Eleven species of the genus Xanthoparmelia were collected from different locations in Xinjiang, China, external morphologies are shown in Supplementary Figure S1. Detailed information regarding their collection sites and accession numbers is provided in Supplementary Table S1. Species identification was conducted through morphological analysis, internal anatomical examination, chemical characterization, and phylogenetic analysis of the Internal Transcribed Spacer (ITS) sequences in Supplementary Figure S2. (1) Morphological observation: Color, cracks, and luster of the upper surface of the thallus; shape of lobes; presence or absence of white spots; observation of structures such as isidia, pruina, and soralia, as well as characteristics of pycnidia; color of the lower surface; color of rhizines and their branching status (if any). (2) Anatomical observation: Color of the upper and lower cortices of the thallus; uniformity of the algal layer distribution; color of the medulla; observation of structural characteristics of reproductive organs including apothecia and pycnidia.(3) Spot test: Application of chemical reagents to the cortex and medulla of the thallus, followed by observation of color changes. All specimens are preserved in the Cryptogamic Herbarium of the College of Life Science and Technology, Xinjiang University (Contact person: Anwar Tumur, email: awartumursk@xju.edu.cn). DNA extraction was performed using a fungal DNA extraction kit (Sangon Biotech, Shanghai, China) following the manufacturer’s instructions.

Whole-genome sequencing of the species was carried out on the DNBSEQ sequencing platform (Shanghai, China). The mitochondrial genomes were de novo assembled using GetOrganelle V1.7.4.1 (Jin et al., 2020), a tool specifically optimized for assembling circular organelle genomes. In addition, NOVOPlasty V4.2 (Dierckxsens et al., 2017) was used as an auxiliary assembler. Both tools have been widely and successfully applied in fungal mitochondrial genome studies (e.g., Wu et al., 2021; Anwar et al., 2023). The determination of the circular structure of our mitochondrial assemblies was supported by the following evidence: (1) Published mitochondrial genomes of closely related lichen-forming fungi within the family Parmeliaceae are almost exclusively circular, with no linear mitogenomes reported to date. This provides a strong expectation that Xanthoparmelia species share a similar genomic architecture (e.g., Funk et al., 2018; Bahenuer et al., 2025). (2) The assembly logs generated by both GetOrganelle and NOVOPlasty explicitly annotated all 11 assembled mitochondrial contigs as “circular.” (3) We remapped the sequencing reads to the final assemblies and used SAMtools V1.19.2 depth to assess coverage. As shown in Supplementary Figure S3, read depth is uniform and continuous across the entire genome, including the inferred circular-junction regions, without any detectable drops or gaps. These results collectively support both the completeness of our mitochondrial genome assemblies and the reliability of their circular structure. Subsequently, genome annotation was performed on the assembled sequences using the online tools GeSeq V2.03 (Tillich et al., 2017), MFannot V1.3.3 (Valach et al., 2014), and MITOS2 (Bernt et al., 2013) (available at https://usegalaxy.eu). The annotation results were then imported into Geneious V2022.1.1 (Kearse et al., 2012) to check the start and stop codons of protein-coding genes. Finally, the mitochondrial genome map was generated using OGDraw V1.2 (Lohse et al., 2007).

To evaluate the completeness and uniformity of the newly assembled mitochondrial genomes, whole-genome sequencing depth was calculated with SAMtools V1.19.2 (Li et al., 2009) after read mapping via BWA-MEM V0.7.17 (Li and Durbin, 2009). Coverage plots were generated in R V4.3.1; raw depth is shown in grey and a 201-bp rolling mean (zoo package; Zeileis and Grothendieck, 2005) in red in Supplementary Figure S3. All assemblies exhibited mean depth ≥ 450 × with no region < 10×, confirming genome completeness.

Repetitive sequences are the main cause of genetic recombination and variation. To evaluate the types and distribution of repetitive sequences in mitochondrial genomes, this study analyzed intragenomic repeats, tandem repeats, interspersed repeats, and simple sequence repeats (SSRs) within the genomes. Local BLASTn (Basic Local Alignment Search Tool, BLASTn) (Chen et al., 2015) was used to align each mitochondrial genome against itself. Tandem repeats in mitochondrial genomes were identified using the online tool Tandem Repeats Finder V4.0 (Benson, 1999) (https://tandem.bu.edu/trf/trf.html). Interspersed repeats in mitochondrial genomes were detected via the online software REPuter (Kurtz, 2001) (https://bibiserv.cebitec.uni-bielefeld.de/reputer) with the following parameters: Hamming Distance = 3, Maximum Computed Repeats = 5,000, and Minimal Repeat Size = 30. MISA (Beier et al., 2017) (https://webblast.ipk-gatersleben.de/misa/) was employed to detect SSRs, with the criteria: 10 repeats for mononucleotides, 5 repeats for dinucleotides, 4 repeats for trinucleotides, and 3 repeats for tetra-, penta-, and hexanucleotides. Finally, TBtools V 2.357 (Chen et al., 2020) was used to visualize the distribution of repetitive sequences.

PhyloSuite V1.2.2 (Zhang et al., 2020) was used to extract PCGs from each mitochondrial genome. Subsequently, MEGA V11 (Tamura et al., 2021) was applied to calculate the relative synonymous codon usage (RSCU) of the PCGs in the mitochondrial genomes.

Introns in fungal mitochondria exhibit significant variation. Most eukaryotic mitochondrial genomes typically contain no introns, while species of the genus Xanthoparmelia (within Parmeliaceae) usually harbor varying numbers of introns (Martin et al., 2007; Zhang et al., 2015; Chen et al., 2021). Following the reported method (Chen et al., 2021), introns in the PCGs of Xanthoparmelia mitochondrial genomes were classified into Pcls (Zhang and Zhang, 2019) using the reference genome of Tolypocladium inflatum (NC036382). The specific steps were as follows: PCG without introns were aligned with the reference mitochondrial sequence using MAFFT. Each Pcl consists of introns inserted at the same position in the coding region of a PCG (Cheng et al., 2021). Introns with the same Pcl generally share high sequence similarity and are considered homologous (Férandon et al., 2010). Most different Pcls have low sequence similarity and are associated with non-homologous mobile genetic elements (Huang et al., 2021). The 14 PCG-containing genomes were named based on the insertion positions of introns in the coding regions of the reference genes.

Comparative genomic analysis was performed on 11 newly sequenced mitochondrial genome sequences from this study and 4 Xanthoparmelia mitochondrial genome sequences downloaded from NCBI in Supplementary Table S2. DNASTAR Lasergene V7.1 (https://www.dnastar.com/software/lasergene/) was used to analyze the base composition of Xanthoparmelia mitochondrial genomes. The calculation formulas were: AT skew = [A - T] / [A + T], GC skew = [G - C] / [G + C]. To evaluate the evolutionary rate of coding genes, DnaSP V6.12.03 (Rozas et al., 2017) was used to analyze 14 PCGs, calculating the non-synonymous substitution rate (Ka), synonymous substitution rate (Ks), and their ratio (Ka/Ks). The Ka/Ks ratio indicates the type of selection pressure acting on a gene: A ratio > 1 indicates positive selection. A ratio = 1 indicates neutral evolution. A ratio < 1 indicates purifying selection. MEGA V11 was used to calculate the genetic distances of the 14 PCGs based on the Kimura 2-parameter (K2P) substitution model. Collinearity was analyzed using Mauve (Darling et al., 2004).

In this study, maximum likelihood (ML) and Bayesian inference (BI) methods were used to construct a phylogenetic tree of Parmeliaceae based on 14 PCGs. Lecanora saxigena (NC_042183) was used as the outgroup. The specific workflow was as follows: Sequences were aligned using MAFFT V7.313 (Katoh et al., 2019). Conserved sequences were filtered using Gblocks V0.91 (Castresana, 2000). The 14 PCGs were concatenated using Sequence Matrix (Vaidya et al., 2011). The most suitable evolutionary model for the dataset was selected using Model Finder (Kalyaanamoorthy et al., 2017). ML analysis was performed using IQ-tree V1.6.8 (Nguyen et al., 2015) with the parameters: -m MFP -bb 1,000 -nt AUTO. Bayesian phylogenetic inference was conducted using MrBayes V3.2.7a (Ronquist et al., 2012) with 2 parallel runs (1,000,000 generations). The initial 25% of run results were discarded (burn-in = 0.25) to construct the BI tree. The phylogenetic trees were visualized and edited using Figtree v1.4.4. (http://tree.bio.ed.ac.uk/software/figtree/).

The mitochondrial genomes of all 11 Xanthoparmelia species consist of circular DNA molecules, with sizes ranging from 81,194 bp to 88,245 bp (Figure 1). Among them, X.viriduloumbrina has the largest mitochondrial genome, while X.conspersa has the smallest. The GC content varies from 30.2 to 30.8%, with an average of 30.6%. The AT skews are all negative (−0.006 to −0.013), and the GC skews are all positive (0.057 to 0.063).

Each mitochondrial genome contains 14 PCGs, including 7 NADH dehydrogenase genes (nad1, nad2, nad3, nad4, nad4L, nad5, nad6), 3 cytochrome oxidase genes (cox1, cox2, cox3), 2 ATP synthase genes (atp6, atp8), 1 cytochrome b gene (cob), and 1 ribosomal protein subunit 3 gene (rps3). Except for rps3, these PCGs are conserved mitochondrial genes involved in the oxidative phosphorylation pathway (Li et al., 2020). In addition, two types of ribosomal RNA genes (rns, rnl) are present in all 11 Xanthoparnelia genomes. The number of tRNA genes ranges from 26 to 27 (Supplementary Table S3).

Through BLASTn alignment of the mitochondrial genomes of 11 Xanthoparmelia species, we identified 29, 36, 32, 20, 32, 25, 25, 36, 23, 29, and 29 repeat elements in the mitochondrial genomes of X.camtschadalis, X.conspersa, X.desertorum, X.durietzii, X.orientalis, X.protomatrae, X.taractica, X.tinctina, X.viriduloumbrina, X.weberi, and X.wyomingica, respectively (Supplementary Table S4). There were 385–457 dispersed duplications (Supplementary Table S5) with repeat sequence sizes ranging from 30 to 281 bp, including 199–256 direct repeats, 71–98 palindromic repeats, 54–96 inverted repeats, and 20–42 complementary repeats. Additionally, there were 51–62 simple sequence repeats (SSRs) (Supplementary Table S6) with sizes ranging from 10 bp to 144 bp. Among these, direct repeats and palindromic repeats exhibited high repetition frequencies across all 11 species. The SSRs included 19–23 mononucleotides, 9–13 dinucleotides, 7–17 trinucleotides, 5–7 tetranucleotides, 2–3 pentanucleotides, and 2–4 hexanucleotides. Furthermore, 48–60 tandem repeats were detected (Supplementary Table S7) with repeat unit sizes of 3–34 bp and copy numbers ranging from 1.9 to 24.2 (Figure 2). SSR markers are commonly used as molecular markers in genetic diversity and evolutionary studies (Ahmad et al., 2018). Analysis of the distribution of repeat sequences across the 11 mitochondrial genomes revealed that they are mainly located in introns and intergenic regions (Figure 3).

The start codons and stop codons of 14 PCGs in the mitochondrial genomes of 11 Xanthoparmelia species were compared (Supplementary Table S8). Among the 14 PCGs: The atp8, cox1, cox3, nad1, nad2, nad3, nad4L, and nad5 genes used ATG as their start codon. Both cox2 and rps3 genes used TTA as the start codon. The atp6 gene used ATG and ATA as start codons. nad4 and nad6 used ATC and ATG as start codons, respectively.

Most species used ATG as the start codon for the cob gene, except X.wyomingica and X.viriduloumbrina, which used ATC and ATA, respectively. For stop codons: The atp6, cob, cox2, cox3, nad2, nad3, nad4, nad4L, nad6, and rps3 genes terminated with TAA. The atp8, cox1, and nad1 genes used TAG as the stop codon. Most species used TAA as the stop codon for the nad5 gene, except X.taractica and X.viriduloumbrina, whose nad5 genes terminated with TAG. Studies have shown that codon usage directly affects translation speed and energy requirements. Genes in mitochondria are thought to prefer specific codons to save time and conserve energy needed for cell growth (Sharp, 2005). The codon usage analysis results revealed that the PCGs in the mitochondrial genomes of 11 Xanthoparmelia species exhibited highly similar codon usage preferences (Figure 2). Among the analyzed amino acids: Leucine (Leu), serine (Ser), and arginine (Arg) were each encoded by 6 codons. Methionine (Met) was encoded by only 1 codon. The codons UUU (for Phe), UUA (for Leu), AUA (for Ile), and UAU (for Tyr) had the highest usage frequency across the 11 mitochondrial genomes (Supplementary Table S9). The mitochondrial genomes of Xanthoparmelia showed a high AT content, mainly due to the high-frequency use of A/T bases in their preferred codons. Arginine (Arg) in Xanthoparmelia mitochondrial genomes was primarily encoded by the AGA codon. The results indicated that, except for AGA, the Relative Synonymous Codon Usage (RSCU) values of all other arginine codons (CGU, CGC, CGA, AGG) were below 1, with CGG having an RSCU value of 0. Meanwhile, the RSCU value of UUA (for Leu) was above 3, which may be associated with environmental stress.

There was a significant correlation between intron size and the size of 15 Xanthoparmelia mitochondrial genomes (p < 0.001). The Pearson and Spearman correlation coefficients were 0.9389 and 0.7964, respectively, indicating that intron length has a significant impact on the variation in the size of Xanthoparmelia mitochondrial genomes (Figure 4). A total of 239 introns were detected across the 15 mitochondrial genomes, with each species containing 11 to 24 introns. Studies have shown that intron loss or gain events occurred throughout the entire evolutionary history of the Xanthoparmelia genus. Among these introns, 95% (228 introns) were located in PCGs, and 5% (11 introns) were located in rRNA genes; thus, PCGs are the main reservoirs of introns. Introns were distributed in the cox1, cox3, cob, nad1, nad4L, and nad5 genes. This uneven distribution of introns indicates gene preference, with most introns targeting PCGs.

Introns were classified into different positional classes (Pcls) corresponding to their reference genes based on their insertion sites within the protein-coding regions of the mitochondrial genome. Introns in the cox1 gene of 15 Xanthoparmelia species were categorized into P282, P386, P492, P615, P720, P731, P807, P1057, P1107, and P1125, among which P282, P807, P1057, and P1107 had the widest distribution. The cob, cox3, nad4L, nad5, and nad1 genes contained 1, 1, 1, 3, and 2 types of Pcls, respectively. The Pcls of cox3, nad5, and nad1 were widely distributed across the 15 Xanthoparmelia species. Except for the cox1 gene, X.taractica contained the most intron types, while X.conspersa, X.coreana, X.orientalis, and X.tinctina showed significant intron loss in their cox1 genes, retaining only 4 to 5 intron types (Figure 5). These findings indicate that among the 14 PCGs, the cox1 gene in the mitochondrial genome of Xanthoparmelia species exhibits relatively high variation, and there is a possibility of intron transfer.

Among the 14 PCGs in the mitochondrial genome of genus Xanthoparmelia, the Ka/Ks values of all 14 PCGs are less than 1 (Figure 6a), indicating that these genes are under purifying selection and their sequences evolve more conservatively. The results of genetic distance (K2P) analysis show (Figure 6b) that the 14 PCGs have different genetic distances, which means their evolutionary rates also vary. Among them, the atp6 gene has the largest genetic distance (with an average value of 0.231), suggesting that it exhibits the fastest mutation rate among the 14 PCGs. In contrast, the cob gene has the smallest genetic distance (with an average value of 0.001), indicating that it has high conservativeness.

A detailed comparison of collinearity among the 15 mitochondrial genomes revealed a conserved core structure punctuated by specific rearrangements. The analysis identified nine homologous blocks, with the order and orientation of most blocks being stable across the genus (Figure 7). However, detailed comparison revealed several specific rearrangements: (1) The third block (containing cox1 and cob) was rearranged exclusively in X.durietzii; (2) The fifth block (intergenic between nad3 and nad4L) was rearranged in X.viriduloumbrina; (3) The ninth block underwent independent rearrangements in both X.weberi and X.viriduloumbrina. This pattern indicates that the sequences of PCGs are highly conserved, while structural variation is confined primarily to intergenic spacer regions, which may facilitate genome evolution without disrupting essential coding functions.

The entire phylogenetic tree was divided into 4 clades, including the genera Alectoria, Bryoria, Hypogymnia, Imshaugia, Parmotrema, Pseudevernia and Usnea. The phylogenetic results were consistent with the topological structure reported in previous studies (Funk et al., 2018; Bahenuer et al., 2025). Fifteen species of the genus Xanthoparmelia formed a monophyletic clade, which clustered with Parmotrema with high bootstrap support (100%) and posterior probability (1.00). This indicates a close phylogenetic relationship between the two genera, or they may share a common ancestor X.conspersa, X.coreana, X.orientalis, and X.tinctina clustered closely into a subclade, while X.viriduloumbrina branched closer to the terminal node (Figure 8).