Adaptations of Pseudoxylaria towards a comb-associated lifestyle in fungus-farming termite colonies

Characterizing ancient clades of fungal symbionts is necessary for understanding the evolutionary process underlying symbiosis development. In this study, we investigated a distinct subgeneric taxon of Xylaria (Xylariaceae), named Pseudoxylaria, whose members have solely been isolated from the fungus garden of farming termites. Pseudoxylaria are inconspicuously present in active fungus gardens of termite colonies and only emerge in the form of vegetative stromata, when the fungus comb is no longer attended (“sit and wait” strategy). Insights into the genomic and metabolic consequences of their association, however, have remained sparse. Capitalizing on viable Pseudoxylaria cultures from different termite colonies, we obtained genomes of seven and transcriptomes of two Pseudoxylaria isolates. Using a whole-genome-based comparison with free-living members of the genus Xylaria, we document that the association has been accompanied by significant reductions in genome size, protein-coding gene content, and reduced functional capacities related to oxidative lignin degradation, oxidative stress responses and secondary metabolite production. Functional studies based on growth assays and fungus-fungus co-cultivations, coupled with isotope fractionation analysis, showed that Pseudoxylaria only moderately antagonizes growth of the termite food fungus Termitomyces, and instead extracts nutrients from the food fungus biomass for its own growth. We also uncovered that Pseudoxylaria is still capable of producing structurally unique metabolites, which was exemplified by the isolation of two novel metabolites, and that the natural product repertoire correlated with antimicrobial and insect antifeedant activity.


INTRODUCTION
The Macrotermitinae are the only termite lineage to have acquired fungal symbionts from the genus Termitomyces (family Lyophyllaceae) as their food source [1][2][3][4]. Termitomyces is cultivated by workers in cork-like structures termed "fungus combs", which are maintained in chambers located within the subterranean colony and are also known as "fungal gardens" [3]. To propagate the food fungus, younger workers ingest plant material alongside with Termitomyces biomass and use excreted lignocellulose and sporeenriched feces to craft new fungus comb on which Termitomyces is able to thrive (Fig. 1A). Termites have several levels of defense measures to protect this obligate nutritional symbiosis, starting with lower individual levels of hygiene measures to a higher collective level, also called social immunity [5][6][7][8]. Despite these preventive measures, fungal gardens inconspicuously host members of a distinct fungal subgenus of Xylaria (Ascomycota: Xylariaceae), commonly referred to as termite-associated Pseudoxylaria [9][10][11][12][13][14][15], which only emerge as vegetative stromata from comb material of deteriorating or inactive termite nests (Fig. 1B) [16]. While a number of studies have provided insights into their co-evolutionary relation with the fungus-farming termite symbiosis, the ecological role of Pseudoxylaria remains debated [1,7]. Although few reports suggested a commensal role supporting biomass degradation within the comb environment [10,17], other studies analyzing Termitomyces-Xylaria co-cultures hinted towards an antagonistic relation. As free-living Xylaria strains inhibited growth of Termitomyces more intensly than their termite-associated relatives [7,18,19], it was postulated that reduced antagonistic behavior might enable Pseudoxylaria to evade the defense mechanisms of a healthy termite colony, and once conditions are favourable to outcompete the fungal mutualist [10][11][12][13][14][15][16].
Driven by the rather anecdotal evidence for the "reduced antagonism hypothesis" and the additional postulation that coevolved Pseudoxylaria strains might have become a fungus garden substrate specialist over evolutionary time [11][12][13], we sequenced the genomes of seven and transcriptomes of two Pseudoxylaria isolates to investigate the genomic, transcriptomic and metabolomic basis for symbiotic associations. Whole genome-based comparison with free-living members of this genus uncovered a substantial reduction in genome size and numbers of proteincoding genes, as well as reduced functional capacities, all of which indicated that Pseudoxylaria might have become a dependent symbiont and comb-substrate specialist. By analyzing the secondary metabolite repertoire as well as co-cultivation studies along with isotope experiments, we were further able to solidify the "reduced antagonism hypothesis". To test if a fungus comb-associated lifestyle of Pseudoxylaria was reflected in differences at the genome level, we sequenced the genomes of all seven isolates using a combination of pairedend shotgun sequencing (BGISEQ-500, BGI) and long-read sequencing (PacBio sequel, BGI or Oxford Nanopore Technologies, Oxford, UK). In addition, we sequenced the transcriptomes (BGI-SEQ, BGI) of two isolates (X802, X170LB). Eleven publicly available genomes of free-living Xylaria ( Fig. 2A, B) were used as reference genomes (Table S4). Hybrid draft genomes were comprised on average of 33-742 scaffolds with total haploid assembly lengths of 33. 2-40.4 Mb, and a high BUSCO completeness of genomes (> 95 %) with a total number of predicted proteins ranging from 8.8 to 12.1 × 10 3 . The GC content was comparable to reference genomes with 49.7-51.6%. To verify the phylogenetic placement of the isolates, different genetic loci encoding conserved protein sequences (α-actin (ACT), second largest subunit of RNA polymerase (RPB2), β-tubulin (TUB) and the internal transcribed spacer (ITS) were used as genetic markers [7,13].
As our seven isolates covered a larger portion of the previously reported phylogenetic diversity of the termite-associated subgenus, we elaborated on genomic characteristics of our isolates to uncover features of the termite-associated ecology of Pseudoxylaria. Indeed, comparative genome analysis of the South African Pseudoxylaria isolates with publicly available genomes of freeliving Xylaria species of similar genome quality revealed significantly reduced genome assembly lengths in Pseudoxylaria with reduced numbers of predicted genes per genome (Table S4). Comparison of the annotated mitochondrial (mt) genomes ( Figure S5, Table S8) also indicated that all seven mt genomes were shorter in length (assembly lengths: 18.5-63.8 kbp) compared to the, albeit few, publicly available mitochondrial genomes of free-living species (48.9-258.9 kbp). The reduction in mitochondrial genome size also corresponded to a significantly reduced mean number of annotated genes (7.6) and tRNAs (14.3) in Pseudoxylaria spp. compared to on average 30.0 (annotated genes) and 25.8 (tRNAs) found in free-living species.
Analysis of the abundance and composition of transposable elements (TEs), which account for up to 30-35% of the genomes of (endo)parasitic fungi due to the expansion of certain gene families [20,21], showed that the mean total numbers of TEs across Pseudoxylaria spp. genomes were comparable (1530), but the numbers were reduced compared to free-living Xylaria species (3690) ( Table S9). We also identified high variation in the TE composition across genomes (1.5-9.9 %), comparable to what was observed in free-living Xylaria spp. (1.3-8.1 %), with reductions in Fig. 1 Natural growth and occurrence of fungal mutualist Termitomyces and termite-associated Pseudoxylaria strains. A Mature fungus comb from a Macrotermes natalensis colony with spore-containing fungal nodules of a Termitomyces strain, B Pseudoxylaria stromata emerging from fungus comb after incubation for eight days in the absence of termites, and C-E axenic fungal cultures of Pseudoxylaria strains isolated from different termite mounds grown on agar plates with, C Pseudoxylaria sp. Mn132, D Pseudoxylaria sp. X3-2, and E Pseudoxylaria sp. Mn153. long terminal repeat retrotransposons (LTRs: Copia and unknown LTRs) in two inverted tandem repeat DNA transposons (TIRs; CACTA, Mutator and hAT). As Pseudoxylaria spp. contained increased numbers of non-ITR transposons of the helitron class and LTRs of the Gypsy class compared to Xylaria strains, we concluded that Pseudoxylaria exhibits no typical features of an (endo)parasitic lifestyle, but that the overall composition and the reduced numbers of TEs could serve as a fingerprint to distinguish the genetically divergent Pseudoxylaria taxa.
Repertoire of carbohydrate-active enzymes indicates specialized substrate use As the fungus comb is mostly composed of partially-digested plant material interspersed with fungal mycelium of the termite mutualist [3], we anticipated that Pseudoxylaria should exhibit features of a substrate specialist similar to the fungal mutualist Termitomyces, which should be reflected in a Carbohydrate-Active enzyme (CAZyme) repertoire distinguishable from free-living saprophytic Xylaria species [22][23][24]. In particular, numbers and composition of redox-active enzymes (e.g., benzoquinone reductase (EC 1.6.5.6/EC 1. 6 Table S10), which catalyze the degradation of lignin-rich biomass, were expected to differ between free-living strains and substrate specialists [22].
Overall, significant differences in the composition of CAZymes were observed [8], most notably in the reduction of auxiliary activity enzymes (AA), carbohydrate esterases (CE), glycosyl hydrolases (GH), and polysaccharide lyases (PL). The most significant reduction was observed in the AA3 family (Fig. 3C), which typically displays a high multigenicity in wood-degrading fungi as many enzymes of this family catalyze the oxidation of alcohols or carbohydrates with the concomitant formation of Geographic and comparative phylogenomic analysis of termite-associated Pseudoxylaria isolates (strains 1-7) and free-living Xylaria (strains [8][9][10][11][12][13][14][15][16][17][18]. A Geographic origins of genome-sequenced free-living Xylaria and termite-associated Pseudoxylaria isolates, B phylogenomic placement based on single-copy ortholog protein sequences, and C comparison of genome assembly length, and numbers of predicted proteins per genome. hydrogen peroxide or hydroquinones thereby supporting lignocellulose degradation by other AA-enzymes, such as peroxidases (AA2). Similarly, although to a lesser extent, reduced numbers within the related AA1 family were detected, which included oxidizing enzymes like laccases, ferroxidases, and laccase-like multicopper oxidases. Along these lines, glycosyl hydrolases of the GH3 and GH5 family, including enzymes responsible for degradation of cellulose-containing biomass and xylose, were less abundant. We also noted that all Pseudoxylaria lacked homologs of the unspecific peroxygenases (UPO; EC 1.11.2.1), while almost all free-living Xylaria spp. and the fungal symbiont Termitomyces harbored at least one or two copies of similar gene sequences.
Pseudoxylaria shows reduced biosynthetic capacity for secondary metabolite production A healthy termite colony is engulfed in several layers of social immunity [5,6], which pose a constant selection pressure on associated and potentially antagonistic microbes. As Pseudoxylaria evolved measures to remain inconspicuously present within the comb environment, we hypothesized that one of the possible adaptations to evade hygiene measures of termites could be reflected in a reduced biosynthetic capability to produce antibiotic or volatile natural products, which often serve as infochemicals triggering defense mechanisms [25][26][27], or as alarm pheromones [4,28].
The biosynthesis of secondary metabolites is encoded in so called Biosynthetic Gene Cluster (BGC) regions. We explored the abundance and diversity of encoded BGCs using FungiSMASH 6.0.0 and manually cross-checked the obtained data set by BLAST to account for possible biases due to varying genome qualities across strains of both groups [29]. Overall, the herein investigated Xylaria genomes harbored on average 90 BGCs per genome, while Pseudoxylaria encoded on average 45 BGCs (Fig. 4, Figure S7).
The nature and relatedness of the BGCs were analyzed by creating a curated similarity network analysis using BiG-SCAPE 1.0 [30]. Overall, 28 orthologous BGCs were shared across all genomes, including the biosynthesis of polyketides like 6-methylsalicylic acid (MSA), chromenes (Chr) and polyketidenon-ribosomal peptide (PKS-NRPS) hybrids like the cytochalasins (Cyt) [31]. Furthermore, five BGC networks, which were shared by Pseudoxylaria and Xylaria, contained genes encoding natural product modifying dimethylallyltryptophan synthases (DMATS). In contrast, and despite the significant reduction in the biosynthetic capacity within Pseudoxylaria genomes [29], about 29 BGC networks were unique to Pseudoxylaria and thus could possibly relate to the comb-associated lifestyle ( Figure S8 and S9). Notably, Pseudoxylaria genomes lacked genes encoding ribosomally synthesized and posttranslationally modified peptides (RiPPs) or halogenases. In comparision, free-living Xylaria spp. harbored at least one sequence encoding a RiPP, and up to two orthologous sequences encoding putative halogenases. In contrast, a reduced average number of terpene synthases (TPS) in Pseudoxylaria (9 TPS) compared to free-living Xylaria (18 TPS) was detected, which included three BGCs encoding TPSs that were unique to Pseudoxylaria. In comparison, genomes of the fungal mutualist Termitomyces were reported to encode for about 20-25 terpene cyclases, but haboured only about two loci containing genes for a PKS and NRPS each [24].
A detailed analysis of the fog-like cluster arrangements within Pseudoxylaria genomes revealed -similar to homologs of the ccs clustervariation in the abundance and arrangement of several accessory genes coding for a cupin protein (pxF), a short chain oxidoreductase (pxB; SDR), and an additional SnoaL-like polyketide cyclase (pxP), which could account for the production of strainspecific structural congeners (vide infra, Fig. 7).

Change of nutrient sources causes dedicated transcriptomic changes in Pseudoxylaria
To further solidify our in silico indications of substrate specialization with comb material as preferred substrate and fungus garden as environment, we analyzed Pseudoxylaria growth on different media (PDA, and reduced medium 1/3-PDA) including comb-like agar matrices (wood-rice medium (WRM), agar-agar or 1/3-PDA medium containing lyophilized (dead) Termitomyces sp. T112 biomass (T112, respectively T112-PDA), PDB covering glass-based surface-structuring elements (GB), Table S11-S14).
Cultivation of Pseudoxylaria on agar-agar containing lyophilized biomass of Termitomyces (T112) as the sole nutrient source allowed Pseudoxylaria to sustain growth, although to a reduced extent compared to growth on nutrient-rich PDA medium (Table S3). Wood-rice medium (WRM) induced comparable growth rates as observed on PDA and also the appearance of phenotypic stromata.
To investigate the influence of these growth conditions on the transcriptomic level, we harvested RNA from vegetative mycelium after growth on comb-like media (WRM, T112, T112-PDA, and GB), PDA, and reduced medium 1/3-PDA (Fig. 5A). The most significant transcript changes (normalized to data obtained from growth on PDA) were observed for genes coding for specific CAZymes including several redox active enzymes (Fig. 5B). The 30 most variable transcripts coded for specific glycoside hydrolases (GH), lytic polysaccharide monooxygenases (AA), ligninolytic enzymes, and a glycoside transferase (GT). Similarly, chitinases (CHT2; CHT4; CHI2; CHI4) were upregulated (up to 243-fold on T112) under almost all conditions compared to PDA, but some of these specific transcript changes were exclusive to growth on Termitomyces biomass or artificial comb material (WRM) suggesting the ability to regulate and increase chitin metabolism if necessary [37].
When the fungus was challenged with lignocellulose-rich WRM medium, higher transcript levels putatively assigned to glutathione peroxidase (GXP2), superoxide dismutase (SOD2), and laccases (LCC5) were observed, which indicated that despite the reduced wood-degrading capacity, Pseudoxylaria activates available enzymatic mechanisms to degrade the provided material and respond to the resulting oxidative stress. T112 biomass (T112). B Heatmap of the most variable transcripts coding for CAZymes (red), redox enzymes (orange), secondary metaboliterelated core genes (green), and more specifically on key genes within the boundaries of cytochalasin (turquoise) and xylasporin/cytosporin BGCs (blue). RNA was obtained from vegetative mycelium after growth on PDA, reduced medium (1/3-PDA), PDB on glass beads (GB), woodrice medium (WRM), 1/3-PDA-medium enriched with Termitomyces sp. T112 biomass (T112-PDA) and agar-agar medium containing lyophilized Termitomyces biomass (T112). Transcript counts are shown as log 10 transformed transcripts per million (top; TPM). Significance of the changes in transcript counts are compared to control (X802 grown on PDA) and depicted in log -10 transformed p values.
To test the hypothesis that the presence of Termitomyces biomass stimulates secondary metabolite production in Pseudoxylaria to eventually displace the mutualist, we also analyzed changes in the transcript levels of core BGC genes that encode the production of bioactive secondary metabolites. Overall, only slight transcript variations were detectable within the most variable expressed genes. (Fig. 5B). Cultivation on GB, WRM, and T112 media caused lower transcript levels of genes coding for terpene synthase TC1, polyketide synthases (PKS7, PKS8), and the NRPS-like1, while an upregulation of NRPS-like2 on WRM (2.5-fold), and of PKS7 (1.7-fold) on reduced 1/3-PDA medium was observed.
Transcript levels of core genes within BGCs assigned to cytochalasines (cyt) or xylasporins/cytosporins (px), e.g., remained nearly constant, while minor transcript level variations of neighboring genes and reduced transcript levels for pxI (flavindependent monooxygenase), pxH (ABBA-type prenyltransferase), pxF (cupin fold oxidoreductase), and pxJ (short-chain dehydrogenase) were detectable. Hence, it was concluded that the presence of Termitomyces biomass only weakly triggers secondary metabolite production in general, but varying transcript levels coding for decorating enzymes could cause substantial structural alterations within the produced natural product composition. It was also notable that transcript levels of the terpene synthase TC1 were downregulated, which could cause a reduced production level of specific volatiles.
Pseudoxylaria antagonizes Termitomyces growth and metabolizes fungal biomass The growth behavior of Pseudoxylaria isolates was also analyzed in co-culture assays with Termitomyces. As expected from prior studies, both fungi showed reduced growth when co-cultured on agar plates, often causing the formation of zones of inhibition (ZOI) between the fungal colonies ( Fig. 6A-D, Table S11-S14) [7]. When fungus-fungus co-cultures were maintained for longer than two weeks on agar plates, Pseudoxylaria started to overcome the ZOI and overgrew Termitomyces via the extension of aerial mycelium. The observation was even more pronounced when co-cultures were performed on wood-rice medium (WRM), where Pseudoxylaria remained the only visible fungus after two weeks.
the isotope pattern of metabolites derived from Pseudoxylaria thriving on living biomass of Termitomyces (T112 ǂ ) was analysed next. Here, an overall positive carbon isotope ( 13 C/ 12 C) fractionation by approximately +0.6 ‰ relative to the control medium was detectable, while the δ 13 C values of lipids remained largely unchanged (Fig. 6F, Table S15). These results suggested that Pseudoxylaria might pursue a preferential uptake of Termitomycesderived carbohydrates.
In a last experiment, Pseudoxylaria was grown on lyophilized (dead) Termitomyces biomass (T112) as sole food source. In this experiment, the isotope fingerprint showed converging δ 13 C values of −1.9 ‰ (relative to the media) for both carbohydrate and lipid fractions, which indicated that Pseudoxylaria is able to simultaneously metabolize and cycle carbohydrates as well as lipids resulting in the equilibration of isotopic levels between carbohydrates and lipids. Thus, it was concluded that in nature, Pseudoxylaria likely harvests nutrients firstly from vegetative Termitomyces, and then-if possible-subsequently degrades dying or dead mycelium.
Pseudoxylaria produces antimicrobial secondary metabolites Based on the observation that Pseudoxylaria antagonizes growth of Termitomyces, we questioned if the formation of a ZOI might be caused by the secretion of Pseudoxylaria-derived antimicrobial metabolites [26,42]. Thus, we performed an ESI(+)-HRMS/MS based metabolic survey using the web-based platform "Global Natural Product Social Molecular Networking" (GNPS) [43] to correlate the encoded biosynthetic repertoire of Pseudoxylaria with secreted metabolites.
A cluster that contained MS 2 signals of molecular ions assigned to the cytosporin/xylasporin family, which was shared by at least four strains, caught our attention as a certain degree of structural diversity of xylasporin/cytosporin family was predicted from the comparison of their respective BGCs. The assigned nodes of this GNPS cluster split into two subclusters with only very little overlap between both regions. Analysis of the mass fragment shifts suggested that both subclusters belong to two different families of xylasporin/cytosporin congeners ( Figure S9). To verify these deductions, we pursued an MS-guided purification of xylasporin/ cytosporins from chemical extracts of Pseudoxylaria sp. X187, which yielded xylasporin G (3.23 mg, pale-yellow solid) and xylasporin I (1.75 mg, pale-yellow solid). The sum formulas of xylasporin G and xylasporin I were determined to be C 17 (Fig. 7B, Figure S10, Table S16-S17). Planar structures were deduced by comparative 1D and 2D NMR analyses, which revealed the presence of an unsaturated polyketide chain that matched the unsaturation degree and the anticipated structural variation from cytosporins (Fig. 7C, Figure S11-S25).
To evaluate if Pseudoxylaria-derived culture extracts and produced natural products (e.g., cytochalasins) are responsible for the observed antimicrobial activity, standardized antimicrobial activity assays were performed (Table S17, S18 and Figure S26). As neither culture extracts nor single compounds exhibited significant antimicrobial activity, they could not be held fully accountable for the antagonistic behavior in co-cultures. Thus, we hypothesized that the observed ZOI might be caused by yet unknown effects like nutrient depletion or bioactive enzymes.
Pseudoxylaria has a negative impact on the fitness of insect larvae Due to the production of structurally diverse and weakly antimicrobial secondary metabolites, we questioned if mycelium of Pseudoxylaria exhibits intrinsic insecticidal or other insectdetrimental activities, which could discourage or ward off grooming behavior of termite workers. Due to the technical challenges associated with behavioral studies of termites, we evaluated instead the effect of Pseudoxylaria biomass on Spodoptera littoralis, a well-established insect model system and a destructive agricultural lepidopterous pest [44,45]. When S. littoralis larvae were fed with mycelium-covered agar plugs of Pseudoxylaria sp. X802, a clear decrease of the relative growth rate (RGR) and decline in survival was observed (Fig. 8: treatment D (green), Table S19, S20) compared to feeding with untreated agar plugs (treatment A (black)). In comparison, when larvae were fed with agar plugs covered with the fungal mutualist Termitomyces sp. T153 (treatment B (blue)) an increased growth rate of larvae was observed.
Additionally, S. littoralis larvae were also fed agar plugs that had been cleaned from fungal mycelium prior to feeding to test if secreted metabolites and/or depletion of nutrients within the agar matrix might have an impact on RGR and survival. Here, it was surprising to note that agar-plugs derived from Pseudoxylaria sp. X802 cultures resulted in the death of all treated caterpillars within six days (treatment E (yellow)). In contrast, feeding with agar plugs previously covered with Termitomyces mycelium (treatment C (red)) caused the survival of almost all caterpillars until the end of Fig. 8 Effect of Termitomyces sp. T153 and Pseudoxylaria sp. X802 mycelia on the relative growth rate and survival of S. littoralis larvae. Insects were fed with either A PDA, B PDA agar plug covered with vegetative Termitomyces sp. T153, C PDA agar plug from which vegetative Termitomyces sp. T153 was removed prior to feeding, D PDA agar plug covered with vegetative Pseudoxylaria sp. X802 mycelium, and E PDA agar plug from which vegetative Pseudoxylaria sp. X802 mycelium was removed prior to feeding. All experiments were performed with 25 replicates per treatment, a duration of 10 days, and larval weights and survival rates were recorded every day. Statistical significances were determined using ANOVA on ranks (p < 0.001, n = 25) followed by Dunn's post-hoc test (indicated by different letters at the alpha = 0.05 level).
the experiment, although a slight decline on RGR was observed compared to treatment B (Fig. 8). Thus, an overall beneficial nutritional effect of Termitomyces was clearly visible, although a minor negative effect of nutrient depletion within the agar environment during fungal growth could not be fully excluded. Overall, we corroborated from these results that Pseudoxylaria exhibits a pronounced negative effect on insect growth and survival, likely due to the combined effect of harmful metabolite secretion, indigestible fungal mycelium and/or nutrient depletion of the growth environment.

CONCLUSION
Symbioses of fungi and social insects have independently evolved multiple times in ants, termites, beetles [3], and bees [46]. While genome reduction, and concomitant gene loss are commonly observed alongside with increased specialization and interdependencies in intracellular symbiotic bacteria during their transition to obligate symbiosis, examples of features that define fungal symbiotic interdependencies are sparse [47]. Characterizing features accompanying the evolution of symbiotic fungi is critical to understand symbiotic adaptations and the diversity of life across kingdoms.
Capitalizing on the availability of viable cultures from South African termite colonies, we tested if Pseudoxylaria shows features of a termite and comb-associated lifestyle on genomic, transcriptomic and metabolomic levels. In this study, we uncovered genomic evidence for a certain degree of substrate specialization in Pseudoxylaria isolates compared to free-living isolates. Similar to termite-associated clades of the fungal genus Podaxis [48] and fungal symbiont of attine ants [49], comparative genome analysis revealed reduced sizes and coding capacities, with a reduced enzymatic capacity to oxidatively degrade recalcitrant plant polymers in all Pseudoxylaria genomes. Although stochastical losses of biosynthetic traits during evolution cannot be excluded, the depletion in specific traits related to saprophytic life styles has likely been driven by a relaxed selection due to the more benign and constant growth conditions (fungus comb) and the availability of fungus-derived carbohydrate and protein-rich biomass. Based on these findings, and analogous to reports from other obligate fungal symbionts of insects [3,48], we conclude that Pseudoxylaria is likely an obligate symbiont adapted to the fungus comb environment of farming termites.
While the association of Pseudoxylaria with termites might have provided several fitness benefits to the fungus (presence of a carbon/nitrogen-rich comb substrate, protection from UV radiation by the termite mound, presence of ambient temperatures and humidity), termite-associated strains also face biotic stressors within the comb environment, such as termite weeding, cooccurring bacterial communities, and competition from and natural products produced by the fungal cultivar Termitomyces.
We hypothesized that Pseudoxylaria adapted to such combspecific stressors by having a reduced but specialized secondary metabolome to reduce triggers that could stimulate alarm responses of the fungal mutualist and termites [50], and the need for specific defense and communication mechanisms to survive. Comparative genome analysis supported the former of these hypotheses as a unique but reduced repertoire of BGCs was present in Pseudoxylaria genomes with a notable reduction in TPSs. In contrast, the co-occurring fungal mutualist Termitomyces has been found to encode above average numbers of TPSs in previous studies, which correlated with the emission of a bouquet of volatile terpenoid products proposed to play roles in the fungal life cycle by exhibiting insect attractant as well as repellant features [51]. These findings aligned with previous reports on behavioral studies showing that worker castes of O. obesus were able to differentiate between their mutualistic crop fungus Termitomyces and vegetative mycelium of Pseudoxylaria by their volatilome [50].
Our metabolic analysis demonstrated that Pseudoxylaria also secretes diffusible bioactive and structurally unique natural products as exemplified by the isolation of two novel metabolites [43]. While neither these or previously identified metabolites had strong antifungal activity, the production of antimicrobial mixtures could still represent a potential benefit in the competition against the co-occurring microbiota and the fungal mutualists to obtain nutrients.
In nature, Pseudoxylaria only emerges from weakened or abandoned comb, where the fungus overgrowths the deteriorating fungus garden. We investigated this phenotypic appearance and documented that Pseudoxylaria exhibited not only moderate antagonistic behavior against the termite mutualist Termitoymces without instantaneously killing the fungus (reduced antagonism), but showed signs of fungal biomass conversion. The hypothesis that Pseudoxylaria might harvest nutrients from vegetative Termitomyces mycelium was supported by comparative genome and RNAseq analyses as well as isotope fractionation results.
Overall, this study also provides a good starting point to address several unanswered questions as it still remains puzzling how and when Pseudoxylaria enters and remains within the fungus comb, why stromata emerge in the absence of termites ("sit and wait strategy"), and what triggers are required to stimulate germination and growth. This study should also encourage scientists to intensify sampling and sequencing studies on these and other fungal genera to enable broader phylogenomic studies that can address factors driving the evolution of insect-associated fungi in general and termite-associated strains specifically.

MATERIAL AND METHODS Cultivation procedures
Isolation. Fungal samples were collected from different termite mounds of Macrotermes natalensis, Odontotermes spp., and Microtermes spp. termite species within the years 2015-2018 (Table S1). Comb material was incubated in boxes at room temperature (rt) in the absence of termites, which resulted in the appearance of Xylaria-like stromata from fungus comb material. These were aseptically transferred to potato-dextrose agar (PDA) plates and cultivated for several cycles until seven axenic, morphologically distinct and viable cultures were obtained.
Whole genome sequencing and phylogenetic placement of Pseudoxylaria DNA extraction. Mycelium was harvested from agar plates, frozen in liquid nitrogen and grounded to a fine powder. DNA was extracted using CTAB and purified by chloroform-isoamylalcohol (24:1) and subsequent alcohol precipitation for PCR amplification. For direct sequencing the DNeasy Plant kit (Quiagen GmbH, Hilden, Germany) was used. Isolated DNA was kept frozen at −78°C until sequencing.
Whole genome sequencing. Whole-genome sequencing was performed using a 150 bp paired-end shotgun (BGIseq) and long-read (PacBio sequel) sequencing at BGI. Additionally, Oxford Nanopore technology (Oxford Nanopore Technologies, Oxford, UK) was employed for long-read sequencing. The MinION sequencing library was prepared using the Rapid DNA sequencing kit (SQK-RAD4) according to the manufacturer. DNA sequencing was performed on a MinION Mk1B sequencing device equipped with a R9.4.1 flow cell, which was prepared and run according to the manufacturer.
Whole genome assembly. Sequencing results were checked for quality using FastQC version 0.11.8 [52] and MultiQC version 1.7 [53]. Kmer depth was calculated using Jellyfish version 2.2.10 [54] and Kmer-based estimates of genome size, heterozygosity, and repeat content generated using GenomeScope [55]. A hybrid de novo genome assembly, combining BGISeq and PacBio data, was performed using SPAdes version 3.13.0 [56]. Nanopore sequencing raw data was generated using MinKNOW software version 4.0.20 (Oxford Nanopore Technologies), and was base-called and trimmed using Guppy version 4.2.2 (Oxford Nanopore Technologies). The resulting fastq files were filtered using Nanofilt [57]. A hybrid de novo genome assembly, combining BGISeq, PacBio and Oxford Nanopore data, was performed using MaSuRCA version 3.4.1 [58]. The resulting draft assembly was then polished with the accurate Illumina reads using the POLCA genome polisher (Table S4) [59].
RNA-Seq analysis. Sequencing of RNA from was performed by BGISeq (BGI, Hong Kong). Data was mapped against the annotated genome of X802 using Geneious Prime v2021.2.2 (Biomatters Ltd). Normalized transcript counts per million (TPM) and significance of the changes compared to normal growth on PDA were calculated using the built-in function in Geneious. Briefly, the TPM value was calculated using the following formula: TPM = (CDS read count * mean read length * 10^6)/ (CDS length * total transcript count). The p value was calculated by multiplication of the normalized mean probabilities that a randomly selected transcript would come from a gene (number of transcripts mapped to that gene/total number of transcripts from that sample) of each sample. The resulting data was sorted by variance and imported into R v4.1 (R Foundation) and the TPM and p values of most variable genes were log 10 and log −10 transformed, respectively. A heatmap for the identified genes was generated using the pheatmap package v1.0.12 in R with color schemes generated by viridis v0.5.1.
Functional gene annotation. Functional gene annotation was performed using InterProScan version 5.40-77.0 [64] with annotation of Gene Ontology (GO) terms, Panther families and KEGG pathways turned on using the "goterms," "iprlookup" and "pathways" options respectively. The InterProScan results were then incorporated into the Orthofinder results using Kinfin version 1.0 [65]. Per-orthogroup Gene Ontology (GO) enrichment analyses for each combination of clade types were performed using Pfam domains from InterPro [66] and Kinfin [65] in dcGO [67].
Phylogenetic analysis. Genetic loci of interest (internal transcribed spacer (ITS)), second largest subunit of RNA polymerase (RPB2), βtubulin (TUB), α-actin (ACT) were identified by BLAST [68] search from groups of reference sequences against the fungal genomes. Reference sequences were chosen from the NCBI RefSeq Targeted Loci database [Accession: PRJNA177353] (Table S4). To proof, that correct sequences were obtained by BLAST search, some sequences were doublechecked by PCR amplification and sequencing. If necessary, results from BLAST search and sequencing were combined. Amplification of the target partial gene sequences (fRPB2, ITS, ACT) [13] was done by PCR with S7 Phusion Polymerase (Biozym, Germany): 98°C 30 s, 35 cycles of 98°C 30 s, 55-66°C for 30 s per kb, and 72°C for 30 s per kb followed by a final denaturation at 98°C 30 s and extension at 72°C for 5 min (Table S6). PCR products were cleaned by gel purification (Zymoclean Gel DNA Recovery Kit, Zymo Research, USA) and sequenced at Eurofins Genomics (Ebersberg, Germany). For phylogenetic analysis, sequences from each target loci were aligned with ClustalW, implemented in MEGAX [69]. Phylogenetic trees were prepared with IQ-TREE [70] using ModelFinder [71] and UFBoot [72] with 1000 bootstrap replicates. For phylogenetic analysis, the Galaxy Eu platform was used [73].
Mitochondrial genome assembly and annotation. An initial mitochondrial genome assembly was performed using Norgal version 1.0.0 [74] with the best candidate being selected based on a BLAST search. The specimen with the longest circular contig from Norgal was set as a reference and the longest sequence from each strain as a seed in NOVOplasty v4.0 [75]. Mitochondrial genomes were annotated using the MITOS web server [76]. Pseudoxylaria mitochondrial genomes were aligned for the visualization of co-linear blocks using Mauve [77] implemented in Geneious Prime (v2021.2.2; Biomatters Ltd.) ( Figure S5) Orthology and phylogenomic analyses. Predicted protein sequences from the Augustus annotation were used as inputs to Orthofinder version 2.3.12 [78] using BLAST [79], Mafft [80] version 7.455, FastTree [81] version 2.1.10 and combined into a single species tree using STRIDE [82] and STAG [83]. All phylogenetically-weighted analyses used the phylogeny produced by Orthofinder. The trees were visualized in the ape and dendextend packages, versions 5.3 [63] and 1.14.0 [84] respectively, in R version 3.6.3 (R Core Team, 2020). Calibration of the phylogenomic tree was performed using the chronos function in ape and visualized using phytools version 0.7-47 [62] in R (R Core Team, 2020) ( Figure S2-S4).
CAZyme analysis. [85] Functional annotation of carbohydrate-active enzymes was performed using HotPep version 1 with default parameters [86,87]. CAZymes for which specific EC numbers could be identified were manually annotated with their substrate using the ExPASy database [88] and classified by the substrates presence in plant, bacterial or fungal cell walls. Comparisons of CAZymes and target substrates between clade types were performed using a phylogenetic ANOVA in the package phytools version 0.7-47. For the comparison of substrate types, adjusted p values were calculated using the false-discovery rate method in the p.adjust function ( Figure S6).
In silico analysis of the biosynthetic gene clusters. To identify the responsible biosynthetic gene clusters (BGCs) the Pseudoxylaria and Xylaria genomes were annotated using the fungal version of antiSMASH 6.0.0 [29]. In case of incomplete or fragmented gene cluster hits, sequences were manually reanalyzed using BLAST. Anticipating that in genomes of ascomycetes fungi core genes of a biosynthetic pathway are co-localized within a certain gene region (< 150 kbp), we counted those hits, which showed sequence homologies higher than 50% to other known sequences and co-localized with sequences relating to the same metabolic pathway. Gen models for Xylaria genomes were predicted with Augustus using the created gene model of Pseudoxylaria sp. X802 [89]. BiG-SCAPE 1.0 was used to create the similarity networks [30]. Two cutoff values were used: 0.3 (default) and 0.5. Other parameters used: --include_singletons --mix --mibig. Networks were visualized with Cytoscape (v3.8.1; https://cytoscape.org/) and manually curated. Homology searches were performed with cblaster v1.3.11 [90] and homologous gene clusters were visualized in Geneious Prime (v2021.2.3; Biomatters Ltd.) ( Figure S8).
Growth studies PDA medium. Pseudoxylaria strains were cultivated on PDA plates for a maximum of four weeks at 25°C and sub-cultured by plating myceliumcontaining agar pieces (0.5 × 0.5 cm) onto PDA. For inoculation of Termitomyces strains, vegetative biomass was scraped from agar plates and suspended in Dulbecco's phosphate buffered saline (PBS, 10 mL/ plate).
Wood medium. Wood medium was prepared by soaking and swelling sawdust for 20 min in warm water, which were then autoclaved twice (24 h apart) in 50 mL glass beaker covered with aluminum containing sawdust (25 g).
Wood-rice medium. A 1:1 mixture of boiled rice and swelled sawdust were kept moist, filled into 100 mL glass beakers (50 mL each), sealed with aluminum foil, and sterilized twice 24 h apart (121°C, 20 min). For Termitomyces cultures, medium was inoculated with a mycelial suspension (1.5 mL). For Pseudoxylaria cultures, agar pieces containing vegetative mycelium were placed on the wood-rice surface. Cultures incubated for up to four weeks at rt and monitored daily. Sterile wood-rice medium was kept as control.
Fungus comb medium. Fungus comb material (5.0 g per beaker including sterile wet tissue paper) was autoclaved twice, 24 h apart. Beakers were individually inoculated with Termitomyces sp. T153 and T112, as well as Pseudoxylaria strain X802 and X170LB. Sterile fungus-comb medium was kept as control.
Fungus-fungus co-culture. Fungal cultures (Pseudoxylaria strains OD126, X802, X187, Mn132, X3-2, X170LB, Termitomyces sp.) were cultivated on standard PDA plates (92 mm × 16 mm) for two weeks at 25°C and used for co-culture set-ups. Method A: Termitomyces was inoculated as a mycelium suspension on a PDA plate and cultivated for eight days; then Pseudoxylaria was inoculated next to (or on top) of the Termitomyces culture. Method B: Agar plugs covered with Pseudoxylaria mycelium were added directly onto agar plates freshly inoculated with Termitomyces. Method C: Termitomyces was pre-grown on wood-rice medium (17 days) and then inoculated next to agar plugs of a Pseudoxylaria culture. All cocultures were incubated at rt for up to 4 weeks and monitored on a daily basis.
Metabolic δ 13 C/ 12 C isotope fractionation analysis Cultivation Method A. Fungal strains (Termitomyces sp. T112, Pseudoxylaria sp. X170LB) were cultured as single strains and fungal co-cultures on both, regular and 13 C glucose-enriched PDA medium. The δ 13 C of glucose in the enriched medium was artificially adjusted to +40‰.
Cultivation Method B. Pseudoxylaria sp. X170LB was cultivated on an agaragar matrix containing natural or isotope-enriched lyophilized Termitomyces sp. T112 biomass (~70 mg biomass, 2% agar-agar, 10 mL), which was obtained after cultivation on either regular or 13 C-enriched PDA (14 d, 25°C). For inoculation, a single agar plug (0.3 cm × 0.3 cm) of Pseudoxylaria sp. X170LB was inoculated on top of the biomass containing plate, and incubated for 21 days at rt. Inoculated plates containing only 2% agar-agar (20 mL) served as control. Samples were derived by homogenization starting with six biological replicates (N = 6). Two plates each were then combined during sample collection resulting in triplicates for extraction and analysis (N = 3).
Sample analysis. Pseudoxylaria sp. X170LB mycelium was carefully separated from the cultivation matrix (either agar surface or from the top of the Termitomyces mycelium) and transferred into sterile, cauterized (500°C, 5 h) glass vials. Samples were dried by lyophilization and weighed. Termitomyces sp. T112 mycelium for comparative controls was collected in a similar manner. Dry fungal biomass was homogenized in a mortar and transferred into glass centrifuge tubes. The resulting powder was subsequently mixed with 20 mL MeOH/H 2 O/DCM (25%/25%/50%), sonicated (30°C, 10 min) and centrifuged (5 min, 2000 rpm). The unpolar, lipid containing lower layer was removed with a glass pipette and collected in a separate glass tube. The extraction was repeated twice via addition of 5 mL fresh DCM to the remaining polar MeOH fraction, and combined extracts were dried over Na 2 SO 4 . The final sugar containing MeOH fraction was collected separately. Samples were concentrated to 0.5 mL in a N 2 stream and in vacuo (30-40°C) and stored at −20°C.
Solid phase extraction and analysis. The unpolar lipid fraction was separated by silica column chromatography (SiO 2 ). Neutral lipids were eluted with two column volumes DCM, followed by glycolipids with two column volumes acetone and phospholipids with four column volumes MeOH. Only the phospholipid containing fraction was further analyzed in this study. Samples (0.01 mg-2.45 mg) were transferred to tin capsules, dried (40°C, 1 h) and submitted for elemental analysis and δ 13 C/δ 12 C measurements. Elemental analysis and carbon isotope ratio analysis was performed using an elemental analysis isotope ratio mass spectrometer (EA-IRMS) fitted with an elemental analysator A (NA 1110, CE Instruments, Mailand) coupled with a ConFlo III and a Delta XL-IRMS (Thermo-Finnigan, Bremen). Acetylanilid (Ali-j3) and caffeine (caf-j3) were used as analytical standards (Table S15).

Metabolomic analysis
Cultivation and analysis. Fungal isolates were cultivated on standard PDA plates (92 mm × 16 mm) for two weeks at 25°C. Mycelium covered agar was cut into small pieces (0.5 cm × 0.5 cm) and extracted with EtOAc overnight (50 mL/plate). Solvent was removed to dryness under reduced pressure. Raw extracts were dissolved in MeOH (50 µg/mL) and submitted for ESI(+)-HRMS/MS analysis on a Dionex Ultimate3000 system (Thermo Scientific) combined with a Q-Exactive Plus mass spectrometer (Thermo Scientific) and an electrospray ion (ESI) source. Metabolite separation was carried out by reverse phase liquid chromatography at 40°C using a Luna Omega C18 column (100 × 2.1 mm, particle size 1.6 μm, 100 A, Phenomenex) preceded by a SecurityGuard ULTRA guard cartridge (2 × 2.1 mm, Phenomenex). Mobile phases were acidified with 0.1% formic acid and consisted of H 2 O (A) and acetonitrile (B). In total, 5 μl of each sample were injected and metabolite separation was achieved using the following gradient: 0-0.8 min, 5% B; 10 min, 97% B; 10-12 min, 97% B; 13 min, 5% B; 13-15 min, 5% B at a constant flow rate of 0.3 mL/min. Metabolites were detected in positive (centroid) ionization mode within a range of m/z 180-1800 with a resolving power of 70,000 at m/z 200. MS 2 measurements were performed by combination of data-dependent MS 2 analysis and Top10 experiments.
Molecular network by GNPS. MS/MS-data was filtered by removing all MS/ MS peaks within +/− 17 Da of the precursor m/z. MS/MS spectra were window filtered by choosing only the top 6 peaks in the +/− 50 Da window throughout the spectrum. The data was then clustered with MS-Cluster with a parent mass tolerance of 0.02 Da and a MS/MS fragment ion tolerance of 0.1 Da to create consensus spectra. Further, consensus spectra that contained less than 2 spectra were discarded. A network was then created where edges were filtered to have a cosine score above 0.9 and more than 6 matched peaks. Further edges between two nodes were kept in the network if and only if each of the nodes appeared in each other's respective top 10 most similar nodes. The spectra in the network were then searched against GNPS' spectral libraries. All matches kept between network spectra and library spectra were required to have a score above 0.7 and at least 6 matched peaks.
Purification and structure elucidation of xylasporin G and I. Raw extracts (54 mg) obtained from EtOAc extraction of Pseudoxylaria sp. X187 (20 plates PDA, static cultivation, 14 d, room temperature) were purified by semi-preparative HPLC on a Phenomenex Luna C18 100 Å LC Column 250 × 10 mm (particle size 5 μm, pore diameter 100 A) using a gradient of (A) acetonitrile and (B) H 2 O + 0.1% formic acid. Xylasporin G (3.23 mg) was obtained as a yellow solid, and xylasporin I (1.75 mg) was obtained as a pale-yellow solid, which decomposed within hours when exposed to air and traces of acid. NMR measurements were performed on a Bruker AVANCE III 600 MHz spectrometer, equipped with a Bruker Cryoplatform. The chemical shifts are reported in parts per million (ppm) relative to the solvent residual peak deuterated solvent.

Bioactivity studies
Testing of extracts obtained from Pseudoxylaria on fungal growth of Termitomyces. Pseudoxylaria strains (X187, Mn132, X802, OD126, X3.2, X170LB) were cultivated on PDA plates at 25°C for up to 14 days. Two mycelium covered agar plates were cut into pieces and extracted with EtOAc or MeOH (100 mL per plate), respectively. Concentrated EtOAc extracts were re-dissolved in MeOH (5 mg/mL) and used in disc-diffusion tests (30 µL = 150 µg extract, Whatman antibiotic assay discs, 6 mm (GE Healthcare)) against Termitomyces sp. T153. Methanolic extracts were concentrated under reduced pressure and crude as well as solid-phase extracts eluted from SPE column using MeOH/H 2 O mixtures.
Standard disc-diffusion antimicrobial activity assays. Culture extracts obtained from agar plate cultivations and commercial cytochalasin D and B (Sigma Aldrich) were tested in disc-diffusion tests. Culture extracts (1 mg/mL) in MeOH were tested for their ability to inhibit growth of the following indicator strains: Bacillus subtilis 6633; Staphylococcus aureus SG511, Escherichia coli SG458, Pseudomonas aeruginosa K799/61, Mycobacterium vaccae 10670, Sporobolomyces salmonicolor 549, Candida albicans C.A., Penicillium notatum JP36. Antimicrobial activity was determined by measuring the inhibition zone in mm (Table S17 and S18). Cytochalasins (75 µg cytochalasin in MeOH) and culture extracts (1 mg/mL in MeOH) were also tested for their ability to inhibit growth of Termitomyces for up to 3 weeks at 25°C. Antimicrobial activity was determined by measuring the inhibition zone in mm (Table S17 and S18).
Broth-dilution assay of Pseudoxylaria sp. extracts against Saccharomyces cerevisiae. Yeast strain BY4741 (uracil deficient) was grown in SD medium (yeast synthetic defined medium) amended with uracil (20 mg/L). Assay was performed in 96 well plates with n = 6. First, 4 µL extract (MeOH, 5 mg/ mL) were added to each well and dried, then 200 µL fresh yeast culture (OD = 0.02) were added per well (inoculated at 2%) and optical density measured every hour for 24 h at 30°C under shaking.
Organisms and artificial diet. Larvae of S. littoralis Boisduval (Egyptian cotton leafworm) were hatched and reared on a modified artificial bean diet [45]. Larvae were kept at 19°C and under a light cycle of 12:12 h. Larvae in the 3rd larval stage (L3) were used for all experiments.
Antifeedant insect performance assays. In all feeding assays, freshly molted L3 larvae (5-10 mg) were placed into polystyrene cups (Solo 3.5 oz) containing an agar piece from a 2-week-old fungus culture on PDA, punched out with a sterile 5 mL pipette tip (~1 cm 2 ), and allowed to feed ad libitum. Larvae were assigned randomly to a feeding treatment and during the experiment, kept under controlled conditions and fresh diet was provided daily. All experiments were performed with 25 replicates per treatment, a duration of 10 days, and larval weights and survival rates were recorded every day. The relative growth rate (RGR) was calculated for each caterpillar that survived until the end of the experiment [45]. Feeding experiments were repeated independently two times for each fungus.
Treatments. The insects were fed with PDA (A), Termitomyces sp. T153 (B) or Pseudoxylaria sp. X802 (D) growing on PDA and PDA after removing the fungal mycelium of the latter fungi, respectively (C, E). The RGR was calculated for each caterpillar that survived until the end of the experiment and is expressed in mg/(mg * days) [45]. Listed are the average RGR values ± standard errors. Calculations were not performed for those treatments where no insects were left in the end of the experiment (N/A). Statistical significances were determined using ANOVA, followed by a suitable posthoc test depending on the pre-requisites of the data set. Two additional replicates each were performed for Termitomyces sp. T153 and Pseudoxylaria sp. X802 (see Table S19), where insects were fed with PDA (A), fungus growing on PDA (B) or PDA after removing the fungal mycelium (C) in a separate experimental setup for each fungus.
Statistical analysis. Statistical analysis was performed using SigmaPlot 12.0 (Systat Software, Inc., San Jose, USA). Data was checked for statistical pre-requisites such as homogeneity of variances and normality. Depending on the data type and its match of pre-requisites a respective statistical test was chosen. The chosen statistical test and its results are mentioned in the subtitles of each table or figure.

DATA AVAILABILITY
Supporting Information of this article is free of charge and contains list of accession numbers of sequences used for analysis, phylogenetic trees, cultivation studies including co-cultivation, analyses of genomic and metabolomic data, NMR and MSdata of isolated metabolites and data of insect feeding studies including statistical analyses.