As the largest group in the family Ancylistanceae (Entomophthorales), the Conidiobolus s.l. has a long list of 80 records in the Index Fungorum¹. Most Conidiobolus s.l. species are saprobic and widely distributed in plant debris and soil, while some members are parasitic to insects (Humber, 2012; Gryganskyi et al., 2013; Nie et al., 2020b), and a few species were proposed as pathogens of mammals and humans (Vilela et al., 2010; Vilela and Mendoza, 2018). A recent multi-locus phylogenetic analysis divided the genus Conidiobolus s.l. into four lineages (Nie et al., 2020b), and consequently, four genera Capillidium, Conidiobolus s.s., Microconidiobolus, and Neoconidiobolus were delimited. The genus Conidiobolus s.s. holds 17 species only and is characterized by microspores arising from conidia (Nie et al., 2020a,b). Currently, four nuclear and four mitochondrial genomes of Conidiobolus s.l. are available in GenBank² and JGI,³ involving only one mitogenome of the Conidiobolus s.s. Remarkably, Capillidium heterosporum was the first fungus whose nuclear genome harbors the gene families CYP5014 and CYP5136 encoding cytochrome P450s, and it contained the second greatest number of introns among basal fungi (Nie et al., 2019; Wang et al., 2020).

Mitochondria of eukaryotes were considered to be evolved from endosymbiotic Alphaproteobacteria (Lang et al., 1999). As an important organelle in the eukaryotic cell, mitochondria are in charge of energy production through oxidative phosphorylation and cellular respiration (Henze and Martin, 2003; Chan, 2006; Martin et al., 2015). Except for the anaerobic mitochondrion-free Neocallimastigomycota, fungal mitogenomes have a wide range of sizes (Aguileta et al., 2014; Sandor et al., 2018). As reported, the smallest is 12,100 bp in the basal fungus Rozella allomycis (James et al., 2013), and the largest is 287,403 bp of an ectomycorrhizal fungus Tuber calosporum (Li X. et al., 2020). Most fungi possess a circular mitochondrial genome (Sandor et al., 2018), and a few are known as linear such as Kluyveromyces marxianus (Inokuma et al., 2015) and Synchytrium endobioticum (van de Vossenberg et al., 2018). Fungal mitogenomes typically harbor indefinite numbers of transfer RNA genes (trn), two ribosomal RNA genes (rnl and rns), and 14 core protein-coding genes (PCGs) for NADH dehydrogenase subunits (nad1, nad2, nad3, nad4, nad4L, nad5, and nad6), cytochrome oxidase subunits (cox1, cox2, and cox3), apocytochrome b (cob), and ATP synthase subunits (atp6, atp8, and atp9) (Lang, 2018). A few of these genes are sometimes lost (James et al., 2013), while two other genes encoding the ribosomal protein S3 (rps3) and the RNA subunit of RNase P (rnpB) are present in certain fungal mitogenomes (Franco et al., 2017).

Aside from genomic size and gene organization, fungal mitogenome comparative analyses have widely paid attention to intron distribution and variation (Nadimi et al., 2016; Nie et al., 2019; Zhang et al., 2019; Li Q. et al., 2020). The number and length of introns vary and contribute to mitogenome size (Friedrich et al., 2012; Torriani et al., 2014). For instance, Clavaria fumosa mitogenome contains 93,365 bp of intronic sequence, accounting for 36.36% (Wang et al., 2020). Two groups (I and II) of mitochondrial introns were recognized based on secondary structure and splicing mechanism (Saldanha et al., 1993). Group I mitochondrial introns frequently exist in fungal mitogenomes, while group II introns dominate in plant mitogenomes (Lang et al., 2007). Group I introns are phylogenetically divided into six main subgroups (IA, IA3, IB, IC1, IC2, and ID), with open reading frames (ORFs) usually coding homing endonucleases (HEs: LAGLIDADG and GIY-YIG). Mitogenomic introns also vary in position among species and even isolates and, consequently, are divided into different position classes (Pcls) (Beaudet et al., 2013; Wang et al., 2018). Recently, a novel nomenclature of mitochondrial introns has been proposed to avoid confusion when comparing different fungal mitogenomes, and the nomenclature system was based on: (1) three-letter abbreviation of host scientific name; (2) host gene name; (3) one capital letter P (for group I introns), S (for group II introns), or U (for introns unclassified); and (4) intron insertion site in the host gene according to the cyclosporin-producing fungus Tolypocladium inflatum (Zhang and Zhang, 2019). In general, mitogenomic introns are highly similar and homologous in the same Pcl (Ferandon et al., 2010) and have an implication for inferring fungal evolution (Sandor et al., 2018). Besides the above characteristics, the codon usage, repetitive elements, and tRNA types have also been used to evaluate the evolutionary dynamics of the fungal mitogenomes (Yuan et al., 2017; Abboud et al., 2018; Li Q. et al., 2020, Li et al., 2021). Mitochondrial genomes are inherited by a single parent and have a higher number of copies, causing a faster evolution than nuclear (Ballard and Whitlock, 2004; Pantou et al., 2006; Cameron, 2014). Therefore, the core conserved PCGs were always used for phylogenetic analyses in both basal and higher fungi (Zhang et al., 2017; Nie et al., 2019; Cai et al., 2021; Li et al., 2021).

According to the NCBI database,⁴ only 29 complete mitogenomes of basal fungi were available. In this study, we reported and performed comparative analyses of six additional basal fungi including Conidiobolus chlamydosporus, Conidiobolus lichenicolus, Conidiobolus mycophagus, Conidiobolus polyspermus, Conidiobolus polytocus, and Conidiobolus taihushanensis. The aims of this study are: (1) to characterize the organization, codon usage, and repetitive element of Conidiobolus s.s. mitogenomes, (2) to reveal the diversity of intron and intronic ORFs among Conidiobolus s.s. mitogenomes, and (3) to investigate the taxonomic status of Conidiobolus s.s. in the subphylum Entomophthoromycotina based on the concatenated amino-acid sequences of 14 PCGs in mitogenomes. This work will promote the understanding of origin, evolution, and phylogeny of entomophthoroid fungi and related species.

Fungal mitogenomes differ greatly in gene content, gene arrangement, repeat elements, and intronic content among species and even closely related ones (Sandor et al., 2018; Li Q. et al., 2020, Li et al., 2021). These differences contain important information about the origin and evolution of species (Sankoff et al., 1992). In this article, we focused on these characteristics to evaluate the evolution of Conidiobolus s.s.

The seven Conidiobolus s.s. mitogenomes (26,612–48,417 bp) were moderate in size compared with those of other basal fungi (12,100–225,604 bp; Nie et al., 2019). The size of mitogenomes was mainly determined by its introns and intergenic regions (Table 1) as denoted elsewhere (Sandor et al., 2018; Liu W. et al., 2020). A much wider range was observed in a single genus of higher fungi. Taking Schizosaccharomyces as an example, its mitogenome sizes ranged from approximately 19,000–80,000 bp (Bullerwell et al., 2003). The wide size range of fungal mitogenomes might imply a variable function in oxidative phosphorylation and cellular respiration.

All the 14 conserved PCGs were all detected in the Conidiobolus s.s. mitogenomes, except that the atp8 gene was absent in the mitogenomes of C. chlamydosporus, C. polyspermus, and C. polytocus. This kind of gene loss is not uncommon. The atp9 gene was lost from the Trichoderma gamsii mitogenome (Kwak, 2021). The nad3 gene was not annotated in the Colletotrichum lindemuthianum mitogenome (De-Queiroz et al., 2018), neither was other NADH genes in some yeast mitogenomes (Friedrich et al., 2012). Most scholars considered that some mitochondrial genes have been transferred to the nuclear genome during the evolution of eukaryotes (Adams and Palmer, 2003). The arrangement of mitochondrial genes in fungal mitogenomes reflected their phylogenetic relationship to certain degrees (Zhang et al., 2018; Wang et al., 2020). As expected, the gene orders were identical in the monophyletic Conidiobolus s.s., which was confirmed by the phylogenomic analyses (Figure 6). The gene direction in C. heterosporum mitogenome was almost opposite to others, implying its distant phylogenetic relationship to other species. As a complement to gene order analysis, the mitogenomic synteny analysis showed more detail about gene length and rearrangement (Figure 6). Basal fungal groups frequently exhibited intra-genus gene rearrangements (Li et al., 2019, 2021; Zhang et al., 2020), similar to the result herein. It is worth to note that the cox1 gene in C. taihushanensis mitogenome was split by the atp8 gene. This phenomenon often appeared in basal fungi, especially in the mitogenomes of Glomeromycetes (Pelin et al., 2012; Nadimi et al., 2016; Arcila-Galvis et al., 2021). It may be a long size and abundant introns of cox1 gene that caused a trans-splicing of cox1 genes located on different RNA molecules and transcribed separately (James et al., 2013; Nie et al., 2019). It is also worth to note that the rps3 gene was exhibited in M. nodosus and N. thromboides mitogenomes. The rps3 gene encodes a key protein component of the ribosome for protein translation, and it is often fused with an rnl intron that harbors an ORF homologous to HEs or a partial sequence of cox1 gene (Bullerwell et al., 2000; Sethuraman et al., 2009; Franco et al., 2017; Zhang et al., 2019).

As shown in Figure 2, the lengths of atp6, cox2, nad1, nad2, and nad6 varied among the six Conidiobolus s.s. mitogenomes species. The GC contents of all PCGs varied across the six mitogenomes, which indicated that the conserved PCGs of Conidiobolus s.s. mitogenomes mutate frequently. The nad4L gene bore the smallest pairwise genetic distance, indicating that it was highly conserved. The Ka/Ks values of the 13 PCGs were all <1, indicating that these genes were subjected to purifying selection (Figure 3).

A total of 41 introns were detected in the cox1 gene that occupied mostly half of the total introns in the Conidiobolus s.s. mitogenomes (Tables 1, 2). Usually, the cox1 gene contains many introns in fungal mitogenomes causing its large variation, and it was not suitable as DNA barcode to identify fungal species (Freel et al., 2015; Nadimi et al., 2016; Sandor et al., 2018). Pcls analysis of cox1 gene and other core PCGs was often used to evaluate the dynamics of introns in basidiomycetes mitogenomes. For example, two large-scale intron loss events were detected in the evolution of Boletales species based on the Pcls analysis of the cox1 gene (Li et al., 2021). The Pcls analysis of several core PCGs that contained introns was conducted in 27 Agaricales species. The result showed that few Pcls were widely distributed indicating that these introns may be inherited from the common ancestors, while the same Pcls were detected in distant species from different lineages providing the evidence of intron loss/gain in the evolution of Agaricales species by intron transfer (Huang et al., 2021). This study revealed that Pcl of Ancylistanceae introns varied greatly. Some introns (e.g., P19 and P30) were widely distributed in the cox1 gene and P6 in the cob gene. These introns may be inherited from the ancestors of Ancylistanceae. This phenomenon indicated the potential transfer of introns in the evolution of Ancylistanceae. However, the intronic comparative analysis was conducted within only 10 Ancylistanceae mitogenomes, so the intron loss/gain events were absent of strong evidences.

Repetitive elements in six Conidiobolus s.s. mitogenomes were obtained by BLAST analysis against themselves, resulting in a length variation of 0.2–5.8%. Strangely, C. chlamydosporus mitogenome was detected in most tandem repeats (50), while it possessed the smallest mitogenome size among six Conidiobolus s.s. mitogenomes. With BLAST to their nuclear genomes, 62–150 aligned fragments were examined between nuclear and mitochondrial genomes of the six Conidiobolus s.s. species. This result indicated that the natural gene fragment transfer events may have occurred between mitogenomes and nuclear genomes during the evolution of six Conidiobolus s.s. species.

Traditionally, the ITS, LSU rDNA sequences, and morphological characteristics were widely used in species identification and evolution (Zheng et al., 2007; Zhao et al., 2021; Zong et al., 2021). In addition, single- or multi-gene loci, SSRs, and single-nucleotide polymorphisms (SNPs) have been widely used to study the evolution and polymorphism of fungi during the last two decades (Tsykun et al., 2017). Phylogenetic analyses based on multi-gene loci and even whole genomic data developed rapidly as the cost of sequencing decreased (Spatafora et al., 2016; Nie et al., 2019). Especially, SNPs called from whole genome were extensively used to study polymorphisms in fungi (Duan et al., 2018; Ju et al., 2020), and SNP-based genome-wide association studies (GWAS) were widely used to explore the control mechanism for wheat leaf rust (Liu F. et al., 2020).

The genus Conidiobolus s.l. was divided into four clades representing four genera: Capillidium, Microconidiobolus, Neoconidiobolus, and Conidiobolus s.s. based on the multiple genetic locus phylogeny (Nie et al., 2020b). Phylogenomic analysis based on the 14 PCGs of mitogenomes revealed that six Conidiobolus s.s. species were grouped together with three related species: C. heterosporum, M. nodosus, and N. thromboides in the Entomophthoromycotina lineage. This phylogram was consistent with our previous study and the phylogenomic analysis based on the 192 orthologous proteins of nuclear genomes (Spatafora et al., 2016; Nie et al., 2019). Also, this phylogram revealed that four lineages among 10 Conidiobolus s.l. species demonstrated that 14 mitochondrial PCGs are suitable to clarify phylogenetic relationships among this fungal groups.

In our previous study, the complete mitogenome of C. heterosporum (synonym with Conidiobolus heterosporus) has been reported, and the comparative analysis with other 22 mitogenomes of basal fungi revealed a high variation in mitogenomes sizes, gene order, GC content, genetic code, etc. (Nie et al., 2019). It was worth to note that the horizontal transfer event of C. heterosporum introns has failed to detect strong evidence due to rarely available mitogenomes of basal fungi. In this study, we have sequenced, assembled, and annotated six additional mitogenomes of basal fungi in the genus Conidiobolus s.s., and the comparative analysis has also been provided to give a comprehensive evolution of mitogenomes in six Conidiobolus s.s. species. The result confirmed that the number and length of intron were the main factors contributing to size variations of Conidiobolus s.s. mitogenomes and confirmed our previous taxonomical treatment of Conidiobolus s.l. Meanwhile, the comparison within six Conidiobolus s.s. mitogenomes showed high intronic variations and gene rearrangement. However, the evidence of intron loss/gain events needs further comparative analysis by using more mitogenomes in the family Ancylistanceae, even in the order Entomophthorales. Fortunately, more and more reported mitogenomes of basal fungi will help us to promote the understanding of its evolution and phylogeny in the future. In addition, mitogenomic sequences from a sufficient number of individuals within a certain species can play essential potential in population heredity, pan-genome, and species definition (Duan et al., 2018; McCarthy and Fitzpatrick, 2019; Nie et al., 2019).

We have sequenced, assembled, and annotated six Conidiobolus s.s. mitogenomes. The result confirmed that the number and length of introns were the main factors contributing to size variations of Conidiobolus s.s. mitogenomes. Phylomitogenomics backed up our previous taxonomical treatment of Conidiobolus s.l., i.e., dividing into four genera. Comparative analysis within Conidiobolus s.s. mitogenomes showed a high gene synteny along with deep intronic variations and gene rearrangement. Potential intron loss/gain and transfer events occurred during the evolution of the family Ancylistanceae.

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 in the article/Supplementary Material.

BH and XL conceived and designed the experiments. YN, HZ, ZW, and ZZ analyzed the data. YN, XL, and BH wrote the manuscript. All authors read and approved the final manuscript.

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.

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.