Complete Plastid Genomes of Nine Species of Ranunculeae (Ranunculaceae) and Their Phylogenetic Inferences

Abstract

The tribe Ranunculeae, Ranunculaceae, comprising 19 genera widely distributed all over the world. Although a large number of Sanger sequencing-based molecular phylogenetic studies have been published, very few studies have been performed on using genomic data to infer phylogenetic relationships within Ranunculeae. In this study, the complete plastid genomes of nine species (eleven samples) from Ceratocephala, Halerpestes, and Ranunculus were de novo assembled using a next-generation sequencing method. Previously published plastomes of Oxygraphis and other related genera of the family were downloaded from GenBank for comparative analysis. The complete plastome of each Ranunculeae species has 112 genes in total, including 78 protein-coding genes, 30 transfer RNA genes, and four ribosomal RNA genes. The plastome structure of Ranunculeae samples is conserved in gene order and arrangement. There are no inverted repeat (IR) region expansions and only one IR contraction was found in the tested samples. This study also compared plastome sequences across all the samples in gene collinearity, codon usage, RNA editing sites, nucleotide variability, simple sequence repeats, and positive selection sites. Phylogeny of the available Ranunculeae species was inferred by the plastome data using maximum-likelihood and Bayesian inference methods, and data partitioning strategies were tested. The phylogenetic relationships were better resolved compared to previous studies based on Sanger sequencing methods, showing the potential value of the plastome data in inferring the phylogeny of the tribe.

1. Introduction

The complete plastid genome (plastome) has become an increasingly popular tool for phylogenetic studies in recent years [1,2,3,4]. Plastid is a common organelle found in plant cells that contains its own genome which is typically circular and relatively conserved across plant species [5]. The plastomes are often uniparentally inherited [6] and typically include about 80 protein-coding genes and more than 30 RNA genes [7,8]. The high degree of evolutionary conservation, large amount of data, uniparental inheritance, ability to identify polymorphisms, and easy availability make the plastome an ideal marker for studying phylogenetic relationships among plant taxa at different taxonomic levels [9].

The tribe Ranunculeae comprises 16 to 19 genera and about 650 species distributed worldwide, making it the most representative and diverse group within the buttercup family (Ranunculaceae) [10,11,12,13]. Among all the genera in this tribe, Ranunculus stands out as the species-rich genus of the family, with about 650 wild species in the world, whereas all the other genera are small or even monotypic [12]. There are four genera: Ranunculus L. (with the inclusion of Batrachium (DC.) Gray), Oxygraphis Bunge, Helerpestes E. L. Greene, and Ceratocephala Moench distributed in China, and Ranunculus is also the largest one of the tribes with more than 120 wild species in China [10,14]. Plants of Ranunculeae include numerous ornamental and medicinal species, with a particularly rich species diversity in temperate and alpine regions [10,15].

In recent years, numerous molecular phylogenetic studies on the tribe Ranunculeae have been published [11,12,13]. However, all of these studies used a small number of DNA fragments for phylogenetic inference, and their results had inevitable limitations such as low resolution and statistical support due to insufficient phylogenetic signal. The complete plastid genomes of the family Ranunculaceae gained more and more attention in the last few years [16,17,18,19]. Both sequence and structural variations (such as IR expansion/contraction, gene inversion, and gene transposition) in the plastomes of Ranunculaceae showed the potential to yield phylogenetic significance when comprehensive data are available [17]. There is an urgent need to incorporate genomic data to deepen our insights into the phylogeny of Ranunculeae. However, a very small number of the plastid genomes of this tribe have been published up to now.

In this study, the complete plastomes of nine species (eleven samples), representing three genera of Ranunculeae, were assembled using the next-generation sequencing method and reference-guided assembly. We described the bioinformatic characteristics of the plastomes, such as gene content, codon usage, RNA editing sites, repeat sequences, and positive selection. We also compared the synteny of the plastid genome sequences across the family to investigate their plastid genome structural variation and gene order. Finally, combining all the currently available plastome sequences of Ranunculeae species in GenBank, we reconstructed the phylogenetic framework to assess the potential value of the plastome sequences across the tribe. The aims of this study are to: understand the variation of the plastomes across Ranunculus and its close allies, to compare the plastome structures (gene order and arrangement) of Ranunculeae with those of the other genera of Ranunculaceae, and to advance the phylogenetic and evolutionary understanding of Ranunculeae.

2. Materials and Methods

2.1. Plant Sampling and Next-Generation Sequencing

Leaf samples of eleven new accessions representing three genera (Ceratocephala, Halerpestes, and Ranunculus) and nine species of tribe Ranunculeae were collected from field (

Table 1. Sample information pertaining to the present study.

Tribe	Species	Collecting Site	Voucher Number	GenBank No.
Ranunculeae	Ceratocephala testiculata *	Altay, Xinjiang, China	L. Xie 2016S3 (BJFC)	OR625574
Ranunculeae	Ce. testiculata *	Altay, Xinjiang, China	L. Xie 2016S46 (BJFC)	OR625575
Ranunculeae	Ranunculus monophyllus *	Altay, Xinjiang, China	L. Xie 2016S2 (BJFC)	OR625578
Ranunculeae	R. polyrhizos *	Altay, Xinjiang, China	L. Xie 2016S47 (BJFC)	OR625579
Ranunculeae	R. tanguticus *	Daocheng, Sichuan, China	W.H. Li WH072 (BJFC)	OR625580
Ranunculeae	R. mongolicus *	Qinghe, Xinjiang, China	C. Shang et al. I-4186 (BJFC)	OR625576
Ranunculeae	R. trichophyllus *	Tingri, Xizang, China	W.H. Li DR008 (BJFC)	OR625577
Ranunculeae	R. bungei *	Xinglong, Hebei, China	L. Xie et al. PL002 (BJFC)	OR625572
Ranunculeae	R. trichophyllus *	Xiangrila, Yunnan, China	L. Xie et al. T-20220808 016 (BJFC)	OR625582
Ranunculeae	R. pekinense *	Yanqing, Beijing, China	L. Xie and Y.K. Luo 20200916001 (BJFC)	OR625573
Ranunculeae	Halerpestes tricuspis *	Zhongba, Xizang, China	W.H. Li ZB006 (BJFC)	OR625581
Anemoneae	Anemoclema glaucifolium			MH205609
Anemoneae	A. tomentosa			NC_039451
Anemoneae	Pulsatilla chinensis			NC_039452
Anemoneae	A. trullifolia			MH205608
Anemoneae	Clematis brevicaudata			MT796620
Anemoneae	Cl. terniflora			KJ956785
Adonideae	Adonis coerulea			MK253469
Delphinieae	Aconitum barbatum			MK253470
Ranunculeae	Ce. falcata			MK253464
Ranunculeae	H. sarmentosa			MK253457
Ranunculeae	Oxygraphis glacialis			MK253453
Ranunculeae	R. austro-oreganus			KX639503
Ranunculeae	R. cantoniensis			NC_045920
Ranunculeae	R. cassubicifolius			OP250948
Ranunculeae	R. chinensis			ON500677
Ranunculeae	R. japonicus			MZ169045
Ranunculeae	R. macranthus			NC_008796
Ranunculeae	R. macranthus			DQ359689
Ranunculeae	R. membranaceus			NC_065303
Ranunculeae	R. occidentalis			NC_031651
Ranunculeae	R. reptans			NC_036977
Ranunculeae	R. sceleratus			MK253452
Ranunculeae	R. silerifolius			ON462450
Ranunculeae	R. ternatus			OQ943173
Ranunculeae	R. yunnanensis			MZ703201

* newly sequenced in this study.

). The identification of the specimens was conducted by LX (Lei Xie) and all the vouchers were deposited in the herbarium of Beijing Forestry University (BJFC). In addition, we retrieved all the available complete plastome sequences of tribe Ranunculeae as well as plastomes of its allies in Ranunculaceae from GenBank for comparative and phylogenetic analyses. In total, 11 genera and 33 species (36 samples) of Ranunculaceae (Table 1) were included for different analyses (see below in detail).

For each new sample, about 50 mg of dried leaf tissue was ground for DNA extraction. We used DNA extraction kits (Tiangen Biotech Co., Ltd., Beijing, China) to obtain total genomic DNA. Extracted DNAs were checked by 1.0% agarose gel electrophoresis and then were sent to BerryGenomics (Beijing, China) for library construction and next-generation sequencing (NGS). NGS was run on the Illumina NovaSeq 6000 platform (Illumina Inc., San Diego, CA, USA) to generate paired-end reads of 2 × 150 bp.

2.2. Plastid Genome Assembling and Annotating

After obtaining raw reads, we used the FASTX Toolkit (http://hannonlab.cshl.edu/fastx_toolkit, accessed on 18 June 2022) to remove the adaptors and low-quality reads. The plastid genome sequences were then de novo assembled according to our previous study [17]. GetOrganelle (https://github.com/Kinggerm/GetOrganelle, accessed on 15 July 2022) was used with SPAdes 3.10.1 as the assembler [20]. Contigs were connected into larger ones using RepeatFinder option in Geneious v. Prime [21], and when necessary, the gaps were bridged using 100 replicates of Fine Tuning in Geneious Prime [21] to generate complete plastome sequences. The gaps and junctions between IRs and LSC/SSC regions were further verified by Sanger sequencing PCR amplifications. The assembled plastome sequences were then annotated using the Plastid Genome Annotator [22]. Plastid genome circles were drawn using the Organellar Genome DRAW v. 1.3.1 [23].

2.3. Comparative Analyses of the Plastomes

The newly sequenced plastomes were aligned and compared with those of the previously published Ranunculaceae species. Geneious Prime [21] and CodonW v. 1.4.2 [24] were used to calculate amino acid frequency and codon usage for the new samples. We checked putative RNA editing sites in protein-coding genes by the PREP-cp suite [25] for the new samples. The plastome sequences across Ranunculaceae were aligned using mVISTA [26] for the synteny analysis. We used LAGAN and Shuffle-LAGAN modes with default parameters to detect possible plastome structural variation. IR expansion/contraction of all the available Ranunculeae samples were checked using IRscope [27]. The nucleotide variability (Pi) of the plastomes of both Ranunculeae and Ranunculus (which have the most species) were calculated using a sliding window analysis implemented in DnaSP v. 5 [28].

We searched plastid microsatellites by using software MIcroSAtellite (MISA) [29] with a minimum threshold of ten nucleotides for mononucleotide repeats, five for di-, four for tri-, and three for tetra-, penta-, and hexanucleotide repeats according to our previous study [17]. We also searched forward (F), reverse (R), complement (C), and palindromic (P) oligonucleotide repeats using the REPuter program [30] with a minimum repeat size of 30 bp and similarity value of 90%.

2.4. Positive Selection Analysis

All the available Ranunculeae samples (25 species, 28 samples) and outgroups (four species from Trib. Anemoneae) were used for CDS extraction using Geneious Prime [21]. The program CODEML implemented in PAML v. 4.10.6 package [31,32] was applied for the positive selection site analysis. We estimated a single dN:dS ratio (ω) of the entire alignment for the null model. Then, the branch model (model = 2; NSsites = 0) was used to estimate a single ω of all the lineages of tribe Ranunculeae as the foreground, and a different ω of the lineages from the outgroups tribe Anemoneae as the background. Finally, a chi-square distribution was applied to assess the significance of the results. On the other hand, the Bayes Empirical Bayes (BEB) method was also applied to identify specific amino acid sites in genes to calculate posterior probability values (PP). High PP values (P > 0.9) of the codon sites were considered to be positive selection sites [33,34]. According to previous studies, we take the genes with a p-value < 0.05 and at least one positively selected site with high PP values as a positive selection gene [35].

2.5. Phylogenetic Analysis

The phylogenetic framework was reconstructed for all the available species (28 samples representing 25 species) of tribe Ranunculeae. Previous studies showed that tribe Anemoneae is sister to Ranunculeae in the family [16,17,19], so we chose four samples from Anemoneae as the outgroups. For phylogenetic tree reconstruction, the IRa region was excluded from the analysis. Inversion and translocation regions in tribe Anemoneae were manually adjusted. To investigate potential differences in phylogenetic reconstruction using different partitions, we divided the complete plastome sequences under the following partition strategies. The complete dataset was first separated into coding regions (CDS), intergenic spacer regions (IGS), and introns. Each dataset was further separated by their positions: LSC, SSC, and IR, respectively. We ultimately obtained 13 datasets for phylogenetic analysis. They are the complete plastome, the complete CDS sequence, the complete IGS, the complete intron, the LSC-CDS, the LSC-IGS, the LSC intron, the SSC-CDS, the SSC-IGS, the SSC-intron, the IR-CDS, the IR-IGS, and the IR-intron datasets. Multiple alignments for all the datasets were conducted by MAFFT v. 6.833 [36]. We removed ambiguous alignments using a Python script written in our previous study [17].

For each dataset, the maximum likelihood (ML) and the Bayesian inference (BI) methods were applied for phylogenetic reconstruction. Substitution models and data partitions of the complete plastome dataset were tested by PartitionFinder v2.1.1 [37]. We tested six partitioning schemes for the complete plastome dataset according to previous studies [38]. They are (1) no partitions, (2) by coding and non-coding regions, (3) by positions of LSC, SSC, and IRs, (4) by genes for the CDS and non-coding region as a separate partition, (5) by genes and codon positions for the CDS and non-coding region as separate partition, (6) by the third codon position for the coding region. The Bayesian information criterion (BIC) was applied to assess the best partitioning scheme. The ML analysis was carried out using RAxML v.8.1.17 [39] with the GTR + G model recommended in the user’s manual. We run 500 replicates of resampling analysis to obtain the ML bootstrap support values. The BI analysis was conducted using MrBayes v3.2.3 [40], with the default priors for tree search. Two Markov chain Monte Carlo (MCMC) chains, each with three heated and one cold chain, were independently run for 2,000,000 generations with tree sampling every 100 generations. The first 25% of the trees were discarded as burn-in, and the remaining 75% of trees were then summarized to yield the Bayesian consensus phylogram.

3. Results

3.1. Plastome Characterization of Ranunculeae Genera and Species

We obtained up to 12 Gb raw NGS data to assemble the plastid genome sequences. By using reference sequences, we filtered out 412,658–651,511 plastid reads from the raw reads for plastome assembly, which was 391 to 626 × coverage of the plastid genome of Ranunculeae. When assembling, we successfully bridged gaps by our previous method [17], and those gaps and IR/SC boundaries were confirmed by PCR amplification. All the newly assembled plastome sequences were deposited in the public online database GenBank under accession numbers from OR625572 to OR625582 (Table 1).

The length of all the newly assembled plastome sequences of Ranunculeae ranged from 150,820 bp (C. testiculata) to 158,344 bp (H. tricuspis) with the overall GC content of 36.7 to 37.4% (Figure 1; Supplementary Table S1). Within the genus Ranunculus, the length of plastome sequences ranged from 155,973 bp (R. monophyllus) to 158,314 bp (R. trichophyllus), with the overall GC content of 36.7 to 36.8%. In Ranunculeae samples, all the plastome sequences contained a LSC (83,575–86,441 bp), an SSC region (17,619–21,735 bp), and a pair of IRs (24,168–27,868 bp) regions and showed a typical structure in Angiosperms. A set of 112 genes were present in the plastomes of Ranunculeae samples, among which 78 are protein-coding genes, 30 are transfer RNAs, and 4 are ribosomal RNA genes (

Table 2. Genes present in the plastid genomes of the 11 newly sequenced Ranunculeae samples.

Gene Type	Gene Name
Ribosomal RNA genes	16S rRNA	23S rRNA	4.5S rRNA	5S rRNA
Transfer RNA genes	trnA-UGC gene	trnC-GCA gene	trnD-GUC gene	trnE-UUC gene	trnF-GAA gene
	trnfM-CAU gene	trnG-GCC gene	trnG-UCC gene	trnH-GUG gene	trnI-CAU gene
	trnI-GAU gene	trnK-UUU gene	trnL-CAA gene	trnL-UAA gene	trnL-UAG gene
	trnM-CAU gene	trnN-GUU gene	trnP-UGG gene	trnQ-UUG gene	trnR-ACG gene
	trnR-UCU gene	trnS-GCU gene	trnS-GGA gene	trnS-UGA gene	trnT-GGU gene
	trnT-UGU gene	trnV-UAC gene	trnV-GAC gene	trnW-CCA gene	trnY-GUA gene
Small subunit of the ribosome	rps2 gene	rps3 gene	rps4 gene	rps7 gene	rps8 gene
	rps11 gene	rps12 gene	rps14 gene	rps15 gene	rps16 gene
	rps18 gene	rps19 gene
The large subunit of the ribosome	rpl2 gene	rpl14 gene	rpl16 gene	rpl20 gene	rpl22 gene
	rpl23 gene	rpl32 gene	rpl33 gene	rpl36 gene
RNA polymerase subunits	rpoA gene	rpoB gene	rpoC1 gene	rpoC2 gene
NADH dehydrogenase	ndhA gene	ndhB gene	ndhC gene	ndhD gene	ndhE gene
	ndhF gene	ndhG gene	ndhH gene	ndhI gene	ndhJ gene
	ndhK gene
Photosystem I	psaA gene	psaB gene	psaC gene	psaI gene	psaJ gene
Cytochrome b/f complex	petA gene	petB gene	petD gene	petG gene	petL gene
	petN gene
ATP synthase	atpA gene	atpB gene	atpE gene	atpF gene	atpH gene
	atpI gene
Large subunit of rubisco	rbcL gene
Maturase	matK gene
Protease	clpP gene
Envelope membrane protein	cemA gene
Subunit of acetyl-CoA-carboxylase	accD gene
Photosystem II	psbA gene	psbB gene	psbC gene	psbD gene	psbE gene
	psbF gene	psbH gene	psbI gene	psbJ gene	psbK gene
	psbL gene	psbM gene	psbN gene	psbT gene	psbZ gene
Copper chaperone for superoxide dismutase	ccsA gene
Conserved open reading frames	Ycf 1,2,3,4

). A total of 16 (Ceratocephala samples) and 17 (other newly sequenced samples) genes were located in a single IR region. A total of 18 (in Ranunculus and Halerpestes samples) and 17 (in Ceratocephala samples) genes have introns (Supplementary Table S2). In Ranunculus and Halerpestes samples, the longest intron is in the clpP gene (1497 bp in R. polyrhizos −1562 bp in H. tricuspis), whereas in Ceratocephala samples, the longest intron is in the ycf3 gene (1442 bp).

3.2. Comparative Results of the Plastomes

Multiple alignments using mVISTA were carried out for Ranunculeae samples to investigate plastid genome structural variations. Species with both normal and specific (in Adonis and tribe Anemoneae) plastome structures were also included. Two methods, LAGAN and Shuffle-LAGAN, were conducted and shown in Figure 2. When using the LAGAN method, Ranunculeae plastomes showed the same gene order as that of the Aconitum samples, but large empty (mismatch) regions were found in the LSC regions of Adonis and tribe Anemoneae samples due to gene inversion or gene translocation events.

Because the IR expansion/contraction may carry important phylogenetic information in Ranunculaceae [17], the IR/SC boundaries of the newly sequenced plastomes were compared with other outgroups in the family. The newly sequenced Ranunculus and Halerpestes samples as well as the published Oxygraphis sample in Ranunculeae have 17 genes in their IR region, which is the same as many other genera in Ranunculaceae (such as Aconitum L., Caltha, L., Coptis Salisb, Delphinium L., and Thalictrum L.) and other angiosperm taxa such as Amborella Baill. and Arabidopsis Heynh. [17,34]. Therefore, this 17-gene IR region of Ranunculus and Halerpestes can be taken as the primitive type in Ranunculaceae [17]. Whereas the IR regions of Ceratocephala samples showed slight contraction with incomplete rpl2 genes on the LSC/IR borders compared to the Ranunculus and Halerpestes samples (Figure 3).

Nucleotide variability was assessed by sliding window analysis, and the results (Figure 4) showed that the IR region has a lower variability than the SC regions in Ranunculus samples. When taking all the Ranunculeae samples into consideration, the trend of lower nucleotide variability in the IR region is also obvious. In Ranunculus samples, our result discovered extremely high variations at the border of the IR/SSC regions.

3.3. Synonymous Codon Usage

This study calculated the relative synonymous codon usage (RSCU) for the newly assembled plastome sequences using all the protein-coding genes. We presented results of amino acid frequency and putative RNA editing sites in Figure 5 and Supplementary Tables S2 and S3. We detected 95 putative RNA editing sites in the 24 protein-coding genes of Ceratocephala, 92 sites in 27 protein-coding genes of Halerpestes, 93 sites in 27 protein-coding genes of R. bungei and R. pekinensis, and 91 sites in 27 protein-coding genes of the other five Ranunculus species. In the Ranunculus samples, ndhF has the most RNA editing sites (10 and 11 sites), and the second was matK (9 sites). In the Ceratocephala samples, rpoC2 gene has the most RNA editing sites (12 sites), and the second was ndhF (11 sites). For the Halerpestes samples, both rpoC2 and ndhF genes have the most RNA editing sites (10 sites), and then was ndhB gene (8 sites).

The substitution from serine to leucine was tested to be the most common type (30.1%) in R. bungei and R. pekinensis, followed by serine to phenylalanine (15.1%), whereas in the other Ranunculus species, serine to leucine was the most common one (29.7%), followed by threonine to isoleucine (14.3%). In Ceratocephala, substitution from serine to leucine accounted for 26.3% of the editing sites, and from serine to phenylalanine was 13.7%. In Halerpestes, 31.5% of editing sites substituted from serine to leucine, and 15.2% from serine to phenylalanine, and among all its RNA editing sites, 23 substitutions appeared at the first nucleotide positions while 71 substitutions occurred at the second nucleotide position. Plastomes of the other two genera showed similar results in the substitution site on the codon positions (Supplementary Table S2).

3.4. SSR, Repetitive Sequences and Positive Selection Analysis

Rich SSRs including mononucleotide to hexonucleotide repeats were detected ranging from 47 to 70 in the newly sequenced plastomes (Supplementary Table S4). Among all the tested species, C. testiculata has the fewest SSRs, whereas H. tricuspis has the most. The most common SSR is mononucleotide repeat (A/T) among the nine species. For the tested species, the least proportion (53.2%) of the mononucleotide repeats was in C. testiculata, whereas the highest proportion (70.0%) was in H. tricuspis. The rare mononucleotide repeat (G/C) was only found in R. mongolicus, R. monophyllus, and R. trichophyllus. The second most common SSR is dinucleotide repeat (AT/TA) with six, eight, and nine replicates, respectively. The third most common SSR is tetranucleotide repeat (AATG/TGAA) with six, seven, nine, and ten replicates, respectively, and its total number was slightly smaller than the dinucleotide repeats. The fourth most common SSR is trinucleotide repeat (AAT/TTA), whereas pentanucleotide repeats were present in all the tested samples but R. mongolicus, and hexanucleotide repeats were only present in the plastomes of H. tricuspis, R. bungei, and R. pekinense. Within the newly sequenced plastomes, the largest proportion of SSR loci were found in IGS, followed by CDS and Intron. Within the plastid genome circle, SSRs are most common in the LSC region, followed by SSC, and the least in IR regions (Supplementary Table S4).

The eleven newly sequenced plastomes had a total of 281 direct, reverse, palindromic, and complement repeats (Figure 6), which may serve as potential molecular markers for further population genetic studies. Direct repeat was tested to be the most common repeat type, which accounted for 54.8% of the total repeats. It was followed by palindromic repeat (38.8%), reverse repeat (5.3%), and complement repeat (1.1%). The only three complement repeats were found in R. pekinense, R. polyrhizos, R. tanguticus, respectively. Repeats were usually short with 30–49 bp in length. We also found several longer direct and reverse repeats up to 82 bp in some Ranunculus samples. The largest proportion of repeats was found in the IGS region (73%), followed by CDS (21%) and Intron (6%) (Figure 6).

Positive selection of 67 CDS was tested for all the available Ranunculeae samples and its close allies. The likelihood ratio analysis showed that p-values of most genes were >0.05 (insignificant), except that atpB, ndhC, ndhG, ndhJ, psaC, rps2, rps15, ycf2 (p < 0.05). Furthermore, the nonsynonymous/synonymous rate ratio (ω = dN/dS) of only one gene, accD, is >1, but its p-value is >0.05. However, the BEB test showed that accD, atpF, ccsA, ndhF, petD, rbcL, rpoA, rpoC2 and ycf2 have high posterior probability values (≥0.9) (Supplementary Table S5). Previous studies considered that a coding region with a high posterior probability value of the BEB analysis can be taken as a positive selection gene [35]. Under this measure, nine genes, accD, atpF, ccsA, ndhF, petD, rbcL, rpoA, rpoC2, and ycf2 can be considered as positive selection genes.

3.5. Partitioning and Phylogenetic Reconstruction Results

We tested the complete plastome dataset by using six data partitioning strategies. The results showed that those six partitioning treatments obtained quite different results, indicating that different partitioning methods may greatly affect phylogenetic reconstruction. It showed that the partitioning strategy by LSC, SSC, and IRs obtained the best results (

Table 3. Partitioning strategy tests for the complete plastid genome dataset using PartitionFinder.

Dataset	Partitioning Strategy	Parameters	Subsets	ln L	BIC
Complete Plastome Dataset	No partition	63	1	−492,025.19	984,804.90
	Coding and non-coding	74	2	−600,890.34	1,202,691.59
	LSC, SSC, IRs	85	3	−489,387.16	979,792.32
	By gene	162	11	−710,387.28	1,422,765.77
	By gene and codon position	219	17	−708,937.30	1,420,566.42
	By the third codon position	84	3	−802,253.60	1,605,550.34

). According to this result, this partitioning strategy was applied for CDS, IGS, introns, and complete datasets, and no partitioning strategy was applied for the rest of the datasets. The GTR model for each partition of CDS, IGS, introns, and complete datasets was applied for BI analysis, while the GTR + I + G model was used for the rest of the datasets for both ML and BI analyses according to our partition results.

Phylogenies reconstructed by both complete and separate datasets and both methods (ML and BI) were generally the same, especially in strongly supported clades (Figure 7, Supplementary Figure S2). The complete plastome dataset generated the most robustly resolved phylogeny of Ranunculeae. For this reason, our discussion was based on the phylogenetic framework inferred from the complete plastome dataset. All four tested genera of the tribe Ranunculeae were strongly supported. Two genera Halerpestes and Oxygraphis were sister groups and formed a clade sister to another clade including Ranunculus and Ceratocephala. Three major clades were resolved in the largest genera Ranunculus. Clade 1 includes species mainly from sect. Auricomus (Spach) Schur [15]. Clade 2 comprises aquatic species from sect. Batrachium DC. and a hydrophilus species R. sceleratus L from sect. Hecatonia. Clade 3 includes species mainly from sect. Flammula and sect. Acris Schur [10].

4. Discussion

Many previous studies have reported that most of the plastid genomes of Ranunculaceae have a set of 112 genes, and this is also the case in Ranunculeae species [17,19,35]. All our sequenced Ranunculeae samples also showed the same results as previous reports. Gene inversions and gene loss in Ranunculaceae plastomes were revealed more than 20 years ago. Johansson [41] studied Adonis species using the restriction site mapping method and found large inversions and gene loss presented in their plastid genomes. In recent years, structural variations including gene inversions, gene translocations, and IR variations have been explicitly reported in Ranunculaceae [17,19]. The plastome structural variations of tribe Anemoneae also have structural variations, but in comparison with Adonis, their variations have differed in many aspects. In genera Anemone s. l. and Anemoclema, there are three large inversions in the LSC region of their plastomes, while in Clematis there are two large inversions and one large transposition in the LSC region of the plastome [17]. In Ranunculeae species, gene orders of the plastome are the same as those of most other genera (such as Aconitum, Thalictrum, and so on), and no gene inversions and translocations are found.

The IR region normally has 17 genes, and this type of IR was considered to be primitive in Ranunculaceae [17]. However, IR expansion/contractions are also very common in Ranunculaceae. He et al. [17] reported that many genera (such as Asteropyrum, Anemone, Anemoclema, Clematis, Dichocarpum, Hepatica, Hydrastis, Naravelia, and Pulsatilla) in Ranunculaceae have expanded IR regions, whereas only two genera, Helleborus and Ceratocephala, were found to have slightly contracted IR regions. Up to 27 genes in Asteropyrum peltatum were found in the plastome of the family, and IR expansion/contractions may carry important phylogenetic information [17]. In tribe Ranunculeae, the majority of tested species have 17 genes in their IR regions except for Ceratocephala, which showed a little contraction in the IR regions (Supplementary Table S1). IR contraction is rare in Ranunculaceae, and only found in Helleborus and Ceratocephala [17]. In these two genera, rpl2 is not completely located in the IR region (Figure 3), and these two generic cases seemed to have no phylogenetic relationship in Ranunculaceae [17]. However, both species of Ceratocephala tested to have the same contracted IR regions, indicating that this IR contraction may be a synapomorphy in the genus Ceratocephala within the Ranunculeae clade. Our results showed that plastome structural variation is not characteristic of Ranunculeae, but IR expansion/contraction may have phylogenetic information.

Simple sequence repeats (SSRs) for microsatellites have been widely applied for population genetics and evolutionary studies of Angiosperm species [42]. However, the use of the plastid SSRs has not been fully developed in Ranunculaceae. Our results showed that 47 to 70 plastid SSRs are found in the 11 new samples (Supplementary Table S4), and pentanucleotide repeats are very common in the plastomes of Ranunculeae species. The rich plastid SSR diversity can provide opportunities for future population genetic studies on Ranunculeae species.

In Ranunculaceae, the tribe Ranunculeae is characterized by its ascending unitegmic ovules (except Myosurus which has pendent ovules), often smaller sepals and larger petals, and petals with one or more nectary glands near the base [10]. Some taxonomists included Callianthemum and Adonis into Ranunculeae [43,44]. However, this treatment was not supported by molecular phylogenetic analysis [45,46]. A large number of molecular phylogenetic studies of Ranunculeae have been published [11,12,13,47,48,49] which helped us understand the delimitation and generic relationship of this tribe. Based on molecular phylogeny and comprehensive sampling, 19 genera were recognized within the tribe Ranunculeae [11,12,13]. However, most of them were based on small numbers of DNA regions (nrITS and plastid DNA fragments), and the phylogenetic relationship within the tribe was still not robustly resolved. In this study, the generic relationship of the tribe inferred from the complete plastome data was congruent with previous studies and more robustly resolved (Figure 7 and Supplementary Figure S2), therefore demonstrating that plastome data may provide the opportunity for the reconstruction of generic phylogeny of Ranunculeae in the future with comprehensive sampling. Our current sampling covered all the generic representatives in China. The results showed the aquatic sect. Batrachium should be included in Ranunculus but as a distinct genus. Generic statuses of Ceratocephala, Halerpestes, and Oxygraphis can be kept.

Ranunculus is the largest genus in both Ranunculeae and Ranunculaceae with about 650 species around the world [10]. Taxonomy of Ranunculus has been considered extremely difficult and there are considerable differences among different classifications [9,15,50,51,52,53]. For this reason, this genus also attracted great attention in its phylogeny using molecular markers [47,54,55,56,57,58,59]. Based on a comprehensive sampling and nrITS and three plastid DNA regions, Emadzade et al. [57] resolved nine major clades in Ranunculus. Although we combined all the available complete plastome sequences of Ranunculus from GenBank, the sampling is still limited. Three major clades were robustly resolved by our plastome data. Clade 1 (Figure 6) corresponded to clade IV of Emadzade et al. [57] which included species of sect. Auricomus. Clade 1 also included R. ternatus from the sect. Tuberifer whose phylogenetic position has never been tested. Sect. Tuberifer is characterized by its tuberous roots. Wang [60] considered that R. ternatus may be closely related to sect. Auricomus and this prediction was supported by our phylogenetic analysis. Clade 2 included an aquatic sect. Batrachium and sect. Hectonia in wetland, and well corresponded to cluster III of Emadzade et al. [57]. The monophyly of cluster III in Emadzade et al. [57] was not supported in their study. However, clade 2 is fully supported showing the advantage of using the plastome data for phylogenetic reconstruction over the small number of DNA regions by Sanger sequencing. Clade 3 was also fully supported. In this clade, the first diverged R. reptans was in clade V of Emadzade et al. [57], and the other two subclades (R. japonicus–R. occidentalis) and (R. cantoniensis–R. chinensis) correspond with clade VI and Clade VIII of Emadzade et al. [57], respectively. Phylogenetic position of R. macranthus has never been inferred, and this species is also nested in clade 3. In general, the phylogenetic relationship within Ranunculus inferred by the complete plastome sequences was fully congruent with previous molecular studies and showed advantages of high resolution. Plastid phylogenomic analysis is needed for future studies with a comprehensive sampling.

5. Conclusions

The complete plastomes of eleven samples representing nine species of tribe Ranunculeae were de novo assembled using a next-generation sequencing method. The plastome sequences from all the Ranunculaceae samples and their allies were compared in various aspects including gene content, nucleotide variability, codon usage, RNA editing sites, simple sequence repeats, and positive selection sites through bioinformatic analyses. The phylogeny of Ranunculeae was reconstructed for the complete and separated datasets using both ML and BI methods to infer generic and specific relationships within the tribe. Our results showed that the majority of the Ranunculeae genera and species have the most common plastid genome type, which is widely shared in the family [17], and there are potential values of the plastome sequences for reconstructing the phylogeny of both the tribe and the genus Ranunculus in future studies.