogenetic tree of sixteen amyrin query sequences and four putative amyrins from Tulsi. Tulsi hits are marked in blue clour, red ones are alpha amyrin synthase, greens are beta amyrin synthase and cyan ones are proteins from other class of amyrin. The presence of motifs and position in the phylogeny indicate that the hits obtained in O. tenuiflorum genome are likely to be alpha-amyrin synthases