Characterization of the Plastid Genomes of Four Caroxylon Thunb. Species from Kazakhstan

The family Chenopodiaceae Vent. (Amaranthaceae s.l.) is known for its taxonomic complexity, comprising species of significant economic and ecological importance. Despite its significance, the availability of plastid genome data for this family remains limited. This study involved assembling and characterizing the complete plastid genomes of four Caroxylon Thunb. species within the tribe Salsoleae s.l., utilizing next-generation sequencing technology. We compared genome features, nucleotide diversity, and repeat sequences and conducted a phylogenetic analysis of ten Salsoleae s.l. species. The size of the plastid genome varied among four Caroxylon species, ranging from 150,777 bp (C. nitrarium) to 151,307 bp (C. orientale). Each studied plastid genome encoded 133 genes, including 114 unique genes. This set of genes includes 80 protein-coding genes, 30 tRNA genes, and 4 rRNA genes. Eight divergent regions (accD, atpF, matK, ndhF-ndhG, petB, rpl20-rpl22, rpoC2, and ycf3) were identified in ten Salsoleae s.l. plastid genomes, which could be potential DNA-barcoding markers. Additionally, 1106 repeat elements were detected, consisting of 814 simple sequence repeats, 92 tandem repeats, 88 forward repeats, 111 palindromic repeats, and one reverse repeat. The phylogenetic analysis provided robust support for the relationships within Caroxylon species. These data represent a valuable resource for future phylogenetic studies within the genus.


Introduction
Chenopodiaceae Vent.(Amaranthaceae s.l.) is one of the largest and most ancient plant families inhabiting desert and semi-desert regions worldwide [1,2].The Chenopodiaceae family encompasses approximately 1700 species distributed among around 110 genera [3].Ecologically, representatives of the Chenopodiaceae family play crucial roles in desert ecosystems, serving as vital food sources for herbivores and contributing significantly to soil stabilization [4].One of the family's largest and most important tribes is Salsoleae s.l.[5,6].The precise number of genera in Salsoleae s.l.worldwide remains uncertain, with estimates ranging from 32 to 98 [5].The species of Salsoleae s.l. are widely distributed across desert and semi-desert regions spanning Central Asia, the Middle East, Africa, and Europe [7,8].Among these members is the genus Caroxylon Thunb., formerly classified as a section within the Salsola L. before being reinstated as a distinct genus [9].According to POWO (Plants of the World Online) [10], the genus comprises 128 species globally, with nine of them found in Kazakhstan.These species are economically significant as forage plants and sources of medicinal compounds [4,7].
Akhani et al. [9] performed an extensive phylogenetic study on the Salsoleae s.l.utilizing sequences from the nuclear ribosomal internal transcribed spacer and the plastid psbB-psbH region.Certain representatives previously classified under Salsola s.l.have been redistributed among reinstated or recently established genera [9].For instance, Salsola canescens (Moq.)Boiss., previously included in the Salsola sect.Belanthera Iljin under the name S. boissieri Botsch.[19] was transferred to the genus Caroxylon as Caroxylon canescens (Moq.)Akhani [9].However, Sukhorukov et al. [20] indicated uncertainty regarding the taxonomic transfer of S. canescens to the genus Caroxylon, and they established a new genus, Akhania, for this species based on molecular phylogeny using ITS and psbB-psbH nucleotide sequences.Furthermore, using three DNA barcodes, Wen and co-authors [6] suggested that all species of tribe Salsoleae s.l. were composed of three monophyletic subunits: Salsolea s.str., the Kali clade, and Caroxylonea.Despite the comprehensive phylogenetic study conducted by Akhani et al. [9], not all species from Central Asia, including Kazakhstan, were included in the analysis.
In addition to the numerous phylogenetic analyses conducted on Salsoleae s.l.species utilizing nuclear and plastid genome markers, population genetics analysis was also performed.The Amplified Length Polymorphism (AFLP) technique was employed to identify and use AFLP markers to study genetic relationships in four Salsola species [21].Inter simple sequence repeat (ISSR) and start codon targeted (SCoT) molecular markers were used to analyze the genetic relationships between the different species of Salsola [22].Seventeen simple sequence repeat (SSR) markers of Beta were employed in the cross-genera amplification of five morphologically distinct invasive Salsola taxa [23].However, only six of them were successfully amplified within the studied Salsola taxa [23].Additionally, several studies were conducted to assess the cross-genera transferability of these SSR markers to Salsola species [24,25].However, there is a notable absence of population structure analyses in representatives of Caroxylon.
The SSR markers are valuable genetic markers extensively employed in population studies [26][27][28][29].These markers, composed of tandemly repeated motifs, offer unique advantages in elucidating genetic variation and evolutionary relationships within and among species [30,31].While SSRs are widespread across the genomes of diverse organisms [32], cpSSRs specifically reside within the chloroplast genome, primarily found in plant cells [26].Despite the rising popularity of SSR markers in plant population studies, there is a noticeable absence of research on economically significant Caroxylon species.
Plastid genome characterization studies of representatives from the tribe Salsoleae s.l.have been limited, with only a few published investigations.Specifically, Li et al. [33] examined Salsola abrotanoides, while Xie et al. [34] investigated Caroxylon passerinum.Comparative analysis of plastid genome data helps reconstruct phylogenetic trees and provides valuable information for understanding the evolutionary relationships among plant species [35,36].The rapid advancement of next-generation sequencing technology has greatly improved the efficiency and accessibility of obtaining complete plastid genome nucleotide sequences [37].Sequencing plastid genomes in plants is crucial for advancing various fields of research, including evolutionary biology [38], taxonomy [39], biogeography [40], breeding [41], and conservation [42].However, despite the widespread use of plastid genome data for comparative analysis, no comparative studies have been conducted on Caroxylon plastid genomes.
Using the nomenclature based on molecular evidence [6,9], we studied the sequencing, assembly, and annotation of plastid genomes of four Caroxylon species collected in Kazakhstan: C. orientale, C. nitrarium, C. dzhungaricum, and C. laricinum.These species thrive in rocky and clay soils, serving as essential forage for herbivores during autumn and winter.C. orientale and C. nitrarium are widely distributed throughout Kazakhstan's territory.Furthermore, we conducted a comparative analysis to characterize these plastomes, comparing them with related taxa plastomes available in GenBank.The objectives of this study encompassed providing plastid genome data for four newly sequenced Caroxylon species; comparing the structure of their plastid genomes and identifying variable regions suitable as potential DNA barcoding markers for species identification and phylogenetic analysis; exploring repeat elements, including simple sequence repeats, tandem repeats, forward repeats, palindromic repeats, and reverse repeats within the analyzed plastomes; and utilizing common protein-coding gene sequences for constructing a phylogenetic tree, aiming to clarify the phylogenetic relationships among the studied species.

Plastome Analysis by Sliding Window
We utilized DnaSP 6 software to perform a sliding window analysis aimed at determining the nucleotide diversity (Pi) value within the 80 protein-coding genes in plastid genomes of the Caroxylon species.Based on the sequence alignment of common proteincoding genes, we identified eight hypervariable regions: accD, atpF, matK, ndhF-ndhG, petB, rpl20-rpl22, rpoC2, and ycf3.Seven of these regions (accD, atpF, matK, petB, rpl20-rpl22, rpoC2, and ycf3) are situated within the LSC region, while only one region (ndhF-ndhG) is found in the SSC region (Figure 2).Among the variable regions identified, rpl20-rpl22 exhibited the highest Pi value at 0.05711 (Table 3).
The nonsynonymous (Ka) and synonymous (Ks) substitutions were calculated using DNASP 6 between studied plastid genomes (Table 3).The results suggested that the ratios of Ka/Ks in six out of ten genes listed in Table 3 were >1, indicating that they were under positive selection.rpoC2, and ycf3) are situated within the LSC region, while only one region (ndhF-ndhG) is found in the SSC region (Figure 2).Among the variable regions identified, rpl20-rpl22 exhibited the highest Pi value at 0.05711 (Table 3).The nonsynonymous (Ka) and synonymous (Ks) substitutions were calculated using DNASP 6 between studied plastid genomes (Table 3).The results suggested that the ratios of Ka/Ks in six out of ten genes listed in Table 3 were >1, indicating that they were under positive selection.

IR and SC Regions Boundary Analysis
Our analysis investigated the boundaries of the IR-SSC and IR-LSC regions in the plastomes of four Caroxylon species, comparing them with the reference sample, C. passerinum.We observed that the gene rps19 flanked the junction between the LSC and IRb regions in the four Caroxylon species, whereas in the reference sample C. passerinum, the rps19 gene, including a pseudogene, was located within the LSC region.Conversely, the

IR and SC Regions Boundary Analysis
Our analysis investigated the boundaries of the IR-SSC and IR-LSC regions in the plastomes of four Caroxylon species, comparing them with the reference sample, C. passerinum.We observed that the gene rps19 flanked the junction between the LSC and IRb regions in the four Caroxylon species, whereas in the reference sample C. passerinum, the rps19 gene, including a pseudogene, was located within the LSC region.Conversely, the ndhF gene in C. passerinum was positioned to flank the junction between the SSC and IRa regions, while in the four Caroxylon samples, this gene was situated within the SSC region.The ycf1 gene exhibited a consistent pattern across all five samples, spanning the boundaries of the SSC and IRb regions.Additionally, a duplicated copy of the ycf1 gene was consistently observed at the junction between the SSC and IRa regions in each sample (Figure 3).ndhF gene in C. passerinum was positioned to flank the junction between the SSC and IRa regions, while in the four Caroxylon samples, this gene was situated within the SSC region.The ycf1 gene exhibited a consistent pattern across all five samples, spanning the boundaries of the SSC and IRb regions.Additionally, a duplicated copy of the ycf1 gene was consistently observed at the junction between the SSC and IRa regions in each sample (Figure 3).

Repeat Sequence Analysis
We identified 814 simple sequence repeats (SSRs) across the four Caroxylon plastomes using the MISA tool.SSRs vary among the four Caroxylon plastomes, ranging from 198 in C. nitrarium to 208 in C. dzhungaricum.The analysis revealed that mononucleotide repeats were the most abundant motifs, representing 72.97% of the total SSRs.Dinucleotide repeats followed, accounting for 19.04%, while tetranucleotide repeats constituted 4.18%.Hexanucleotide repeats, representing 0.25% of the total SSRs, were uniquely identified within the C. laricinum plastome.The vast majority of mononucleotide repeats consisted of A/T sequences (71.38%), with a minor portion (2.19%) comprising C/G sequences.Regarding dinucleotide repeats, AT/AT repeats were predominant, representing 54.84%, while AC/GT repeats constituted only 6.45%, and AG/CT repeats made up 38.71% (Table 4).Most of the identified SSRs were located in the non-coding and LSC regions of Caroxylon plastomes (Supplementary File S1).

Repeat Sequence Analysis
We identified 814 simple sequence repeats (SSRs) across the four Caroxylon plastomes using the MISA tool.SSRs vary among the four Caroxylon plastomes, ranging from 198 in C. nitrarium to 208 in C. dzhungaricum.The analysis revealed that mononucleotide repeats were the most abundant motifs, representing 72.97% of the total SSRs.Dinucleotide repeats followed, accounting for 19.04%, while tetranucleotide repeats constituted 4.18%.Hexanucleotide repeats, representing 0.25% of the total SSRs, were uniquely identified within the C. laricinum plastome.The vast majority of mononucleotide repeats consisted of A/T sequences (71.38%), with a minor portion (2.19%) comprising C/G sequences.Regarding dinucleotide repeats, AT/AT repeats were predominant, representing 54.84%, while AC/GT repeats constituted only 6.45%, and AG/CT repeats made up 38.71% (Table 4).Most of the identified SSRs were located in the non-coding and LSC regions of Caroxylon plastomes (Supplementary File S1).The lengths of the identified simple sequence repeats varied from 6 bp in C. laricinum to 20 bp in C. dzhungaricum plastid genomes.The majority of SSRs were eight base pairs in length, with 93, 97, 100, and 97 repeats identified in the plastid genomes of C. laricinum, C. orientale, C. dzhungaricum, and C. nitrarium, respectively.The categorization of repeats with different lengths is presented in Figure 4. Furthermore, our analysis identified tandem, forward, palindromic, and reverse repeat types in the plastomes of the four Caroxylon species.A total of 292 repeats were detected, comprising 92 tandem repeats, 88 forward repeats, 111 palindromic repeats, and just one reverse repeat.The reverse repeat was found solely within the plastome of C. dzhungaricum (Figure 5).Furthermore, our analysis identified tandem, forward, palindromic, and reverse repeat types in the plastomes of the four Caroxylon species.A total of 292 repeats were detected, comprising 92 tandem repeats, 88 forward repeats, 111 palindromic repeats, and just one reverse repeat.The reverse repeat was found solely within the plastome of C. dzhungaricum (Figure 5).

Phylogenetic Analysis
To elucidate the phylogenetic relationships among the ten species from the Salsoleae tribe, we reconstructed the phylogenetic tree using the Maximum Likelihood (ML) and Bayesian Inference (BI) methods.We utilized nucleotide sequences of 80 protein-coding genes, commonly found in 12 chloroplast genomes, including two outgroup species.Two

Phylogenetic Analysis
To elucidate the phylogenetic relationships among the ten species from the Salsoleae tribe, we reconstructed the phylogenetic tree using the Maximum Likelihood (ML) and Bayesian Inference (BI) methods.We utilized nucleotide sequences of 80 protein-coding genes, commonly found in 12 chloroplast genomes, including two outgroup species.Two datasets, (1) nucleotide sequences derived from protein-coding genes and (2) nucleotide sequences encompassing the entire plastid genome, were employed to construct the phylogenetic trees.The phylogenetic analyses based on the ML and BI methods grouped all ten samples into a single clade with strong bootstrap support.The species examined in this study (C.orientale, C. nitrarium, C. dzhungaricum, and C. laricinum) formed a distinct subclade alongside species obtained from GenBank (C.passerinum), thus constituting a subgroup within the Caroxylon clade (Figure 6).

Phylogenetic Analysis
To elucidate the phylogenetic relationships among the ten species from the Salsoleae tribe, we reconstructed the phylogenetic tree using the Maximum Likelihood (ML) and Bayesian Inference (BI) methods.We utilized nucleotide sequences of 80 protein-coding genes, commonly found in 12 chloroplast genomes, including two outgroup species.Two datasets, (1) nucleotide sequences derived from protein-coding genes and (2) nucleotide sequences encompassing the entire plastid genome, were employed to construct the phylogenetic trees.The phylogenetic analyses based on the ML and BI methods grouped all ten samples into a single clade with strong bootstrap support.The species examined in this study (C.orientale, C. nitrarium, C. dzhungaricum, and C. laricinum) formed a distinct subclade alongside species obtained from GenBank (C.passerinum), thus constituting a subgroup within the Caroxylon clade (Figure 6).The phylogenetic analyses performed on the entire plastid genome sequences revealed results consistent with those obtained from the dataset of protein-coding gene sequences.The ML and BI trees generated from the complete plastid genome sequences are included in Supplementary File S2.

Discussion
Data from plastid genomes offer valuable insights for taxonomic studies aimed at assessing evolutionary relationships and conducting comparative analyses across various taxonomic levels [43][44][45][46].In this investigation, we obtained and analyzed four Caroxylon plastid genomes using Illumina sequencing technology.The comparative analysis revealed consistent genome structure and gene count across these examined genomes.The study of plastid genomes of C. orientale, C. nitrarium, C. dzhungaricum, and C. laricinum revealed a consistent presence of 133 genes in each species, with 114 genes being unique, including 80 protein-coding genes, 30 tRNA genes, and 4 rRNA genes (Table 1).
Compared with the plastome annotation of C. passerinum [34], variations were observed in the number of protein-coding and tRNA genes.Specifically, the total number of protein-coding genes in the C. passerinum plastome was reported as 89, including duplicated genes.However, in the four newly sequenced plastomes, this count was 88, as the pseudogene rps19 was not annotated due to its short length.Conversely, while the trnG-GCC and trnK-UUU genes were not annotated in the C. passerinum plastome [34], these genes were identified in the four Caroxylon species analyzed in this study, resulting in a total of 37 tRNA genes, including duplicated genes.
Our sliding window analysis investigation further identified eight relatively variable regions, encompassing accD, atpF, matK, ndhF-ndhG, petB, rpl20-rpl22, rpoC2, and ycf3 (Figure 2), where six out of ten genes in those regions were under positive selection (Table 3).Noteworthy among these findings is the recognition of the matK gene as a core plant barcode by the CBOL Plant Working Group [47].Moreover, our examination of nucleotide diversity demonstrated similar variability in other species.For instance, Liu et al. [48] and Almerekova et al. [49] identified variability in the accD region among Quercus and Juniperus species, respectively.Rodda and Niissalo [50] observed variability in the accD and ndhF regions within Hoya species plastomes.Additionally, variability in the rpl20 region among Aerides species' plastid genome was noted by Chen et al. [51], while Ding et al. [52] reported variability in the rpl22 region among the Clethra plastid genome.The identified relatively variable regions have the potential to serve as molecular markers for phylogenetic studies of Caroxylon species.
The boundaries between IR and SC regions are conserved in studied Caroxylon species (Figure 3).The IRb/LSC boundary was consistently found within the rps19 gene among the four newly sequenced species analyzed in this study.Similarly, the IRb/SSC boundary was located within the ycf1 gene, while the IRa/SSC boundary was identified within the duplicated copy of the ycf1 gene.These results are in accordance with previous studies conducted on different species [53,54].Notably, no major changes in the position of IR regions were observed in the studied plastid genomes, suggesting the absence of substantial structural rearrangements [55,56].Among more closely related species, any observed shifts in IR boundaries tended to be relatively minor [57].
SSRs are widespread throughout plastid genomes across different species and are extensively employed in plant population studies [58][59][60].In this study, the identified SSRs varied in number, with counts ranging from 198 in C. nitrarium to 208 in C. dzhungaricum plastomes, resulting in a total of 814 SSRs (Table 4).Similar to the previous studies [43,61], mononucleotide repeats were the most prevalent motifs, comprising 594 of the total identified SSRs.Polyadenine (poly-A) or polythymine (poly-T) repeats were the most abundant in the plastid genome of four Caroxylon species collected in Kazakhstan, a common phenomenon observed in the plastid genomes of higher plants [62][63][64].According to the previous findings [65,66], most of the identified SSRs were located in the non-coding and LSC regions of the four examined plastomes.The SSRs revealed in this analysis offer valuable resources for investigating the population genetics of Caroxylon species, thereby filling the gaps in population studies within this genus.
The plastid genome significantly conserves structure and gene composition [57], making it a valuable resource for analyzing phylogenetic relationships across diverse taxonomic levels [67].Before this study, phylogenetic relationships within the tribe Salsoleae had been assessed using only a limited number of genes, and the precise taxonomy of the tribe remains unresolved.In this analysis, nucleotide sequences from common protein-coding genes of ten representatives of the Salsoleae tribe were employed, comprising four newly sequenced Caroxylon species.The phylogenetic tree was reconstructed using the Maximum Likelihood (ML) method.The resulting phylogenetic tree exhibited a topology with highresolution values at the clades.Caroxylon was initially considered a section of the genus Salsola [19,68,69], but later it was recognized as a separate genus [9,70].The monophyly of Caroxylon reported in previous studies [6,20] remained consistent with the findings of this study.The assessment of the ML phylogenetic tree suggests that C. passerinum (syn.C. gemmascens) is the oldest species in the five analyzed taxa.In the four Caraxylon species reported in this study, C. laricinum, C. dzhungaricum, and C. orientale have formed a distinct subclade with a high bootstrap value (Figure 6).The fourth species, C. nitrarium, seems to have evolved from a common ancestor in the earlier stage of speciation, which is well agreed with the results obtained based on using universal DNA barcodes [6,9].
The robust support values obtained for the phylogenetic relationships inferred from plastid genome data closely mirrored those derived from nuclear gene data, suggesting the reliable resolution of phylogenetic relationships within this genus by plastid genome data.However, further plastid genome data are needed to comprehensively assess the phylogenetic relationships within the Caroxylon clade.This study represents the first attempt to evaluate phylogenetic relationships using genomic data in this genus and may serve as a valuable resource for future phylogenetic studies of the genus.

Plant Materials and DNA Extraction
Plant leaves of four Caoxylon species were collected from Kyzylorda (C.orientale and C. nitrarium), Almaty (C.dzhungaricum), and West Kazakhstan (C.laricinum) regions of Kazakhstan (Table 5).The voucher herbarium specimens of C. orientale, C. nitrarium, C. dzhungaricum, and C. laricinum were deposited in the Herbarium (AA) of the Institute of Botany and Phytointroduction under accession numbers AA0003263, AA0003264, AA0003265, and AA0003266, respectively.The collected fresh leaves were preserved in silica gel and used for DNA extraction.Total genomic DNA was extracted using the cetyltrimethylammonium bromide (CTAB) protocol [71], and DNA samples were stored at −80 • C until sequencing.

Genome Sequencing, Assembly, and Annotation
The total genomic DNA that passed quality control analysis was used for library preparation using the TruSeq Nano DNA Kit manufactured by Illumina Inc. (San Diego, CA, USA).The plastomes of four Caroxylon species were sequenced using an Illumina NovaSeq 6000 platform (Illumina Inc., USA), which was conducted at Macrogen Inc. (Seoul, Republic of Korea).FastQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc, accessed on 26 January 2024) was used to conduct initial quality control checks on raw sequence data coming from high throughput sequencing pipelines.The Trimmomatic 0.36 software [72] removed adapter sequences from the raw reads.The reads with a quality score over 20 were accepted as good-quality reads.Subsequently, the clean reads were assembled using the NOVOPlasty 4.3.3program [73].Further annotation of the assembled sequences was conducted using the published plastome of Caroxylon passerinum (MW192441) as a reference.The annotation of protein-coding, rRNA, and tRNA genes was performed using GeSeq [74] with manual corrections.The OrganellarGenomeDRAW 1.3.1 tool (OGDRAW) [75] generated a circular gene map of the four Caroxylon species.Finally, the annotated plastome sequences of four Caroxylon species were deposited into the GenBank.

Plastome Analysis by Sliding Window, Ka/Ks Calculation, IR Regions Contraction, and Expansion
To evaluate the nucleotide diversity (Pi) of the plastome sequences, we conducted sliding window analysis using DnaSP 6 software [76].The window length was configured to 600 bp, with a step size of 200 bp.The synonymous (Ka), nonsynonymous (Ks), and

Figure 1 .
Figure 1.Plastid genome maps of the four Caroxylon species.(a) C. orientale, (b) C. nitrarium, (c) C. dzhungaricum, and (d) C. laricinum.Genes from various functional categories are colorized accordingly.The darker gray on the inner circle represents GC content, while the lighter gray denotes AT content.

Figure 1 .
Figure 1.Plastid genome maps of the four Caroxylon species.(a) C. orientale, (b) C. nitrarium, (c) C. dzhungaricum, and (d) C. laricinum.Genes from various functional categories are colorized accordingly.The darker gray on the inner circle represents GC content, while the lighter gray denotes AT content.

Figure 2 .
Figure 2. Sliding window analysis of the common protein-coding genes (window length: 600 bp; step size: 200 bp) in plastid genomes of the Caroxylon species.The vertical axis indicates the nucleotide diversity for each window, while the horizontal axis denotes the midpoint position.

Figure 2 .
Figure 2. Sliding window analysis of the common protein-coding genes (window length: 600 bp; step size: 200 bp) in plastid genomes of the Caroxylon species.The vertical axis indicates the nucleotide diversity for each window, while the horizontal axis denotes the midpoint position.

Figure 3 .
Figure 3. Comparisons of the borders of LSC, IR, and SSC regions among the five plastid genomes JLB (junctions between LSC and IRb regions), JSB (junctions between IRb-SSC regions), JSA (junctions between SSC-IRa regions), and JLA (junctions between IRa-LSC regions).

Figure 3 .
Figure 3. Comparisons of the borders of LSC, IR, and SSC regions among the five plastid genomes JLB (junctions between LSC and IRb regions), JSB (junctions between IRb-SSC regions), JSA (junctions between SSC-IRa regions), and JLA (junctions between IRa-LSC regions).

Figure 4 .
Figure 4.The length of the identified simple sequence repeats in plastomes of C. laricinum, C. orientale, C. dzhungaricum, and C. nitrarium.

Figure 4 .
Figure 4.The length of the identified simple sequence repeats in plastomes of C. laricinum, C. orientale, C. dzhungaricum, and C. nitrarium.

Plants 2024 , 16 Figure 5 .
Figure 5. Number of tandem, forward, palindromic, and reverse repeats in the plastomes of the four Caroxylon species.

Figure 5 .
Figure 5. Number of tandem, forward, palindromic, and reverse repeats in the plastomes of the four Caroxylon species.

Figure 5 .
Figure 5. Number of tandem, forward, palindromic, and reverse repeats in the plastomes of the four Caroxylon species.

Figure 6 .
Figure 6.Phylogenetic tree based on nucleotide sequences from protein-coding genes using the Maximum Likelihood (a) and Bayesian Inference (b) methods.ML bootstrap value and BI posterior probability are given at each node.The species from Kazakhstan were highlighted with the letters "KZ".

Figure 6 .
Figure 6.Phylogenetic tree based on nucleotide sequences from protein-coding genes using the Maximum Likelihood (a) and Bayesian Inference (b) methods.ML bootstrap value and BI posterior probability are given at each node.The species from Kazakhstan were highlighted with the letters "KZ".
The complete plastid genome in four Caroxylon species ranged in size from 150,777 bp in C. nitrarium to 151,307 bp in C. orientale (Table

Table 2 .
Genes identified in plastomes of four Caroxylon species.

Table 3 .
Regions with highly variable sequences and Ka/Ks ratios of genes in plastid genomes of Caroxylon species.

Table 3 .
Regions with highly variable sequences and Ka/Ks ratios of genes in plastid genomes of Caroxylon species.

Table 4 .
Types and numbers of simple sequence repeats in the plastomes of four Caroxylon species.

Table 4 .
Types and numbers of simple sequence repeats in the plastomes of four Caroxylon species.The lengths of the identified simple sequence repeats varied from 6 bp in C. laricinum to 20 bp in C. dzhungaricum plastid genomes.The majority of SSRs were eight base pairs in length, with 93, 97, 100, and 97 repeats identified in the plastid genomes of C. laricinum, C. orientale, C. dzhungaricum, and C. nitrarium, respectively.The categorization of repeats with different lengths is presented in Figure 4.

Table 5 .
Information on collected places of four Caroxylon species.