Comparative and Adaptive Analyses of the Complete Chloroplast Genome Diversity in Sium serra Authors: SeongJun Park, SeonJoo Park Published: 2024 ABSTRACT Background/Objectives: Sium serra is distributed in Korea, China, and Japan. It was first identified as the genus Pimpinella and then reclassified as Sium by Kitagawa. Some Sium species are used as herbal medicine and are often confused with the similar form Ligusticum sinense. In this study, we analyzed the cp genome of S. serra and conducted comparative analyses with the cp genomes of related taxa. Methods: We extracted gDNA from fresh leaves and sequenced it using Illumina HiSeq2500. For the chloroplast genome assembly, de novo assembly was performed using Velvet v1.2.07. For the annotation, GeSeq and NCBI BLASTN were used. Afterwards, related taxa were analyzed using programs such as DnaSP and MISA. Results: S. serra was excluded from the study on the chloroplast (cp) genome in Sium because it was classified as Pimpinella in China. Therefore, this study aimed to analyze the cp genome of S. serra for the first time and its location within the genus Sium. The complete cp genome of S. serra was 154,755 bp in length, including a pair of inverted repeats, each 26,255 bp, a large single-copy region of 84,581 bp, and a small single-copy region of 17,664 bp. The cp genome comprised 79 protein-coding, 30 tRNA, and 4 rRNA genes. Furthermore, six regions of high nucleotide diversity were identified in the genus Sium. In the genus Sium, 1630 repeats that can serve as markers were also identified. Eight protein-coding genes with high KA/KS values were under positive selection in the Sium. Our phylogenetic analyses suggest that S. serra was positioned with high bootstrap support within the Sium of the tribe Oenantheae, specifically in the southern Palearctic subclade. Conclusions: In this study, the S. serra chloroplast genome was sequenced and assembled. The genus Sium formed a monophyletic group; however, as not all the Sium species were included in this study, further research is necessary. This study can serve as foundational data not only for Sium but also for the tribe Oenantheae. 1. Introduction According to the APG Ⅳ classification system, the genus Sium L. is classified within the tribe Oenantheae, in the family Apiaceae and subfamily Apioideae [1]. There are approximately 11 taxa of the genus Sium, which is primarily native to Eurasia and North America. Sium is a perennial herb that thrives in shady and wet environments [2,3]. Sium is characterized by its hairless appearance and its leaves, which are trifoliate or pinnate. In addition, it features compound inflorescences, and its fruit is ovate with distinct primary ribs and milk ducts [4,5]. The genus Sium is closely related to the genera Berula W.D.J. Koch and Pimpinella L. Some species of Sium have been reclassified into the genus Berula (B. burchellii, B. bracteata, B. repanda) [6,7]. Pimpinella serra has been reclassified into the genus Sium. However, in Flora of China, S. serra is still recognized as a synonym for Pimpinella serra [5]. In East Asia, some Sium species are employed in herbal medicine, primarily using their roots. These species are commonly used to treat headaches and high blood pressure. They contain numerous essential oils and chemical compounds. However, their dried roots are similar to those of “Chinese Gao ben” (Ligusticum sinense) and are often mistaken for and used interchangeably with it [8]. S. serra (Franch. & Sav.) Kitag. is known to be distributed in Korea, China, and Japan [4,5,9]. This species was first identified as P. serra by Franchet and Savatier in 1876 [10]. In 1941, Kitagawa reclassified the species into the genus Sium [11]. In Korea, it was first discovered by Lee and Ko in 2009 and named S. ternifoilum [12]. Later, in 2014, Chang renamed it S. serrum, which is now recognized as S. serra [13]. Plants, unlike other organisms, possess organelles called chloroplasts, which are responsible for photosynthesis and various metabolic processes [14,15]. The chloroplast (cp) genome, typically inherited maternally, ranges in size from 120 to 180 kb. The cp genome exists in a circular form and is divided between large single copy (LSC) and small single copy (SSC) by inverted repeats (IRs), which usually occur in two copies [16,17,18]. The cp genome contains approximately 79 protein-coding, 30 transfer RNA, and 4 ribosomal RNA genes [19]. This genome has a relatively slow mutation rate and high polymorphism between species, making it widely used in plant genetic identification and phylogenetic analysis [20]. In this study, the complete chloroplast genome of S. serra was constructed and annotated through de novo next-generation sequencing (NGS) and assembly. Therefore, this study aims to examine the complexity of the Sium cp genome structure, including the gene content, repeats, nucleotide diversity, and adaptive evolution. 2. Materials and Methods 2.1. Genomic DNA Isolation and Genome Sequencing Fresh leaves of S. serra were collected from Wonju-Si, Gangwon State, Republic of Korea. Genomic DNA was isolated from the leaves using the Exgene Plant SV kit (GeneAll Biotechnology Inc., Seoul, Republic of Korea). The quality of the gDNA was measured using a Nano-200 spectrophotometer (Hangzhou Allsheng Instruments Inc., Hangzhou, China), and the quantity was verified using 1% agarose gel. NGS was conducted using the Illumina HiSeq2500, provided by Phyzen Ltd., Seongnam, Republic of Korea. The paired-end library (2 × 150 bp) was constructed using the Illumina platform. Sequencing was performed on paired reads with a 550 bp insert size, resulting in ~3 GB of raw data. Low-quality reads were identified using FastQC v0.11 and subsequently filtered out with Trimmomatic 0.39 [21,22]. 2.2. Assembly and Annotation of the S. serra Chloroplast Genome For the de novo cp genome assembly, plastid-like reads were extracted from the clean reads using the Velvet v1.2.07 [23]. The cp genome sequence was annotated using the GeSeq tool with NCBI BLASTN searches [24]. The architecture of the S. serra cp genome was visualized using the OGDRAW program [25]. 2.3. Chloroplast Genome Sequence Divergence and Comparison The chloroplast genome of S. serra, newly sequenced in this study, was compared with 6 publicly available Sium cp genomes (Supplementary Table S1) to determine the synteny of the cp genome structures. The IR boundary comparison diagram was manually created based on their plastome structures to access the expansion and contraction of the LSC, SSC, and IR junctions in the 7 Sium plastomes. The complete plastome sequences of all the Sium were analyzed for sequence similarity using mVISTA in Shuffle-LAGAN mode with default parameters, using the S. serra cp genome as a reference [26]. 2.4. Genetic Divergence The genetic divergence of the seven Sium cp genomes was calculated using the nucleotide diversity (Pi) to identify highly variable sites with DnaSP v6.12.03 [27]. The cp genomes were aligned using MAFFT implemented in Geneious Prime (Biomatters, Auckland, New Zealand) [28,29]. In the nucleotide diversity analysis, the window length was set to 800 bp and the step size to 200 bp using the DnaSP v6.12.03 [27]. 2.5. Repeat Sequence The simple sequence repeats (SSRs) motifs in the seven Sium plastomes were analyzed using MISA v2.1, with the minimum number of repeats thresholds set to ten repetitions for mononucleotide SSRs, six for dinucleotide SSRs, and five for tri-, tetra-, penta-, and hexanucleotide SSRs [30]. Tandem repeats were identified using the Phobos Tandem Repeats Finder v1.0.6 with parameters set to 1 for the matches, −5 for mismatches and gaps, and 0 for N positions [31]. Additionally, the forward, reverse, complement, and palindromic repeats were detected using REPuter with a Hamming distance of 3, minimum sequence identity of 90%, and minimum repeat size of 30 bp [32]. A single copy of the IR region was used for all these analyses. 2.6. Substitution Rate The complete cp genome of S. serra was compared with those of six Sium species. The nonsynonymous (KA) and synonymous (KS) substitution rates were analyzed by extracting DNA sequences of the identical specific 79 functional protein-coding genes, translating them into amino acids, and aligning them independently using Geneious Prime (Biomatters, New Zealand). The KA and KS substitution rates were calculated using DnaSP v6.12.03 [27]. 2.7. Phylogenetic Tree Thirty-one cp genomes from Apioideae were selected to construct a phylogenetic tree, with four Araliaceae species chosen as an outgroup (Supplementary Table S1). This approach was used to determine the position of S. serra within Apioideae and analyze the phylogenetic relationship of the Sium genus. The cp genome sequences of 34 species from the Apiaceae and Araliaceae families were downloaded. The 79 protein-coding genes shared by 35 cp genomes were concatenated, aligned, and saved in FASTA format using MAFFT. The maximum likelihood (ML) tree was constructed using RAxML v7.2.6 with the general time reversible + γ + invariant model. One thousand nonparametric bootstrap replicates were performed to estimate the support for each internal branch of the phylogeny [33]. 3. Results 3.1. General Features of the S. serra Chloroplast Genome The chloroplast genome sequence of S. serra was 154,755 bp and featured a representative circular structure, including an LSC region of 84,581 bp, an SSC region of 17,664 bp, and a pair of 26,255 bp IR (IRa/IRb) (Figure 1). The genome contained 113 genes, comprising 79 protein-coding, 30 tRNA, and 4 rRNA genes. In addition, the genes duplicated by IR included 6 protein-coding, 7 tRNA, and 4 rRNA genes, resulting in 130 genes. Eighteen genes contained introns, including six tRNA and ten protein-coding genes with a single intron, while clpP and ycf3 had two introns (Supplementary Table S2). The GC content of the S. serra cp genome was 37.4%, consistent with other Sium plastomes (Table 1). 3.2. Comparative Analysis of Sequence Divergence in the Sium cp Genomes The sequence variation across the cp genomes of seven Sium species was plotted using the mVISTA program, with the annotated S. serra cp genome serving as the reference (Figure 2). For the overall sequence identity, as indicated by the peaks and valleys across all seven Sium species, the results showed that the LSC and SSC regions were divergent, while the IR region pairs were highly conserved. Three noncoding regions (rps16-trnQ, trnE-trnY, and ycf4-cemA) in the SC regions exhibited low similarity owing to indel events exceeding 350 bp, which only occurred in the northern Holarctic clade (S. suave, S. medium). The IR and SC junctions in seven Sium species were compared (Figure 3). The longest LSC was observed in S. ninsi (85,036 bp), SSC in S. ventricosum (18,705 bp), and IR in S. medium (26,472 bp). In the cp genomes of all the Sium species, rps19 was located in JLB, ndhF in JSB, and ycf1 in JSA. In the case of S. serra, unlike the other Sium species, trnH was located 2 bp away from JLA. In S. ventricosum, the ycf1 gene present in IRa was shortened as a result of IR contraction and SSC expansion of approximately 1000 bp. The nucleotide diversity analysis performed with DnaSP showed highly variable regions across the seven Sium cp genomes (Figure 4). The mean nucleotide diversity (Pi) across the entire cp genome was 0.00607. Three highly variable regions were identified, each exhibiting a markedly higher Pi value of >0.02. The most variable region was the ndhF-trnL intergenic spacer in SSC, which had a Pi value of 0.3417. Polymorphism was detected in seven species due to substitution and short-region indels. Other highly variable regions included in trnE-trnT (Pi = 0.0265) and rps16 intron (Pi = 0.0201) within the LSC regions. 3.3. Comparative Analysis of the Repeat Sequences in the Sium cp Genomes The results showed that the total number of SSRs ranged from 39 (S. ninsi) to 49 (S. suave). The distribution of the SSRs varied across the seven cp genomes of Sium (Figure 5, Supplementary Table S3). In all the Sium species, only two types were found: mononucleotide and dinucleotide. Mononucleotides comprised 74.8%, while dinucleotides accounted for 25.2%. Among the mononucleotides, 93% of the SSRs in all the cp genomes were of the A and T types. For the dinucleotides, only the AT and TA types were observed. The pattern of tandem repeats in the Sium cp genomes ranged from 142 to 169 (Figure 6, Supplementary Table S3). S. tenue had the highest number of hexanucleotides, while S. crispulifolia and S. ventricosum had the lowest. In the Sium cp genomes, the observed repeat types were forward (57%), complementary (27.2%), reverse (12.3%), and palindrome (3.5%). Palindromic repeats were absent from S. crispulifolia (Figure 7, Supplementary Table S3). 3.4. Adaptive Evolution Analysis in the Sium cp Genomes Seventy-nine shared protein-coding genes from all seven Sium cp genomes were used to calculate the KS and KA substitution rates (Figure 8, Supplementary Table S4). The results showed that most protein-coding genes had a low average KS value (<0.05), except for the matK, accD, ycf2, ndhF, rpl32, and ycf1 genes. Similarly, most protein-coding genes exhibited a low average KA value (<0.05), with the exceptions of matK, accD, rpl22, ycf2, ndhF, and ycf1 genes. The average KA/KS value was 0.26, with a range of KA/KS from 0 to 1.35 (Figure 8). Eight genes (matK, rps14, accD, infA, rpl22, ycf2, ndhF, and ycf1) exhibited high average KA/KS values (>1). Among these, matK, accD, ycf2, ndhF, and ycf1 genes had KA and KS values that were also higher than those of other genes. 3.5. Phylogenetic Analysis of the Genus Sium Phylogenetic analysis was performed using 79 protein-coding genes from the Sium and related Apioideae taxa (Figure 9). The genus Sium was found to be monophyletic, clustering into two lineages (Southern Palearctic and Northern Holarctic Clade). The Southern Palearctic clade included S. crispulifolia, S. ventricosum, S. serra, S. ninsi, and S. tenue, which are biogeographically native to the southern Palearctic. The Northern Holarctic clade included S. suave and S. medium, which are species native to the Northern Holarctic region. All the Sium species were supported with strong bootstrap values (BS = 100). These findings indicate that Sium forms a sister group to Cryptotaenia. 4. Discussion In this study, we sequenced and analyzed Sium serra (Franch. & Sav.) Kitag. distributed in Korea, China, and Japan. Compared to previous studies, the length and gene contents of the cp genome in the genus Sium were conserved [34]. Events such as gene loss or rearrangement did not occur in the genus Sium. However, some noncoding regions in the SC exhibited low similarity, which could result from indel events (Figure 2). The total cp genome size of Sium species varied between 153 kb in S. ventricosum and 155 kb in S. ninsi. Among the seven Sium cp genomes, S. serra exhibited the third-largest genome size (Figure 3). The average size of Sium cp genomes is 154.4 kb, similar to those of previously published Apioideae cp genomes [35,36]. Among the seven Sium cp genomes, S. ventricosum had the smallest size, which could result from an IR contraction and SSC expansion of approximately 1000 bp. This indicates that the ycf1 gene at the SSC/IRa junction (JSA) is predominantly located in the SSC (Figure 3). A previous study reported that the ndhF gene is entirely located in the SSC region for three taxa (S. ninsi, S. tenus, and S. ventricosum) [34]. However, in this study, the ndhF gene was found to be located at the SSC/IRb junction (JSB) in all the taxa. This discrepancy is likely due to an error in the previous search. Furthermore, the Southern Palearctic clade exhibited a shorter IR region compared to the Northern Holarctic clade, likely due to IR contraction. In addition, the clade containing S. serra showed a slightly larger IR region than those of the other Southern Palearctic clades. Consequently, nucleotide diversity analysis revealed approximately six highly variable sites across seven Sium species (Figure 4). This site is located in the SC, indicating that the IR regions are more conservative. Additionally, this finding is almost similar to that of a previous study [34]. The highest nucleotide diversity (Pi = 0.03417) is present in the ndhF-trnL region, which exhibits significant variation. These highly variable sites could function as effective molecular markers for shallow-level taxonomic studies between and within specific taxa of the genus Sium [37,38]. According to our knowledge, we analyzed the repeat sequences in the plastomes of the genus Sium for the first time. Repeat sequences play a crucial role in homologous recombination in plant genomes [39]. The distribution of the SSRs in the cp genome of Sium ranged from 39 (S. ninsi) to 49 (S. suave) (Figure 5, Supplementary Table S3). Most SSRs are mononucleotides, comprising 74.8%, while the others are dinucleotides. The C repeat sequence is only present in S. serra and S. medium. The distribution of tandem repeats in the plastomes of Sium ranged from 142 (S. crispulifolia) to 169 (S. tenue) (Figure 6, Supplementary Table S3). In this study, S. tenue had the highest number of hexanucleotide tandem repeats in the genus Sium. The distribution of dispersed repeats in the plastomes of Sium ranged from 27 (S. crispulifolia, S. ventricosum) to 45 (S. ninsi) (Figure 7, Supplementary Table S3). No palindrome-type repeats were present only in S. crispulifolia, S. tenue, and S. ninsi. These species are primarily distinguished by differences in the leaf width [4]. This repeat sequence may serve as a powerful species identification marker within the genus Sium [40,41]. Furthermore, it can help distinguish Sium species from Gao-ben, which is used as an herbal medicine [8]. Vascular plants adapt to various ecological niches. Factors such as the light intensity, moisture levels, and temperature influence their photosynthesis. Owing to adaptation to these factors, genes undergo selection [42]. Several genes undergo positive selection in this study (Figure 8, Supplementary Table S4). The matK gene is involved in group Ⅱ intron splicing. This gene is also positioned in the intron of trnK and exhibits a high substitution rate [43]. In this study, the KA/KS ratio of the matK gene was confirmed to be 1.298, indicating positive selection. The rpl22 gene encodes the chloroplast ribosomal protein CL22, which is associated with the large ribosomal protein subunit [44]. In the genus Sium, the KA and KS values of the rpl22 gene were approximately 0.5, while the KA/KS ratio was the highest at 1.356. This result could be attributed to the fact that most Sium species grow in shaded, humid environments. In addition, a similar pattern is also observed in the genus Chrysosplenium, which grows in a similar environment [45]. Furthermore, genes subjected to positive selection are distributed across the LSC and SSC regions, showing a similar trend to that observed in the nucleotide diversity. Phylogenetic studies of the genus Sium have been conducted in the past. Analyses using nuclear rDNA ITS and a few cp DNAs reveal that Sium does not form a monophyletic group but is polyphyletic with the genus Berula [46,47,48]. In addition, a study using the cp genome was also conducted; however, S. serra was excluded because it is classified as Pimpinella in China [5,34]. In this study, all the species within the genus Sium, including S. serra, form a single clade with high bootstrap values. Moreover, as in previous studies, two subclades are formed, supporting the results of biogeographic studies on the genus Sium [2,34]. 5. Conclusions In this study, the chloroplast genome of S. serra was sequenced and assembled for the first time. The detailed characteristics of the S. serra chloroplast genome were identified through an analysis of its features. All the Sium species have typical circular structures, are 153–155 kb long, and contain 130 genes. Hotspots and repeat sequences such as SSRs and tandem repeats can be used as molecular markers at the intraspecific level. The substitution rates analyzed in this study can be applied to other plants living in similar environments. Finally, the genus Sium formed a monophyletic group. Furthermore, these genetic resources serve as a reference for further molecular phylogenetic studies on the Oenantheae tribe and, more generally, the Apiaceae family.