Comparative chloroplast genome analyses of Oxytropis DC. species: new insights into genome evolution and phylogenomic implications

Abstract

The genus Oxytropis DC. comprises about 310 species distributed in Asia, Europe, and North America. Previous studies based on evidences from morphology or a few molecular markers are helpful for understanding the classification and systematic evolution of Oxytropis. However, a scarcity of chloroplast genomic resources for Oxytropis has hindered the understanding of the genus’s systematic classification and chloroplast genome evolution. Here comparative genomic analyses were conducted on chloroplast genomes of 24 Oxytropis species. Chloroplast genomes of Oxytropis species showed the triad structure due to the loss of one copy of the IR, with the size range from 121854 bp to 125271 bp. The Oxytropis cp genomes encoded a total of 110 genes, including 76 protein-coding genes (PCGs), 30 transfer RNA (tRNA) genes, and four ribosomal RNA (rRNA) genes. It was found that the atpF intron, one clpP intron, one rps12 intron, rpl22 gene, rps16 gene, and infA gene were lost in the Oxytropis cp genomes. Seven regions (5’-rps12-clpP, clpP intron, psbM-petN, rpl23-trnI-CAU, ndhJ-trnF-GAA,trnQ-UUG-accD, trnL-UAA-trnT-UGU) were chosen as potential molecular markers, which will contribute to species identification, population genetics and phylogenetic studies of Oxytropis. The phylogenetic relationships among Oxytropis species provided some implications for the classification of Oxytropis. Congruent with studies based on the morphological evidence, the close relationships between O. neimonggolica and O. diversifolia, as well as O. filiformis and O. coerulea were revealed. The results supported the treatment of O. daqingshanica as a separate species and refuted the inclusion of O. daqingshanica in O. ochrantha as conspecific taxa. In addition, it was suggested that O. chiliophylla should be considered as a separate species rather than its inclusion in O. microphylla. The 16 positively selected genes (rps3, rps4, rps7, rps11, rps12, rpl2, rpl20, rpl32, rpoC2, psbC, rbcL, atpF, clpP, accD, ycf1, ycf2) are related to important biological processes for instance self-replication, photosynthesis and metabolite biosynthesis, which may contribute to the adaptation of Oxytropis to its habitats. This study will lay a solid foundation for further studies on species identification, taxonomy, and systematic evolution of Oxytropis.

1. Introduction

As the third largest flowering plant family after Asteraceae Bercht. & J.Presl and Orchidaceae Juss., the Fabaceae Lindl. comprises about 751 genera and 19,500 species worldwide (LPWG (The Legume Phylogeny Working Group), 2013). Combined with a series of diagnostic characteristics, phylogenetic analyses based on matK sequences supported the classification system with six monophyletic subfamilies within the Fabaceae, with Papilionoideae DC. (503 genera, ca. 14,000 species) and Caesalpinioideae DC. (148 genera, ca. 4400 species) as the largest two subfamilies, followed by Detarioideae Burmeist. (84 genera, ca. 760 species), Cercidoideae LPWG (12 genera, ca. 335 species), Dialioideae LPWG (17 genera, ca. 85 species), and Duparquetioideae LPWG (one genus, one species) (LPWG (The Legume Phylogeny Working Group), 2017). Phylogenetic analyses based on plastomes and nuclear genes strongly support the classification system of six Fabaceae subfamilies (Zhang et al., 2020; Zhao et al., 2021), which has currently gained widespread acceptance and consensus among scholars. Oxytropis DC. belongs to the Astragalean clade under the inverted-repeat-lacking clade (IRLC) of the subfamily Papillonoideae (LPWG (The Legume Phylogeny Working Group), 2017; Zhao et al., 2021; Duan et al., 2024). The genus Oxytropis has about 310 species distributed in Asia, Europe, and North America, with a concentrated distribution in Central Asia (Zhang, 1998; Zhu et al., 2010). Oxytropis is an important component of the flora in the alpine and arid regions of the Northern Hemisphere temperate zones, and is one of the common groups in alpine, desert, and semi desert regions (Li and Ni, 1985). The Oxytropis plants have certain feeding, medicinal, and ornamental value (Kholina et al., 2021a; Sandanov et al., 2023; Wang B. et al., 2024). Due to the extremely similar morphology between Astragalus and Oxytropis, the Oxytropis species were included in Astragalus defined by Linnaeus (1753). De Candolle (1802) first separated Oxytropis from Astragalus based on the characteristics of keel petal shape and pod septum shape. Delimitation of the subgenera and sections were conducted by taxonomist since the establishment of Oxytropis, and although there is a certain consensus, different perspectives also exist (e.g., Bunge, 1874; Vasil’chenko et al., 1948; Pavlov, 1961; Zhang, 1998; Zhu et al., 2010). Micromorphological evidence has been applied to the classification of Oxytropis and some insights have been gained (Karaman et al., 2009; Ceter et al., 2013; Erkul et al., 2015; Zhao et al., 2022, 2023). With the development of sequencing technology, molecular markers have been used to address the questions on systematic evolution of Oxytropis, however, most studies involved a few molecular markers and referred species sampling with limited geographical ranges due to the focus on regional treatments (e.g., Archambault and Strömvik, 2012; Tekpinar et al., 2016a, 2016b; Kholina et al., 2016, 2021a, 2022). In addition, the cp genome sequences of the Oxytropis that can be used for study on systematic evolution are still scarce (Su et al., 2019; Liu et al., 2021; Bei et al., 2022; Tavares et al., 2022). Some progress has been achieved in the phylogenetic study of Oxytropis, but there is still a long way to uncover the systematic evolutionary questions for this complex taxonomic group with a large number of species, wide distribution, diverse morphology, and a relatively recent diversification history (Shavvon et al., 2017). The lack of effective molecular markers has to some extent hindered the phylogenetic study of Oxytropis, thus employing highly variable molecular markers coupled with increased taxon sampling promise advances in the issues on its taxonomy and evolution. The adaptation of Oxytropis species to special habitats makes it an excellent model for studying adaptive evolution, which is still an open issue for Oxytropis.

Chloroplast (cp) is a vital organelle in green plants, having a crucial role in photosynthesis and a myriad of metabolic activities (Neuhaus and Emes, 2000; Daniell et al., 2016; Wang J. et al., 2024). In angiosperms, the cp genomes are mostly a quadripartite structure: a large single-copy (LSC) region and a small single-copy region (SSC) separated by two inverted repeats (IRs) (Wicke et al., 2011), however, losses of the IR exist in a few angiosperm families, such as Geraniaceae (Guisinger et al., 2011; Ruhlman et al., 2017), Cactaceae (Sanderson et al., 2015), Arecaceae (Barrett et al., 2016), Fabaceae (Choi et al., 2019), Lophopyxidaceae and Putranjivaceae (Jin et al., 2020), and Passifloraceae (Cauz-Santos et al., 2020). Chloroplast genome has been widely used in studies on taxonomy, phylogeny and evolution of angiosperm (e.g., Kan et al., 2024; Li et al., 2024a; Yan et al., 2024; Wang et al., 2025; Yan et al., 2025), due to its own advantages such as uniparental inheritance, small size, lack of recombination, and moderate nucleotide substitution rate (Palmer, 1985; Wicke et al., 2011; Mower and Vickrey, 2018).

Through comparative genomics analyses of the cp genomes of 24 species of Oxytropis, this study aims to (1) explore the basic characteristics of Oxytropis cp genomes, (2) screen the hotspot regions as potential molecular markers of Oxytropis, (3) provide preliminary insights into the current classification of some Oxytropis species, and (4) understand the adaptation of Oxytropis species to the environment at the molecular level. Our study will lay a solid foundation for future studies on cp genome evolution, species identification, genetic diversity, and systematic evolution of Oxytropis.

2. Materials and methods

2.1. Plant material, DNA extraction and sequencing

Materials for the 19 Oxytropis species in the present study were collected during field trips, with the collected Oxytropis plants pressed into herbarium specimens, and fresh and tender leaves dried in silica gel without affecting identification. The collected specimens were identified by referring to relevant reference books (e.g., Li and Ni, 1985; Zhang, 1998; Zhu et al., 2010; Zhao et al., 2019), and all the voucher specimens were preserved in the herbarium of the Inner Mongolia Normal University (NMTC) ( Table 1 ). Total genomic DNA was isolated from the silica-dried leaves according to the protocol of Doyle and Doyle (1987). The extracted DNA was fragmented by sonication and then used for construction of short-insert library (insert size, 300 bp) by NEBNext^® Ultra™ II DNA Library Prep Kit for Illumina^®. Finally, the pooled libraries were sequenced by the Illumina NovoSeq platform in Novogene (Beijing, China).

Table 1.

Summary of chloroplast genome features of Oxytropis species.

Species	Voucher	GenBank accession	Size (bp)	Number of genes				GC content (%)	References
				Total	Protein-coding	tRNA	rRNA
O. aciphylla Ledeb. 1	Li QQ 20230602057 (NMTC)	PV684027	122411	110	76	30	4	34.2	This article
O. aciphylla Ledeb. 2		MW794135	122173	110	76	30	4	34.3	Unpublished
O. arctobia Bunge		MT409175	125271	110	76	30	4	34.0	Tavares et al. (2022)
O. bicolor Bunge 1	Li QQ 20230520004 (NMTC)	PV684034	122387	110	76	30	4	34.2	This article
O. bicolor Bunge 2		MN255323	122461	110	76	30	4	34.2	Su et al. (2019)
O. ciliata Turcz.	Li QQ 20230620045 (NMTC)	PV684026	122272	110	76	30	4	34.2	This article
O. chiliophylla Royle ex Benth.	Li QQ 20230720015 (NMTC)	PV694277	122480	110	76	30	4	34.2	This article
O. coerulea (Pall.) DC.	Li QQ 20230807007 (NMTC)	PV684031	122121	110	76	30	4	34.3	This article
O. daqingshanica Y.Z.Zhao & Zong Y. Zhu	Li QQ 20220820044 (NMTC)	PV694279	122181	110	76	30	4	34.3	This article
O. diversifolia E. Peter 1	Li QQ HLT001 (NMTC)	PV684028	122012	110	76	30	4	34.3	This article
O. diversifolia E. Peter 2		MT780271	122210	110	76	30	4	34.2	Unpublished
O. falcata Bunge		OR491708	122781	110	76	30	4	34.3	Unpublished
O. filiformis DC.	Li QQ 20230805042 (NMTC)	PV684033	122321	110	76	30	4	34.2	This article
O. glabra DC.		MW349014	122094	110	76	30	4	34.3	Liu et al. (2021)
O. hirta Bunge	Li QQ 20230805058 (NMTC)	PV684032	122356	110	76	30	4	34.2	This article
O. holanshanensis H. C. Fu	Li QQ 20230804029 (NMTC)	PV694276	123621	110	76	30	4	34.1	This article
O. latibracteata Jurtzev	Li QQ 20230805031 (NMTC)	PV694278	121854	110	76	30	4	34.3	This article
O. microphylla (Pall.) DC.	Li QQ 20230528014 (NMTC)	PV684029	122453	110	76	30	4	34.2	This article
O. myriophylla (Pall.) DC.	Li QQ 20220830082 (NMTC)	OR911498	122251	110	76	30	4	34.2	This article
O. neimonggolica C. W. Chang & Y. Z. Zhao	Li QQ 20230803032 (NMTC)	PV684030	122195	110	76	30	4	34.2	This article
O. ochrantha Turcz.	Li QQ 20220828007 (NMTC)	PV684024	122228	110	76	30	4	34.2	This article
O. oxyphylla (Pall.) DC.	Li QQ 20230807023 (NMTC)	PV684036	122284	110	76	30	4	34.2	This article
O. proboscidea Bunge	Li QQ 20230718011 (NMTC)	PV694280	122363	110	76	30	4	34.3	This article
O. racemosa Turcz.	Li QQ 20230531062 (NMTC)	PV684037	122172	110	76	30	4	34.3	This article
O. sericopetala Prain ex C. E. C. Fisch.	Li QQ 20230718021 (NMTC)	PV684035	122405	110	76	30	4	34.3	This article
O. splendens Douglas		MT409174	122318	110	76	30	4	34.2	Tavares et al. (2022)
O. squammulosa DC.	Li QQ 20230620004 (NMTC)	PV684025	122352	110	76	30	4	34.3	This article

2.2. Chloroplast genome assembly and annotation

Trimomatic v. 0.33 (Bolger et al., 2014) was used to remove adapters in the obtained raw sequencing data. The filtered raw reads for each species were then used to assemble the cp whole genome sequence by NOVOPlasty v. 4.3.1 (Dierckxsens et al., 2017), with the cp genome sequence of O. bicolor (GenBank accession no. MN255323) (Su et al., 2019) as the reference and its rbcL sequence as the seed. The cp genome of O. myriophylla obtained from our previous study (Niu et al., 2024) was used as the reference to conduct annotations of the 19 Oxytropis cp genomes in the present study. The brief procedure for annotating the cp genome of O. myriophylla was as follows: following the annotation method of Zhang et al. (2022), GeSeq (Tillich et al., 2017) and CPGAVAS2 (Shi et al., 2019) were used to annotate the cp genome of O. myriophylla, with the cp genomes of O. bicolor (MN255323), O. arctobia (MT409175) and O. spelendens (MT409174) designated as custom reference genomes. The annotation results obtained from GeSeq and CPGAVAS2 were imported into Geneious Prime (Kearse et al., 2012) to check the intron/exon boundaries and the start and stop codon positions. If necessary, manual corrections were performed to obtain the elaborated annotated cp genome of O. myriophylla. The brief workflow for annotations of the 19 Oxytropis cp genomes was as follows: in Geneious Prime, MAFFT (Katoh and Standley, 2013) alignment was performed between the cp genome sequence of O. myriophylla with the complete annotation information and the cp genome sequence of other Oxytropis species. Based on the alignment results, transferring annotations function in Geneious Prime was used for annotation, and the annotation results were manually checked and proofread to finally generate the complete annotated cp genome of other Oxytropis species. The cp genome sequences of Oxytropis species with annotation information in gb format were imported into OrganellarGenomeDRAW (Greiner et al., 2019) to draw their cp genome circular maps. Moreover, annotation of other cp genomes obtained from GenBank were checked before being used for analysis.

2.3. Comparative chloroplast genome analyses

Comparative analysis was conducted on the basic characteristics of cp genome lengths, GC contents, and gene quantities in 27 cp genomes of 24 Oxytropis species using Geneious Prime ( Table 1 ). The 27 Oxytropis cp genomes were aligned in MAUVE ver. 2.4.0 under the progressiveMauve algorithm (Darling et al., 2004, 2010). Due to cp genomes of O. falcata and O. arctobia with inversion, the two cp genome sequences were not used in molecular marker identification. The coding and noncoding regions in 25 cp genomes of 22 Oxytropis species were extracted by Geneious Prime, and all the homologous sequences were aligned one by one in MAFFT v. 7.490 (Katoh and Standley, 2013). The final aligned homologous sequences in fasta format were imported into DnaSP v. 6.12.03 (Rozas et al., 2017) and their nucleotide variability (Pi) values were calculated, and finally candidate molecular markers were screened based on the Pi values and sequence lengths.

2.4. Phylogenetic analyses

To reconstruct the phylogenetic relationships among Oxytropis species under the phylogenetic background of the Astragaglean clade, a total of 46 cp genome sequences from 43 species under the IRLC of the subfamily Papilionoideae were selected for phylogenetic analysis based on Zhao et al. (2021) ( Supplementary Table S1 ). Considering the phenomena of gene/intron loss and inversion in the cp genomes of the IRLC of Papilionoideae (Jansen et al., 2008), only protein coding genes (PCGs) were selected for phylogenetic tree construction. Seventy-six PCGs were extracted from the cp genomes by Geneious Prime and each PCG was aligned separately using MAFFT v. 7.490 (Katoh and Standley, 2013). Alignments of genes that were not common to all species and genes with significant length differences (accD, clpP, psbL, ycf1, and ycf2) were removed. Finally, alignments of the remaining 71 PCGs were concatenated to form the phylogenetic dataset. Bayesian inference (BI) and the maximum likelihood (ML) methods were employed to construct the phylogenetic trees. GTR+I+G was recommended as best-fit model by PartitionFinder2 (Lanfear et al., 2017), and MrBayes v. 3.2.7a (Ronquist et al., 2012) was then used to construct BI tree based on the method of Zhang et al. (2022). The ML tree was constructed by RAxML v. 8.2.12 (Stamatakis, 2014) following the method of Li et al. (2024b). Finally, these two phylogenetic trees were visualized using FigTree v. 1.4.4 (Rambaut, 2018), and Alhagi sparsifolia, Caragana arborescens, Chesneya acaulis, Corethrodendron multijugum, and Tibetia liangshanensis were designated as the outgroup to root the trees.

2.5. Adaptive evolution analyses

Based on the results of phylogenetic analyses, 31 cp genomes involving 24 species of Oxytropis and its four related taxa (Carmichaelia australis, Lessertia frutescens, Phyllolobium chinense, and Sphaerophysa salsula) were selected for selection pressure analyses. The CodeML program in the PAML software package (Yang, 2007) is currently the most widely used bioinformatics tool for selection pressure analyses. EasyCodeML (Gao et al., 2019) can offer a user-friendly graphical interface for executing CodeML. Site models from CodeML were performed in EasyCodeML with the purpose of detecting positive selection sites of PCGs in Oxytropis cp genomes. Seventy-six PCGs shared by Oxytropis and its related taxa were firstly extracted from the cp genomes using Geneious Prime. MAFFT (Katoh and Standley, 2013) was then employed to perform multiple alignment of each PCG according to its codons and stop codons were manually deleted in the final alignment matrix. The final alignment of each PCG was concatenated into a supermatrix, which was exported into fasta format as an input file for EasyCodeML. The ML tree constructed based on the supermatrix by RAxML v. 8.2.12 (Stamatakis, 2014) was as an input tree in EasyCodeML ( Supplementary Figure S1 ). The likelihood ratio test (LRT) was used to detect positive selection sites with four comparison models: M0 vs. M3, M1a vs. M2a, M7 vs. M8, and M8a vs. M8. With LRT threshold p<0.05, Bayesian empirical Bayes (BEB) (Yang et al., 2005) or Naïve empirical Bayes (NEB) (Nielsen and Yang, 1998) analysis was adopted to detect positive selection sites with posterior probabilities ≥0.95.

3. Results and discussion

3.1. Features of Oxytropis chloroplast genome

The size range of cp genomes of 24 Oxytropis species was from 121854 bp (O. latibracteata) to 125271 bp (O. Arctobia) ( Table 1 ; Figure 1 ). Compared with the cp genomes of some Papilionoideae taxa such as Cyamopsis (Kaila et al., 2017), Ormosia (Liu et al., 2019), and Campylotropis (Feng et al., 2022) with typical quadripartite structure, the cp genomes of Oxytropis species showed the triad structure due to the loss of approximately 25 kb IR. The cp genomes of the IRLC groups in Papillonoideae have lost one IR copy and exhibit the triad structure (Wojciechowski et al., 2004; Jansen et al., 2008). The GC content in the cp genomes of 24 Oxytropis species (34.0%–34.3%) was roughly equivalent to that in cp genomes of other IRLC taxa such as Glycyrrhiza, Astragalus, and Galega (Duan et al., 2020; Su et al., 2021; Feng et al., 2023). The cp genomes of Oxytropis species encoded a total of 110 genes, including 76 protein-coding genes (PCGs), 30 transfer RNA (tRNA) genes, and four ribosomal RNA (rRNA) genes ( Tables 1 , 2 ). The 110 genes can be classified into four categories according to their functions: 57 genes related to self-replication, 46 associated with photosynthesis, five for other genes, and two genes with unknown function. Moreover, 15 genes contained one intron (ndhB, clpP, ndhA, rpl16, petB, rpoC1, rpl2, petD, rps12, trnI-GAU, trnG-UCC, trnL-UAA, trnK-UUU, trnA-UGC, trnV-UAC), while gene ycf3 possessed two introns. Gene rps12 had trans-splicing in the Oxytropis cp genome, like in most other angiosperms.

Figure 1.

Chloroplast genome map of the 19 Oxytropis species generated in this study. Genes outside the circle are transcribed counterclockwise, and those inside are transcribed clockwise.

Table 2.

Genes contained in the Oxytropis chloroplast genomes.

Category of genes	Group of genes	Name of genes
Self-replication	Ribosomal RNAs	rrn4.5, rrn5, rrn16, rrn23
	Transfer RNAs	trnA-UGC *, trnC-GCA, trnD-GUC, trnE-UUC, trnF-GAA, trnfM-CAU, trnG-GCC, trnG-UCC *, trnH-GUG, trnI-CAU, trnI-GAU *, trnK-UUU *, trnL-CAA, trnL-UAA *, trnL-UAG, trnM-CAU, trnN-GUU, trnP-UGG, trnQ-UUG, trnR-ACG, trnR-UCU, trnS-GCU, trnS-GGA, trnS-UGA, trnT-GGU, trnT-UGU, trnV-GAC, trnV-UAC *, trnW-CCA, trnY-GUA
	Small subunit of ribosome	rps2, rps3, rps4, rps7, rps8, rps11, rps12 *a, rps14, rps15, rps18, rps19
	Large subunit of ribosome	rpl2 *, rpl14, rpl16 *, rpl20, rpl23, rpl32, rpl33, rpl36
	DNA dependent RNA polymerase	rpoA, rpoB, rpoC1 * , rpoC2
Photosynthesis	Subunits of ATP synthase	atpA, atpB, atpE, atpF, atpH, atpI
	Subunits of photosystem I	psaA, psaB, psaC, psaI, psaJ, ycf3** , ycf4
	Subunits of photosystem II	psbA, psbB, psbC, psbD, psbE, psbF, psbH, psbI, psbJ, psbK, psbL, psbM, psbN, psbT, psbZ
	Subunits of cytochrome b/f complex	petA, petB *, petD *, petG, petL, petN
	Subunits of NADH-dehydrogenase	ndhA *, ndhB *, ndhC, ndhD, ndhE, ndhF, ndhG, ndhH, ndhI, ndhJ, ndhK
	Subunit of Rubisco	rbcL
Other genes	Subunit of Acetyl-CoA-carboxylase	accD
	C-type cytochrome synthesis	ccsA
	Envelop membrane protein	cemA
	Protease	clpP *
	Maturase	matK
Unknown function	Conserved open reading frame	ycf1, ycf2

*Genes containing one intron, **genes containing two introns; ^atrans-splicing gene.

There are phenomena such as intron loss, gene loss, and inversion in the cp genome evolution of Papilionoideae (Jansen et al., 2008). Our study found the absence of the atpF intron, clpP intron, and rps12 intron in the cp genomes of Oxytropis and its closely related species. The atpF intron is lost in the cp genomes of Oxytropis, Lessertia, and Phaerophysa species, while it is present in Phyllopium, Carmichaelia, and Astragalus species. The absence or presence of the atpF intron in the cp genome could be used as a potential molecular marker for distinguishing the morphologically highly similar genera Oxytropis and Astragalus. Loss of one clpP intron was detected in the cp genomes of Oxytropis and its closely related taxa, including Astragalus, Carmichaelia, Lessertia, Phyllolobium, and Sphaerophysa species. The absence of clpP intron has also occurred in other IRLC groups of Papillonoideae, for example, one clpP intron was lost in Alhagi, Caragana, and Vicia species; and two clpP introns were lost in Tibetia, Corethrodendron, and Glycyrrhiza species (Jansen et al., 2008; Li et al., 2020; Lee et al., 2021), while the two clpP introns are present in other angiosperms genera such as Uncaria (Dai et al., 2023), Alisma (Lan et al., 2024), and Argentina (Li et al., 2024b). The rps12-3′-end intron was lost in the cp genomes of Oxytropis and its closely related taxa, and the loss of rps12 intron is a common phenomenon in the IRLC group (Jansen et al., 2008; Lee et al., 2021). Most angiosperm cp genomes contain genes rpl22, rps16, and infA, all of which are lost in the Oxytropis cp genomes. The gene rpl22 is absent in the cp genomes of all legumes (Jansen et al., 2008). We detected that genes rps16 and infA are also not present in the cp genomes of related taxa of the Oxytropis, including Alhagi, Astragalus, Caragana, Carmichaelia, Chesneya, Corethrodendron, Lessertia, Phyllolobium, Sphaerophysa, and Tibetia species. Analysis by Mauve showed that among the 27 cp genomes of Oxytropis, cp genomes of O. falcata and O. arctobia had inversion, while the remaining 25 cp genomes have the same gene order with no obvious reorganization ( Supplementary Figure S2 ). The gene order rearrangement did not affect the sequences of any of the involving genes in cp genomes of O. falcata and O. arctobia.

3.2. Divergence hotspots

DnaSP v. 6.12.03 (Rozas et al., 2017) was employed to calculate the Pi values of a total of 253 regions of Oxytropis cp genome with the aim to screen the highly divergent regions. The Pi values ranged from 0%-4.244%, with a mean value of 0.436%, showing Oxytropis cp genomes with a high level of similarity ( Supplementary Table S2 ; Figure 2 ). As a whole, 42 regions with Pi =0, 132 regions with 0%<Pi ≤ 0.5%, 50 regions with 0.5%<Pi ≤ 1%, 20 regions with 1%<Pi ≤ 1.5% (petG-trnW-CCA, atpF, trnT-GGU-trnE-UUC, rps8-rpl14, atpA-trnR-UCU, trnF-GAA-trnL-UAA, trnQ-UUG-accD, rpl33-rps18, trnV-GAC-rrn16, rps11-rpl36, rps4-trnS-GGA, atpH-atpF, trnL-UAA-trnT-UGU, rpoA-rps11, trnR-ACG-trnN-GUU, psbK-trnQ-UUG, psaB-rps14, ycf1-rps15, trnH-GUG-psbA, trnP-UGG-psaJ), and nine regions with Pi>1.5% (5’-rps12-clpP, trnR-UCU-trnG-UCC, clpP intron, psbM-petN, trnfM-CAU-trnG-GCC, trnI-CAU-ycf2, ndhI-ndhG, rpl23-trnI-CAU, ndhJ-trnF-GAA). Among the 29 regions with Pi>1%, 27 regions (excluding clpP intron and atpF) were located in the intergenic region, indicating that the non-coding regions exhibited higher variation compared to the coding regions, and regions located in the intergenic spacers (IGS) with greater potential for development of molecular markers.

Figure 2.

The nucleotide diversity (Pi) values of shared regions in 25 Oxytropis chloroplast genomes.

In order to screen molecular markers with potential for development, 19 regions with Pi >1% and alignment lengths >300bp were identified as candidate molecular markers for Oxytropis, namely 5’-rps12-clpP, clpP intron, psbM-petN, ndhI-ndhG, rpl23-trnI-CAU, ndhJ-trnF-GAA, atpF, trnT-GGU-trnE-UUC, rps8-rpl14, trnQ-UUG-accD, rpl33-rps18, rps11-rpl36, atpH-atpF, trnL-UAA-trnT-UGU, trnR-ACG-trnN-GUU, psbK-trnQ-UUG, ycf1-rps15, trnH-GUG-psbA, and trnP-UGG-psaJ. Among these 19 markers, taking into account both Pi value and sequence alignment length, seven regions (5’-rps12-clpP, clpP intron, psbM-petN, rpl23-trnI-CAU, ndhJ-trnF-GAA,trnQ-UUG-accD, trnL-UAA-trnT-UGU) were selected as potential molecular markers for Oxytropis. The cp molecular markers used in previous phylogenetic studies of Oxytropis (Kulshreshtha et al., 2004; Wojciechowski et al., 2004; Artyukova et al., 2011; Tekpinar et al., 2016a, 2016b; Kholina et al., 2016, 2018a, 2018b, 2020, 2021a, 2021b, 2021c, 2022; Chen et al., 2020; Kozyrenko et al., 2020; Sandanov et al., 2023) included matK, rpoC1, rpoC2, trnL intron, trnV intron, trnL-trnF, trnH-psbA, petG-trnP, and trnS-trnG. All other markers except trnH-psbA were not among the developed candidate molecular markers for Oxytropis, which suggested the significance of developing molecular markers for specific taxonomic groups. Overall, our newly screened potential molecular markers will contribute to species identification, population genetics and phylogenetic studies of Oxytropis.

3.3. Phylogenetic analyses

Overall, compared to previous phylogenetic studies of Oxytropis using several molecular markers obtained with Sanger sequencing (e.g., Kholina et al., 2016; Tekpinar et al., 2016b; Shavvon et al., 2017), relatively high phylogenetic resolution was obtained in our study by utilizing the plastid genome data. Phylogenetic trees inferred from BI and ML analyses were almost identical in topology, and the difference mainly lied in the relative position of O. squammulosa versus O. filiformis and O. coerulea ( Figure 3 ; Supplementary Figures S3, S4 ). Both the BI and ML trees showed that the outgroup species were robustly separated from the Astragaglean clade (PP = 1.00, ML BS = 100%). Within the Astragaglean clade, there are three major clades, namely Oxytropis, Astragalus, and Coluteoid clades. Phylogenetic tree showed that Oxytropis species were well clustered together (PP = 1.00, ML BS = 100%), which corroborated the previous studies that Oxytropis was monophyletic (e.g., Archambault and Strömvik, 2012; Tekpinar et al., 2016a; Kholina et al., 2016; Shavvon et al., 2017). Consistent with studies of Su et al. (2021); Tian et al. (2021) and Moghaddam et al. (2023) based on cp genome, our result indicated that Oxytropis was sister to Coluteoid clade and Oxytropis+Coluteoid clade had a sister relationship with Astragalus. However, studies of Moghaddam et al. (2016) using ITS, matK and rpl32-trnL data and Zhao et al. (2021) based on low-copy nuclear genes revealed that Oxytropis+Astragalus clade was sisters to Coluteoid clade. The nuclear-cytoplasmic conflict on the phylogenetic position of Oxytropis may reflect the complex evolutionary history of this genus. Although the current taxon sampling was still limited, the systematic relationships among Oxytropis species showed in our study still provided some insights into the classification of Oxytropis. In the phylogenetic trees, O. neimonggolica was clustered together with O. diversifolia (PP = 1.00, ML BS = 100%), which supported their close relationship based on morphological study (Zhu et al., 2010; Zhao et al., 2019). Oxytropis ochrantha and O. myriophylla were clustered together and was sister to O. daqingshanica, which supported the treatment of O. daqingshanica as a separate species (Zhao et al., 2019) and disapproved the inclusion of O. daqingshanica in O. ochrantha as conspecific taxa (Zhu et al., 2010). Oxytropis filiformis was clustered with O. coerulea (PP = 1.00, ML BS = 99%), suggesting their close affinity, which was congruent with studies based on the morphological evidences (Zhu et al., 2010; Zhao et al., 2019). Oxytropis microphylla and O. ciliata grouped together and O. chiliophylla was distantly related to these two species, which suggested that O. chiliophylla should be considered as a separate species rather than including it in O. microphylla. Further taxonomic treatments in Oxytropis should be conducted by combining evidence from morphology, anatomy, ecology and palynology.

Figure 3.

Phylogenetic trees of Oxytropis and its related taxa based on the dataset of 71 concatenated protein-coding genes (PCGs) of the chloroplast genomes. (A) Bayesian inference (BI) tree, (B) maximum likelihood (ML) tree. Values along branches indicate Bayesian posterior probabilities (only PP < 1.00 are shown) and ML bootstrap percentages (only values < 100% are shown), respectively.

3.4. Adaptive evolution

The p-values of LRTs for compared models M0 vs. M3, M1a vs. M2a, M7 vs. M8, and M8a vs. M8 is below threshold 0.05, suggesting adaptation signatures within Oxytropis cp genomes ( Supplementary Tables S3, S4 ; Table 3 ). According to the manual of PAML (Yang, 2007), M0 vs. M3 was not suggested as a test of positive selection but as a test of variable ω among sites. In addition, M1a vs. M2a seems to be more stringent compared with M7 vs. M8 which has been confirmed in our results. Therefore, we relied on result under model M8 to discuss positive selection sites in Oxytropis cp genomes. Sixteen genes with positive selection sites were detected according to BEB analysis under model M8. The number of positive selection sites in these genes ranged from 1 to 47: nine genes (rps4, psbC, rpl20, rps12, rps11, rps3, rpl2, rps7, rpl32) with one site, two genes (rpoC2, atpF) having two sites, clpP possessing three sites, rbcL with five sites, ycf2 containing eight sites, accD with 12 sites, and ycf1 harboring the largest number of sites. According to their functional category, nine genes (rps3, rps4, rps7, rps11, rps12, rpl2, rpl20, rpl32, rpoC2) were associated with self-replication, three genes (psbC, rbcL, atpF) were responsible for photosynthesis, genes clpP and accD belonged to other genes, and genes ycf1 and ycf2 are functionally unknown.

Table 3.

Positively selected sites (*: P>95%; **: P>99%) detected in the Oxytropis chloroplast genomes in comparisons of M7 vs. M8 and M8a vs. M8 under Bayes empirical Bayes (BEB) analysis.

Gene	Positive selected sites	Pr value(ω>1)	Number of sites
rbcL	907 A/1091 A/1142 S/1303 E/1333 E	0.985*/0.988*/0.988*/0.959*/1.000**	5
rps4	2615 E	0.985*	1
psbC	4929 A	0.999**	1
rpoC2	7690 T/7830 R	0.968*/0.963*	2
atpF	9257 Q/9297 -	0.986*/0.978*	2
accD	9923 S/9968 D/10002 Y/10088 D/10089 E/10090 -/10092 -/10098 -/10106 -/10120 T/10129 E/10143 V	0.996**/0.952*/0.968*/0.996**/0.979*/0.969*/0.961*/0.953*/0.971*/0.960*/0.965*/0.980*	12
rpl20	11836 L	0.952*	1
rps12	12017 -	0.967*	1
clpP	12031 V/12093 D/12155 G	0.986*/0.997**/0.995**	3
rps11	13670 A	0.967*	1
rps3	14240 K	0.990*	1
rpl2	14612 L	0.974*	1
ycf2	14838 Q/15489 L/15715 H/15790 F/15952 S/16027 Q/16074 R/16172 Q	0.990**/0.963*/0.992**/0.995**/0.962*/0.957*/0.988*/0.965*	8
rps7	17335 Q	0.958*	1
ycf1	17495 W/17683 H/17690 V/17693 S/17716 N/17747 -/17751 H/17760 Y/17775 L/17797 N/18085 S/18087 V/18088 Q/18110 Y/18111 S/18115 K/18116 P/18120 Y/18136 F/18141 Q/18142 D/18145 I/18176 F/18202 L/18205 Y/18329 T/18364 K/18366 K/18375 N/18376 V/18377 K/18399 F/18489 Y/18612 L/18974 D/19011 -/19033 S/19037 F/19042 G/19050 D/19051 W/19052 A/19055 S/19075 Y/19103 R/19150 R/19155 T	0.967*/0.965*/0.999**/0.982*/0.989*/0.963*/0.996**/0.998**/0.985*/0.981*/1.000**/0.991**/0.998**/0.967*/0.996**/1.000**/0.980*/0.998**/0.996**/0.992**/0.990*/0.999**/0.992**/0.984*/0.981*/0.999**/0.998**/0.986*/0.998**/1.000**/0.984*/0.964*/0.992**/0.977*/0.976*/0.979*/0.954*/0.968*/0.951*/0.992**/0.998**/0.982*/0.984*/1.000**/0.986*/1.000**/0.997**	47
rpl32	21547 T	0.980*	1

Amino acids refer to sequence of O. racemosa.

The adaptive evolution of these 16 genes may help Oxytropis species adapt to their habitats. Among them, rps3, rps4, rps7, rps11, rps12, rpl2, rpl20, and rpl32 encoded ribosomal subunit proteins. Chloroplast ribosomal proteins are essential for cp ribosome assembly, which plays an important role in plant survival, acclimation and adaptation (Schmid et al., 2024). DNA dependent RNA polymerase subunit beta’’ encoded by rpoC2, is one of the components of the core of plastid-encoded polymerase (PEP) which acts as the major transcription machinery of mature chloroplasts (Zhelyazkova et al., 2012; Kindgren and Strand, 2015). The 43-kDa chlorophyll a-binding protein (CP43) encoded by psbC, together with CP47 encoded by psbB, binds chlorophyll, as an inner light-harvesting complex of photosystem II (PSII) (Landi and Guidi, 2022). The large subunit of Rubisco was encoded by rbcL (Wicke et al., 2011). Rubisco mediates the fixation of inorganic carbon from CO2 into organic compounds during photosynthesis (Wilson and Hayer-Hartl, 2018). In most lineages of terrestrial land plants, rbcL is under positive selection (Kapralov and Filatov, 2007). ATP synthase CF0 B subunit encoded by atpF is one of the important constituents of chloroplast ATP synthase, which using the proton gradient produces ATP that is indispensable for photosynthesis and plant growth (Hahn et al., 2018; Yamamoto et al., 2023). Gene clpP in chloroplast is essential for plant development, with an indispensable function for cell viability (Shikanai et al., 2001; Kuroda and Maliga, 2003). The chloroplast gene clpP together with a nuclear multi gene family encodes the Clp protease that degrades damaged proteins during environmental stresses (Clarke, 1999; Adam and Clarke, 2002). The key enzyme acetyl-CoA carboxylase (ACCase) regulates de novo synthesis of fatty acids in plants (Rawsthorne, 2002). The accD gene encodes one of the four subunits of ACCase, which is essential for cell viability, leaf development, and seed development (Madoka et al., 2002; Kode et al., 2005; Caroca et al., 2021). Products encoded by essential genes ycf1 and ycf2 of higher plants are essential for plant cell survival (Drescher et al., 2000).

Previous work suggested that Oxytropis arctobia accD gene was under positive selection, which might be related to its adaptation to the cold environment in the Arctic (Tavares et al., 2022). Positively selected genes in Oxytropis detected in our study were also found under positive selection in some other Fabaceae genera. For example, rpl2, rpoC2 and accD were under positive selection in Pueraria (Zhou et al., 2023), and so were rps11, clpP, accD and ycf1 in Astragalus (Moghaddam et al., 2023), rps4, rpl32, accD and ycf2 in Pterocarpus (Hong et al., 2020), rps4, rps7, rpl32 and clpP in Vicia (Li et al., 2020), rps7, rpl20, atpF, ycf1 and ycf2 in Caragana (Cui et al., 2024), and rps3, rps12, rpoC2, psbC, rbcL, clpP, accD, ycf1 and ycf2 in Dalbergia (Li et al., 2022). Oxytropis species spread in temperate and cold regions of the Northern Hemisphere in Asia, Europe, and North America, usually thriving in harsh environments such as the Arctic areas and alpine ecosystems (Zhu et al., 2010; Archambault and Strömvik, 2012; Kholina et al., 2016; Tavares et al., 2022). Oxytropis species grow in various habitats such as mountains, steppes, prairies, meadows, deserts, semi deserts, forest-steppes, and forests (Zhu et al., 2010; Sandanov et al., 2023; Welsh, 2023). The origin of Oxytropis was dated to about 5.6 million years ago, with 95% highest posterior density intervals ranging from 3.61 to 8.07 Ma, which coincides with climate modifications around the Miocene-Pliocene boundary (Shavvon et al., 2017). It was inferred that Oxytropis experienced a recent rapid radiation based on its recent age estimates, short interior branch on gene tree, little genetic differences, and diverse morphology and ecological habitats (Shavvon et al., 2017). The 16 positively selected genes in the Oxytropis cp genome are related to important biological processes for instance self-replication, photosynthesis and metabolite biosynthesis, which may contribute to the adaptation of Oxytropis to diverse habitats, especially under extreme arid and cold conditions. The adaptation of Oxytropis to diverse habitats may have to some extent promoted the rapid diversification of Oxytropis in its relatively recent evolutionary history.

4. Conclusion

In this study, comparative analysis of cp genomes of 24 Oxytropis species revealed that their cp genomes exhibited a triad structure, and the cp genome size, GC content, and gene content were conserved. Seven highly divergent regions (5’-rps12-clpP, clpP intron, psbM-petN, rpl23-trnI-CAU, ndhJ-trnF-GAA,trnQ-UUG-accD, trnL-UAA-trnT-UGU) identified in this study may be potentially utilized as high-resolution DNA barcodes, which will facilitate species identification and phylogenetic and phylogeographic studies of Oxytropis. Phylogenetic analysis based on the cp genome sequences supported the monophyly of Oxytropis and provided some new insights into the classification of Oxytropis. The results indicated that the cp genome can be utilized as an informative molecular marker for enhancing our understandings of evolutionary diversification in Oxytropis. Sixteen protein-coding genes (rps3, rps4, rps7, rps11, rps12, rpl2, rpl20, rpl32, rpoC2, psbC, rbcL, atpF, clpP, accD, ycf1, ycf2) showed evidence for positive selection, which may contribute to the adaptation of Oxytropis to its diverse habitats. Overall, our study improved the understanding of cp genome features, phylogenetic relationships, and adaptive evolution in Oxytropis. Employing single-copy nuclear genes coupled with more detailed taxon sampling will facilitate future work on the phylogeny, biogeography, and adaptive evolution of Oxytropis.

Funding Statement

The author(s) declare financial support was received for the research and/or publication of this article. This work was supported by the Fundamental Research Funds for Inner Mongolia Normal University (No. 2022JBBJ012), the National Natural Science Foundation of China (No. 32260053), and the Natural Science Foundation of Inner Mongolia, China (No. 2022LHQN03004).

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: GenBank of NCBI (https://www.ncbi.nlm.nih.gov/genbank/), PV684024-PV684037, and PV694276-PV694280.

Author contributions

Q-QL: Conceptualization, Formal Analysis, Investigation, Project administration, Writing – original draft. YN: Formal Analysis, Investigation, Writing – original draft. Z-PZ: Formal Analysis, Investigation, Writing – original draft. JW: Conceptualization, Writing – review & editing. C-YL: Conceptualization, Writing – original draft.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declare that no Generative AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fpls.2025.1645582/full#supplementary-material