Plastid Phylogenomics Provide Evidence to Accept Two New Members of Ligusticopsis (Apiaceae, Angiosperms)

Peucedanum nanum and P. violaceum are recognized as members of the genus Peucedanum because of their dorsally compressed mericarps with slightly prominent dorsal ribs and narrowly winged lateral ribs. However, these species are not similar to other Peucedanum taxa but resemble Ligusticopsis in overall morphology. To check the taxonomic positions of P. nanum and P. violaceum, we sequenced their complete plastid genome (plastome) sequences and, together with eleven previously published Ligusticopsis plastomes, performed comprehensively comparative analyses. The thirteen plastomes were highly conserved and similar in structure, size, GC content, gene content and order, IR borders, and the patterns of codon bias, RNA editing, and simple sequence repeats (SSRs). Nevertheless, twelve mutation hotspots (matK, ndhC, rps15, rps8, ycf2, ccsA-ndhD, petN-psbM, psbA-trnK, rps2-rpoC2, rps4-trnT, trnH-psbA, and ycf2-trnL) were selected. Moreover, both the phylogenetic analyses based on plastomes and on nuclear ribosomal DNA internal transcribed spacer (ITS) sequences robustly supported that P. nanum and P. violaceum nested in Ligusticopsis, and this was further confirmed by the morphological evidence. Hence, transferring P. nanum and P. violaceum into Ligusticopsis genus is reasonable and convincing, and two new combinations are presented.


Introduction
Ligusticopsis Leute, a small flowering plant genus of Apiaceae, was established by Gerfried Horand Leute in 1969 with L. rechingeriana Leute as the type species, recognizable by conspicuous calyx teeth and strongly dorsally flattened mericarps with numerous vallecular vittae [1]. However, these characteristics can also be detected in some Ligusticum L. members [2]; therefore, the morphological delimitation between Ligusticopsis and Ligusticum is historically unclear. Furthermore, several Ligusticopsis species described by Leute do not have prominent calyx teeth [3], which further blurred the boundaries of this genus. Hence, the genus Ligusticopsis has been merged into the genus Ligusticum by some authors [2,[4][5][6]. However, Li et al. [7] recently confirmed that the genus Ligusticopsis is a natural unit based on molecular and morphological evidence, and confirmed the presence of nine "true Ligusticopsis species". Subsequently, the phylogenetic analyses based on plastome data performed by Ren et al. [8] also recovered Ligusticopsis as a monophyletic group and recognized two additional species. To date, the genus Ligusticopsis contains eleven validated species.
Peucedanum nanum R.H.Shan and M.L.Sheh and P. violaceum R.H.Shan and M.L.Sheh are species endemic to China, which grow on dry mountain slopes and in sparse forests or grassy places on riverbanks, respectively [9,10]. Both species are placed in Peucedanum L. owing to their dorsally compressed mericarps with slightly prominent dorsal ribs and narrowly winged lateral ribs [11]. However, P. nanum and P. violaceum are not similar to the type species of Peucedanum (P. officinale L.) [12] but resemble Ligusticopsis species in overall morphology ( Figure 1). Additionally, the genus Peucedanum is not monophyletic [13][14][15][16][17][18][19], and its taxonomy has faced extreme challenges. Therefore, the taxonomic positions of P. nanum and P. violaceum need to be re-evaluated.
overall morphology (Figure 1). Additionally, the genus Peucedanum is not monoph [13][14][15][16][17][18][19], and its taxonomy has faced extreme challenges. Therefore, the taxonomic tions of P. nanum and P. violaceum need to be re-evaluated. A robust molecular phylogenetic framework could provide valuable informati resolving the taxonomic positions of P. nanum and P. violaceum. Unfortunately, mole data for both species are limited, and these species have not been included in pre phylogenetic studies. Hence, it is necessary to identify molecular markers to inves the phylogenetic position of P. nanum and P. violaceum.
The plastid genome (plastome) sequence, possessing highly variable chara gives us the potential to obtain a robust phylogenetic framework at low taxonomic [20][21][22][23][24][25][26][27][28][29]. With the development of next-generation sequencing, plastome sequences been applied extensively and successfully to resolve the phylogenetic position of nomically difficult taxa [8,[30][31][32][33][34][35]. In this study, we sequenced and assembled the tomes of P. nanum and P. violaceum for the first time. Together with the previously lished eleven Ligusticopsis plastomes, we carried out comprehensively comparative yses to reveal the plastome features for P. nanum, P. violaceum, and Ligusticopsis sp Subsequently, we performed phylogenetic analyses based on the plastome data an clear ribosomal DNA internal transcribed spacer (ITS) sequences to investigate the p genetic positions of P. nanum and P. violaceum. Finally, by combining evidence fro A robust molecular phylogenetic framework could provide valuable information for resolving the taxonomic positions of P. nanum and P. violaceum. Unfortunately, molecular data for both species are limited, and these species have not been included in previous phylogenetic studies. Hence, it is necessary to identify molecular markers to investigate the phylogenetic position of P. nanum and P. violaceum.
The plastid genome (plastome) sequence, possessing highly variable characters, gives us the potential to obtain a robust phylogenetic framework at low taxonomic levels [20][21][22][23][24][25][26][27][28][29]. With the development of next-generation sequencing, plastome sequences have been applied extensively and successfully to resolve the phylogenetic position of taxonomically difficult taxa [8,[30][31][32][33][34][35]. In this study, we sequenced and assembled the plastomes of P. nanum and P. violaceum for the first time. Together with the previously published eleven Ligusticopsis plastomes, we carried out comprehensively comparative analyses to reveal the plastome features for P. nanum, P. violaceum, and Ligusticopsis species. Subsequently, we performed phylogenetic analyses based on the plastome data and nuclear ribosomal DNA internal transcribed spacer (ITS) sequences to investigate the phylogenetic positions of P. nanum and P. violaceum. Finally, by combining evidence from the comparative plastome analyses, molecular phylogeny, and morphology, taxonomic revisions for P. nanum and P. violaceum were conducted.

Plastome Features
Illumina sequencing obtained 31,564,816 and 33,651,636 paired-end clean reads for P. nanum and P. violaceum, respectively (Table S1); among these reads, 1,025,204 and 543,725 reads were mapped to the assemblies, respectively. Based on these data, two high-quality plastomes for P. nanum and P. violaceum were generated with 1022.351× and 536.171× coverage, respectively.
The plastome features of eleven Ligusticopsis taxa and two Peucedanum species were comprehensively investigated. The overall size ranged from 146,900 bp (P. nanum) to 148,633 bp (L. brachyloba (Franch.) Leute) in the thirteen plastomes (  Figure 2). The total GC content of the thirteen plastomes ranged from 37.3% to 37.5%, and 113 unique genes were identified, including 79 protein-coding genes, 30 tRNA genes, and four rRNA genes (Tables 1 and S2). The 79 protein-coding genes typically shared by the thirteen plastomes were extracted and connected for each species. These sequences were 67,566-67,896 bp in length and harbored 22,522-22,632 codons (Table S3). Among these codons, the least number of codons were used to encode the Cys, while the highest number of codons were used to encode the Leu. Additionally, the relative synonymous codon usage (RSCU) values of all codons ranged from 0.34 to 2.00 in the thirteen plastomes ( Figure 3). Among them, the RSCU values of 30 codons were greater than 1.00 in all plastomes. All of these codons ended with A/U, except for UUG. The 79 protein-coding genes typically shared by the thirteen plastomes were extracted and connected for each species. These sequences were 67,566-67,896 bp in length and harbored 22,522-22,632 codons (Table S3). Among these codons, the least number of codons were used to encode the Cys, while the highest number of codons were used to encode the Leu. Additionally, the relative synonymous codon usage (RSCU) values of all codons ranged from 0.34 to 2.00 in the thirteen plastomes ( Figure 3). Among them, the RSCU values of 30 codons were greater than 1.00 in all plastomes. All of these codons ended with A/U, except for UUG.  A total of 57-59 potential RNA editing sites were identified in the thirteen plastomes (Table S4). All detected RNA editing sites were Cytosine to Uracil (C-U) conversion, and most of them occurred in the second codon position (43)(44)(45) followed by the first codon position (14), but no site was located in the third codon position (Table S5). Moreover, the ndhB gene contained the highest number of RNA editing sites (10) in all plastomes (Table S6).
The total number of simple sequence repeats (SSRs) ranged from 67 to 84 among the thirteen plastomes ( Figure 4). Among these, mononucleotide repeats were the most abundant (34-43) followed by dinucleotides (17)(18)(19)(20)(21)(22)(23)(24). In addition, bases A and T were dominant for all the identified SSRs in the thirteen plastomes (Table S7).  A total of 57-59 potential RNA editing sites were identified in the thirteen plastomes (Table S4). All detected RNA editing sites were Cytosine to Uracil (C-U) conversion, and most of them occurred in the second codon position (43)(44)(45) followed by the first codon position (14), but no site was located in the third codon position (Table S5). Moreover, the ndhB gene contained the highest number of RNA editing sites (10) in all plastomes (Table S6).
The total number of simple sequence repeats (SSRs) ranged from 67 to 84 among the thirteen plastomes ( Figure 4). Among these, mononucleotide repeats were the most abundant (34-43) followed by dinucleotides (17)(18)(19)(20)(21)(22)(23)(24). In addition, bases A and T were dominant for all the identified SSRs in the thirteen plastomes (Table S7).  A total of 57-59 potential RNA editing sites were identified in the thirteen plastomes (Table S4). All detected RNA editing sites were Cytosine to Uracil (C-U) conversion, and most of them occurred in the second codon position (43)(44)(45) followed by the first codon position (14), but no site was located in the third codon position (Table S5). Moreover, the ndhB gene contained the highest number of RNA editing sites (10) in all plastomes (Table S6).

Plastome Comparison
The borders between the IR and SC among the thirteen plastomes were compared ( Figure 5). The junctions of IRa/LSC and IRb/LSC fell into the ycf 2 gene and intergenic region of trnL-trnH, respectively. The borders of IRb/SSC fell into the ycf 1 gene in all species, whereas the overlap between the ycf 1 gene and the ndhF gene in the IRa/SSC junctions was only detected in L. capillacea (H.Wolff) Leute.

Phylogenetic Analyses
The analyses of maximum likelihood (ML) and Bayesian inference (BI) based on the plastome data generated identical tree topologies. As shown in Figure 9, the eleven Ligusticopsis taxa, P. nanum, and P. violaceum Although phylogenetic analyses based on ITS sequences yielded topologies with low support and resolution, the results also indicated that the sister group of P. nanum and P. violaceum clustered with the Ligusticopsis species (PP = 1.00, BS = 97), and this clade was relatively distant from other Ligusticum taxa ( Figure S1).

Phylogenetic Analyses
The analyses of maximum likelihood (ML) and Bayesian inference (BI) based on the plastome data generated identical tree topologies. As shown in Figure 9, the eleven Ligusticopsis taxa, P. nanum, and P. violaceum Although phylogenetic analyses based on ITS sequences yielded topologies with low support and resolution, the results also indicated that the sister group of P. nanum and P. violaceum clustered with the Ligusticopsis species (PP = 1.00, BS = 97), and this clade was relatively distant from other Ligusticum taxa ( Figure S1).

Plastome Features
In this study, we conducted comprehensively comparative analyses for the plastomes of P. nanum, P. violaceum, and Ligusticopsis species. The thirteen plastomes showed typical quadripartite structures, including a pair of inverted repeat regions divided by a large single-copy region and a small single-copy region, which is the same as the other plastomes of Apiaceae [7,8,19,[36][37][38][39][40][41]. Although gene loss and rearrangement have been reported in the plastomes of Apiaceae [19,38,39], the gene content and order in the thirteen studied plastomes were identical. All these plastomes also shared similar genomic size, total GC content, and IR borders. Furthermore, the patterns of codon bias, RNA editing sites, and SSR were extremely similar and have also been detected in the plastomes of Ligusticum and Peucedanum within Apiaceae [19,42]. These results indicated that the thirteen plastomes were highly conserved. Meanwhile, the conserved and similar plastome

Plastome Features
In this study, we conducted comprehensively comparative analyses for the plastomes of P. nanum, P. violaceum, and Ligusticopsis species. The thirteen plastomes showed typical quadripartite structures, including a pair of inverted repeat regions divided by a large single-copy region and a small single-copy region, which is the same as the other plastomes of Apiaceae [7,8,19,[36][37][38][39][40][41]. Although gene loss and rearrangement have been reported in the plastomes of Apiaceae [19,38,39], the gene content and order in the thirteen studied plastomes were identical. All these plastomes also shared similar genomic size, total GC content, and IR borders. Furthermore, the patterns of codon bias, RNA editing sites, and SSR were extremely similar and have also been detected in the plastomes of Ligusticum and Peucedanum within Apiaceae [19,42]. These results indicated that the thirteen plastomes were highly conserved. Meanwhile, the conserved and similar plastome characters among P. nanum, P. violaceum, and Ligusticopsis species also implied that P. nanum and P. violaceum may be members of Ligusticopsis.

Phylogenetic Inference
Since the establishment of the genus Ligusticopsis, its taxonomy has been controversial. Pu [2], Pu and Watson [5], Zhang [4], and Wang et al. [54] did not recognize Ligusticopsis as a distinct genus but merged it into Ligusticum based on morphological characteristics. However, based on carpoanatomical evidence, Pimenov et al. [55] accepted the establishment of Ligusticopsis. Subsequently, Pimenov [56] recognized 18 Ligusticopsis species in his checklist of Chinese Umbelliferae based on reviews of the type specimens and morphological evidence. Recently, plastome phylogenetic analyses performed by Li et al. [7] and Ren et al. [8] robustly confirmed the monophyly of Ligusticopsis, although limited samples of Ligusticopsis and Ligusticum were used in both studies. In the present study, twelve Ligusticum species and eleven Ligusticopsis taxa were included in the phylogenetic analyses. Both the phylogenies based on plastome data and on ITS sequences revealed that eleven Ligusticopsis species clustered as a clade and belonged to the Selineae tribe. Although the type species of Ligusticum (Ligusticum scoticum L.) was absent in our analyses, the phylogenetic position of this species, located in the Acronema Clade, was revealed by a previous study [18], which was obviously distant from the clade formed by the Ligusticopsis species. Our results with more extensive taxa sampling provided additional evidence to accept Ligusticopsis as a distinct genus.
Additionally, all our phylogenetic analyses based on plastome data and ITS sequences robustly supported that P. nanum and P. violaceum nested within Ligusticopsis. The type species of Peucedanum (P. officinale) was not included in our analyses; however, a previous study revealed its phylogenetic location was distant from Ligusticopsis [18]. These results implied that P. nanum and P. violaceum were distant from P. officinale but closely related to Ligusticopsis. Furthermore, the affinity between both species and Ligusticopsis was supported by the high similarity of their plastome sequences and also supported by the shared morphological features: stem base clothed in fibrous remnant sheaths, conspicuous calyx teeth, and strongly compressed dorsally mericarps with slightly prominent dorsal ribs, winged lateral ribs, and numerous vittae in the commissure and in each furrow [7,9,10]. However, hispid mericarps can easily distinguish P. nanum and P. violaceum from the glabrous mericarps of other Ligusticopsis species [7]. Moreover, P. nanum has densely hispid mericarps with slightly prominent dorsal ribs and six vittae in the commissure, whereas sparsely hispid mericarps with filiform dorsal ribs and eight vittae in the commissure are observed in P. violaceum [9,10]. Therefore, we could reasonably transfer P. nanum and P. violaceum into Ligusticopsis as two new members of this genus.
The sister relationship between P. nanum and P. violaceum was strongly supported by both the phylogenetic analyses based on plastome data and ITS sequences. The hispid mericarps shared by both species could further support this relationship [9,10]. Unfortunately, the relationship between this sister group and other Ligusticopsis species was not clearly resolved in our phylogenetic analyses. To confirm the phylogenetic position of the sister group of P. nanum and P. violaceum within Ligusticopsis, additional molecular sequences such as additional nuclear DNA fragments are required in future studies.

Plant Sample, DNA Extraction, Sequencing, and Assembly
Fresh leaves of P. nanum and P. violaceum were collected from their type localities and dried with silica gel. Voucher specimens were deposited in the herbarium of Sichuan University (Chengdu, China) (Table S1). Genomic DNA was extracted from the silica-geldried leaves using the modified CTAB method [57] and then fragmented into 400 bp to create a pair-end library according to the manufacturer's protocol (Illumina, San Diego, CA, USA). Subsequently, the libraries were sequenced on the Illumina NovaSeq platform at Personalbio (Shanghai, China). The raw data yielded by Illumina sequencing were filtered with fastP v0.15.0 [58] to obtain high-quality reads with -n 10 and -q 15. The plastomes were then assembled based on the high-quality reads using NOVOPlasty v2.6.2 [59] with the default parameters and rbcL sequence extracted from the plastome of L. rechingeriana (MZ491175) as the seed. In addition, the ITS sequences were assembled using the GetOrganelle pipeline [60] with the ITS sequence of L. rechingeriana (MZ497220) as the reference.
Eleven plastomes of Ligusticopsis, which we have previously reported, were downloaded from the NCBI database (Table S8). In conjunction with two newly sequenced plastomes, codon usage of the thirteen plastomes was detected using CodonW v1.4.2 (Nottingham, UK). Subsequently, the potential RNA editing sites of the protein-coding genes for the thirteen plastomes were predicted using the online program Predictive RNA Editor for Plants suite with a cutoff value of 0.8 [64]. We also detected simple sequence repeats (SSRs) in the thirteen plastomes using MISA (http://pgrc.ipk-gatersleben.de/misa/, accessed on 11 September 2022). The minimum number of repeat units for mono-, di-, tri-, tetra-, penta-, and hexa-nucleotides was set to 10, 5, 4, 3, 3, and 3, respectively.

Phylogenetic Analyses
To resolve the phylogenetic positions of P. nanum and P. violaceum, 48 plastomes and 48 ITS sequences were used to reconstruct the phylogenetic tree (Tables S1 and S8). Among them, Chamaesium mallaeanum Farille and S.B.Malla and Chamaesium viridiflorum (Franch.) H.Wolff ex R.H.Shan were chosen as the outgroup based on a previous study [39]. The two datasets were aligned using MAFFT v7.221 [67]. Alignments were used for the maximum-likelihood analyses (ML) and Bayesian inference (BI). For the ML analyses, RAxML v8.2.8 [69] was used to reconstruct the phylogenetic tree with 1000 replicates and the GTRGAMMA model as suggested by the RAxML manual. The BI analyses were performed using MrBayes v3.2.7 [70]. The best-fit substitution models for plastome data (TVM+I+G) and ITS data (SYM+I+G) were tested using Modeltest v3.7 [71]. Two independent Markov chains were run for 1,000,000 generations with sampling every 100 generations and discarding the first 25% of the trees as burn-in.

Conclusions
The whole plastomes of P. nanum and P. violaceum were reported for the first time in the present study. The plastome comparisons among P. nanum, P. violaceum, and eleven Ligusticopsis species revealed that these plastomes were highly conserved and similar in terms of structure, size, GC content, gene content and order, IR borders, and the patterns of codon bias, RNA editing, and SSR. Nevertheless, 12 mutation hotspot regions (matK, ndhC, rps15, rps8, ycf 2, ccsA-ndhD, petN-psbM, psbA-trnK, rps2-rpoC2, rps4-trnT, trnH-psbA, ycf 2-trnL) were identified, which could serve as potential DNA markers for species identification and phylogenetic analysis of Ligusticopsis. Moreover, the phylogenetic analyses based on plastome data and ITS sequences robustly supported that P. nanum and P. violaceum nested in the genus Ligusticopsis. Considering also the morphological affinities, we transferred P. nanum and P. violaceum into Ligusticopsis and proposed two new combinations.  Data Availability Statement: Plastomes and ITS sequences of P. nanum and P. violaceum generated in the current study are available at the NCBI database (https://www.ncbi.nlm.nih.gov, accessed on 19 December 2022).

Conflicts of Interest:
The authors declare no conflict of interest.