Cladobotryum rhodochroum sp. nov. (Hypocreales, Ascomycota): A New Fungicolous Species Revealed by Morphology, Phylogeny, and Comparative Genomics

Abstract

Species of the ascomycetous genus Cladobotryum (Hypocreales, Hypocreaceae) are ecologically and economically important mycoparasites that cause cobweb disease in cultivated and wild mushrooms. Despite their significance as fungal pathogens and producers of bioactive metabolites, the taxonomy of Cladobotryum remains unresolved due to extensive morphological plasticity, complex teleomorph–anamorph connections, and the presence of cryptic species. This study employs an integrative approach combining micro- and macromorphological characterization, multi-locus phylogeny (ITS, rpb2, and tef-1a), and comparative genomics to clarify the taxonomic position of the Greek isolate Cladobotryum sp. ATHUM 6904, previously designated as an unclassified red-pigmented (URP) strain. Phylogenetic analyses demonstrated that URP strains form a distinct, well-supported clade closely related to C. tenue and C. rubrobrunnescens, yet genetically and morphologically distinct from both. Comparative genomic analyses of isolate ATHUM 6904 and the ex-type strains of C. tenue and C. rubrobrunnescens revealed pronounced divergence in transposable element content, mitochondrial genome architecture, gene order, orthologous gene composition, secondary metabolite biosynthetic potential, and overall genomic distance. Micro- and macromorphological comparisons further supported the differentiation of isolate ATHUM 6904 from both reference species. Based on the combined molecular, morphological, and genomic evidence, the Greek isolate ATHUM 6904 is described as a novel species, Cladobotryum rhodochroum sp. nov.

1. Introduction

Species of the ascomycetous genus Cladobotryum Nees (Hypocreales, Hypocreaceae) are of considerable ecological and applied importance. They are the causal agents of cobweb disease, a mycoparasitic infection that affects both cultivated and wild mushrooms. Hosts include commercially important genera, such as Agaricus, Pleurotus, and Lentinula, as well as numerous wild taxa including Lactarius, Russula, Tricholoma, Laccaria, Mycena, Inocybe, Polyporus, and Helvella [1,2,3,4,5,6]. The mycoparasitic lifestyle of Cladobotryum species is associated with the production of diverse enzymes and secondary metabolites [7,8,9,10], emphasizing their dual importance as both fungal pathogens and sources of valuable biotechnological compounds. These metabolites and enzymes have been explored for applications in biocatalysis, biocontrol, pest management, and the discovery of bioactive natural products [10,11,12,13].

Despite this ecological and applied significance, the taxonomy of Cladobotryum remains challenging. The genus comprises anamorphic species (with known asexual states) that are fungicolous (mycophilic) due to their ability to colonize and form trophic associations with other fungi [4,14]. Several anamorphic species are linked to the teleomorphic genus Hypomyces when a sexual state is present, complicating morphology-based taxonomy and nomenclature [1,15,16,17,18]. The high degree of morphological plasticity, the intricate teleomorph–anamorph connections, and the occurrence of cryptic species complexes necessitate polyphasic approaches that integrate morphology, molecular phylogeny, and increasingly, genomics and metabolomics, to achieve robust species delimitation [6,17].

Currently, the genus Cladobotryum includes approximately 66 described species (Index Fungorum, Available online: http://www.speciesfungorum.org/Names/Names.asp, accessed on 2 October 2025). Several studies have employed molecular markers for species identification and delimitation, particularly the ITS, tef-1a, and rpb2 loci [6,9,19], while only a few species have been analyzed at the whole-genome level, including C. dendroides, C. mycophilum, and C. protrusum [20,21,22].

A recent study on Greek isolates of Cladobotryum [6] revealed a distinct cluster of strains that could not be confidently attributed to any known species according to chemotaxonomic, morphological, and ITS-based phylogenetic evidence. These isolates, provisionally designated as “Unidentified Red-Pigmented” (URP) strains, formed a well-supported clade closely related to C. tenue, with C. rubrobrunnescens appearing basal to this lineage based on ITS phylogeny and chemical profiling. However, it remained uncertain whether the URPs represented novel, independent species or geographically isolated variants of an existing taxon, given the distinct origin of the Greek strains compared to the ex-type cultures of C. tenue and C. rubrobrunnescens.

To address these taxonomic and phylogenetic uncertainties and to further investigate the genomic characteristics of the novel species, the present study integrates multi-locus phylogenetic analysis (ITS, rpb2, and tef-1a markers) with morphological characterization and comparative genomic analysis. The ex-type strains of C. tenue and C. rubrobrunnescens were sequenced and compared to the representative URP isolate Cladobotryum ATHUM 6904. These analyses collectively reveal that the leading URP strain analyzed in this work represents a distinct, previously undescribed species, which is proposed as Cladobotryum rhodochroum sp. nov.

2. Materials and Methods

2.1. Fungal Material/Strains

The strain of Cladobotryum rhodochroum ATHUM 6904 was isolated from a basidioma of Lactarius sp., a species belonging to the order Russulales, that was collected from a forest of Fagus sylvatica on Mt. Zigourolivado, Ag. Nikolaos, Karditsa, Greece, in 2009 [6]. Additionally, strains of Cladobotryum and Hypomyces (anamorph) used in the phylogenetic analysis were isolated from basidiomata of different species collected in various habitats and regions of Greece (Supplementary File S1: Table S1). Furthermore, the ex-type cultures of Cladobotryum tenue CBS152.92 and Cladobotryum rubrobrunnescens CBS176.92 were obtained from the Fungal and Yeast Collection (CBS) of the Westerdijk Fungal Biodiversity Institute (WI-KNAW) Biobank (Utrecht, The Netherlands). Strains of the aforementioned species were used for reference and comparison reasons. All culture isolates under study are safeguarded at the Fungal Culture Collection of the Mycetotheca ATHUM (National and Kapodistrian University of Athens, Athens, Greece).

2.2. Morphological Study

For the morphological study of the fungal colonies, the isolates were cultured in Petri dishes (Ø 90 mm) with Potato Dextrose Agar (PDA, BD Difco) used as the growth medium. Incubation was carried out at a temperature of 23 °C under natural day/night conditions. The macro- and micromorphological characters were studied from the early state of growth until full maturity of the colonies. Mycelium color, texture, and growth, as well as the color of the medium reverse, were considered. In addition, the color reaction of the immature and mature mycelium after the application of KOH (3%) was recorded. Color coding follows Kornerup and Wanscher (1978) [23]. The studied micromorphological characters concern the vegetative and reproductive states and the structures formed during the development of these phases, e.g., the type of hyphae, resistant formations, conidiophores, conidiogenous cells, and conidia. Conidia sizes are given as (MIN)[mean-stdev] − [mean + stdev](MAX) followed by the number of spores measured (n), their length–width ratio (Q), and the mean values of spore length (Lm), width (Wm), and length-width ratio (Qm). Microscopical characters were studied and photographed using a Zeiss AxioImager A1 Differential Interference Contrast (DIC) microscope. Microscopical mounts were stained using either an aqueous solution of Phloxine B 2% with the addition of KOH 3% or plain KOH 3%.

2.3. Molecular Phylogenetic Analyses

Mycelia, which were cultivated as described above, were collected from 30 Greek strains isolated by our group and previously described in depth [6]. The total DNA isolation was performed using 100 mg of ground fungal material using the HigherPurity™ Plant DNA Purification Kit (Canvax, Spain). The isolated DNA samples were used for PCR amplification of the nuclear genes of the second-largest subunit of RNA polymerase II (rpb2) and the translation elongation factor 1-alpha partial gene (tef-1a) [primers: rpb2-5F and rpb2-7cR [24] and EF1-983f and EF1-2218r [17]].

PCR amplification reactions were performed with a KAPA Taq PCR Kit (KAPA Biosystems, Wilmington, MA, USA) in a PTC-200 Gradient Peltier Thermal Cycler (MJ Research, Waltham, MA, USA), according to the manufacturer’s instructions. The amplification protocol for the rpb2 region was as follows: 3 min at 95 °C; 35 cycles of 30 s at 95 °C, 60 s at 55 °C, 2 min at 72 °C; and a final extension of 5 min incubation at 72 °C. PCR amplicons were purified and cleaned using the PCR cleanup kit (NEB, Monarch PCR and DNA Cleanup Kit). All amplicons were sequenced in both directions and assessed using the program SeqMan of Lasergene Suite 11 (DNASTAR Inc., Madison, WI, USA). The final sequences were deposited into GenBank. (Acc. Nos. rpb2: PX610068-PX610097, tef-1a: PX842961-PX84981).

A concatenated dataset of the internal transcribed spacer (ITS), RNA polymerase II second largest subunit (rpb2), and translation elongation factor 1-alpha partial gene (tef-1a) regions was assembled for phylogenetic reconstruction of Cladobotryum/Hypomyces. For the 30 Greek strains, rpb2 and tef-1a sequences were newly generated in this study as mentioned above, while their corresponding ITS sequences had been published previously [6]. ITS, rpb2, and tef-1a sequences of the remaining taxa were retrieved from the NCBI nucleotide database to provide a broader representation of the genus (Supplementary File S1: Table S2). The alignments were created using the E-INS-i method, as implemented in the multiple sequence alignment program MAFFT [25]. Alignment parameters were set to default. The alignments were manually trimmed to remove poorly aligned or ambiguous regions on the sequence ends (Supplementary File S2). Maximum-likelihood (ML) phylogenetic analysis was performed in IQ-TREE v2.1.3 [26,27]. Substitution model selection was conducted with ModelFinder [28], and the best-fit model according to the Bayesian Information Criterion (BIC) was TN + F + R3, which was applied for tree inference.

Branch support was assessed using 1000 ultrafast bootstrap replicates (UFBoot) and 1000 SH-aLRT replicates [29,30]. The resulting ML tree was rooted with Fusarium sp. as the outgroup and visualized in FigTree v1.4.4 (http://tree.bio.ed.ac.uk/software/figtree/).

2.4. DNA Extraction and Whole Genome Sequencing

Mycelia of Cladobotryum rhodochroum ATHUM 6904, Cladobotryum tenue CBS152.92, and Cladobotryum rubrobrunnescens CBS176.92 were collected, and the total DNA isolation was performed using 100 mg of fungal material using the HigherPurity™ Plant DNA Purification Kit (Canvax Reagents S.L., Valladolid, Spain) according to the manufacturer’s instructions. The extracted DNA was checked for quality and quantity using a Nanodrop (ThermoFisher Scientific™, Waltham, MA, USA) and the Qubit broad range DNA assay kit (ThermoFisher Scientific™, Waltham, MA, USA), respectively.

For each of the three strains, 1 μg of the total genomic DNA was used for Nanopore library preparation using the ligation sequencing DNA V14 (SQK-LSK114, Oxford Nanopore Technologies, Oxford, UK). Sequencing was performed using the R10.4.1 flow cell (FLO-MIN114) on a MinION device (Oxford Nanopore Technologies, Oxford, U.K.). Base calling was performed offline with ONT Guppy basecalling software version 6.4.8 + 31becc9 and minimap2 version 2.24-r1122, enabling the—pt_scaling flag and setting the—trim_strategy flag to DNA.

2.5. Long Read Filtering, Correction, and Assembly

Adapter trimming of the raw sequences of all genomes was performed by Porechop version 0.2.4 (www.github.com/rrwick/Porechop), setting the—adapter_threshold to 96 and enabling the—no_split flag. Setting the genome size to 40 MB, the trimmed reads were further trimmed and corrected using Canu version 2.3 [31], enabling the -trim and -correct flags, respectively.

The genome assemblies were created using Flye version 2.9.5 [32] using the—nano-hq flag, setting the genome size to 40Mb and the—min-overlap to 1500, and were refined by two rounds of polishing in Flye. Bandage v.0.9.0 [33] was used to visualize assembly graphs and search for telomere sequences by using the built-in blast function to search for the telomere sequence (TTAGGGT)n_5–15. To evaluate the completeness of the final genome assembly, Benchmarking Universal Single-Copy Orthologs (BUSCO) analyses were performed with BUSCO version 5.8.2, using fungal_odb10 and hypocreales_odb10 lineage gene sets [34]. Genomes of C. rhodochroum ATHUM 6904, C. tenue CBS152.92, and C. rubrobrunnescens CBS176.92 were deposited at GenBank (Bioproject PRJNA1366398, Acc. Nos. JBTXJK010000000, JBTVZP010000000, and JBTVZQ010000000 for the whole genomes of C. rhodochroum, C. rubrobrunnescens, and C. tenue, respectively).

2.6. Gene Prediction and Functional Annotation

Assembly annotations were performed using GenSAS v.6.0 [35], unless otherwise stated. Interspersed repeats, low-complexity DNA sequences, and transposable elements were detected and masked by RepeatModeler v.2.0.1 and RepeatMasker v.4.1.1 (http://www.repeatmasker.org/), with the DNA source set to fungi. RNAmmer version 1.2 [36] and tRNAscan-SE version 2.0.7 [37] were used to detect the ribosomal RNA and tRNA genes, respectively. In order to primarily identify genomic regions with putative protein genes, transcript alignments were performed with the BLAST nucleotide (blastn) tool version 2.12.0 using the transcript database NCBI refseq fungi [38] and the BLAT tool version v2.5 using the transcripts FASTA file from NCBI refseq fungi [39], and protein alignments were performed using DIAMOND proteins version 2.0.11 against Protein Data Set—NCBI refseq fungi (Protein) [40]. De novo gene prediction was performed using the following tools with default parameters: (a) AUGUSTUS tool version S3.4.0 with the reference gene dataset of Fusarium graminearum, (b) GeneMarkES version 4.48, and (c) GlimmerM tool version 2.5.1 selecting Aspergillus reference genus [35]. The tool EvidenceModeler was used to create a consensus gene set using the output files of the previously described tools [41]. The ab initio official gene set (OGS) was evaluated with BUSCO analysis. Functional analysis of the OGS was performed using BLAST protein vs protein (blastp) against Protein Data Set: NCBI refseq fungi (Protein) [38], DIAMOND Functional version 2.0.11 against Protein Data Set: NCBI refseq fungi (Protein) [40], and InterProScan version 5.53–87.0 [42]. The presence and location of signal peptide cleavage sites in the amino acid sequence were identified using SignalP version 5.0b, setting the -org flag to eukaryote [43]. The annotation of the OGS was performed using local BLASTp (e-value 1 × 10⁻⁵⁰) against Non-Redundant Protein Sequence (NR), Swiss-Prot, KEGG (The UniProt Consortium 2015), Gene Ontology (GO), Clusters of orthologous groups for eukaryotic complete genomes (COG), pathogen–host interaction (PHI), carbohydrate active enzymes (CAZy), MEROPS, PredGPI prediction server and Transporter Classification (TCdb) databases [44,45,46,47,48,49,50]. TMHMM v2.0 was used to identify transmembrane proteins based on a hidden Markov model for transmembrane helices [51].

2.7. Mitochondrial DNA Characterization

The mitochondrial contig was annotated as follows: the protein-coding, ribosomal (rRNA), and tRNA genes were identified using BLASTx, BLASTn [52], and tRNAscan-SE version 2.0.9 [53], respectively. The genetic code employed was “The Mold, Protozoan, and Coelenterate Mitochondrial Code and the Mycoplasma/Spiroplasma Code” (NCBI transl_table = 4). The mitochondrial genomes and plasmids were visualized using OrganellarGenomeDRAW (OGDRAW) version 1.3.1 [54].

2.8. Comparative Genomics

To study the evolution and genetic diversity of Cladobotryum genomes, the OrthoVenn3 tool [55] was used to identify and annotate orthologous clusters among selected species representing the Hypocreaceae family, i.e., Trichoderma harzianum, Escovopsis sp., and Hypomyces perniciosus, with the endophytic-entomopathogenic species Metarhizium anisopliae used as an outgroup (Supplementary File S1: Table S3). The OrthoFinder algorithm [56] was selected with enabled annotation, protein similarity analysis, and cluster relationship network options, using default parameters. Maximum-likelihood phylogenetic analysis of the selected species was also conducted in OrthoVenn3 inference using the program FastTree2 and the evolution model JTT + CAT [57], and the reliability of each node was determined by the SH test. CAFE5 software was used to calculate the contraction and expansion in gene family size [58]. The divergence times were set at 155 MYA for M. anisopliae and T. harzianum and 76 MYA between Escovopsis sp. and T. harzianum, as found in the TimeTree 5 web portal [59]. Genome synteny among Cladobotryum species was analyzed using a gene-based comparative approach. Orthologous relationships were inferred with OrthoFinder based on predicted protein sequences [56]. Quantitative synteny detection was performed using MCScanX with gene coordinate information extracted from genome annotations and default parameters to identify collinear gene blocks across species. Synteny statistics were derived from MCScanX collinearity outputs [60]. Chromosome-level visualization of genome rearrangements was carried out using OrthoVenn3, which also employs OrthoFinder for orthology inference and was used exclusively for graphical representation of syntenic relationships [55].

2.9. Biosynthetic Gene Clusters and Secondary Metabolism Comparative Analysis

To examine the secondary metabolism variability, the biosynthetic gene clusters (BGCs) of the six available Cladobotryum genomes (along with the genomes of Trichoderma aggressivum, Escovopsis sp., and Hypomyces perniciosus, which were used as outgroups) were identified using antiSMASH 7.0 [61], enabling the cluster-border prediction based on transcription factor binding sites (CASSIS) selection. All genes that were identified as part of BGCs were analyzed in OrthoFinder v2.5.5 [56] to examine the shared and unique secondary metabolism-related protein-coding genes (PCGs) and BGCs in the Hypocreaceae family. The similarity network analysis and the exploration of BGC diversity among species were conducted using the BiG-SCAPE program [62].

2.10. Average Nucleotide Identity/Genomic Distance Analysis

The recently developed tool FungANI [63], which adapts the Average Nucleotide Identity (ANI) methodology for fungal genomes, was employed. The analysis involves segmenting each genome under study (e.g., genome A and genome B) into 1000 bp fragments. Each fragment of genome A was aligned against genome B using BLAST, producing a percent identity value (forward analysis). Conversely, each fragment of genome B was aligned against genome A to obtain a reverse analysis. According to Lalanne and Silar [63], in a test case involving Sordariomycetes, an ANI threshold of ≥99% was proposed as indicative that two genomes belong to the same species, whereas values below 99% suggest distinct taxonomic units. This threshold considers the higher intraspecific diversity typically observed in fungi compared to bacteria [63,64]. To assess the genetic distance of Cladobotryum rhodochroum ATHUM 6904 relative to C. tenue and C. rubrobrunnescens, pairwise ANI analyses were conducted. The resulting values were compared against the proposed 99% species threshold, allowing evaluation of whether ATHUM 6904 represents a distinct species or shows close affiliation with the examined taxa.

3. Results and Discussion

3.1. Cladobotryum Rpb2-Based Phylogeny

The rpb2 gene has been considered a particularly robust phylogenetic marker for species identification and evolutionary studies within Cladobotryum [9,17]. Accordingly, a phylogenetic tree of the genus Cladobotryum (including its teleomorph Hypomyces) was constructed based on 90 concatenated ITS-rpb2-tef-1a sequences (Figure 1).

The ITS-rpb2-tef-1a matrix provided substantially greater intraspecific resolution. The resulting tree is mostly similar to the respective one of the ITS-based phylogeny described by Milic et al. [6], particularly with respect to overall species clustering. The URPs formed a distinct and well-supported clade, hereafter referred to as the C. rhodochroum clade. In accordance with ITS-based phylogeny, in the ITS-rpb2-tef-1a-based tree, C. rubrobrunnescens is basal to C. tenue and C. rhodochroum (Figure 1). The isolates belonging to the three species form well-supported monophyletic clades, with bootstrap values of 99% for C. rhodochroum, 99% for C. rubrobrunnescens, and 100% for C. tenue. Although C. rhodochroum shares high sequence identity with both related species (99.6%, 99.06%, and 99.3% with C. tenue; 99.8%, 99.44% and 99.6 with C. rubrobrunnescens in ITS, rpb2, and tef-1a, respectively), monophyletic placement of the URPs provides the first evidence that they represent a distinct cluster, with strain ATHUM 6904 representing a new species. High ITS similarity alone does not necessarily indicate conspecificity, since interspecific ITS divergence varies substantially across Cladobotryum and related genera in the Hypocreales [6,65].

Thus, the ITS-rpb2-tef-1a-based tree (Figure 1) provides greater resolution than the single-region ITS-based counterpart [6]. Nevertheless, the precise placement of the novel species C. rhodochroum in relation to C. tenue and C. rubrobrunnescens is still unclear. To validate these results and achieve a more comprehensive understanding of the evolutionary relationships within this complex, additional evidence was integrated from macro- and micromorphological examinations along with comparative genomics.

3.2. Morphological Study

Cladobotryum rhodochroum Christinaki, Kouvelis, and Gonou-Zagou sp. nov.

Mycobank No.: MB861264

Figure 2, Figure 3, Figure 4 and Figure 5

Etymology

The species epithet refers to the reddish coloring of the mature mycelium of the colony, as well as of the pigment diffusing into the medium. It derives from the Greek “rhodon” (ῥόδον) referring to “rose = pink-red color” and the suffix “-chrous/-chroa/-chroum” (-χρους/-χρόα/-χρους), here used in its neuter form “-chroum” which denotes “color”.

3.2.1. Diagnosis

The colony macro- and micromorphology of the species differs from that of its closest phylogenetic species studied, i.e., C. tenue and C. rubrobrunnescens, as well as from its related C. dendroides and C. mycophilum. The colony of C. rhodochroum, differing from all others, is characterized by the felty to slightly cottony texture and the reddish color of the mature mycelium as well as that of the metabolite diffused to the substrate. The conidia are typically two-celled, ellipsoidal to cylindrical, produced from elongated conidiogenous cells.

3.2.2. Types

GREECE, Karditsa, Afchenas Agiou Nikolaou, Mt. Zigourolivado; Fagus sylvatica forest, on basidioma of Lactarius sp.; Coll. P. Delivorias (PD3248), November 2009; Isol. A. Liakouri (AL6), November 2009. Holotype Fungal Culture Collection of Mycetotheca ATHUM 6904, stored in a metabolically inactive state (deep frozen, lyophilized); Ex-holotype Fungal Culture Collection of Mycetotheca ATHUM 6904 (living, in paraffin oil); GenBank Acc. No ITS: OM993297; RPB2: PX610068; whole genome: JBTXJK010000000.

3.2.3. Description

Colonies on PDA covering the Petri dishes within 7–10 days. Mycelium felt-like to slightly cottony, whitish at first gradually turning from center to margin grayish yellow to pale yellow (4C5–4A4), progressing as amber/dark yellow to grayish/dark red (4B6, 4C8–8C4-5), becoming brownish red to rose/dull red (11A4-B4–8C6) after 20 days, finally brownish-grayish red (10D5-11C5) (Figure 2A–F); watery droplets appearing on the surface of the marginal area while maturing (Figure 2H,I). KOH reaction faintly positive when immature, mature positive turning in a while to strongly positive (Figure 2G,H). Colony reverse from center to margin, butter yellow to golden brown (4A5–5D7), progressively turning reddish golden to brownish red to violet-brown (6C6-7–10D7–10E8), dull red to brownish red (10C4–10D6), and finally violet-brown 10F8 (Figure 2A–F). Conidiophores: very long, delicate, sparsely to verticillately branched mainly at conidiophore apices, often in a unilateral manner; septate; hyaline (Figure 3A,B). Conidiogenous cells elongated, sometimes very elongated, delicate, narrow-cylindrical to almost filiform; arranged in verticils, 1–4(5) per verticil, mainly 2–4 at the apex, sometimes emerging solitary, orthogonally to conidiophore axis or in a dichotomous manner at the apex; aseptate; hyaline (Figure 3A–G). Conidiogenesis blastic; a single to multiple conidiogenous loci per conidiogenous cell, commonly a single conidium or up to two conidia in a V-shaped arrangement (Figure 3D,G). Conidia in Petri dish culture: ellipsoidal to cylindrical, usually with distinct truncate apiculus; (13.3)17.4–25.2(31.3) × (2.9)5.1–7.7(8.8) μm, n = 100, Q 1.9-6.1, Lm 21.3 μm, Wm 6.4 μm, Qm 3.5, mainly two-celled, rarely one- or three-celled; basal cell often swollen (Figure 4A–G); when mature, sometimes restricted at septum, several times degradation of one of the two conidial cells; (Figure 4H,I). Mature mycelium hyphae mainly rose-red to brownish-red colored, few hyaline, with variable width, typical cylindrical, often with encrusted material, many irregularly multiseptated, forming short and swollen cells with granular content; hyphal bridges also formed (Figure 5A–C). Chlamydospores usually not observed; very rare, apical, up to three-celled, and detached (Figure 5D). No production of teleomorphic state.

3.3. Ecology-Distribution

Only known from the holotype. Mycelium covering various parts (lamellae, surface of pileus) of Lactarius sp. basidiomata.

3.3.1. Strains Examined

GERMANY: Cladobotryum rubrobrunnescens: on Inocybe sp., 1989, A. Resinger, CBS 176.92, ex-type strain. Cladobotryum tenue: Regesburg-Keilberg, on agaric, 1986, H. Besl, CBS 152.92, ex-type strain. GREECE: Cladobotryum mycophilum: Attiki, on cultivated basidioma of Agaricus bisporus, 2010, coll./isol. Z.Gonou-Zagou, ATHUM 8001; Eurytania: Ag. Nikolaos, in Platanus orientalis forest with Quercus sp., Abies cephalonica, and Castanea sativa (sporadically), on basidioma of Mycena sp., 2010, coll./isol. Z. Gonou-Zagou, ATHUM 8000; Karditsa: Belakomitis, in Abies borisii-regis forest, on basidiome of Inocybe sp., 1999, coll./isol. Z. Gonou-Zagou, ATHUM 7994; Magnisia, Mt. Pilio, in Castanea sp. forest, on basidioma of Hypholoma sp., 2009, coll. P. Delivorias, isol. Liakouri, ATHUM 6906. Hypomyces rosellus (anamorph: Cladobotryum dendroides): Attiki, Athens “National Gardens”, on basidioma of Flammulina velutipes, 2008, coll./isol. Z. Gonou-Zagou, ATHUM 6847; Mt. Parnitha, in Abies cephalonica forest, on basidioma of Tricholoma sp., 2008, coll./isol. Z. Gonou-Zagou ATHUM6849; Mt. Parnitha, in Abies cephalonica forest, on basidioma of Hohenbuehelia sp., 2000, coll./isol. Z. Gonou-Zagou ATHUM 7998; Eurytania, Ag. Nikolaos, in Platanus orientalis forest with Quercus sp., Abies cephalonica, and Castanea sativa (sporadically), on ascoma of Helvella lacunosa, 2010, coll./isol. Z. Gonou-Zagou, ATHUM 7999; Karditsa, Ag. Nikolaos, Mt. Zigourolivado, in Fagus sylvatica forest, on polypore basidioma, 2009, coll. P. Delivorias, isol. A. Liakouri ATHUM 6909; Xanthi, Mt. Leivaditis, in Fagus sylvatica forest with Juniperus communis, on basidioma of Polyporus varius, 2009, coll. A. Sergentani, isol. Z. Gonou-Zagou, ATHUM 6848.

3.3.2. Notes

According to this study and the results of the extensive work of Milic et al. 2022 [6] on Greek strains of Cladobotryum spp., the newly described species C. rhodochroum ATHUM 6904 is phylogenetically closer to C. tenue and C. rubrobrunnescens and related to C. mycophilum and C. dendroides. Nevertheless, there are diagnostic morphological features that support the independence of all species. The different colony color and texture, the longer and narrower conidia, and the simpler conidiophores of C. tenue and C. rubrobrunnescens set them apart from C. rhodochroum. Though strains of C. mycophilum (teleomorph Hypomyces odoratus) exhibit certain variability in culture morphology, there are features that differentiate them from those of C. rhodochroum. The color and texture of the colony, the longer and broader, two- to four-celled conidia produced retrogressively from gradually shortening conidiogenous cells, and the production of secondary conidia [6] of C. mycophilum clearly distinguish it from C. rhodochroum. Additionally, strains of the C. mycophilum can produce the teleomorphic state, forming perithecia with asci and ascospores. Moreover, the species C. dendroides (teleomorph Hypomyces rosellus) has isolates that are distinct due to the characteristic texture and color of the mycelium, the sympodial conidiogenesis, and the long and broad three- to four-celled conidia. It is worth mentioning that the distribution of the species C. tenue and C. rubrobrunnescens is restricted to very few countries: Germany and Estonia in Northern Europe for both, and Mediterranean Spain and Israel for each one, respectively [7,9]. Both species, C. mycophilum and C. dendroides, seem to have a rather cosmopolitan appearance [4,66].

3.4. Comparative Genomics

3.4.1. Genome Features and Gene Prediction

Although additional isolates cluster near C. rhodochroum in ITS-based analyses, only the holotype isolate ATHUM 6904 could be confidently assigned to this species based on combined molecular and morphological evidence [6]. Therefore, genomic analyses were restricted to ATHUM 6904 to ensure taxonomic consistency. Genome sequencing of the species C. rhodochroum ATHUM 6904, C. tenue CBS152.92, and C. rubrobrunnescens CBS176.92 yielded 1985 Mb, 2487 Mb, and 1824 Mb of clean data, respectively. These were assembled into high-quality draft genomes of 41 Mb (12 contigs, N50: 5.65 Mb), 39.7 Mb (45 contigs, N50: 1.73 Mb), and 38.6 Mb (60 contigs, N50: 1.76 Mb), with mean coverage of 36×, 74×, and 35×, respectively (

Table 1. Whole genome sequencing statistics and genomic features of C. rhodochroum ATHUM 6904, C. tenue CBS152.92, and C. rubrobrunnescens CBS176.92.

Genome Statistics and Features	C. rhodochroum ATHUM 6904	C. tenue CBS152.92	C. rubrobrunnescens CBS176.92
Genome size (bp)	41,067,690	39,686,118	38,610,303
Number of fragments	12	45	60
N50 (bp)	5,646,276	1,727,773	1,766,473
Largest fragment (bp)	8,188,398	4,373,917	4,397,480
Mean genome coverage	36	74	35
Number of predicted tRNA genes	271	269	249
Number of predicted rRNA genes	60	57	52
Number of predicted PCGs	11,737	11,654	11,555
Transposable elements (%)	6.04	4.62	2.08
Mitochondrial genome size (bp)	82,745	115,194	103,881
Mitochondrial plasmid(s) size (bp)	15,129|8925	7786	4532
BUSCO hypocreales_odb10 (%):	99.1	99.2	98.8
BUSCO fungi_odb10 (%):	99.7	99.8	99.4

). Genome completeness, assessed using BUSCO with the fungi_odb10 and hypocreales_odb10 datasets, showed high levels of completeness across all three assemblies, C. rhodochroum (99.7% and 99.1%), C. tenue (99.8% and 99.2%), and C. rubrobrunnescens (99.4% and 98.8%) for fungal and hypocrealean BUSCOs, respectively.

Regarding the protein-coding genes (PCGs), all three strains exhibit a comparable number, ranging from 11,555 in CBS176.92 to 11,737 in ATHUM 6904 (Table 1). A similarly high degree of conservation is observed in the tRNA genes, with their counts ranging from 249 (CBS176.92) to 271 (ATHUM 6904), reflecting the overall stability of the transcriptional machinery across strains. A consistent pattern is also seen in the rRNA genes: ATHUM6904 harbors 60, CBS152.92 contains 57, and CBS176.92 carries 52. Taken together, these findings highlight a high level of genomic uniformity across fundamental gene categories, reinforcing the evolutionary relatedness of the strains. Minor variations in gene counts may reflect either genuine genetic differences or discrepancies in genome assembly quality.

3.4.2. Transposable Elements

The analysis of transposable elements (TEs) in the genomes of C. rhodochroum ATHUM 6904, C. tenue CBS152.92, and C. dendroides CBS176.92 reveals notable variation in both the diversity and abundance of these elements. These differences may reflect underlying evolutionary adaptations and are likely to influence the structure and functionality of their respective genomes.

In the genome of C. rhodochroum ATHUM 6904, approximately 6.04% consists of repeat sequences and transposable elements (Table 1). Among these, 3% corresponds to retrotransposons, including 567 LINEs, which make up 2.04% of the genome, and 446 LTR elements, representing 0.96%. The LTR elements are further subdivided into BEL/Pao (0.01%) and Gypsy/DIRS1 (0.95%) families. The remaining 3.04% comprises unclassified transposable elements. It is worth noting that no elements of class I DNA transposons or Helitron-type rolling-circle transposons were identified in this strain (Supplementary File S1: Table S4).

The genome of C. tenue CBS152.92 displays a slightly lower TE content, with 4.62% of its sequence occupied by repeat regions and transposable elements. Of this, 0.56% represents retrotransposons, composed of 152 LTR elements that are further classified into retroviral (0.01%) and Gypsy/DIRS1 (0.55%) elements. An additional 0.20% is made up of “hobo-Activator” type DNA transposons, while the remaining 3.86% includes unclassified elements. As with ATHUM 6904, no Helitron-type rolling-circle transposons were detected in this genome.

In contrast, the genome of C. rubrobrunnescens CBS176.92 contains only 2.08% repeat sequences and potential transposable elements. Interestingly, no known categories of transposable elements, such as retrotransposons, DNA transposons, or Helitrons, were found. Instead, the genome comprises 1010 unclassified elements, accounting for the entire 2.08%, along with repeat regions. This distinct lack of identifiable TE families differentiates CBS176.92 from the other strains examined. The notably lower number of such elements may be associated with the smaller size of this genome (38.6 Mb), compared to C. tenue (39.67 Mb) and C. rhodochroum ATHUM 6904 (41.07 Mb).

The observed differences in TE abundance and diversity among these strains may have important evolutionary and functional implications. Prior studies have linked the accumulation of transposable elements to genome expansion across a variety of fungal species, as repeated sequences and TE insertions contribute significantly to genome size [67]. This trend is evident in the current data, where ATHUM 6904, with the highest proportion of TEs (6.04%), also exhibits the largest genome, which is approximately 2.5 Mb larger than that of CBS176.92, where TEs make up only 2.08%. Furthermore, the presence and diversity of different TE categories, including LINEs, LTRs, and DNA transposons, may not only influence genome architecture but also impact its evolutionary dynamics, e.g., in the metabolic pathways employed by these species. Overall, transposable elements can increase genetic variability, generate novel regulatory sequences, and induce structural rearrangements [68], potentially accelerating genetic divergence among species.

3.4.3. Mitochondrial Genomes and Plasmids

During the assembly of the genomes, the mitochondrial (mt) genomes and plasmids were identified, and all genes within them were annotated. The mt genomes of the three Cladobotryum strains studied are circular molecules varying in size and gene content. C. rhodochroum ATHUM 6904 has the smallest mt genome at 82,745 bp, containing the 14 core mt genes involved in oxidative phosphorylation and ATP production, two rRNA genes, 25 tRNA genes, 40 open reading frames (ORFs), and 29 introns (Figure 6A;

Table 2. Intron content by gene in the mt genomes of C. rhodochroum ATHUM 6904, C. tenue CBS152.92, and C. rubrobrunnescens CBS176.92. When multiple introns occur within a gene, the intron types are listed in the order they appear in the gene, from the first to the last intron.

Gene	ATHUM 6904	CBS 152.92	CBS 176.92
rns	II	II	II
rnl	IC1, IC1, IC2, IA	II, IC1, IC1, IC2, IA	II, IC1, IC1, IC2, IA
atp6	IB	IB	IB
atp9	IA	IA	IA
cob	I (unclassified), ID, IA, IB	I (unclassified), ID, IB, IA, IB	I (unclassified), ID, IB, IA, IB
cox1	IB, IB, ID, IB, IB, IB, IB, IB	IB, IB, ID, IB, IB, IB, IB, IB	IB, ID, IB, IB, IB, IB, IB
cox2	IC2, IC2, IB, IC1	IC2, IC2, IB	IC2, IC1
cox3	IC2, ID	IC2, ID, IA	IB, IC2, ID, IA
nad1	I (unclassified), IB, IA	I (unclassified), IB, IA, IB	I (unclassified), IB, IA
nad2	—	IC2	IC2
nad3	—	IC2	—
nad4L	—	—	IB
nad5	IC2	ID, IC2	ID, IC2, ID
Total	29	35	34

). C. tenue CBS 152.92 features the largest mt genome at 115,194 bp, with the same core gene set plus 26 tRNA genes, 64 ORFs, and 35 introns (Figure 6B; Table 2). C. rubrobrunnescens CBS 176.92 has a 103,881 bp genome including the 14 core genes, 25 tRNA genes, 62 ORFs, and 34 introns (Figure 6C; Table 2). As described, the mitogenome of C. rhodochroum is markedly smaller than those of the ex-type strains CBS152.92 and CBS176.92. This size increase in the latter is primarily due to the number and length of introns, as well as the higher number of ORFs, consistent with previous studies showing that introns significantly contribute to the mitogenome size variation in fungi [69,70].

Despite these differences in ORF and intron content, the synteny of the remaining genes is largely conserved among the mitogenomes examined. The only notable exception is within the gene order of the C. tenue mitogenome, which contains an additional tRNA gene (trnR) inserted between trnM–trnH–|trnR|–trnQ–trnL (Figure 6).

The analysis of the mt plasmids in the three species sequenced in this study showed that each species contains distinct mt plasmids with different origin and gene content. In detail, the first mt plasmid identified in the genome of C. rhodochroum ATHUM 6904 is a circular molecule of 15,129 base pairs and includes seven protein-coding genes (PCGs), three of which are identical copies encoding GIY-YIG endonucleases, while the remaining four have unknown functions (Figure 7A). The second mt plasmid found in the genome of ATHUM 6904 is a circular molecule of 8925 base pairs, containing four PCGs, including one reverse transcriptase (RVT_1) and one group II intron reverse transcriptase (RT_G2_intron), while the remaining two represent hypothetical genes (Figure 7A). The mt plasmid identified in the genome of C. tenue CBS 152.92 is a circular molecule of 7786 base pairs and contains two identical PCGs that encode mitochondrial type B DNA polymerases (Figure 7B). A similar plasmid containing three copies of type B DNA polymerases has also been found in the genome of C. mycophilum ATHUM 6906 [21]. These genes have high sequence identity to the respective gene of C. parasitica’s pCRY1 plasmid [71]. This is a plasmid that reduces pathogenicity in C. parasitica when present [71]. Nevertheless, its role in C. tenue and C. mycophilum is still unknown. The mt plasmid found in the genome of CBS 176.92 is a circular molecule of 4532 base pairs and contains two PCGs, one encoding a protein with an ICP4 transcriptional regulator domain and another reverse transcriptase with RVT_1 (commonly found in retrotransposable elements, retroviruses, group II introns, and other mobile DNA elements) and H-like domains with RNase activity (Figure 7C). The association of the RVT_1 domain with group II introns suggests that this plasmid may have originated from the mobilization or excision of a type II mitochondrial intron.

Although these strains are evolutionarily related, there is substantial variability in the presence, size, and gene content of their mt plasmids, reflecting the evolutionary dynamics and diversity of mitochondrial DNA in fungi. Overall, these observations indicate that mt genomes and plasmids may dynamically influence genome structure and function [72]. The origin of these plasmids is currently unclear; however, their uniqueness and limited distribution within Hypocreales are consistent with the hypothesis that they may have been acquired via horizontal gene transfer, potentially from fungal hosts.

3.5. Phylogenomics

To elucidate the evolutionary and taxonomic position of C. rhodochroum ATHUM 6904, a comparative genomic analysis was conducted, extending beyond the newly sequenced genomes generated as part of this work. The analysis incorporated all publicly available genomes of Cladobotryum species to assess the phylogenetic proximity of C. rhodochroum to the ex-type strains of C. rubrobrunnescens and C. tenue. Additionally, representatives from other genera within the family Hypocreaceae, specifically Trichoderma harzianum, Escovopsis sp., and Hypomyces perniciosus (anamorph Mycogone perniciosa), were included to provide a broader phylogenetic perspective.

In the resulting phylogenetic tree, the genus Cladobotryum forms a well-supported monophyletic clade, with Escovopsis identified as its sister group, diverging approx. 51.5 MYA (Figure 8). Within Cladobotryum, C. rhodochroum clusters with C. tenue, and C. rubrobrunnescens occupies a basal position relative to them. These relationships are in strong agreement with the topology recovered from the multi-locus phylogeny (Figure 1) and single-locus ITS phylogeny [6]. Molecular dating suggests that the novel species C. rhodochroum ATHUM 6904 and C. tenue diverged 1.2 MYA, whereas C. rubrobrunnescens branched off earlier, approximately 2.7 MYA. Despite their distinct phylogenetic topologies, the relatively short branch lengths indicate close evolutionary relationships among these three species.

3.6. Synteny and Genome Rearrangements in the Genus Cladobotryum

Comparative synteny (or gene order) analysis across the six Cladobotryum genomes identified 699 syntenic blocks, encompassing 64,768 genes (93.2% of the total gene set). Pairwise comparisons revealed 11–33 syntenic blocks per species pair, with median block sizes ranging from ~190 to >800 genes. This indicated extensive conservation of gene content, accompanied by varying levels of gene order rearrangement (Figure 9).

Pairwise comparisons involving C. rhodochroum showed substantial but heterogeneous conservation across the genus (Supplementary File S1: Table S5). The proportion of syntenic genes between C. rhodochroum and other species ranged from ~53% to ~71%, with corresponding differences in the number and size of conserved blocks. The strongest conservation was observed between C. rhodochroum and C. dendroides, which shared approximately 70–71% of genes in syntenic blocks distributed across a small number of large regions (median block size > 700 genes). Consistently, the seven chromosomes of C. rhodochroum ATHUM 6904 display extensive collinearity with those of C. dendroides, with rearrangements primarily affecting specific genomic regions rather than the overall chromosomal framework.

More extensive rearrangements were observed in the respective comparisons with the rest of the Cladobotryum species. Slightly lower but still substantial syntenic variability was detected between C. rhodochroum and C. mycophilum, with ~65% syntenic genes, indicating increased fragmentation of conserved regions (Supplementary File S1: Table S5). In contrast, comparisons between C. rhodochroum, C. protrusum, C. tenue, and C. rubrobrunnescens retained only ~53–56% syntenic genes, accompanied by a higher number of smaller syntenic blocks.

It should be noted that assembly contiguity differs among the analyzed genomes and may influence the resolution of synteny statistics. In C. tenue and C. rubrobrunnescens, genome assemblies are fragmented into 45 and 60 scaffolds, respectively, precluding chromosome-level reconstruction and potentially limiting the detection of large-scale genomic rearrangements. In contrast, the genome of C. rhodochroum is assembled at the chromosome level, allowing more accurate delineation of extended syntenic blocks. Consequently, synteny metrics involving C. tenue and C. rubrobrunnescens may reflect both biological rearrangements and technical fragmentation, whereas comparisons involving C. rhodochroum likely provide a more accurate representation of underlying genome architecture.

3.7. Orthologous Analysis

In general, orthologous gene analysis aids in deciphering evolutionary relationships, the conservation and divergence of gene functions, and the adaptations of species to their environment and lifestyle. In this study, orthologous genes were examined to assess potential genome expansions and contractions, which may have arisen through gene loss and/or duplication during evolution. The comparative analysis of orthologous genes across Hypocreaceae showed that the six Cladobotryum species share 9067 orthologous clusters, representing the genus’s core genome. These conserved gene clusters include both essential housekeeping genes involved in basic cellular metabolism and genes with genus-specific functions. Among them, Cluster 1 (325 predicted coding genes, PCGs) and Cluster 2 (142 PCGs) were the largest, harbors genes encoding polyketide synthases (PKSs) and nonribosomal peptide synthases (NRPSs), respectively (Supplementary file 1: Table S6), highlighting the metabolic diversity and secondary metabolite potential within the genus.

Consistent with this observation, genome-wide analysis of the 9067 orthologous gene clusters indicates a notable expansion in genes associated with secondary metabolite biosynthesis and pathogenicity. Of these clusters, 565 are unique to the six Cladobotryum species—absent from other Hypocreaceae genera—and most of them appear to contribute to the genus’s rich secondary metabolism. This is in accordance with the CAFÉ analysis, which shows that at the base of the genus Cladobotryum, there is a significant expansion of gene families (Figure 8).

The species C. rubrobrunnescens, C. tenue, and C. rhodochroum exclusively share a significant set of PCGs, organized into 86 orthologous clusters (Figure 10A). Such “shared exclusive orthologs” have been linked in previous studies to common ancestry and potential functional convergence [56,73]. The majority of these PCGs have uncharacterized functions, suggesting the presence of novel biological mechanisms or specialized adaptations. Indeed, the accumulation of hypothetical proteins within specific evolutionary lineages is indicative of rapid specialization or even early stages of speciation [74,75]. Notably, 10 of them show homology to known functional genes associated with (a) secondary metabolic pathways (NRPS, P450 monooxygenase, efflux pump), (b) regulation of growth and cellular processes (sporulation regulator, transcription factor), (c) protein and metabolic regulation (proteases, esterase, ubiquitin-E2, RNA helicase), and (d) genes involved in the biosynthesis of the mycotoxin fusarin C. The data suggest that these species share a unique set of genetic elements that may play a significant role in their metabolic activity.

Despite their close genetic relatedness, C. rubrobrunnescens, C. tenue, and C. rhodochroum each possess unique PCGs (singletons). In contrast, C. dendroides has fewer singletons (only 60) compared to the strains under study (Figure 10A). Such genetic uniqueness is an important criterion for defining distinct species, particularly in microorganisms, where the concept of a “biological species” may be insufficient [76,77]. It also highlights a level of genetic differentiation that contrasts with their subtle morphological and molecular distinctions [6]. Whether these species-specific genes contribute to their ecological role as mycoparasites remains unclear, especially since C. dendroides, despite having the fewest singletons, is regarded as one of the most aggressive species in the genus [9,78].

3.8. Secondary Metabolism and Biosynthetic Gene Clusters (BGCs)

The study of secondary metabolism through the analysis of biosynthetic gene clusters (BGCs) is an essential tool for understanding the evolutionary diversity and ecological adaptation of fungi. The diversity and organization of these clusters can vary considerably even among closely related species, reflecting phylogenetic differentiation and metabolic specialization. In Cladobotryum, the analysis of genes and clusters associated with secondary metabolite production revealed a marked expansion compared to other fungicolous fungi (Figure 10B). Notably, C. rhodochroum harbors a high number of BGCs (107), comparable to C. tenue (119 clusters) and exceeding other species like C. dendroides (105 clusters). This enrichment in BGCs, considerably higher than that observed in other fungal pathogens of cultivated mushrooms, such as Hypomyces perniciosus (anamorph Mycogone perniciosa, 64 clusters) and Trichoderma aggressivum (63 clusters), highlights the exceptional biosynthetic potential of C. rhodochroum and may reflect an adaptive strategy linked to its mode of life.

Differences among C. rubrobrunnescens, C. tenue, and C. rhodochroum in the total number of BGCs per genome primarily involve NRPS(-like), terpene, and fungal RiPP(-like) clusters (

Table 3. Detailed overview of the abundance of synthase/synthetase PCGs associated with secondary metabolism in C. rhodochroum, C. tenue, and C. rubrobrunnescens, along with the total number of BGCs identified in their genomes.

	C. rhodochroum	C. tenue	C. rubrobrunnescens
Total synthase/synthatase PCGs	148	161	146
T1PKS	56	56	56
NRPS (-like)	46	50	39
terpene	19	26	21
fungal-RiPP (-like)	19	21	23
NI-siderophore	2	2	2
indole	2	2	1
isocyanide	2	2	2
NRP-metallophore	1	1	1
NAPAA	1	1	1
Total number of BGCs	107	119	105

), illustrating both genetic and metabolic divergence within the genus.

Mapping of BGCs in C. rhodochroum, C. mycophilum, C. protrusum, and C. dendroides (teleomorph Hypomyces rosellus), whose genomes are assembled at the chromosome level, revealed that secondary metabolism-related genes are predominantly located at chromosome ends (Supplementary File S3) [21]. This subtelomeric localization, consistent with previous studies in other ascomycetes [79], may facilitate the generation of new BGCs through rearrangements and expansion, potentially contributing to the metabolic specialization of C. rhodochroum.

To compare the genes involved in secondary metabolite production, a network analysis was conducted, including the BGCs from six Cladobotryum species, with Escovopsis sp., H. perniciosus, and T. aggressivum as outgroups. A total of 827 BGCs were analyzed and initially grouped into families according to their broader class (e.g., T1PKS, NRPS, etc.). Figure S2 illustrates the presence of these BGCs in relation to both the phylogenetic relationships of the species and the specific class and family of each cluster (Figure 11). BGCs belonging to T1PKS, NRPS, NRPS-like, terpenes, and fungal-RiPP categories show a broad distribution across the species, whereas clusters for isocyanides, isocyanide-NRP, NRP-metallo, indole compounds, and β-lactones display a more patchy, discontinuous distribution. Notably, NAPAA, NI siderophore, and T3PKS BGCs were found exclusively in Cladobotryum species, including C. rhodochroum, reinforcing the species’ unique secondary metabolite potential [80].

From the distribution patterns and network analysis of BGCs across Hypocreaceae genomes, it is evident that the evolution of secondary metabolism in C. rhodochroum, as in other Cladobotryum species, involves the formation of new BGCs, the loss of others, and the acquisition of certain secondary metabolite-related genes via horizontal gene transfer. The latter is consistent with previously reported evidence for horizontal gene transfer in Cladobotryum [21]. The presence of numerous unique single-copy BGCs (singletons) in C. rhodochroum suggests that these clusters may contribute to specialized metabolic pathways (Figure 11), highlighting the ongoing rapid evolution of secondary metabolism and the emergence of entirely novel and distinct BGCs. Consequently, considerable diversity exists even among the species C. rubrobrunnescens, C. tenue, and C. rhodochroum, which, alongside their shared core BGCs, possess species-specific clusters. The fact that these lineages diverged only 1–2 million years ago further emphasizes how rapidly these clusters can evolve, reflecting strong genomic plasticity within the genus (Figure 8).

Orthologous groups associated with secondary metabolism, growth regulation, metabolism, and pathogenicity point to potential functional divergences in C. rhodochroum. In particular, differentiation in genes involved in host or environmental interactions may reflect ecological specialization, an important criterion for species delimitation [81].

3.9. Genomic Distance

Comparing genomic distances is a key criterion for delineating species. In bacteria, the Average Nucleotide Identity (ANI) test is widely used to quantify genetic relatedness between strains. Recently, the tool FungANI was developed to apply the same methodology to fungal genomes [63]. Studies in Sordariomycetes suggest that genomes with ANI values greater than 99% belong to the same species. Nevertheless, their broader applicability across fungi remains uncertain and warrants cautious interpretation. In this analysis, C. rhodochroum ATHUM 6904 exhibited ANI values below this threshold when compared to C. tenue and C. rubrobrunnescens (Figure 12). This provides supportive evidence that it represents a distinct species. The genomic distance and the placement of C. rhodochroum as a sister clade to C. tenue, rather than within the same species, are further supported by concordant results across phylogenetic trees constructed from ITS, rpb2, and whole-genome data (Figure 1 and Figure 8).

4. Conclusions

In conclusion, the combined evidence derived from the analysis of molecular markers, macro/micromorphological data, and comparative genomics, including genomic distance, orthologous gene comparisons, metabolic variability, and mitochondrial organization, clearly demonstrates that C. rhodochroum is substantially differentiated from its closely related species C. tenue and C. rubrobrunnescens. Collectively, these convergent findings strongly support the recognition of the fungicolous C. rhodochroum as a distinct, hitherto unclassified species within the genus Cladobotryum.