Unraveling Culturable Microfungal Communities Associated with Colonies of the Fungus-Farming Ant Mycetomoellerius urichii (Forel, 1893) Authors: Mateus Oliveira da Cruz, Quimi Vidaurre Montoya, Raquel Lima de Sousa, Gabriel Giorgio Pressuto Pennachioni, Andre Rodrigues Published: 2026 ABSTRACT AbstractMicrofungal communities inhabit the fungus gardens of fungus-growing ants (Formicidae: Myrmicinae: Attini: Attina, the “attines”) and may play cryptic yet important ecological roles within this symbiosis. While the diversity and composition of these microorganisms are relatively well characterized in leaf-cutting ant colonies, they remain poorly understood in non-leaf-cutting attine species, including Mycetomoellerius urichii. To address this gap, we investigated the microfungal communities in colonies of M. urichii using culture-dependent methods. Based on analyses of four independent molecular loci, we identified 94 microfungal species, with Trichoderma spirale, Syncephalastrum sp., and Cladosporium sp. as the most abundant taxa. Several of the microfungi found in this study have also been reported from leaf-cutting ant colonies. The microfungal communities were dominated by fungi exhibiting a multitrophic lifestyle (pathotroph–saprotroph–symbiotroph). Community composition showed considerable variation among colonies, with no consistent species co-occurrence patterns detected. Together, these findings provide the first community-level characterization of culturable microfungi inhabiting the fungus gardens of M. urichii and offer new insights into the microbial communities associated with the fungus-farming ant symbiosis.Supplementary InformationThe online version contains supplementary material available at 10.1007/s00248-026-02776-5.Keywords: Insect-fungi interaction, Hypocreales, Pleosporales, Trophic fungal types, Fungal lifestyles Introduction Fungal communities are integral components of symbiotic systems, even when fungi are not the primary symbionts [1–3]. This is the case for microfungi associated with the symbiosis of fungus-growing ants (Formicidae: Myrmicinae: Attini; the “attines”) [4–7]. In this association, the ants provide substrate for the cultivation of mutualistic basidiomycete fungi, which is dispersed by ant over generations [4]. In turn, the fungus is the sole food source for the colony, characterizing this association as an obligatory mutualism that originated 66 million years ago [8]. Fungus cultivation is carried out on a structure known as the “fungus garden”, usually composed of plant substrate and fungal cultivar mycelium [4, 9]. In addition to the ant cultivar, a wide diversity of microfungi inhabit fungus gardens (e.g. Aspergillus, Cladosporium, Escovopsis, Trichoderma, and Syncephalastrum), particularly on the extensively studied leaf-cutting ants [10–12]. Although microfungi are intrinsic components of this symbiosis, our understanding of the microfungal communities remains limited, especially for non-leaf-cutting attines [13]. Microfungi usually access the fungus garden through ant foraging, along with the substrate gathered by ants to nourish the fungus cultivar [4]. In contrast to leaf-cutting ants (e.g. Atta and Acromyrmex), non-leaf-cutting species primarily collect dry plant material and other organic substrates [9, 14]. Although essential for maintaining the cultivar, input of foraged material may also introduce potentially harmful microfungi in fungus gardens with different lifestyle, including antagonists, mycoparasites, and entomopathogens [6, 14, 15]. Thus, the type of substrate foraged by ants can influence the composition of their associated microfungal communities [5, 16, 17]. In addition, microfungi can also gain access to the gardens by being carried on the workers’ bodies, likely soil-borne fungi [18–20]. Although several ecological factors, including colony location and sampling periods, have been shown to influence fungal community composition [6, 12, 13, 21, 22], microfungal communities associated with many attine ants remain poorly characterized. This gap is even greater for non-leaf-cutting species such as Mycetomoellerius urichii (Forel, 1893). This ant species maintains a mutualistic symbiosis with Leucocoprinus sp. and is distributed across the Neotropics [8, 23, 24]. Despite its broad geographic range, species of Mycetomoellerius are poorly studied regarding its biology and associations with other microbes [25]. The structure and ecological profiles of microfungal communities inhabiting the fungus gardens cultivated by M. urichii, as well as those associated with non–leaf-cutting ants in general, remain underexplored. Here, we used a culture-dependent approach to analyze the structure (diversity and trophic models) of culturable microfungal communities associated with colonies of M. urichii. We then interpreted these patterns in the context of microfungal communities previously described for leaf-cutting ants. Culturable microfungal communities associated with M. urichii were dominated by species of Cladosporium, Syncephalastrum, and Trichoderma, as previously reported for leaf-cutting ants. The communities were primarily composed of fungi exhibiting a combined trophic mode (pathotrophic–saprotrophic–symbiotrophic). Nevertheless, we observed considerable variation in community composition among ant colonies. Materials and Methods Sampling A total of 20 colonies of M. urichii were collected at São Paulo State University (UNESP), Campus Rio Claro, São Paulo, Brazil (coordinates: 22°23’46.91"S 47°32’30.79"W) located in the Cerrado biome (Fig. 1). As all colonies were found within a short distance (less than 500 square meters), thus they were co-existing in the same habitat. This spatial scale was selected based on field observations indicating homogeneous vegetation and soil cover, as well as the absence of physical and biological barriers among colonies, such that all colonies occupied the same local habitat. Colonies were sampled over a two-month (Dec. 2022 to Jan. 2023) period during the summer. Collections were avoided in the days immediately following rainfall events, reducing short-term environmental variation. Colonies were excavated following the methods described by Rodrigues et al. [26]. In some cases, it was necessary to track foraging worker ants to locate the nests as described in Sosa-Calvo et al. [27]. Fungus gardens were collected using flamed forceps and spoons, placed in sterilized plastic containers with moistened plaster at the bottom, and transported to the laboratory. Environmental characteristics of the collection area. (A) Geographic location of the study area. (B) Spatial arrangement of the sampled colonies (each circle represents a colony). (C) Vegetation features of the collection area. (D) Fungus garden of Mycetomoellerius urichii Fungal Isolation Five fragments (approx. 3 mm3 each) of the fungus gardens were placed, equidistant from each other, on a Petri dish containing Potato Dextrose Agar (PDA; Neogen® Culture Media, Lansing, USA) supplemented with 150 µg mL⁻¹ of chloramphenicol (Sigma-Aldrich, St. Louis, MO, USA). Three plates were used for isolation totalling 15 fungus garden fragments per ant colony. Plates were kept in the dark at 25 °C and monitored daily for up to 15 days. When mycelial growth of microfungi was observed, which generally happened within two days after plating, the fungi were transferred to new Petri dishes containing PDA [5, 13, 21]. For each ant colony and for each plate, all emerging fungal colonies (isolates) were recorded, and the number of isolates per morphotype was quantified. Each isolate corresponded to a distinct fungal colony growing from a fragment. As this study focused on the community of microfungi inhabiting the fungus garden, which is largely composed of the mycelium of the mutualistic fungus, the ant fungal cultivar was not isolated and not included in subsequent analyses. Identification and Phylogenetic Analysis of Fungi Initially, isolates were grouped into morphotypes based on macroscopic and microscopic characteristics, including colony color, diameter, sporulation patterns, pigmentation on PDA, conidiogenesis, and general conidiophore morphology. Morphotyping was used to assign isolates to putative genera and to guide the selection of the most appropriate molecular markers for taxonomic identification, following established literature. One representative of each morphotype was selected for sequencing and the number of isolates per morphotype is then used as a proxy for species abundance in our analysis. Accordingly, morphotypes preliminarily assigned to Aspergillus, Penicillium, and Talaromyces were barcoded using the β-tubulin (benA) gene. Morphotypes assigned to Trichoderma and Escovopsis were barcoded using the translation elongation factor 1-alpha (tef1). Morphotypes preliminarily assigned to Mucoromycota were barcoded using the LSU rDNA region, whereas all remaining morphotypes were barcoded using the ITS rDNA region. For total DNA extraction, fungi isolates were grown for 4–6 days on PDA. This time frame varied depending on fungal growth rate, which differed between morphotypes. DNA was extracted following a modified CTAB protocol [28]. Briefly, fungal mycelium was mechanically disrupted using 500–600 μm diameter glass beads (Sigma-Aldrich, St. Louis, MO, USA) in 600 µL of lysis buffer. Subsequently, 10 µL of 20 mg mL⁻¹ proteinase K was added, then incubated at 65 °C for 30 min. to ensure protein digestion. The organic phase separation was achieved by adding 600 µL of a chloroform: isoamyl alcohol mixture (24:1), followed by centrifugation at 12,000 RPM for 10 min. The aqueous phase containing DNA was carefully recovered and subjected to 70% ethanol precipitation for purification. Finally, the DNA pellet was resuspended in 30 µL of Tris-Edta buffer (10 mM Tris and 1 mM EDTA) and stored at − 20 °C. The gene/genic region was amplified in 25 µL reactions containing: 5 µL of 5× buffer, 2 µL of 25 mM MgCl2, 4 µL of 1.25mM dNTPs, 1 µL of 10 µM forward primer, 1 µL of 10 µM reverse primer, 1 µL of 10 mg mL− 1 Bovine Serum Albumin, 0.2 µl of 5 U mL− 1 Taq polymerase, 2 µL DNA (diluted 1:10), and 8.8 µL of ultrapure water. The primers and conditions used for each marker were according to Nasri et al. [29] for benA; Taerum et al. [30] for tef1; Nguyen et al. [31] for LSU; White et al. [32] and Schoch et al. [33] for ITS. Amplicons were cleaned up with the Wizard SV Gel and PCR Clean up System (Promega, Madison, WI, USA) following the manufacturer’s protocol for subsequent sequencing. For DNA sequencing, the purified amplicons were quantified using NanoDrop® (Thermo Scientific, Waltham, MA). Then, samples were subjected to cycle sequencing reactions with BigDye Terminator® v.3.1 (Thermo Scientific, Waltham, MA), following the manufacturer’s instructions. The sequencing reaction was performed in 10 µL volume containing: 0.5 µL of BigDye, 2 µL of 5× Buffer, 0.32 µL of 10 µM primer, 1 µL of 10–20 ng µL− 1 DNA template, and 6.18 µL of ultrapure water. The sequencing conditions for all molecular markers were: 95 °C for 1 min followed by 28 cycles at 95 °C for 15 s, 50 °C for 45 s, and 60 °C for 4 min. Forward and reverse sequences were obtained using an ABI 3500 Genetic Analyzer (Life Technologies). Sequences were assembled in contigs in Geneious v.6.0 [34] and deposited in GenBank (www.ncbi.nlm.nih.gov) under accessions (ITS: PX762964–PX763056; LSU: PX763694–PX763702; tef1: PZ051250–PZ051264; benA: PZ051261– PZ051264). These sequences were compared with sequences in GenBank using Nucleotide Blast (https://blast.ncbi.nlm.nih.gov/Blast.cgi). Phylogenetic analyses were conducted separately for each marker and included only sequences generated in this study. Single-marker phylogenetic trees were used to assess whether sequences corresponding to distinct morphotypes clustered into the same or well-supported clades, thereby supporting morphotype-based species delimitation. The sequences were aligned in MAFFT v. 7 [35]. The nucleotide substitution model for alignment was determined using jModelTest v. 2 [36], based on the Akaike Information Criterion (AIC) with 95% confidence intervals. Based on the preliminary identification of fungi through sequence comparison using Nucleotide Blast, one representative sequence per species was selected to compose the dataset for each of the four markers: benA (10 sequences), tef1 (11 sequences), LSU (10 sequences), and ITS (62 sequences). Phylogenetic tree reconstruction was carried out using Bayesian Inference (BI) in MrBayes v. 3.2.2 [37], with two independent runs (each consisting of three heated chains and one cold chain) under the GTR model. A total of two million generations of the Markov Chain Monte Carlo (MCMC) were sufficient to achieve convergence (standard deviation of split frequencies < 0.01), and the first 25% of the trees were discarded as burn-in to infer the consensus tree. The resulting trees were visualized in FigTree v. 1.4.4 [38] and edited in Corel Draw v. 24.5. Analysis of Fungal Community All analyses were conducted using R version 4.3 [39]. First, we inferred a rarefaction curve to evaluate our sampling effort by interpolating and extrapolating species richness found in the colonies of M. urichii. For this, we used the iNEXT package [40] and tidyverse [41], considering the number of species obtained from the 20 ant colonies. For alpha diversity analyses we performed plate-level analyses. Three isolation plates per colony were treated as technical replicates. For each plate, fungal abundance was quantified as the number of isolates assigned to each species. The relative abundance of each genus per ant colony (also considering the three isolation plates) was calculated and plotted. To visualize fungal abundance within colonies, species richness, Shannon diversity, and evenness were calculated separately for each plate. Mean values were then obtained for each ant colony and used for statistical comparisons, considering the colony as the unit of analysis (Supplementary material, Table S1). First, we tested for data normality using the Shapiro–Wilk test. After confirming the non-normal distribution of the data, we used Kruskal–Wallis test to assess differences. These analyses were conducted using the tidyr [42], dplyr [43], and vegan [44] packages. For descriptive analyses, the relative frequency of each species across colonies was calculated as the proportion of colonies in which a given species was recorded (i.e., number of colonies with at least one isolate of the species divided by the total number of colonies). To determine whether species composition differed among colonies (beta diversity), we applied permutational multivariate analysis of variance (PERMANOVA) [45] using the vegan [44] and dplyr [43] packages, using abundance data. For this analysis, isolates obtained from the three plates of each colony were pooled, and colony-level community matrices were used as independent units. Dissimilarity in fungal community composition across colonies was graphically represented through Non-metric Multidimensional Scaling (NMDS) based on Bray–Curtis dissimilarity (permutations = 999), implemented with the vegan [44] and tidyverse [41] packages. Bray–Curtis dissimilarity was selected because community data were based on isolate abundance per colony, allowing comparison of relative differences in community composition among colonies. All community composition plots were generated using the ggplot2 package [46] in R. To visualize possible occurrences of exclusive species in the colonies and shared occurrence combinations among the colonies, we selected species with an abundance greater than two isolates and represented their presence and absence in a heatmap using the tidyr [42] and dplyr [43] packages. We also generated an UpSet plot using the package ComplexUpset [47]. Presence indicates that at least one isolate of a given species was detected in a colony, whereas absence indicates that no isolates of that species were recorded in that colony. The trophic guild of the fungal community associated with colonies of M. urichii was evaluated using FUNGuild [48], a tool designed to assign ecological guilds based on taxonomic data. For this, we used the list of taxa identified through molecular analysis and matched them using the script available in the FUNGuild database hosted on GitHub (https://github.com/UMNFuN/FUNGuild), running in Python v. 3.13 [49]. Guilds were not assigned to taxa identified only at taxonomic levels above genus. The proportion of each trophic mode was calculated and plotted. Results Fungi Found in Colonies of Mycetomoellerius urichii Fungi were successfully isolated from all the 20 colonies analyzed using PDA, yielding a total of 274 isolates (Table S1). These isolates were assigned to 164 morphotypes, each represented by one isolate selected for barcode sequencing. Although all morphotypes were sequenced, molecular identification revealed that multiple morphotypes corresponded to the same species, resulting in a total of 94 identified species. The 94 species comprise 59 genera and 17 orders distributed in the phyla Ascomycota (n = 13 orders), Basidiomycota (n = 3 orders), and Mucoromycota (n = 1 order, Fig. 2). In addition to the orders, two taxa, Scolecobasidium and Toxicocladosporium sp., was classified as incertae sedis. The most speciose orders were Pleosporales (13 genera, 17 species) and Hypocreales (11 genera, 24 species, Fig. 2). The most species-rich genera were Trichoderma (T. atroviride, T. harzianum, T. koningiopsis, T. lentiforme, T. paratroviride, Trichoderma sp., T. spirale, and T. virens), Penicillium (P. ortum, P. pulvillorum, P. tanzanicum, and P. wotroi), and Mucor (M. janssenii, M. lusitanicus, M. racemosus, and Mucor sp. One species of Cunninghamella one of Diaporthe were identified just using morphology due to difficulties in obtaining high-quality sequences (Fig. 2; Table S2 and Figs. S1–S4). Phylogeny revealing the relationships among isolates of different fungal orders obtained from colonies of Mycetomoellerius urichii. The tree was inferred from 62 ITS sequences and 1.584 sites. Numbers on branches indicate ML bootstrap support values > 75%. Different colors represent different fungal orders. The 62 ITS sequences were sufficient to represent all fungal orders identified in this study. However, the number of genera and species reported for each order is based on the complete dataset, including taxa identified using additional molecular markers (LSU, tef1, and BenA) Analysis of Fungi Community The rarefaction curve suggests that sampling was insufficient to capture the full fungal diversity in the colonies of M. urichii. Extrapolation indicates that sampling an additional 20 colonies would be required to reach a plateau, suggesting these communities are species rich (Fig. S5). The number of fungal isolates per colony ranged from 3 to 28 (mean ± SD = 13.9 ± 6.3). For alpha diversity analyses, diversity indices were calculated for each of the three technical replicates (plates) per colony and averaged per colony prior to statistical testing. Species richness per colony ranged from 3 to 17, Shannon diversity index from 1.0 to 2.4, and evenness from 0.7 to 0.9. Despite these ranges, there were no significant differences in richness (Kruskal-Wallis test, p = 0.1), Shannon diversity (p = 0.1), or evenness (p = 0.3), revealing no detectable variation in fungal alpha diversity (Fig. 3). Distribution of fungal alpha diversity across colonies of Mycetomoellerius urichii A total of 32 taxa had an absolute abundance greater than two isolates. The most abundant were T. spirale (19 isolates), Syncephalastrum sp. (17), and Cladosporium sp. (16) (Fig. S6). Among the 32 most abundant taxa, eleven species occurred at least in more than 20% of ant colonies: Cladosporium sp.1, T. spirale (40% each), Curvularia sp.1 (35%), T. virens, Schizophyllum commune, Fusarium equiseti, Epicoccum sorghinum, Diaporthe sp.2 (25% each), Aspergillus udagawae, Cladosporium tenuissimum and Microsphaeropsis sp.1 (20% each) (Fig. S7). Each colony harbour fungi from at least three of the 15 most abundant genera. Marked discrepancies were observed among colonies. The genus Trichoderma reached 50% relative abundance in one colony, but it was absent from five others. Syncephalastrum exhibited a restricted distribution, comprising approximately 50% of the fungal community in only two colonies that, coincidentally lacked Trichoderma. Fusarium accounted for approx. 70% of isolates in one colony but was absent from 14 colonies (Fig. 4). No co-occurrence combination among the 32 most abundant species was observed, as each occurrence combination was exclusive to a single colony. No species was shared among all colonies, and no species was restricted to a single colony (Fig. 5 and Fig. S8). This pattern was corroborated with PERMANOVA that revealed significant differences in species composition among colonies (p = 0.001), and this pattern was visually evident with high compositional dissimilarity among fungal communities (Fig. S9). Relative abundance of fungi genera by colony of Mycetomoellerius urichii Heatmap representing the presence and absence of the 32 species with total abundance greater than two isolates (i.e., species from which more than two isolates were recovered from Mycetomoellerius urichii colonies). Presence (in orange) indicates that at least one isolate of a given species was detected in a colony, whereas absence (in grey) indicates that no isolates of that species were recorded in that colony Functional annotation using FUNGuild identified 12 functional categories spanning symbiotrophic, pathotrophic, and saprotrophic modes. Representatives of all three trophic modes were present in all colonies. Microfungal communities associated with M. urichii were primarily composed of fungi encompassing multiple trophic strategies, with pathotroph–saprotroph–symbiotroph being the most prevalent category. In contrast, saprotroph–symbiotroph and pathotroph–symbiotroph categories consistently represented the smallest proportions across colonies (Fig. 6). Trophic modes of culturable microfungi from colonies of Mycetomoellerius urichii Discussion Here, we characterize the structure of the culturable microfungal community inhabiting colonies of M. urichii. As an inherent limitation of culture-dependent surveys, the total microfungal community is not captured; instead, our results represent only those fungi able to grow under standard cultivation conditions on nutrient-rich medium such as PDA. Nevertheless, the fungi isolated here are metabolically active on culture media, and culture-based surveys provide a valuable framework for subsequent functional, ecological, and taxonomic analyses [50–52]. A total of 94 species, representing 17 fungal orders, were identified using DNA sequence data. Our analyses focused on the microfungal community associated with the fungus garden and therefore did not include the primary fungal mutualist (Leucocoprinus sp.). To our knowledge, this study provides the first community-level characterization of culturable microfungi associated with M. urichii, a non–leaf-cutting fungus-growing ant. To date, research on fungi from non–leaf-cutting attines has been largely restricted to taxonomic studies of Hypocreaceae (e.g., Escovopsis and its sister groups) [53, 54] and one diversity study focused on a lower attine (Cyphomyrmex wheeleri) and a higher attine (Trachymyrmex septentrionalis) [13]. The microfungi isolated from M. urichii colonies encompass diverse taxonomic groups that are also found in a variety of other environments. Members of the orders Pleosporales and Hypocreales, which were the most species-rich in our study, occur in both terrestrial and aquatic environments, functioning as saprobes or as symbionts of plants and other animals [55–57]. The most species-rich genera, Trichoderma, Penicillium, and Mucor, are primarily soil-borne saprobes [58–60]. While foraging is considered the primary source for introducing fungi into the fungus garden, the presence of these genera suggests that contact with soil may represent a potential route of entry, although this pathway was not directly evaluated in the present study. This assumption is plausible for non-leaf-cutting ants, as they frequently collect plant material from the ground [9]. Trichoderma was the most abundant and representative genus in M. urichii colonies, a pattern similar to that observed in leaf-cutting ants of the genera Atta and Acromyrmex [5, 6, 12, 21, 22, 61]. This genus is known to have an antagonistic effect on the mutualistic fungus Leucocoprinus gongylophorus cultivated by leaf-cutting ants [15]. Trichoderma is ecologically versatile, occurring both in soil and as an endophyte in plant tissues [62, 63], which may facilitate its introduction into ant colonies via foraged substrates, representing a plausible but untested source. Future studies could investigate the role of Trichoderma in M. urichii colonies and assess whether its interaction resemble those documented in leaf-cutting ant systems. Among the other abundant fungi in M. urichii colonies, Syncephalastrum has previously been reported as one of the most prevalent genera in colonies of Atta sexdens [22]. Evidence suggests that it negatively affect leaf-cutting ant colonies [64, 65]. However, its abundance in non-leaf-cutting ant colonies raises questions about the pathogenic potential of Syncephalastrum in these systems. Here, Trichoderma and Syncephalastrum displayed contrasting abundance patterns across colonies, with each genus being low in colonies where the other was abundant. Although our study was not designed to formally test fungal–fungal interactions, this contrasting pattern suggests potential ecological interactions that merit targeted experimental investigation. Cladosporium was among the most abundant fungi in M. urichii colonies , as well as in colonies of C. wheeleri, a genus of non-leaf-cutting ants [13]. In contrast, this genus is not typically dominant in colonies of Atta and Acromyrmex [5, 6, 21], a difference that may be partly explained by variation in foraging behaviour among attine ants. Interestingly, Cladosporium is frequently found on the integument of leaf-cutting ants, suggesting a preference for this microhabitat rather than the fungal garden [20]. Future studies could explore whether the integument of non-leaf-cutting ants provides a favourable habitat for Cladosporium, as observed in the fungal gardens of M. urichii. Colonies of M. urichii exhibit similar fungal alpha diversity and richness. Curiously, it’s a pattern also observed in other attine ants such as Atta and Acromyrmex [6, 21]. However, studies investigating the underlying causes of this pattern in the attine symbiosis remain limited. Although we identified a total of 94 species, Shannon diversity values are relatively low [66, 67] with most values falling below two. This value range is intermediate between those reported for Atta cephalotes (1.7–1.8) [21] and Acromyrmex crassispinus colonies (2.7–2.8) [6], for example. Such variation may be attributed to biotic factors, particularly ant species identity, which plays a significant role in shaping fungal diversity within attine colonies [12]. On the other hand, fungal richness varies in colonies of Atta sexdens depending on the sampling period [5]. Since M. urichii colonies were collected during the same period, we cannot infer whether this abiotic factor influences fungal richness in our study. Although our analyses do not identify the drivers of community differentiation, the observed compositional differences among co-existing colonies indicate substantial colony-level heterogeneity in microfungal communities. However, this pattern should be interpreted cautiously, since the PERMANOVA analysis was performed at the colony level without replication. Although sampling was conducted over a relatively narrow temporal window, short-term temporal variation over the two-month sampling period cannot be entirely excluded as a potential source of among-colony variation. In this context, comparisons with allopatric colonies from different geographical areas could provide additional insight into the generality of the patterns observed here. Workers of M. urichii transport fungi with diverse trophic profiles into their colonies. This may pose potential risks, as they can carry animal-pathogenic fungi (Cladosporium, Fusarium, and Trichoderma) and mycoparasites (Escovopsis and Trichoderma). To our knowledge, only Camargo et al. [6] discriminated the trophic profiles of fungi associated with attine ants. Their results showed that colonies of A. crassispinus are primarily composed of antagonistic and antagonistic-entomopathogenic fungi, providing insights for potential biological control strategies. Perhaps, the diversity of fungal guilds found in M. urichii colonies and prevalence of multitrophic lifestyle pathotroph–saprotroph–symbiotroph, is likely related to the variety of substrates provided for fungus cultivation. Interestingly, M. urichii maintains associations with patotrophic fungi and future studies could further elucidate this symbiosis. The fungal garden cultivated by M. urichii hosts a fungal community with a multitrophic lifestyle (pathotroph–saprotroph–symbiotroph), and some of these taxa have also been reported from leaf-cutting ant colonies. While our study does not include a direct, method-matched comparison with leaf-cutting ants, this overlap with previous studies indicates that these taxa are commonly associated with fungus-growing ant colonies across different attine (non-leaf-cutting and leaf-cutting ants) fungiculture systems. On the other hand, the pronounced variation observed among co-occurring colonies underscores colony-level heterogeneity in culturable microfungal community composition. Together, these findings provide a baseline for understanding microfungal community in non–leaf-cutting attine ants. Supplementary Information Below is the link to the electronic supplementary material. Supplementary File 1 (XLSX 28.9 KB) Supplementary File 2 (DOCX 2.93 MB) Acknowledgements We would like to thank Conselho de Gestão do Patrimônio Genético (CGEn) for providing permits for access to genetic heritage # AA39A6D issued to AR. The authors are grateful to Instituto Chico Mendes para Conservação da Biodiversidade (ICMBio) for the collecting permit issued to AR (permit # 74585). We also thank Raphael Pereira da Silva for his assistance with LSU sequencing for some isolates, and members of the Laboratory of Fungal Ecology and Systematics (LESF), São Paulo State University (UNESP), Rio Claro, Brazil, who helped in colony collection. We would like to thank Dr. Rodrigo Machado Feitosa for ant identification. Author contributions MOC: conceptualization, investigation, data analysis, and writing. QVM: conceptualization, investigation. RLS: conceptualization, investigation. GGPP: investigation. AR: project administration, supervision, funding acquisition, and writing. All authors revised and approved the final manuscript. Funding This study was financed in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior – Brasil (CAPES) – Finance Code 001. This study was supported by the Thematic Project “Collaborative Research: Dimensions US-São Paulo: Integrating phylogeny, genetics, and chemical ecology to unravel the tangled bank of the multipartite fungus-farming ant symbiosis” (FAPESP grant # 2019/03746-0) of AR, for scholarships to MOC (FAPESP #2022/16087-7, CAPES #88887.706554/2022-00), QVM (FAPESP #21/04706-1), and GGPP (Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq # 126804/2024-9). Data Availability All the data that support the findings of this study are available on GenBank or electronic supplementary material indicated in the main text. Declarations Competing interests The authors declare no competing interests. No ethical approval was required for this paper. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary File 1 (XLSX 28.9 KB) Supplementary File 2 (DOCX 2.93 MB) Data Availability Statement All the data that support the findings of this study are available on GenBank or electronic supplementary material indicated in the main text.