Complete chloroplast genome of the medicinal plant Evolvulus alsinoides: comparative analysis, identification of mutational hotspots and evolutionary dynamics with species of Solanales

Abstract

Evolvulus alsinoides, belonging to the family Convolvulaceae, is an important medicinal plant widely used as a nootropic in the Indian traditional medicine system. In the genus Evolvulus, no research on the chloroplast genome has been published. Hence, the present study focuses on annotation, characterization, identification of mutational hotspots, and phylogenetic analysis in the complete chloroplast genome (cp) of E. alsinoides. Genome comparison and evolutionary dynamics were performed with the species of Solanales. The cp genome has 114 genes (80 protein-coding genes, 30 transfer RNA, and 4 ribosomal RNA genes) that were unique with total genome size of 157,015 bp. The cp genome possesses 69 RNA editing sites and 44 simple sequence repeats (SSRs). Predicted SSRs were randomly selected and validated experimentally. Six divergent hotspots such as trnQ-UUG, trnF-GAA, psaI, clpP, ndhF, and ycf1 were discovered from the cp genome. These microsatellites and divergent hot spot sequences of the Taxa ‘Evolvulus’ could be employed as molecular markers for species identification and genetic divergence investigations. The LSC area was found to be more conserved than the SSC and IR region in genome comparison. The IR contraction and expansion studies show that nine genes rpl2, rpl23, ycf1, ycf2, ycf1, ndhF, ndhA, matK, and psbK were present in the IR-LSC and IR-SSC boundaries of the cp genome. Fifty-four protein-coding genes in the cp genome were under negative selection pressure, indicating that they were well conserved and were undergoing purifying selection. The phylogenetic analysis reveals that E. alsinoides is closely related to the genus Cressa with some divergence from the genus Ipomoea. This is the first time the chloroplast genome of the genus Evolvulus has been published. The findings of the present study and chloroplast genome data could be a valuable resource for future studies in population genetics, genetic diversity, and evolutionary relationship of the family Convolvulaceae.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12298-021-01051-w.

Introduction

E. alsinoides, commonly known as shankhpushpi belonging to the Convolvulaceae family, is an essential plant in the Indian traditional medicine system for treating various ailments (Sethiya et al. 2010). It can be seen in regions with tropical and subtropical climates. The plant is extensively found in Kerala’s western Ghats region, where it is one of the ten sacred plants known as ‘Dashapushpam’ (Austin 2008; Priya 2017; Uthaman and Nair 2017). Jeevanyadi ghrita, Brahma rasayana, Vachadi ghrita, Brahmi ghrita, Naladi ghrita and Agastya rasayana are renowned preparations in Ayurveda that have Dashapushpam as the vital ingredient (Madhavan et al. 2008). The various phytochemicals present in the plant possess antibacterial, antifungal, antiviral, anticancer, neurodegenerative, antidiabetic, antioxidant, and immunomodulatory activities (Priya 2017; Ambika and Nair 2019; Biémont and Vieira 2006; Sethiya et al. 2019; Singh 2008; Siraj et al. 2019). The green plants have chloroplast, which is a vital organelle responsible for photosynthesis, carbon fixation, biosynthesis of amino acids, fatty acids, starch, and pigments (Jansen et al. 2007; Kim et al. 2015; Sugiura 1992). Chloroplast DNAs (cpDNAs) are double-stranded molecules, existing in a circular model, ranging it’s size from 35 to 217 Kbp, but the size varies from 115 to 165 Kbp in terrestrial plants, with a structure that is quadripartite comprising small single copy (SSC) and large single copy (LSC) region that is parted by two inverted repeats (Sugiura 1992, 2005; Ravi et al. 2008; Wicke et al. 2011). The chloroplast genome (cp genome) has been categorized into introns, protein-coding genes, and intergenic spacers based on functions. The introns and intergenic spacers that do not encode proteins are referred to as non-coding regions. It could be used for phylogenetic, molecular evolution, and population genetic studies (Shaw et al. 2007). The first complete cpDNA was sequenced in 1986 for Marchantia polymorpha; after that, several other cpDNA have been sequenced and characterized from several plants to understand the evolutionary and functional information (Supriya and Priyadarshan 2019). Next-generation sequencing techniques allow the generation of massive sequence data at a low cost. It has paved the way for molecular ecology research which has made cp genome sequencing more feasible (Henry et al. 2014). The cp genome sequencing has developed as an alternative solution for marker identification, DNA barcoding, species identification, and phylogenetic studies. In general, the nuclear genome is less conserved than cp genome (Kim et al. 2015; Park et al. 2018; Erixon and Oxelman 2008; Erixon and Oxelman 2008; Sloan et al. 2014).

Although cp genome of other genera belonging to the Convolvulaceae family has been sequenced and used in phylogenetic studies, no cp genome of the genus Evolvulus has been sequenced and published so far. Hence, in the current study, the complete cp genome of E. alsinoides was sequenced and characterized using Illumina and PacBio sequencing technologies. The denovo assembled sequence was annotated and compared with the previously reported cp genome of Solanales to understand the gene content, genome organization, and its evolutionary history.

Materials and methods

Plant material and DNA extraction

The plant E. alsinoides used for the study was gathered from Kariavattom North campus, 8.5674° N; 76.8879° E, elev. 60 m. University of Kerala, Trivandrum, Kerala, India. Under the voucher number KUBH9910, the voucher specimen was deposited at the Herbarium of the Department of Botany, University of Kerala, India. CTAB (cetyltrimethylammonium bromide) method (Healey et al. 2014) was used to isolate whole genomic DNA from leaf tissues. Nanodrop (Nanodrop Technologies, Wilmington, DE, US) and 1% (w/v) agarose gel electrophoresis were used to estimate the purity of the sample, and the quantification was done using the Qubit system (Thermo Fisher Scientific, Waltham, MA, USA).

Library preparation and genome sequencing

A total amount of 1 µg of high-quality genomic DNA was used for library preparation. The genomic DNA was randomly fragmented to 300 bp using the S220 Focused‐ultrasonicator system (Covaris, Woburn, MA, USA). The NEBnext Ultra DNA library prep kit for Illumina (E7370L; New England Biolabs, Ipswich, MA, USA) was used to prepare sequencing libraries as per manufacturer’s instructions. Paired-end sequencing was done in Illumina HiSeqX platform (2 × 150 bp; Illumina, San Diego, CA, USA). For PaCBio sequencing, 5 µg of mechanically sheared genomic DNA [using Covaris g‐TUBE (Covaris, Woburn, MA, USA)] of size range 10–20 Kb were enzymatically repaired and converted into Single-Molecule Real-Time bell templates (SMRTbell) according to the manufacturer’s protocol. The SMRTbell template libraries were prepared using SMRTbell Template Prep Kit 1.0 (100-259-100; Pacific Biosciences, Menlo Park, CA, USA). The resulting SMRTbell templates were then size-selected by Blue Pippin electrophoresis (Sage Sciences). The SMRTbell templates ranging from 15 to 20 Kb were sequenced on a PacBio Sequel Sequencing chemistry 2.1 (Pacific Biosciences, Menlo Park, CA, USA) instrument with 3 SMRT cells.

Assembly and annotation

The FastQC was used to perform the quality of the raw reads (v0.10.0) (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). The preprocessing step was performed using high quality Illumina raw reads. Cutadapt v1.8 (Martin 2011) were used to remove the adaptor sequences, and Sickle v 1.33 (https://github.com/najoshi/sickle) were used to perform quality trimming of the reads. To extract the chloroplast reads, preprocessed Illumina and PacBio reads were compared to a reference set of previously published cp genomes from 5 Convolvulaceae species obtained from the National Center for Biotechnology Information (NCBI) database (Supplementary Table S1). The reference genome set was mapped to Illumina reads using Bowtie2 v2.2.6 (Langmead and Salzberg 2012), with default parameters, and PacBio reads were aligned to the reference genome set using minimap2-2.17 (Li 2018). All mapped PacBio reads were extracted, filtered, and the error rectified with aligned Illumina reads using Proovread v2.12 (Hackl et al. 2014) and Canu v1.8 was used to assemble the resulting data (Koren et al. 2017). The raw reads were remapped to the final cp genome assembly using Bowtie2 v2.2.6. to verify the coverage and quality of the assembled genome. Genome annotation was carried out using GeSeq (Tillich et al. 2017) with reference species of the family Convolvulaceae (Supplementary Table S1). The protein-coding regions were confirmed using BLAST (Altschul et al. 1990) analysis. The codon positions (start and stop) and the borders of introns and exons, were manually adjusted using the cp genome reference sets. Predicted tRNAs were confirmed by tRNAscan-SE 1.21 (http://lowelab.ucsc.edu/tRNAscan-SE/) (Chan and Lowe 2019; Lowe and Chan 2016). Organellar Genome DRAW (OGDRAW) (https://chlorobox.mpimp-golm.mpg.de/OGDraw.html) (Lohse et al. 2013) was used to generate a map of the genome with a circular shape.

Codon usage analysis and RNA editing site prediction

CodonW v1.4.4 (http://codonw.sourceforge.net/) was used to carry out synonymous codon usage analysis of the protein-coding genes. PREPsuite (http://prep.unl.edu/) (Du et al. 2009) was used to identify the RNA editing sites in the protein-coding region of E. alsinoides with default settings.

Analysis of microsatellites and oligonucleotide repeats

Simple sequence repeats (SSRs) referred as microsatellites were identified using the MISA‐Web (https://webblast.ipk-gatersleben.de/misa/) (Beier et al. 2017) with the default parameters. A comparison of SSRs was also performed among the cp genome of 5 species belongs to the family Convolvulaceae (Supplementary Table S1). The minimum number of repeats used for the analysis was set as default. REPuter v2.74 (https://bibiserv.cebitec.uni-bielefeld.de/reputer/) (Kurtz et al. 2001) was used to identify four different kinds of repeats, such as forward repeat, reverse repeat, complement repeat, and palindromic repeat.

Experimental validation of SSR markers

DNA isolation was performed using above mentioned method. SSR primers flanking the microsatellites were designed using Primer3 (https://primer3.ut.ee/) (Untergasser et al. 2012) and Perl script provided by the MISA (https://webblast.ipk-gatersleben.de/misa/) (Beier et al. 2017). Fifteen SSR primer pairs were synthesized, and gradient PCR was performed to determine the annealing temperature. Different temperatures variations were checked for each primer. DNA amplification was performed in a 25 μL reaction mixture containing 1 × Buffer (NEB, USA), 200 μM dNTPs (NEB, USA), 1 μL of two primers (forward and reverse primers) (10 Pico Moles, Europhins, Luxembourg), 1U Taq DNA polymerase (NEB, USA) and 50 ng DNA template. SSR-PCR amplification was performed in a C1000 Touch Thermal Cycler (Bio-Rad Laboratories, Hercules, CA, USA) with the following program: initial denaturation at 94 °C for 5 min, followed by 35 cycles of 94 °C for 40 s, primer annealing with specific temperature for 40 s, 72 °C for 1 min 30 s, final extension for 5 min at 72 °C. On a 3% agarose gel the amplified products were separated, stained with ethidium bromide, visualized and photographed using Chemi Doc image documentation analysis system (BioRad, USA). Fragment sizes were estimated with 1 kb DNA Ladder sizing markers (NEB, USA). A similarity search was executed with amplified SSRs using BLAST with an e-value of 1E−5 to identify the functional homologs.

Genome comparison, sequence variation, IR contraction, and expansion

Genome comparison was performed using cp genome of E. alsinoides and 5 species, which are closely related to the family Convolvulaceae representing three genera (Cressa, Ipomoea, and Cuscuta) (Supplementary Table S1). Multiple sequence alignment was performed using mVISTA program (available online: http://genome.lbl.gov/vista/index.shtml) which compares the cp genome of E. alsinoides with other selected species using Shuffle‐LAGAN mode to estimate the divergence of sequence. Using the IRscope tool (https://irscope.shinyapps.io/irapp/), a comparison of junction sites among small single copy (SSC), inverted repeat (IR), and large single copy (LSC) were done.

Synonymous and non-synonymous substitutions rates and divergent hotspot identification

The individual protein-coding genes (Katoh and Standley 2013) MAFFT v7 (https://mafft.cbrc.jp/alignment/server/) were used to map the individual protein-coding genes of E. alsinoides and 18 species of the order Solanales, and translated to protein sequence. DNAsp v5.10.01(Rozas et al. 2003) was used to identify nonsynonymous (Ka) and synonymous (Ks) substitution rate and Ka/Ks ratio to identify the genes under selection pressure. The above sequence alignment was used to calculate the nuclear diversity (Pi) using DnaSP v5.10.01 software (Rozas et al. 2003).

Phylogenetic analysis and divergence time estimation

For phylogenetic analysis, Convolvulaceae and Solanaceae families were included within the order Solanales as among the families, Solanaceae is closely related to Convolvulaceae. The cp genomes of 33 species were selected (Supplemental Table S2). The in‐group contains the genomes of 32 species from the order Solanales, including fourteen Convolvulaceae (five Cuscuta species, seven Ipomoea species, C. cretica, and E. alsinoides) and eighteen Solanaceae (five Capsicum species, six Solanum species, and six Nicotiana species and Petunia exserta). Arabidopsis thaliana was selected as the outgroup since it is a well-studied model organism and it has been used as an outgroup in several studies involving identification, characterization, and annotation of organelle genome (Zhao et al. 2019; De-la-Cruz and Núñez-Farfán 2020). Sequences of all the cp genomes were downloaded from the NCBI Organelle Genome Resources database. Conserved protein-coding genes were extracted using local Perl script and aligned using MAFFT 7 (Katoh and Standley 2013). The alignment was adjusted manually and concatenated to get a super alignment for constructing the phylogenetic tree. The best substitution model for constructing phylogenetic trees was predicted using jModelTest version 3.7. Maximum Parsimony (MP), Maximum Likelihood (ML), and Bayesian Inference (BI) were employed for creating the phylogenetic tree. The ML analysis was performed by MEGA v7.0.26 (Kumar et al. 2016) with 1000 bootstrap replicates. The MP analysis were done using PAUP* v4.0a165 (Wilgenbusch and Swofford 2003). 1000 replicates of random taxon addition, tree bisection-reconnection branches wrapping, multitree on, the collapse of zero-length branches, and numerous tree options were used in a heuristic search, with each replicate saving a maximum of 100 trees. BI analysis was performed using the MrBayes program v3.2.7a (Huelsenbeck and Ronquist 2005).

BEAST v1.10.1 (Yoder and Yang 2000; Li and Drummond 2012) was used to estimate the divergence times based on the Bayesian method with the GTR + GAMMA substitution model, with a strict molecular clock and Yule model. The tree model used was a random starting tree. The super alignment was used to analyze the divergence time. The calibration points for the species (divergence between Ipomoea nil and Arabidopsis thaliana was 121 million years ago) were obtained from Tree time webserver (Kumar et al. 2017), based on previously published calibration time. BEAST Monte Carlo Markov Chain (MCMC) simulations were run for 10,000,000 generations to generate the tree file. The tree file was annotated using Tree Annotator v1.6.1 software. FigTree v1.4.3 was used to create the final tree with node ages.

Results and discussion

Assembly and annotation

A total of 496,329,876 Illumina (74 Gb) and 3,576,937 PacBio (20.99 Gb) raw reads were generated after sequencing of leaf tissues (SRA ID: PRJNA551197). The pre-processing step resulted in 450,360,918 cleaned high quality Illumina reads. Out of these, 38,132,788 (7.68% of total reads) Illumina reads and 203,338 (5.68% of total reads) PacBio reads were mapped to the whole cp genome of closely related species. The de novo assembly using error corrected high-quality Illumina reads and PacBio reads created a circular cp genome of E. alsinoides with a size of 157,015 bp. With the entry number MN548282, the whole cp genome sequence of E. alsinoides has been deposited in the NCBI GenBank database. The cp genomes of all angiosperm possess a typical quadripartite structure with an LSC region, SSC region, and repeat regions (IRa/IRb) (Wicke et al. 2011; Jones and Kang 2015). The LSC region (85,820 bp) and SSC region (13,691 bp) of E. alsinoides’ cp genomes are separated by a couple of inverted repeats, IRa and IRb (28,752 bp) (Fig. 1). The junction regions between IR, LSC, and SSC of the cp genome may vary among all angiosperms (Wang et al. 2013; Xu et al. 2012; Zhang et al. 2011). The cp genome of E. alsinoides consists of 114 unique genes, including 80 PCGs, 30 tRNAs, and 4 rRNAs (Table 1). The SSC region has 11 PCGs and 1tRNA, while LSC region has 63 PCGs and 23tRNAs (Supplementary Table S3). Of 114 genes, 6 PCGs (ycf1, ycf2, ycf15, rps7, rps12, ndhB), 6 tRNA genes (trnL-CAA, trnA-UGC, trnI-GAU, trnN-GUU, trnV-GAC, trnR-ACG) and 4 rRNA genes (rrn16, rrn23, rrn4.5, rrn5) in the IR regions were duplicated. The cp genome of E. alsinoides contains 13 intron-containing genes [7 PCGs and 6 tRNA (Table 1)]. The trans-spliced gene ‘rps12’ is seen in the LSC region’s 5′ end and repeated in the IR region’s 3′ end. The matK gene was positioned inside the intron of trnK-UUU. In the border zones between IR and SSC, one pseudogene (ycf1) was found. The GC content of the cp genome was 37.3%, which was similar to the published cp genome of other species belonging to the family Convolvulaceae (Funk et al. 2007; Park et al. 2018, 2019). IR regions had a GC content of 41.2%, which was higher than the GC content of LSC (39.5%) and SSC (32.2%) regions.

Fig. 1.

cp genome map of E. alsinoides. Genes in the inner circle are transcribed clockwise, and genes in the outer circles are transcribed in counter-clockwise direction. The IR/LSC/SSC boundaries are represented in the inner circle. The inner grey circle represent the GC and AT content of the cp genome

Table 1.

Genes present in E. alsinoides cp genomes

Gene category	Gene group	Gene name
Photosynthesis related genes	Photosystem I	psaA, ycf4, psaI, psaB, psaC, ycf3c, psaJ
	Photosystem II	psbH, psbI, psbA, psbD, psbB, psbC, psbF, psbE, psbJ, psbT, psbZ psbK, psb, psbN, pbf1, psbM
	Cytochrome b/f complex	petBb, petDb, petA, petG, petL, petN
	ATP synthase	atpH, atpA atpE, atpB, atpFb, atpI
	NADH dehydrogenase	ndhG, ndhAb, ndhE, ndhD, ndhC, ndhBab, ndhH, ndhF, ndhI, ndhK, ndhJ
Transcription and translation-related genes	RNA polymerase	rpoC2, rpoA, rpoC1b, rpoB
	Large subunit ribosomal proteins	rpl14, rpl2, rpl16, rpl20, rpl23, rpl32, rpl22, rpl36, rpl33
	Small subunit ribosomal proteins	rps2, rps8, rps4, rps7a, rp3, rps14, rps12abc, rps15, rps11, rps16b, rps19, rps18
RNA genes	Ribosomal RNA genes	rn4.5a, rrn23a, rrn5a, rrn16a
	Transfer RNA genes	trnA-UGCac, trnC-GCA, trnD-GUC, trnE-UUC, trnF-GAA, trnG- GCC, trnG-UCC, trnI-CAU, trnfM-CAU, trnK-UUU, trnI-GAUac, trnL-CAAa, trnL-UAAb, trnL-UAG, trnM-CAU, trnN-GUUa, trnP-UGG, trnR- ACGa, trnQ-UUG, trnR-UCU, trnS-CGAb, trnS-GCU, trnS-GGA, trnS-UGA, trnT-GGU, trnT-UGU, trnV-GACa, trnV-UACb, trnW- CCA, trnY-GUA
Other genes	Fatty acid synthesis	accDb
	Carbon metabolism	cemA
	Cytochrome c synthesis	ccsA
	Proteolysis	clpPc
	RNA processing	matK
	Rubisco	rbcL
Genes of unknown function	Conserved reading frames	ycf1ab, ycf2a, ycf15a

^aSimilar gene in IRs

^bSingle intron gene

^cTwo intron gene

Codon usage analysis

In plants, codon usage bias refers to synonymous codons having different usage frequencies. Several evolutionary mechanisms that may lead to gene mutation and selection influenced the occurrence of codon bias (Ermolaeva 2001; Wong et al. 2002). Usage patterns of Codon may differ among different genes and species. An adequate number of codons (Nc) of 80 PCGs of E. alsinoides were calculated to study codon usage analysis (Supplementary Table S4). Results show that Nc values varied between 30.55 (rpl36) and 61 (psbM) in all the identified PCGs. The majority of Nc values in the E. alsinoides cp genome were greater than 44, indicating a fragile gene codon bias. With a mean Nc value of 30.55, the rpl36 gene was found to have the most biassed codon use. (Supplementary Table S4). The 79 unique PCGs consisted of 66,448 bp that encoded 22,026 codons. The most frequently used AA in the E. alsinoides cp genome was leucine, which was encoded by 2366 (10.74%) of these codons (Table 2). Only 312 (1.42%) codons encode cysteine, the least frequent amino acid. It was reported that in angiosperms leucine is the frequently used AA while cysteine is the rarely used AA. (Chen et al. 2015; Dong et al. 2012; Kaila et al. 2016). The Relative Synonymous Codon Usage was used to calculate the codon usage bias (Sharp and Cowe 1991). The frequency of a codon divided by the expected frequency gives Relative Synonymous Codon Usage. In the E. alsinoides cp genes, thirty codons showed codon use bias due to RSCU values of > 1. Twenty-nine of the thirty codons had A (adenine) or U (uracil) endings. The G (guanine) or C (cytosine) ending codons (RSCU values 1) showed the contrary pattern, indicating that they are less prevalent in E. alsinoides cp genes. The use of stop codons was discovered to be skewed toward UAA. Poplar, rice, and other plants were found to have similar codon usage rules of bias for A (adenine) or T (thymine) (Shaw et al. 2007; Wang et al. 2013; Qingpo and Qingzhong 2004).

Table 2.

The Relative synonymous codon usage of the E. alsinoides cp genome

Amino acid	Codon	Number	RSCU	AA frequency	Amino acid	Codon	Number	RSCU	AA frequency
Phe	UUU	857	1.33	5.86	Tyr	UAU	630	1.61	3.56
	UUC	434	0.67			UAC	155	0.39
Leu	UUA	721	1.83	10.74	TER	UAA	106	1.39	1.04
	UUG	494	1.25			UAG	56	0.73
	CUU	480	1.22			UGA	67	0.88
	CUC	142	0.36
	CUA	365	0.93		His	CAU	383	1.46	2.37
	CUG	164	0.42			CAC	140	0.54
Ile	AUU	923	1.48	8.49	Gln	CAA	595	1.43	3.79
	AUC	382	0.61			CAG	239	0.57
	AUA	565	0.91
					Asn	AAU	695	1.47	4.28
Met	AUG	517	1.00	2.35		AAC	249	0.53
Val	GUU	427	1.43	5.42	Pro	CCU	324	1.44	4.09
	GUC	180	0.6			CCC	205	0.91
	GUA	420	1.41			CCA	244	1.08
	GUG	167	0.56			CCG	129	0.57
Lys	AAA	886	1.48	5.45	Trp	UGG	368	1	1.67
	AAG	315	0.52
					Ala	GCU	525	1.74	5.47
Cys	UGU	220	1.41	1.42		GCC	213	0.71
	UGC	92	0.59			GCA	329	1.09
						GCG	139	0.46
Ser	UCU	418	1.54	5.31
	UCC	269	0.99		Glu	GAA	791	1.49	4.83
	UCA	317	1.17			GAG	273	0.51
	UCG	165	0.61
					Thr	ACU	425	1.54	5
Ser	AGU	345	1.27	2.07		ACC	208	0.75
	AGC	111	0.41			ACA	342	1.24
						ACG	128	0.46
Arg	AGA	400	1.76	2.49
	AGG	149	0.66		Asp	GAU	660	1.58	3.8
						GAC	176	0.42
Arg	CGU	252	1.11	3.7
	CGC	105	0.46		Gly	GGU	472	1.26	6.78
	CGA	327	1.44			GGC	161	0.43
	CGG	129	0.57			GGA	556	1.49
						GGG	305	0.82

The significance of relative synonymous codon usage (RSCU) > 1 is denoted in bold

RNA editing sites prediction

E. alsinoides cp genome contains 69 RNA editing sites, which was distributed among 20 PCGs (Table 3). The organelle genome of most plants has RNA editing phenomenon (Bock 2000). Out of 69 RNA editing sites, 33 are C- to -T conversion which is the most common type in Organellar transcripts of plants (Jansen et al. 2007; Duruvasula et al. 2019; Maier et al. 1995). The psaB (16) gene has maximum RNA editing sites trailed by psbB (12), rpoC1 (7), rpoC2 (6), rpoB (5), petD (4), accD (3), psaI, psbF, petB, rps14 (2 each), rps16, atp1, rps2, ycf3, psbL, psbE, petG, rpoA Meanwhile, the genes ndhD and rpoB (1 each). The highest transition in cp genome is the Leucine to Phenylalanine, trailed by Proline to Leucine and Serine to Leucine. The location of RNA editing sites present in the cp genome was 1st codon (36), 2nd codon (33), and 3rd codon (0), respectively (Supplementary Table S5). The RNA prediction site in the first codon of the first nucleotide was absent in the genes such as atp1, rps2, rpoB, rps14, ycf3, accD, PsaI, and PetG. The following genes atp1, rps2, rpoB, rps14, ycf3, accD, PsaI, and PetG do not have an RNA predicting site in their first nucleotide codon.

Table 3.

The predicted RNA editing site in the E. alsinoides cp genes

Gene	Codon conversion	Amino acid conversion	Gene	Codon conversion	Amino acid conversion
rps16	CTT =  > TTT	L =  > F	ycf3	ACT =  > ATT	T =  > I
atpI	TCA =  > TTA	S =  > L
rps2	TCA =  > TTA	S =  > L	accD	GCA =  > GTA	A =  > V
rpoC2	CAT =  > TAT	H =  > Y		GCT =  > GTT	A =  > V
	CAT =  > TAT	H =  > Y		CTT =  > TTT	L =  > F
	CAC =  > TAC	H =  > Y
	CGG =  > TGG	R =  > W	psaI	TCA =  > TTA	S =  > L
	CGC =  > TGC	R =  > C
	CCG =  > TCG	P =  > S	psbL	CTT =  > TTT	L =  > F
rpoC1	ACA =  > ATA	T =  > I
	CCC =  > TCC	P =  > S	psbF	CCT =  > CTT	P =  > L
	ACT =  > ATT	T =  > I		CTT =  > TTT	L =  > F
	TCG =  > TTG	S =  > L
	CCT =  > CTT	P =  > L	psbE	CCT =  > TCT	P =  > S
	CAT =  > TAT	H =  > Y
	ACA =  > ATA	T =  > I	petG	TCT =  > TTT	S =  > F
rpoB	TCA =  > TTA	S =  > L	psaI	TCA =  > TTA	S =  > L
	TCA =  > TTA	S =  > L
	TCG =  > TTG	S =  > L	psbB	CTT =  > TTT	L =  > F
	TCT =  > TTT	S =  > F		CTT =  > TTT	L =  > F
	GCA =  > GTA	A =  > V		CTT =  > TTT	L =  > F
				CTT =  > TTT	L =  > F
rps14	TCA =  > TTA	S =  > L		CCT =  > CTT	P =  > L
	CCA =  > CTA	P =  > L		CTT =  > TTT	L =  > F
				CTT =  > TTT	L =  > F
psaB	CTT =  > TTT	L =  > F		CTT =  > TTT	L =  > F
	CTT =  > TTT	L =  > F		CTT =  > TTT	L =  > F
	CCT =  > CTT	P =  > L		CTT =  > TTT	L =  > F
	CTT =  > TTT	L =  > F		CTT =  > TTT	L =  > F
	CCG =  > TCG	P =  > S		CCT =  > CTT	P =  > L
	CCT =  > CTT	P =  > L
	CCT =  > CTT	P =  > L	petB	CCA =  > CTA	P =  > L
	CTT =  > TTT	L =  > F		ACC =  > ATC	T =  > I
	CTT =  > TTT	L =  > F	petD	CTT =  > TTT	L =  > F
	CTT =  > TTT	L =  > F		CTT =  > TTT	L =  > F
	CTT =  > TTT	L =  > F		CCT =  > CTT	P =  > L
	ACA =  > ATA	T =  > I		CCT =  > CTT	P =  > L
	CAC =  > TAC	H =  > Y
	TCT =  > TTT	S =  > F	rpoA	CTT =  > TTT	L =  > F
	CCT =  > CTT	P =  > L
	CTT =  > TTT	L =  > F

Analysis of microsatellites and oligonucleotide repeats

Microsatellites also referred as SSRs are short nucleotides stretches that are widely present in the genomes. 49 SSRs were identified in the cp genome of E. alsinoides. Out of it, the majority were mononucleotide repeats (44) trailed by dinucleotide (1), trinucleotide (2) tetranucleotide (1) and hexanucleotide (1) repeats (Fig. 2A). Of the 49 SSRs, 88% of the mononucleotide SSRs were (A/T) Poly T (polythymine) or poly-A (polyadenine) repeats in the cp genome of E. alsinoides (Fig. 2B) similar to preceding studies of other species (Saina et al. 2018; Zhou et al. 2016; Qian et al. 2013) (Supplementary Table S6). About 42 SSRs were situated in the non-coding regions (IGS intergenic spacers) and 7 SSRs, including ycf3, accD, clpP and ndhA were located in the genic region. The majority of SSRs were found in the LSC region with 73%, and the SSC and IR regions contain 22% and 5% respectively. Two mononucleotides are present in the IR region, which was the duplicated SSR. In total, 44 non-redundant SSRs were present in the cp genome of E. alsinoides. SSRs are used as molecular indicators in genetic heterogeneity studies, plant breeding programs, evolutionary studies, and species identification (Chan and Lowe 2019; Bryan et al. 1999; Ebert and Peakall 2009; Ebert and Peakall 2009; Yang et al. 2016; Powell et al. 1995; Dong et al. 2013). A comparison of SSRs was performed among the cp genome of 5 species belongs to the family Convolvulaceae (Supplementary Table S6). Among all the species, E. alsinoides and Ipomoea purpurea have the maximum number of SSRs, and Cuscuta exaltata has the least number of SSRs (Fig. 3). The length of the repeats in all the species ranges from 10 to 115 bp. I. nil and I. purpurea have only mononucleotide repeats. Cuscuta reflexa and C. exaltata have mono, and di nucleotide repeats, whereas Cressa cretica has mono and tri nucleotide repeats. C. cretica and I. purpurea have the shortest length of repeats. The A/T mono- repeats were profusely existing in all the species chloroplast genome used for the study.

Fig. 2.

A Number of SSRs in cp genome, coding region and non-coding region. B Frequency of SSR motifs in the cp genome. The X-axis represent the type of repeats and the Y-axis represents the No of SSRs

Fig. 3.

Comparison of SSRs with the closely related species of E. alsinoides

In addition to the SSRs, repeat analysis identified 86 repeats, which include 24 reverse repeats, 26 palindromic repeats, 34 forward repeats, and 2 complementary repeats. Among these repeats 53% (forward 22, palindrome 17, reverse 6) of the repeats are present in the IR region, 37% (reverse 18, forward 9, palindrome 3, complementary 2) in the LSC region, and 10% (palindrome 6, forward 3) in the SSC region (Supplementary Table S7). The size of reverse and complementary repeats ranges from 30 to 47, whereas the size of forward and palindrome repeats are > 100. Based on the comparison of repeats among the closest species of the genus (Park et al. 2018; Park et al. 2018; Park et al. 2018) of E. alsinoides, it is evident that the size of the repeats is larger (> 100). Also, the reverse and complementary repeats are very less (Park et al. 2018). Hence, it could be concluded that repetitive regions are present in higher amounts in the family Convolvulaceae than other angiosperm species.

Experimental validation of SSR markers

Out of 44 SSRs predicted by MISA, 15 representative SSRs were taken randomly for the functional validation based on GC% (> 50), primer dimer formation and hairpin secondary structure formation. The annealing temperature was fixed according to the intensity of amplified product generated with gradient PCR (Supplementary Fig. 1). Among the 15 SSR primers synthesized (Supplementary Table S8), 13 primers showed amplification with single amplified products with sizes ranging between 300 and 400 bp (Fig. 4). Based on the similarity search, 8 SSRs showed sequence homology with PCGs of chloroplast genome (Supplementary Table S9). The amplified SSRs showed a clear banding pattern in agarose gel. Among the 13 SSRs, 8 SSRs showed similarity with chloroplast PCGs such as psbl, pasl, atpA and rps2. Out of the 8 SSRs, 4 SSRs showed similarity with psbl (SSR3,4,5) and pasl (SSR6) genes, photosynthetic related genes involved in photosystem I and II. Three SSRs (SSR8,10,13) exhibited similarity with ATP synthase, and one has similarity with rps2 (SSR 12), which is involved in transcription and translation machinery. The SSR sequences were submitted to NCBI GenBank (Supplementary Table S10). Gene diversity studies could use these genes as specific markers. The SSR primers developed were highly authenticated SSR markers that could be used in genetic diversity studies in the future.

Fig. 4.

Representative agarose gel image of SSR amplified regions in E. alsinoides. Gel image showing amplification with 3% agarose. Lane1 represents 1 kb DNA ladder and Lane2 to 14 represents SSR1 to SSR13

Genome comparison, sequence variation, IR contraction, and expansion

Genome comparison establishes that the LSC region is more conserved than the IR and SSC region due to gene rearrangement (Park et al. 2018; Delannoy et al. 2011; Wicke et al. 2013). Many of the protein-coding regions of the family are more conserved than the non-coding regions (Fig. 5). The IR-LSC and IR-SSC boundaries comparison of the five cp genome of the family Convolvulaceae were represented in (Fig. 5). It is evident from the results that the size of IR, SSC, and LSC regions was highly variable among the five species of the family Convolvulaceae, but the genes present in the cp genome’s boundaries are not much highly variable. The genes rpl2, rpl23, ycf1, ycf2, ycf1, ycf15, ndhH, ndhF, ndhA, ndhD, ccsA, rpl32, psbA, psbK, matK are present in the IR-LSC and IR-SSC boundaries of cp genome (Fig. 6). The rpl2, rpl23, and ycf2 genes are present in the LSC-IRa boundaries. The rpl2 and rpl23 genes are present in all the species. ycf2 gene was located in the LSC –IRb and LSC-IRa border in all the species. The ndhF are present in 4 species except for the genus Cuscuta whereas the ndhD gene is present in two species of Cuscuta. ycf1 genes that are located in the SSC-IR boundaries are present in all the species except I. nil. The ccsA gene is present in the SSC region of all the species. psbA genes are present in all the five species while E. alsinoides has psbK in the IRa-LSC boundary. Many genes are common in all the species that are considered for the study. It is observed that the position and presence of few genes vary because of variation in size among the cp genomes (Huang et al. 2014; Hansen et al. 2007; Raubeson et al. 2007; Khakhlova and Bock 2006).

Fig. 5.

Comparison of cp genomes of five species within the family Convolvulaceae with reference to the cp genome of E. alsinoides. The name of the genes, their position and orientation are shown above the alignment. X-axis shows the cp genome coordinates while Y-scale shows percent identity between 50 and 100%

Fig. 6.

Cp genome comparison of LSC, IRb, SSC and IRa border region of E. alsinoides with five species belongs to the family Convolvulaceae

The size of the complete genome, the number of protein-coding genes, LSC, IR, and SSC regions are highly variable among all the Convolvulaceae species that are considered for the study. Among all the species, Ipomoea has the largest genome size, and Cuscuta has the least genome size. The LSC and SSC region was larger, and the IR region was found to be less in C. cretica than in all other species (Table 4). The GC content and rRNAs are almost the same in all the species, whereas the protein-coding regions and tRNAs are highly variable among all the family species. These could be the reasons for variability among genes present in LSC, IR, and SSC regions and boundaries. From the comparative analysis, it is evident that the cp genomes are highly variable among all the species of family Convolvulaceae.

Table 4.

Comparison table showing chloroplast genome annotation of five species of the family Convolvulaceae

Species	E. alsinoides	C. cretica	C. exaltata	C. reflexa	I. nil	I. purpurea
Accession id	MN548282	MF067398	NC_009963	NC_009766	AP017304	NC_009808
Size (bp)	157,015	141,419	125,373	121,521	161,897	162,046
LSC (bp)	85,820	102,334	82,721	79,468	88,118	88,172
SSC (bp)	13,691	38,115	9250	8571	12,087	12,110
IR (bp)	57,504	970	33,402	33,482	30,846	61,764
GC%	37.3	38.6	38.1	38.2	37.5	37.5
Protein	80	78	70	61	78	78
rRNA	4	4	4	4	4	4
tRNA	30	17	19	19	30	29
Total genes	114	99	93	84	112	111

Synonymous and non-synonymous substitutions rates

To estimate the selection pressure of 72 protein-coding genes in the cp genome of E. alsinoides, the nonsynonymous (Ka), synonymous (Ks) substitution rate and Ka/Ks ratios were calculated. For the analysis, 18 species belong to the order Solanales (seven Ipomoea species, E. alsinoides, and C. cretica) and nine species belong to the family Solanaceae (five Capsicum species, three Solanum species, and N. tabacum) (Supplemental Table S2). If Ka/Ks ratio > 1 the gene is influenced by positive selection, if Ka/Ks ratio < 1 the gene is influenced by negative selection, and If Ka/Ks ratio = 1 the gene is influenced by neutral selection (Fiz-Palacios et al. 2011). The mean Ka/Ks ratio indicates that, out of 72 protein-coding genes (Fig. 7), 54 genes were under negative selection pressure. 14 genes (petG, petD, psbL, rpl16, atpA, petB, ndhD, rpl22, ycf3, ndhH, ccsA, clpP, ycf2, rpl23) under positive selection pressure (Supplementary Table S11) and 4 genes (psaB, psbE, petN, psaC) have the Ka/Ks ratio 0. The negative selection pressure indicates that 54 genes were relatively well conserved and undergoing purifying selection. Positive selection showed that 14 genes undergo adaptive evolution in response to their environment among the species of order Solanales (Ivanova et al. 2017; Kimura 1989; Raman and Park 2016). The Ka/Ks ratio of 0 indicates that there were no nonsynonymous (Ka) and synonymous (Ks) substitution sites in the genes, which suggests that no significant gene evolution was observed.

Fig. 7.

Comparison of nonsynonymous (Ka) and synonymous (Ks) substitution rates among eighteen species of Solanales. Protein-coding genes name is shown in X-axis, and the mean Ka/Ks values are shown in Y-axis

Divergent hotspot identification

To identify the sequence divergence level, nuclear diversity was calculated for E. alsinoides and species of the order Solanales (Supplementary Table S12). Sequences with Nuclear diversity values > 1.5% have been described as highly variable regions, and they can be leveraged for evolutionary studies and identification of species in seed plants (Korotkova et al. 2014; Särkinen and George 2013). It is evident from previous reports that non-coding regions were more divergent than coding regions with few exceptions of ycf1, matK, and rbcL genes, commonly used in plant DNA barcoding (Shaw et al. 2007; Xu et al. 2017). The nuclear diversity values ranged from 0 to 0.20, which reveals that the genomes are highly variable. Among the 18 species considered for the study, few genes such as trnQ-UUG, trnF-GAA, PsaI, ClpP, ndhF and ycf1 were found to be highly variable. These divergent hot spot regions can be leveraged as phylogenetic markers for species identification, population genetics and evolutionary studies of the genus Evolvulus (Fig. 8). The ndhF gene in the cp genome was highly divergent among the species used for the present study. The ycf1 gene also has a higher divergence than other genes found in other cp genomes’ coding regions (Liu et al. 2013; Nie et al. 2012; Song et al. 2015; Yukawa et al. 2006). These divergent hotspots of cp genomes can be used for species identification of the family Convolvulaceae. The IR region was less divergent than LSC and SSC regions of the cp genome (Fig. 8). The sliding window method revealed that the majority of the variations occurred in the cp genome’s LSC and SSC regions (Fig. 8) (Park et al. 2018).

Fig. 8.

Nuclear diversity (Pi) analysis of eighteen species of the order Solanales. The X-axis denotes the midpoint position of the window while the Y-axis denotes nucleotide diversity (Pi) of window. The thick blue line specifies the threshold for variation hotspots (Pi threshold = 0.13). The highly variable regions are represented in the graph

Phylogenetic analysis and divergence time estimation

For the present study, only Convolvulaceae and Solanaceae families are included within the order Solanales. The complete cp genome provides valuable information regarding the phylogenetic position of Angiosperms. Hence it could be used as an essential tool to understand the phylogeny (Austin 1978, 2008; Kim et al. 2015; Huang et al. 2014; Moore et al. 2007; Wu et al. 2010; Eserman et al. 2014). A phylogenetic tree was constructed using two datasets (1) complete cp genome and (2) concatenated alignment of 72 shared genes (Supplementary Table S13) for better understanding the phylogenetic position of E. alsinoides. The three methods (MP, ML, and BI) that constructed the phylogenetic tree generated the same topology tree for each dataset. The selected species formed five clades (Clade I–V), supported with 100% bootstrap (BP) values in the phylogenetic tree (Fig. 2). The clade I composed of Convolvulaceae, formed the basal position with three subclades. Subclade A included all Cuscuta species, subclade B comprised of E. alsinoides, and C. cretica and subclade C included all Ipomoea species. Clade II has Petunia exserta (Fig. 9), Nicotiana species formed clade III, Clade IV includes Capsicum species, and Clade V included all Solanum species. The only difference in the constructed tree using three methods (ML, MP&BI) is that the bootstrap value for a clade that belongs to the genus Ipomoea is 98% in the MP constructed tree. In ML analysis, the bootstrap value is 100 for all the clades. In BI analysis, the Bayesian posterior probabilities values are 1 for all the clades (Supplementary Fig. 2).

Fig. 9.

Phylogenetic tree of complete cp genome constructed using BI, ML and MP methods. The values above the nodes are Bayesian posterior probabilities, MP and ML bootstrap values. The values inside the clades represent the divergence times and the divergence dates in Mya are represented in the nodes

The divergence analysis was performed with the previously calibrated time between I. nil and Arabidopsis thaliana, and it showed they diverged 121 million years ago (Mya) (Christin et al. 2011; Fiz-Palacios et al. 2011). Our findings revealed that Clade I (Convolvulaceae) had diverged from its ancestor, Solanales, some 90 million years ago. The second speciation event happened around 68 Mya, where the family Solanaceae had evolved (Fig. 2). The Convolvulaceae species belong to Clade I and its divergence happened about 49 Mya (Fig. 2), the Ipomoea species diverged 39 Mya from the common ancestor of Evolvulus and Cressa. The divergence of Evolvulus and Cressa happened around 28 Mya, and the Cuscuta species divergence occurred about 15 Mya. Later, Nicotiana species (Clade III) diverged approximately 10 Mya (Fig. 2). Next, speciation occurred around 8 Mya (Clade IV) between the Capsicum and Solanum species, and finally, the Solanum species evolved about 6 Mya (Clade V). The present phylogenetic analysis showed that E. alsinoides and C. cretica form a separate sub-clade, and they diverged from the genus Ipomoea around 39 Mya.

Conclusion

The cp genome of E. alsenoides is the first completely sequenced cp genome of the genus Evolvulus. The cp genome had quadripartite structure with 157,015 bp in length similar to the land plant’s cp genome size. The experimentally validated 15 SSR markers could be reliable species-specific markers in upcoming studies in the genus Evolvulus. Based on the comparison studies, 72 shared PCGs were present among the species of Solanales. Six divergent hotspots trnQ-UUG, trnF-GAA, psaI, clpP, ndhF, and ycf1 identified from the cp genome could act as molecular markers for genetic diversity studies. Fifty-four genes were conserved in the cp genome and were undergoing purifying selection. From the phylogenetic studies, it was apparent that E. alsinoides is closely related C. cretica and diverged from the genus Ipomoea about 39 Mya. The overall analysis of the cp genome shows that the size, gene content, gene orientation, and gene number of the cp genome were not highly conserved among the species of the family. The cp genome and its annotation are beneficial resources for species identification, the taxonomic position of the species, genetic breeding, and evolutionary studies of the species of the family Convolvulaceae.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

AA

Amino Acid

Cp

Chloroplast

IR

Inverted repeat

LSC

Large Single Copy

PCGs

Protein Coding Genes

RSCU

Relative Synonymous Codon Usage

SMRT

Single Molecule Real Time

SSC

Small Single Copy

SSR

Simple Sequence Repeat

Author contributions

SPR and ASN designed the study. SPR, VCB, SV, AS, VCL, VR, AJ and AS conducted the experiments and performed the analysis. NF conducted the experimental validation of SSR. SPR coordinated the project and wrote the manuscript.

Funding

We acknowledge the funds received from the University of Kerala (Plan fund) and also the facility under SIUCEB project, DBT –BIF centre, MHRD-FAST (AiCADD center) in the Department of Computational Biology and Bioinformatics, University of Kerala.

Declarations

Conflict of interest

No conflict of interest declared by authors.