Nine new species of black lichenicolous fungi from the genus Cladophialophora (Chaetothyriales) from two different climatic zones of China

Lichenicolous fungi are parasites of lichens. Many of these fungi are referred to as “black fungi”. A diversity of these black fungi include species that are pathogenic to humans and plants. A majority of black fungi reside in the phylum Ascomycota within the sub-classes Chaetothyriomycetidae and Dothideomycetidae. To explore the diversity of lichenicolous “black fungi” associated with lichens in China, we conducted several field surveys in the Inner Mongolia Autonomous Region and Yunnan Province between 2019 and 2020. We recovered 1,587 fungal isolates from the lichens collected during these surveys. During the preliminary identification of these isolates using the complete internal transcribed spacer (ITS), partial large subunit of nuclear ribosomal RNA gene (LSU), and small subunit of nuclear ribosomal RNA gene (SSU), we identified 15 fungal isolates from the genus Cladophialophora. However, these isolates had low sequence similarities with all known species from the genus. Therefore, we amplified additional gene regions, such as, translation elongation factor (TEF) and partial β-tubulin gene (TUB), and constructed a multi-gene phylogeny using maximum likelihood, maximum parsimony, and Bayesian inference. In our datasets, we included type sequences where available for all Cladophialophora species. Phylogenetic analyses revealed that none of the 15 isolates belonged to any of the previously described species in the genus. Therefore, using both morphological and molecular data, we classified these 15 isolates as nine new species within the genus Cladophialophora: C. flavoparmeliae, C. guttulate, C. heterodermiae, C. holosericea, C. lichenis, C. moniliformis, C. mongoliae, C. olivacea, and C. yunnanensis. The outcome from this study shows that lichens are an important refugia for black lichenicolous fungi, such as those from Chaetothyriales.


Introduction
"Black fungi" are characterized by dark-colored mycelia due to the buildup of melanin in their cell walls (Ametrano et al., 2019). This melanization of the cell wall allows them to colonize extreme habitats, such as rock surfaces and lichen thalli (Gostinčar et al., 2012). These groups of fungi can also have an assortment of thallus morphologies, such as yeast, pseudo-filamentous, and filamentous (Gostinčar et al., 2012). When co-cultured with compatible algae, several black fungi show an earlier stage of lichenization, such as the formation of lichen-like fungal plectenchyme (Brunauer et al., 2007;Muggia and Grube, 2018). A majority of black fungi reside in the phylum Ascomycota within the sub-classes Chaetothyriomycetidae and Dothideomycetidae (Harutyunyan et al., 2008). One such genus is Cladophialophora, from the order Chaetothyriales.
To explore the fungal diversity associated with lichens, we conducted several field surveys in two climatic zones of China, the Inner Mongolia Autonomous Region and Yunnan Province, between 2019 and 2020. During these surveys, we collected various species of lichens from these regions. We isolated an assortment of fungi from the medullary tissues of these lichens. Preliminary morphological and molecular identification of these fungi using the ITS gene region, we identified 15 isolates of black fungi. Further molecular and phylogenetic analyses revealed that these isolates belonged to the genus Cladophialophora, but did not represent any of the previously described species. Therefore, we classified these 15 isolates as nine new species within the genus Cladophialophora.
All lichen thalli were repeatedly rinsed with sterile denoised water and surface sterilized for 10 min under UV lamp. Using a Leica Zoom 2000 stereo microscope, the upper cortex of the thallus was scraped off and pieces of medullary tissues were rinsed in sterile deionized water. These tissue pieces were placed on potato dextrose agar (PDA, Qingdao Hope Bio-Technology Co., Ltd., China) amended with 0.05% streptomycin (Sangon Biotech (Shanghai) Co., Ltd., China).
All Petri plates were incubated at 25°C for 10 days in darkness. Mycelia emerging from medullary tissue pieces were sub-cultured onto fresh PDA plates. After that, pure cultures were obtained using single hyphal tip technique. All the ex-holotype cultures of new fungal species described in this study were deposited in the China General Microbiological Culture Collection Center (CGMCC), Beijing, China. The holotype specimens were deposited in the Institute of Microbiology (HMAS), Beijing, China (Accession numbers are listed in Table 1). The lichen thalli were deposited in the Collection of Shandong Normal University (SD).
All PCR products were sequenced by the Sangon Bioengineering (Shanghai) Co., Ltd. The resulting sequences were assembled using Geneious v.10.2.2 (Biomatters, Auckland, New Zealand). Preliminary identification of the isolates was done using BLAST (Altschul et al., 1990). All sequences of novel fungal species obtained in this study were deposited in the NCBI Gene Bank (Table 1).

Phylogenetic analyses
For phylogenetic analyses, ex-type sequences of ITS, LSU, SSU, TUB, and TEF for all Cladophialophora species were retrieved from the NCBI GenBank if available. The final datasets included sequences generated in this study and those retrieved from GenBank. All the datasets were individually aligned using MAFFT v. 7 (Katoh and Standley, 2013) and manually adjusted using MEGA v. 10.2.0 (Kumar et al., 2018). Individual gene regions were phylogenetically analyzed using the method described below, followed by the compilation and analysis of a concatenated dataset.
Phylogenetic analyses of single gene and concatenated datasets were done using three approaches. These were maximum likelihood (ML), maximum parsimony (MP), and Bayesian inference (BI). Software for ML and BI analyses were accessed through the CIPRES Science Gateway v. 3.3 1 (Miller et al., 2011). The best evolution models of each dataset were estimated using jModelTest2 (Darriba et al., 2012). ML analyses were conducted using RAxML-HPC2 with GTR + GAMMA as the substitution model and 1,000 bootstrap replications (Stamatakis, 2014). MrBayes v. 3.2.7 (Ronquist et al., 2012) was used for the BI analyses with the best substitution model TIM1 + I + G for SSU, TIM2ef + I + G for ITS, TrN + I + G for LSU, TPM2uf + I + G for TUB, TrNef+G for TEF, and SYM + I + G for concatenated data set. Four MCMC chains were run from a random starting tree. Twenty million generations were run with trees sampled at every 100 generations, resulting in 200,000 trees. Tracer v. 1.7 2 was used to determine the chain convergence and the effective sample size values. We discarded a quarter of sampled trees (50,000) during burn-in. Posterior probabilities were calculated from the remaining trees. MEGA v. 10.2.0 (Kumar et al., 2018) was used to for the MP analyses with 1,000 bootstrap replicates, gaps were treated as a fifth state character. FigTree v. 1.4.3 was used to visualize and edit trees. All the datasets and trees were submitted to the TreeBase (Study ID 29976).

Morphology and culture characteristics
For morphological studies, all isolates of the new fungal species identified in this study were used. Colony morphologies of potentially new fungal species were described from 21-day-old cultures growing at 25°C on PDA. Micro-morphological structures of the isolates were visualized and photographed using a Leica DFC95 camera attached to a Leica DM6 microscope. ImageJ v. 1.53u 3 was used for measuring taxonomically relevant features. At least 40 measurements were recorded for each morphological feature, and the statistics were presented as (minimum -) mean -standard deviationmean + standard deviation (−maximum).
Twenty-one-day-old fungal isolates grown on PDA at 25°C were utilized for the growth study. Agar plugs measuring 5 mm in diameter were obtained from the culture and placed in 90 mm Petri dishes containing PDA. Three replicate plates were prepared for each fungal isolate, spanning temperatures from 5 to 35°C at 5°C intervals (± 0.5°C). Following a 28-day period, the colony diameters of each isolate were measured.

Collections of lichens and isolations of fungi
A preliminary identification of 1,587 fungal isolates using ITS, LSU, and SSU sequences revealed that 15 isolates recovered from lichens were Cladophialophora species. Among these, five fungal isolates were recovered from Inner Mongolia Autonomous Region whereas ten from the Yunnan Province (Table 1). For a majority of the isolates, the ITS sequence had a similarity lower than 98% with all previously described Cladophialophora species. This suggested that among those isolates were some potentially unknown species.

Phylogenetic analyses
The topologies of the trees derived from the phylogenetic analyses of single gene and concatenated datasets were not identical due to unequal taxon sampling ( Figure 1, Supplementary Figures S1−S5). Henceforth, we used the ML tree emerging from the analyses of concatenated dataset for delineation of Cladophialophora species recovered in this study ( Figure 1). This concatenated data included 63 taxa (Table 1)  In the concatenated tree, Cladophialophora isolates CGMCC3.20360, CX08B15, CGMCC3.20359, and CGMCC3.24219 formed a monophyletic lineage that emerged as the sister to a clade that included Cladophialophora tengchongensis and Taxon 4 (see below). Among these four isolates, the monophyly of CGMCC3.20360 and CX08B15 (Taxon 1) received significant statistical support (ML/ MP/BI: 97/99/0.99; Figure 1). Isolate CGMCC3.20359 (Taxon 2) emerged as the sister taxon to Taxon 1 ( Figure 1). CGMCC3.24219 (Taxon 3) was sister to the clade that included Taxon 1 and Taxon 2 ( Figure 1). Even though the monophyly of these four isolates obtained considerable branch support in the concatenated tree, there was still a significant difference in the gene regions between the taxa ( Figure 2).
Isolates CX19Da and CGMCC3.20428 (Taxon 4) emerged as the sister taxa to C. tengchongensis in concatenated tree ( Figure 1). However, this relationship varied between the single gene trees (Supplementary Figures S1−S5). We were unable to amplify the TEF of the isolate CGMCC3.20428, as well as the SSU, LSU, TUB, and TEF of CX19Da. In ITS sequences, there were three base pairs differences between isolates of Taxon 4.
In the concatenated tree, CGMCC3.20361 (Taxon 8) did not group with any known species of Cladophialophora (Figure 1). This relationship was echoed in the LSU, SSU, and TEF trees (Supplementary Figures S2, S3, S5), but in the ITS and TUB trees, Taxon 8 formed a monophyletic clade with C. bromeliacearum (Supplementary Figures S1, S4). None of these relationships received significant branch support.

Taxonomy
Culture characteristics: The colony morphology on the PDA after 21 days was compact, dark gray in the center with a dark olivaceous gray margin, surface velutinous, convex, and margin entire ( Figure 3A). The colony grows slowly on PDA medium, reaching 17 mm in diameter after 4 weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: Each isolate of this fungus was recovered from two separate thalli of the lichen H. vexans collected in Yunnan Province of China.

Frontiers in
Culture characteristics: The colony morphology on the PDA after 21 days was compact, dark olive green in color, convex, surface villose, and margin entire ( Figure 3E). The colony grows slowly on PDA medium, reaching 15 mm in diameter after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: A single isolate of this fungus was recovered from Heterodermia pseudospeciosa collected in Yunnan Province of China.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, off-white in color, with thin dark entire margin, convex, surface villose, dark striation extending from the colony margin toward the center ( Figure 3I). The colony grows slowly on PDA, reaching 11 mm in diam after 4 weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: The isolates of this fungus was recovered from Hypotrachyna sinuosa collected in Yunnan Province of China.
Note: Cladophialophora guttulata is distinguished from C. holosericea and C. olivacea by its distinctive colony morphology and abundance of oil bodies (guttules) in hyphal compartments and conidia (Table 2, Figure 2).  The sequence variability between Cladophialophora holosericea, Cladophialophora olivacea and Cladophialophora guttulata.
Frontiers in Microbiology 09 frontiersin.org Diagnosis: Cladophialophora heterodermiae differs from its closely related species, C. tengchongensis, in regards to growth rates and colony morphology.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, smoky gray in color, with a dark-gray wide entire margin, surface villose, with distinct striation, light brown diffusible pigment around the colony ( Figure 3M). Colony grows slowly on PDA, reaching 9 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: The isolate of this fungus was recovered from Heterodermia pseudospeciosa collected in Yunnan Province of China.
Note: Cladophialophora heterodermiae is phylogenetically close to C. tengchongensis. However, these two species have substantial differences in growth rate and colony morphology, septal diameter, conidial morphology, and dimensions (Sun et al., 2020). These two species also have some differences between the ITS, LSU, and TUB gene regions: 36-40 bps in ITS, 10 bps in LSU, and 63 bps in TUB.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, dark gray in color, margin entire, slightly raised in the center, and surface smooth ( Figure 3Q). The colony grows slowly on PDA, reaching 10 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: Associated with three species of lichen, Flavopunctelia flaventior, Punctelia borreri, and Parmotrema reticulatum, in the Yunnan Province of China.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, gray to dark gray in color with distinct black entire margin center convex ( Figure 4A). The colony grows slowly on PDA medium, reaching 15 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5 and 35°C.
Habitat: Medullary tissue of the lichen of Parmelia sp. collected from Inner Mongolia Autonomous Region.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, dark pinkish brown in color with a prominent brown irregular margin, surface villose with distinct protuberances in the center ( Figure 4E). The colony grows slowly on PDA medium, reaching 16 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5 and 35°C.
Habitat: Medullary tissue of the lichens Xanthoparmelia tinatina and Heterodermia pseudospeciosa collected from Inner Mongolia Autonomous Region of China.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, dark gray in color, lobbed, with dark gray margin, striation emerging from the fissures extending to the center of the colony, surface villose, flat ( Figure 4I). The colony grows slowly on PDA medium, reaching 15.5 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, 30, and 35°C.
Habitat: A single isolate of this fungus was recovered from the medullary tissue of Flavoparmelia caperata collected in the Inner Mongolia Autonomous Region of China.
Culture characteristics: The colony morphology on the PDA after 21 days was compact, smoky gray in color with a diffusing dark gray margin, glossy around the margin, surface velutinous, flat, and margin entire with faint striations ( Figure 4M). The colony grows slowly on PDA medium, reaching 27.5 mm in diam after four weeks at 25°C. The optimal growth temperature is 25°C. No growth was observed at 5, and 35°C.
Habitat: A isolate of this fungus was recovered from the medullary tissue of Heterodermia pseudospeciosa collected in the Inner Mongolia Autonomous Region of China.

Discussion
In this study, 15 Cladophialophora isolates were recovered from nine lichen species collected from Yunnan Province and the Inner Mongolia Autonomous Region of China. Analyses of morphological characteristics and molecular data using phylogenetic approaches revealed that these 15 isolates belonged to nine novel species from the genus. Consequently, these species were classified as C. flavoparmeliae, C. guttulate, C. heterodermiae, C. holosericea, C. lichenis, C. moniliformis, C. mongoliae, C. olivacea, and C. yunnanensis.
The phylogenetic analyses in this study were carried out using five gene regions, namely ITS, SSU, LSU, TUB, and TEF. For the majority of previously identified Cladophialophora species, only sequences for ITS and LSU are available. Sequences of SSU, TUB, and TEF were available for 23, 8, and 15 species, respectively. This is why the tree topologies of the single-gene and the concatenated datasets differed substantially. During the phylogenetic analyses, we also realized that ITS and LSU were sufficient for discriminating a majority of Cladophialophora species. This is most likely why several newly described species, such as C. aquatica (Boonmee et al., 2021), C. cabanerensis (Crous et al., 2020), and C. tumbae (Kiyuna et al., 2018), were based on these two gene regions. However, it was challenging for us to discriminate closely related species, such as C. holosericea, C. olivacea, and C. guttulate, without using sequences from protein-coding gene regions. For example, the ITS sequence was insufficient for distinguishing C. olivacea from C. holosericea. Similarly, LSU alone was unable to differentiate between C. olivacea and C. guttulate. TEF, on the other hand, was able to differentiate all three species. In the future, considering to amplify protein-coding gene regions, such as TEF and TUB, in addition to ITS and LSU would positively influence the taxonomy of Cladophialophora.  (Shen et al., 2020), Pseudocladosporium hachijoense (Braun, 1998), and Alternaria malorum (Braun et al., 2003), respectively. We compared the micro-morphological characteristics of C. cladoniae [MB 800397], C. hawksworthii [MB 800398], C. megalosporae [MB 800399], and C. normandinae [MB 800400] with those of our nine species. These four Cladophialophora species have conidiomata (Diederich et al., 2013), a characteristic that was absent from our species. However, we were unable to compare the morphology of C. bennettii [MB 491905] to our species due to the unavailability of publication information for this species in any of the recognized databases, such as MycoBank and Index Fungorum. This evidence shows that all nine new species identified in this study were previously undocumented.
The species of Cladophialophora formed three distinct clades in the phylogenetic tree constructed using the concatenated dataset. Among these, C. hostae, C. scillae, and C. behniae formed the basal clade. Based on some previous studies, the phylogenetic position of this clade within Cladophialophora is controversial because these species might represent a new family (Gueidan et al., 2014;Quan et al., 2020). Similarly, the clade that included C. proteae, C. eucalypti, C. pucciniophila, C. modesta, C. sylvestris, C. humicola, and C. minutissima might not also include fungi from Cladophialophora. In a recent study, fungi from this clade grouped into various other families within Chaetothyriales (Quan et al., 2020). We opted to include these species in our phylogenetic studies because they are still classified as Cladophialophora. However, based on the data from this study and previous research, we propose a full taxonomic revision of the genus Cladophialophora.
A majority of Cladophialophora species, including those described in this study, emerged as a monophyletic clade with significant branch support values. This clade also included the type species, C. carrionii. Previously, based on SSU, LSU, and RPB1 sequence data, this clade included two phylogenetic groups: the Carrionii-clade and the Bantiana-clade (Badali et al., 2008;de Hoog et al., 2011). However, in a recent phylogeny of Chaetothyriales using ITS and LSU data, the existence of these clades was disputed (Quan et al., 2020). In this study, a phylogenetic tree constructed using a concatenated dataset showed the existence of the Carrionii-clade and the Bantiana-clade, but without significant branch support. Aside from that, the species composition of these clades did not completely overlap with that of Badali et al. (2008). This suggests that Carrionii-clade and Bantianaclade are not well defined and should be used with caution.
Among the isolates of the species identified in this study, polymorphism was detected between sequences for the same gene region, for example, C. lichenis and C. yunnanensis. Sun et al. (2020) and Badali et al. (2009) while describing C. nyingchiensis and C. carrionii, respectively, also reported this trend. There are two possible explanations for this phenomenon. These fungi have multiple copies of the ITS and TUB genes, as do many other fungi (Fourie et al., 2014;Zhao et al., 2014). Alternatively, these sequence differences between the isolates might also be the product of cryptic species. This is because multiple copies of TEF in fungi have yet to be reported. However, TEF sequences varied by 20 bp among C. nyingchiensis isolates (Sun et al., 2020). This dilemma may be resolved in the future by species discovery and the recovery of additional isolates of identified species.
Lichenicolous fungi can exclusively parasitize the mycobiont (mycoparasites), the photobiont (phycoparasites), or sometimes both by forming haustoria (Lawrey and Diederich, 2003;Diederich et al., 2018). Irrespective of their host choice, lichenicolous fungi can be further classified into two types. These are slow-growing species that cause no or minor symptoms in their hosts, and fast-growing ones that are highly pathogenic to lichens (Harutyunyan et al., 2008). Based on these facts, we hypothesize that all nine Cladophialophora species described in this study were of the former type. The relationship between these nine species and their lichen host is either biotrophic or commensal (Lawrey and Diederich, 2003). This explains why we saw no symptoms on the lichen from which we isolated these fungi. However, infection trials are necessary to confirm their interaction with their hosts.
Results of this study indicated that Cladophialophora is an important lineage of lichenicolous fungi. However, a majority of lichen-associated species are yet to be discovered. We still do not know the precise role of these fungi and black fungi as a whole in lichen thallus. The lack of symptoms in the lichen thalli where these fungi were isolated suggests that these fungi can also be saprophytes or stress-associated latent pathogens that only exhibit symptoms when the host lichen is subjected to stress (Lücking et al., 2021). In addition, a large number of guttles in the hyphae of all Cladophialophora species identified in this study, especially C. guttulate. Therefore, there is scope for evaluating the potential to use these fungi in the production of microbial oil.

Data availability statement
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.

Author contributions
RC and HS contributed to the conceptualization. YCW, XZ, and GZ performed the methodology and conducted the formal analysis. RC and TB was written the original draft preparation. YCW, YRW, YL, and SZ was performed the experiment. SL and MD carried out the resources. RC, HS, and TB wrote, reviewed, and edited the manuscript and directed the data. RC and GZ were responsible for project management and funding access. All authors contributed to the article and approved the submitted version.

Funding
This study was supported by the Qingchuang Talents Induction Program of Shandong Higher Education Institution in 2021, Open Fund for Instruments and Equipment of Shandong Normal University, the "Startup Fund" awarded to Runlei Chang by the Shandong Normal University and National College Students Innovative Entrepreneurship Training Programs [202210445022].