Chloroplast Genomes and Phylogenetic Analysis of Three Carthamus (Asteraceae) Species

Abstract

The genus Carthamus Linnaeus, which belongs to the tribe Cardueae in the Asteraceae family, originated in the Mediterranean region and consists of approximately 20 species worldwide. Understanding the phylogeny of the Carthamus is crucial for the cultivation of C. tinctorius. Although chloroplast genomes are widely used for species identification and evolutionary studies, there have been limited investigations on the chloroplast genomes of Carthamus species. In this study, we assembled the chloroplast genomes of C. persicus, C. tinctorius × C. persicus, and C. lanatus and combined them with the five chloroplast genomes of C. tinctorius for comparative genomic analysis. The sizes of the chloroplast genomes of C. lanatus, C. persicus, and C. tinctorius × C. persicus were 152,602 bp, 153,177 bp, and 153,177 bp, respectively. Comparative analysis showed that the chloroplast genome structures of the four Carthamus species were highly conserved. Additionally, the phylogenomic analysis demonstrated that the plastid genome and angiosperms353 dataset significantly improved the phylogenetic support of Carthamus species. This analysis supported Carthamus as a monophyletic taxon and its internal division into the sect. Carthamus and sect. Atractylis. The Carthamus was closely related to Carduncellus, Femeniasia, Phonus, and Centaurea. In conclusion, this study not only expands our understanding of the cp genomes of Carthamus species but also provides support for more comprehensive phylogenetic studies of Carthamus.

1. Introduction

Chloroplasts, which are organelles exclusive to plants, play an essential role in indirectly providing significant amounts of nutrients and oxygen to humans through photosynthesis [1,2]. Moreover, they have the ability to effectively bind carbon dioxide, thus aiding in mitigating global warming. The chloroplast (cp) genome has served as a valuable source of molecular data for investigating plant phylogeny and evolution; this is attributed to its predominantly maternal inheritance, although patrilineal [3,4,5] or biparental inheritance [6,7,8,9] can occur in certain species and its relatively conserved nature, which differs from plant mitochondrial genomes [10]. In most plants, the cp genomes possess a cyclic quadripartite structure with two single-copy regions separated by two inverted repeat regions, although linear structures are found in some species [11,12,13]. Recent advancements in sequencing technologies have facilitated whole-genome sequencing and large-scale phylogenetic analyses of numerous plants, resulting in an enhanced understanding of phylogeny [14,15]. Consequently, an increasing number of plant cp genomes have been reported, significantly contributing to our knowledge of plant phylogeny and aiding in plant taxonomy.

Targeted enrichment of nuclear genes addresses the potential limitation of relying solely on organelle genome data for estimating phylogeny. Unlike large nuclear genomes, angiosperm organelle genomes are predominantly maternally inherited and are prone to high rates of genetic drift in small populations, resulting in potentially less accurate estimates of lineage divergence [16]. The Angiosperms353 dataset has gained widespread usage in angiosperm phylogeny [17]. This dataset comprises 353 targeted low-copy nuclear genes that are highly representative and have become a valuable tool for molecular systematics and population genetics [17,18,19]. Additionally, researchers can easily access the dataset through enrichment sequencing (Hyb-Seq), transcriptome sequencing (RNA-Seq), and genome skimming to obtain the corresponding data [20]. Previous studies have demonstrated the feasibility of applying the angiosperms353 dataset in phylogenetic studies of Asteraceae [21].

The Carthamus Linnaeus, belonging to the tribe Cardueae of the family Asteraceae, originated in the Mediterranean region and comprises approximately 20 species worldwide [9,10,11]. These Carthamus species are predominantly found in arid and semi-arid environments and exhibit high tolerance to drought stress [22]. Among them, C. tinctorius Linnaeus is the sole domesticated oilseed crop of economic importance, while the remaining species, including C. lanatus Linnaeus and C. persicus Desf. ex Willd., are wild weeds distributed from northwestern India to the Mediterranean Sea and its surrounding regions [23]. Previous phylogenetic studies on the Carthamus primarily utilized plastid and nrITS (nuclear ribosomal internal transcribed spacer) markers. Bowles et al. conducted an investigation of the phylogenetic relationships within Carthamus using nrITS, plastid, and microsatellite markers [24]. Their results revealed that the Carthamus consisted of two sections: sect. Carthamus and sect. Atractylis. To objectively evaluate the Carduncellus-Carthamus complex, Vilatersana et al. performed a phylogenetic analysis using nrITS and plastid markers, highlighting the subdivision of the complex into four genera: Carduncellus, Carthamus, Phonus, and Femeniasia [25]. Despite the widespread use of cp genomes in species identification and evolutionary studies, limited studies have been carried out on the cp genome of Carthamus species. Currently, only the cp genome of C. tinctorius has been reported [26,27], which hinders phylogenetic studies of Carthamus.

Here, we assembled the cp genomes of C. persicus, C. tinctorius × C. persicus, and C. lanatus and performed a comparative genomic analysis with the known cp genome of C. tinctorius to explore differences in these cp genomes. In addition, we investigated the phylogenetic relationships between these four Carthamus species and other Cardueae species. This study not only lays the foundation for understanding the cp genomes of Carthamus but also provides data support for more comprehensive phylogenetic studies of Carthamus.

2. Results

2.1. Basic Characteristics of Chloroplast Genome

The sizes of the cp genome of C. lanatus, C. persicus, C. tinctorius × C. persicus, and C. tinctorius were determined to be 152,602 bp, 153,177 bp, 153,177 bp, and 152,938–153,114 bp, respectively (Figure 1a, Table S1). It is worth noting that C. tinctorius × C. persicus, which was a hybrid of C. tinctorius (male parent) and C. persicus (female parent), had the same cp genome size as its female parent at 153,177 bp (Figure 1b). Their cp genome structures displayed a typical quadripartite structure, consisting of a large single copy region (LSC), a small single copy region (SSC), and two inverted repeats (IR) regions. The differences in GC content among the four species were relatively small. However, it was observed that the IR region had a higher GC content compared to the LSC and SSC regions. In terms of gene composition, the number and type of genes were consistent among the four Carthamus species. A total of 131 genes were identified, including 86 protein-coding genes (PCGs), 8 rRNAs, and 37 tRNAs. Among these genes, eighteen genes had two copies. These cpPCGs were mainly classified into five categories (Table S2): five photosystem I genes, fifteen photosystem II genes, six subunits of ATP synthases, nine large ribosomal proteins, and twelve small ribosomal proteins. In addition, 14 genes (ndhA, ndhB, atpF, petB, petD, rpoC1, rpl2, rpl16, trnI-GAU, trnG-GCC, trnL-UAA, trnV-UAC, trnA-UGC, trnK-UUU) contained one intron, and three genes (clpP, rps12, ycf3) contained two introns.

2.2. Repeats Identification

Based on the comparison of intraspecific and interspecific repetitive sequences in the cp genomes of four Carthamus species, it was found that they had similar types of repeats but with some differences in terms of their numbers (Figure 2a, Table S3). In the four species, namely C. tinctorius, C. persicus, C. lanatus, and the hybrid species C. tinctorius × C. persicus, the differences in the number of simple sequence repeats (SSRs) were minimal. For instance, only mono-, di-, and trinucleotides were detected in C. tinctorius, C. persicus, and C. tinctorius × C. persicus, while C. lanatus had the additional presence of one tetranucleotide. Both C. persicus and C. tinctorius × C. persicus contained 36 SSRs, including 20 mononucleotides, 11 dinucleotides, and 5 trinucleotides. The C. lanatus contained 40 SSRs, including 21 mononucleotides, 11 dinucleotides, 7 trinucleotides, and 1 tetranucleotide. Further analysis was conducted specifically on the five C. tinctorius species. It was found that the differences in mononucleotides were more pronounced compared to other repeat types. Additionally, the distribution of SSRs revealed that they were primarily located in the LSC region, with a smaller number found in the SSC region. These SSRs were mainly distributed in the intergenic spacer regions. Moreover, the analysis of long repeats indicated that they predominantly consisted of forward, reverse, and palindromic repeats. These long repeats were mainly distributed in the LSC and Inverted Repeat (IR) regions. Among these long repeats, the largest proportion was accounted for by forward and palindromic repeats, while reverse repeats were the least common (Figure 2a, Tables S4 and S5).

2.3. Sequence Variation and IR Boundary Analysis

The genomic collinearity analysis revealed that the four cp genomes exhibited a relatively conserved pattern with no significant rearrangements (Figure S1); this suggests that the organization of genes and other functional elements in the cp genomes of four Carthamus species was similar. To further understand the differences in the cp genomes of four Carthamus species, we performed a sequence difference visualization analysis (Figure 3a, Figure S2). In interspecific comparisons, we found that C. lanatus had more sequence variation compared to C. tinctorius and C. persicus. On the other hand, C. tinctorius and C. persicus had relatively few sequence differences, indicating a closer relationship between these two species. In intraspecific comparisons of C. tinctorius, different C. tinctorius species had some sequence variation, all of which occurred in intergenic spacer regions. IR boundary analysis revealed that the chloroplast boundaries of these Carthamus species were highly conserved, and the differences between gene locations and boundaries were smaller. For example, trnN and rpl2 genes were located within the IR regions, ndhF genes were located within the SSC regions, and the LSC-IRb and SSC-IRa boundaries were located on the ycf1 and rps19 genes, respectively.

2.4. Codon Usage Analysis

We selected 53 PCGs with a length greater than 300 base pairs for codon analysis. The analysis revealed variations in various aspects of codon composition, such as the number of codons, as well as the GC1, GC2, GC3, and overall GC content, among the 19 species of the tribe Cardueae (Figure S3 and Table S6). Our results indicated that the content of GC1 and GC2 was higher compared to that of GC3 and the overall GC content; this suggested that the distribution of GCs among codons was not uniform. The trend in GC1, GC2, and GC3 remained consistent across the 19 Cardueae species. Specifically, leucine was found to be the most abundant amino acid in these cp genomes, while cysteine was the less frequent amino acid. Within the 59 codons, the more common isoleucine was encoded by AUU, and the less common cysteine by UGC (Table S7). Additionally, leucine encoded by CUC had the lowest RSCU value, whereas leucine encoded by UUA had the highest RSCU value. Both Tryptophan (UGG) and methionine (AUG) were encoded by a single codon and had RSCU values of 1. Furthermore, 29 codons had RSCU values greater than 1, indicating biased use (Figure 4).

2.5. Phylogenetic Analysis

To determine the phylogenetic position of the four Carthamus species, we constructed phylogenetic trees by combining cp genome, nrITS, and angiosperm 353 data for Cardueae species. Of these, four species (Phonus arborescens, Femeniasia balearica, Carduncellus pinnatus, Carduncellus rhaponticoides) acquired only about 70 PCGs. Additionally, we also assembled the angiosperms353 data from eight species (P. arborescens, F. balearica, C. pinnatus, C. rhaponticoides, C. persicus, C. tinctorius, C. lanatus, C. tinctorius × C. persicus). The average angiosperms353 gene number was 333, with a minimum gene number of 303. We performed informative loci and tree-building modeling statistics for the three sequence matrices (Table S8). The AP353 matrix had a length of 425,889 bp, with 103,034 parsimony informative sites, 191,242 variable sites, and 147,787 conserved sites. The cpPCGs matrix length was 79,671 bp, consisting of 13,032 parsimony informative sites, 33,016 variable sites, and 39,417 conserved sites. The nrITS matrix length was 830 bp, with 196 parsimony informative sites, 310 variable sites, and 402 conserved sites. The phylogenetic trees of cpPCGs and AP353 matrixes exhibited high support, while the nrITS phylogenetic tree showed some branches with low support (Figure 5). Nevertheless, the overall structure remained similar across most branches of the three phylogenetic trees. These trees indicated that the Carthamus-Carduncellus complex was monophyletically sister to Centaurea in the Centaureinae branch. The Carthamus was closely related to Phonus, Femeniasia, and Carduncellus. Our results were consistent with previous studies, which divided Carthamus into two sections, C. persicus, and C. tinctorius, along with their hybrids in sect. Carthamus, and C. lanatus in sect. Atractylis. Based on three phylogenetic trees, we observed conflicts between the nucleoplasm. In the plastid phylogeny, C. rhaponticoides did not cluster with C. pinnatus but showed a closer relationship to F. balearica. In the AP353 and nrITS phylogenetic trees, C. rhaponticoides clustered with C. pinnatus. Additionally, nucleoplasmic differences were also identified in other genera.

3. Discussion

3.1. Chloroplast Genome Structure Variation within Three Carthamus Species

In this study, we assembled and analyzed the cp genomes of C. persicus, C. tinctorius × C. persicus, and C. lanatus for the first time. We found that these cp genomes shared a similar size and structure with the previously reported C. tinctorius, all exhibiting a typical quadripartite structure. While there have been instances of evolutionary events such as genome rearrangements, gene deletions, IR contractions, and expansions, plastids generally demonstrate a high level of conservation in terms of genome size, structure, and gene content [28]. Our study revealed that the cp genomes of four Carthamus species showed minimal variability in size and were conserved in terms of genome structure, gene composition, and gene order. In terms of sequence differences, C. lanatus exhibited significant variation compared to the other three Carthamus species, displaying more sequence variation. In addition, due to maternal inheritance, the cp genome of the female parent (C. persicus) was inherited by C. tinctorius × C. persicus, resulting in identical cp genomes. The selfed progeny of this hybrid also possessed the same chloroplast genomes. The GC content in the IR region was significantly higher than that in the LSC and SSC regions, which was a common characteristic in the cp genomes of most species [29,30]. Furthermore, the four Carthamus species showed consistency in the number of total genes, PCGs, rRNA, and tRNA genes, suggesting that their cp genomes were highly similar, which partly explains the similarity in their morphological features. The best synonymous codons in the four Carthamus species mostly end in A or U, resulting in increased AT content in the genomes; this supports the A/T codon bias prevalent in the cp genomes of higher plants [29]. Plant cp genomes have a high number of SSRs, which provide rich genetic information and can be used as important molecular markers to study species’ genetic evolution [31]. Among the four Carthamus species, pentanucleotides and hexanucleotides were not detected, while tetranucleotides were only found in C. lanatus. Dynamic changes in the IR region may lead to cp genomic variation in angiosperms. The IR boundaries were highly conserved in the four Carthamus species, suggesting that they are closely related. However, the complete cp genomes of Carduncellus, Phonus, and Femeniasia were still missing in this study for comparison with Carthamus. In the future, more material from the Carthamus and its relatives should be collected to obtain genomic data and further investigate the differences between them.

3.2. Phylogeny of Carthamus Species

Phylogenetic trees for Cardueae species were constructed using Maximum Likelihood and Bayesian Inference methods. The topology of the phylogenetic trees obtained by these two methods was similar. The phylogenetic trees of the cpPCGs and AP353 matrices showed high support. However, the use of nrITS sequences alone did not provide strong support for some species, possibly due to the limited number of informative sites in nrITS sequences. Previous phylogenetic studies have suggested that the Carthamus-Carduncellus complex was divided into four genera (i.e., Carthamus, Carduncellus, Phonus, and Femeniasia) and that the branch of the genus Carthamus could be divided into sect. Carthamus and sect. Atractylis. Our phylogenetic trees indicated that the complex was closely related to Centaurea. Within the complex, the four Carthamus species were clustered into a single clade, suggesting that Carthamus was monophyletic. Additionally, the branching structures between the cpPCGs and the AP353 phylogenetic trees were not completely consistent. The plastid phylogeny showed that C. rhaponticoides was more closely related to F. balearica, which was not consistent with the nuclear phylogenetic trees. The nuclear phylogenetic trees showed C. rhaponticoides clustered with C. pinnatus and F. balearica clustered with P. arborescens, which was consistent with previous studies [24,25,31,32]. Within the Carthamus, our results also show that C. persicus clustered with C. tinctorius × C. persicus and C. tinctorius, with the three having a closer relationship, all belonging to sect. Carthamus, while C. lanatus belonged to sect. Atractylis.

Differences between the plastid phylogenetic tree and the nuclear phylogenetic tree may arise from various factors, including hybridization, incomplete lineage sorting (ILS), chloroplast capture, and genetic differences in plastids [33,34,35,36]. ILS is a common occurrence in phylogenies, where nucleoplasmic conflicts in Asteridae could be attributed to ancient chloroplast capture or ILS. Different DNA fragments exhibit varying evolutionary rates, leading to disparities between gene trees and species trees—the phenomenon of nucleoplasmic conflicts in Asteraceae warrants further investigation in the future. Hybridization is a frequent natural occurrence and is relatively common in Asteraceae, as exemplified by the hybridization and polyploid evolutionary radiation observed in Achillea [37]. Additionally, Phalacrocarpum has demonstrated susceptibility to frequent hybridization [38]; this is one of the reasons for the rich diversity of Asteraceae species. Nucleoplasmic conflicts can be affected by interspecific hybridization in angiosperms; this is due to the prevalence of maternal inheritance in angiosperms, resulting in the offspring inheriting the cp DNA of the female parent. Over time, the offspring of hybridization may evolve into a new species while still retaining the cp DNA of the female parent. Although most eukaryotic organelle genomes are maternally inherited, there are cases of biparental inheritance of plastids. Previous studies have indicated that around 20% of angiosperm pollen germ cells contain chloroplasts, suggesting the possibility of biparental inheritance of plastids [6,7]. For example, the Oenothera [9] and Pelargonium [8] exhibit biparental inheritance of cp genomes. Additionally, paternal inheritance of plastids is more common in gymnosperms [3] and relatively less common in angiosperms, such as Nicotiana tabacum and Arabidopsis thaliana [4,5]. Previous studies have suggested that mild low-temperature stress promotes paternal inheritance of plastids [39], but this has not been reported in Asteraceae. Carthamus and its relatives are known for their preference for light and tolerance to drought, suggesting that low temperatures may not be the main factor contributing to paternal plastid inheritance. However, the specific influences and causes of paternal plastid inheritance are still unknown in the Carthamus and its relatives. Therefore, it is necessary to collect more Carthamus species and study the plastid inheritance pattern, hybridization, and ILS in order to explore a more accurate understanding of the phylogenetic network in the future.

4. Materials and Methods

4.1. Sampling, DNA Extraction, and Sequencing

The C. lanatus (specimen number: MH001) was cultivated in the experimental field at South-Central Minzu University, Wuhan, China. Fresh leaves of this species were collected for genome sequencing. Genomic DNA was extracted using the modified CTAB method [40]. A short insertion library was constructed, and then 2 × 150 bp paired-end reads were obtained from the BGI-Seq platform. Adaptors and low-quality reads were removed using Trimmomatic v. 0.39 [41]. The filtered reads were quality-controlled using Fastqc v. 0.11.9 [42]. In addition, short-read sequencing data of C. tinctorius (male parent, SRR3418170-SRR3418175), C. persicus (female parent, SRR3418173-SRR3418276), and C. tinctorius × C. persicus (hybrid offspring, SRR3418176-SRR3418167, SRR3418176-SRR3418269) [43] for cp genome assembly. Then, we also collected target-enriched genome sequencing data for C. pinnatus (SRR12917348), C. rhaponticoides (ERR5033745), F. balearica (ERR9230214), P. arborescens (ERR9230215), Synurus deltoides (SRR12849144), Dolomiaea baltalensis (SRR11926473), and Xanthopappus subacaulis (SRR12917358) from the NCBI database.

4.2. Chloroplast Genome Assembly and Annotation

To assemble the cp genomes of C. tinctorius, C. lanatus, C. persicus, and C. tinctorius × C. persicus, we used GetOrganelle v.1.7.5 [44] with the parameters: “-R 15 -t 20 -k 21, 45, 65, 85, 105 -F embplant_pt”. Subsequently, we assessed the integrity of the genomes using Bandage v. 0.8.1 software (http://rrwick.github.io/Bandage/ (accessed on 10 August 2023)). The cp genome annotation was performed using CPGAVAS2 [45] and PGA [46]. Additionally, we constructed the cp genome map using CPGview (http://www.1kmpg.cn/cpgview/ (accessed on 12 August 2023)) [47]. For the nrITS sequences, we also used GetOrganelle v.1.7.5 [44] with the parameter ‘-R 7 -t 20 -k 35, 85, 115 -F embplant_nr’ to assemble them. Then, these nrITS sequences were annotated using Geneious Prime v. 2022.2.2 [48], with C. tinctorius (KY397481) serving as the reference genome. Based on the target-enriched genome sequencing data of the C. pinnatus, C. rhaponticoides, F. balearicus, and P. arborescens, we used Geneious Prime v. 2022.2.2 [48] to assemble some chloroplast genes.

4.3. Repeat Structure Identification

Simple sequence repeats (SSRs) were identified using MISA v.2.1 software [49]. The minimum repeat numbers were set at 10, 6, 5, 5, 5, and 5 for mono-, di-, tri-, tetra-, penta-, and hexanucleotide, respectively. We used REPuter (https://bibiserv.cebitec.uni-bielefeld.de/reputer (accessed on 12 August 2023)) to predict the palindromic and complement, forward, and reverse repeats [50]. The minimum repeat size and Hamming distance were set to 30 and 3, respectively.

4.4. Codon Usage Analysis

To reduce sampling error, we excluded protein-coding genes (PCGs) shorter than 300 bp [51,52,53,54]. A total of 53 PCGs were used for codon usage analysis. We used CodonW v.1.4.4 [53,55] (https://codonw.sourceforge.net/culong.html#CodonW (accessed on 12 August 2023)) to calculate the GC content of the silent 1st, 2nd, and 3rd codon positions (GC1, GC2, and GC3), as well as the relative synonymous codon usage (RSCU) value. The RSCU value greater than 1 indicates a higher frequency of codon usage, while a value less than 1 indicates a lower frequency.

4.5. Sequence Variation Analysis

To analyze sequence variation in the cp genomes of the four Carthamus species, we performed multiple sequence alignments using MAFFT v.7.4 [56]. Then, aligned sequences were visualized using Geneious Prime v.2022.2.2 [48]. Additionally, We used mVISTA [57] to analyze sequence variation, with C. tinctorius (OR538395) serving as the reference genome.

4.6. Phylogenetic Analysis

We selected the cp genomes of the 19 species of the tribe Cardueae and C. indicum for the plastid phylogenetic analysis and used Chrysanthemum indicum as the outgroup (Table S9). We extracted the 78 common cpPCGs from these 21 cp genomes and aligned them using MAFFT v.7.4 [56]. The aligned sequences were then concatenated using PhyloSuite v.1.2.3 [58] to create a cpPCGs matrix. Additionally, we compiled a nrITS matrix by combining the nrITS sequences of all 25 species. We also used easy353.py [20] to assemble the angiosperms353 dataset of the 11 species (C. tinctorius, C. tinctorius × C. persicus, C. persicus, C. lanatus, C. pinnatus, C. rhaponticoides, F. balearica, P. arborescens, S. deltoides, D. baltalensis, X. subacaulis). Then, we incorporated data from the public dataset (Kew Tree of Life Explorer, https://treeoflife.kew.org/ (accessed on 14 August 2023)) to create the angiosperms353 (AP353) matrix. Finally, we constructed phylogenetic trees for each of the three matrices (cpPCGs matrix, nrITS matrix, AP353 matrix) and performed comparative analysis.

The phylogenetic trees were constructed using Maximum Likelihood (ML) and Bayesian Inference (BI) methods. ModelFinder [59] was utilized to determine the best-fitting model for ML analysis. Use IQ-TREE v. 2.1.2 [60] to construct the ML tree with the parameter “-m MFP -bb 1000 -nt 30”. For the BI tree, MrBayes v. 3.2.6 [61] was used to obtain a maximum clade credibility (MCC) tree. The parameters for the BI analysis were set as follows: nst = 6, rates = invgamma. Bayesian inference was performed with the concatenated sequence, using one million generations, two runs, four chains, and 25% of trees were discarded as burn-in. Trees were sampled every 1000 generations. The resulting trees were visualized using Figtree v.1.4.3 (https://github.com/rambaut/figtree/releases (accessed on 16 August 2023)).

5. Conclusions

In this study, we conducted comparative genomic and phylogenetic analyses of the cp genomes of three Carthamus species. By conducting comparative genomic analyses, we were able to identify several findings. Firstly, our analysis revealed that the Carthamus species exhibited a high level of conservation in terms of genome structure, gene content, IR boundaries, repeats, and codon preference; this suggests that these species have maintained a relatively stable genomic framework throughout their evolution. However, we also observed some sequence variation in C. lanatus, suggesting that C. lanatus was more distantly related to C. tinctorius and C. persicus. To better understand the phylogenetic relationships of the three Carthamus species within the tribe Cardueae, we performed a phylogenomic analysis using the cp genomes, nrITS, and angiosperms353 dataset. This analysis greatly improved the support and resolution of the phylogenetic trees. Our phylogenomic analysis provided insights into the relationships between Carthamus and other closely related genera. We found that Carthamus was closely related to species such as Carduncellus, Femeniasia, Phonus, and Centaurea. This information contributes to our understanding of the relationships of these species. Overall, our study enriches the knowledge of the cp genomes of Carthamus species and provides a valuable reference for future research on their evolution and origin.