Ants/Nest-Associated Fungi and Their Specialized Metabolites: Taxonomy, Chemistry, and Bioactivity

Microscopic fungi occupy a vast number of habitats, are taxonomically diverse, degrade complex substrates, and have stood out for their capacity to biosynthesize a plethora of specialized metabolites. Such molecules are structurally diverse, and many have applications in fundamental and applied sciences, for example, in medicine, material sciences, food chemistry, textile and pharmaceutical industries, and agronomy, among other fields. However, despite the tremendous biotechnological value of fungi, these organisms are understudied, limiting the knowledge to their taxonomy, chemistry, and some putative applications. Notably, some specific habitats remain unexplored in terms of their mycobiota. Based on these considerations, this review describes the known fungal diversity associated with ants/nests, their metabolic potential, and the possible applications of their specialized metabolites in drug discovery programs focused on developing treatments for human diseases. According to this revision, fungal diversity has been studied by applying conventional methodologies such as isolation and morphological identification of soil fungi from mounds and nest chambers, and indirectly from ants’ cuticles and glands. The subfamilies, genera, taxonomical information, and geographical origin of ants from which filamentous fungi and yeast are commonly isolated are also described. Furthermore, some important information is presented concerning the difference between the ant-associated mycobiota and that in the surroundings, discarding the extrapolation of the chemical and biological information known for soil fungi. Altogether, this review evidenced the lack of information regarding the chemical composition of ant-associated fungi, encouraging research focused on exploring the chemistry biosynthesized from ants’ mycobiota, as well as the elucidation of their allelopathic potential inside the nests.


Introduction
Since ancient times, fungi have been used by humankind as food and through bioprocesses to produce beer and wine (Ghorai et al. 2009). Currently, technological advances have allowed the development of several biotechnological areas, for example, biocontrol, bioremediation, biomass degradation, biofuel, organic acids, enzymes, agrochemical, and pharmaceutical production industries, to name but a few (Berasategui et al. 2016;Meena and Siddhardha 2019;Hyde et al. 2019). The fungi occupy a vast number of habitats, are taxonomically diverse, degrade complex substrates, and have stood out for their capacity to biosynthesize a plethora of specialized metabolites. Moreover, such molecules are structurally diverse, and many have applications in fundamental and applied sciences, for example, in medicine, material sciences, food chemistry, textile and pharmaceutical industries, and agronomy, among other fields (Schueffler and Anke 2014;González-Medina et al. 2016Aldholmi et al. 2019). However, despite recent technological advances and the high biotechnological value of fungi, these organisms are understudied, and specific habitats remain unexplored. In this regard, the chemical constituents of a small number of species are known, and consequently, the use of their chemical heritage is limited (Schueffler and Anke 2014;Chassagne et al. 2019;Pham et al. 2019).
Given their biological diversity and cosmopolitan presence, fungi are the second most diverse kingdom on Earth, bringing together among 2.2-3.8 million species. However, so far, just over 200,000 have been described 1 3 (Hawksworth and Lücking 2017). Macroscopic fungi are recognized due to their fruiting bodies; nevertheless, a considerable part of the biodiversity is composed of those with microscopic reproductive structures. Microscopic fungi are further divided into yeasts and filamentous fungi, and multicellular and unicellular eukaryotic organisms, respectively (De-Wei 2016). Consequently, the study through bioprospecting projects of areas with unknown mycobiota can lead to the description of new species and valuable chemical-biological information about the microorganisms and the specialized molecule produced. Some of these underexplored habitats are the ants' nests, whose importance from the chemical-pharmaceutical point of view was evidenced by the study of their bacteriobiota, which have led to the discovery of various antifungals, for example, the cyclic depsipeptide dentigerumycin (1), involved in interspecific interactions among ants of the Attini tribe and the mycobiota of their nests Pathak et al. 2019;Agarwal et al. 2022). The importance of microbiota associated with the nests and materials of other arthropods such as termites (Bignell et al. 1991;Yan et al. 2011;Chen et al. 2022), bees (Poulsen et al. 2011;Madden et al. 2013;Promnuan et al. 2021), and spiders (Iwai et al. 2009) has also been explored.
Based on these considerations, this review describes the known fungal diversity associated with ants of various species widely distributed globally, their nest, and the metabolic potential of isolated taxa, as well as the possible applications of their specialized metabolites. In a complementary way to some previous works, this compilation provides information concerning the variety of ant species from which fungi have been isolated and chemically studied. Additionally, the mycobiota of the nests is compared with that of the surroundings. However, a similar work highlighting the study of yeast diversity in several ants of the Attini tribe was published (Bizarria et al. 2022). Analysis of the retrieved information shows that literature on fungal diversity associated with nest focuses on the mounds and chambers of a few ant species, mainly the so-called leafcutters, generally distributed in South and North America (Fig. 1). Likewise, the most common ants/nest-associated fungal genera are listed, highlighting the underrepresented ones for their bioactivity. Although the chemical composition of ants/nest-associated fungi remains almost unexplored, compounds with antibacterial, antifungal, antidiabetic, and antiviral activity in vitro isolated from fungi have been described. These data will uncover the potential of the chemical and biological space of unexploited fungal-specialized metabolites. Additionally, they will demonstrate further evidence of their ability to synthesize complex structures applicable in various scientific and industrial fields. Finally, how the known interspecific relationships support the search for antibacterial chemicals from the nest mycobiota is also explained.

Search Strategy
This review began with an independent search for pairwise combinations, using "and" or " + " operators, of the English words "fungus," "mycobiota," "filamentous fungi," "yeast," "molecule," "compound," "isolation," and "natural product" with "ant nest," "mound," and "ant." The search engine Google Scholar and the databases PubMed, ScienceDirect, SciFinder, SciELO, Redalyc.org, JSTOR, Web of Science, and EBSCO were used. In the case of SciELO, the same words were searched for in Spanish and Portuguese. The American Chemical Society, Wiley, Springer, SAGE, Thieme, Royal Society of Chemistry, Canadian Science Publishing, and Taylor & Francis publishers were also consulted directly. Also, tracking additional publications in listed references was also useful. In summary, literature prior to December 2022 on fungal diversity associated with ants, their cuticle, and nests, as well as its chemical composition, were considered for discussion. Then, the publications that did not detail the origin of the mycobiota or the method to determine the fungal species, as well as those that did not provide sufficient evidence for the identification or elucidation of compounds, were excluded; the rest is discussed in this review; laboratory colony-based studies were also considered. The analysis of the mycobiota is restricted to that from healthy terrestrial ants and nests, whose identification of ants and fungi is at least at the genus level. The names of ant and fungal species were verified in the Integrated Taxonomic Information System, as well as AntCat, AntWeb databases, or directly in the report of their discovery (ITIS 2022; Bolton 2023; California Academy of Science 2023).

Ants and Their Nests
Ants are a diverse, taxonomically functional group of insects classified in the Formicidae family. These organisms have existed for at least 100 million years and were the first social insects with predatory habits on Earth (Fernández 2003). Ants live in almost all terrestrial environments, underground or in trunks, branches, or subcortical cavities; only Antarctica, Iceland, Greenland, and the central Pacific islands do not have native species (Rojas Fernández 2001). Ants' abundance is difficult to measure, but they are always in the most copious group of the edaphic macrofauna. The number of species and how they are distributed are unknown. However, their abundance in tropical, low-lying subtropical forests, and hot deserts is well established; temperature is the most critical factor limiting its distribution (Rojas Fernández 2001;Fernández 2003). Currently, more than 12,500 species belonging to 21 subfamilies and 290 genera have been described; nevertheless, it is suggested that the number of species could be up to 30,000, with some awaiting discovery. Among the known taxa, Myrmicinae and Formicinae subfamilies are the most studied (Ríos-Casanova 2014; Boudinot 2015; Bolton 2023).
Ants live in hierarchical communities within which it is possible to distinguish a worker and a reproductive caste. The number of individuals can range from dozens to hundreds of thousands, with specimens ranging from less than a millimeter to 1.5 cm in length (Ríos-Casanova 2014). Considering their diet, ants are highly specialized and generalist and could be classified as omnivorous, mycophagous, granivorous, and predatory (Rojas Fernández 2001). A variety of ant species occasionally feed on fungi, but Attini tribe are remarkably interesting as they grow fungal "gardens" within their nests and feed themselves almost exclusively with mycelium. Species of the Atta, Acromyrmex, and Sericomyrmex genus are called "leafcutters" or "foragers" for their skills to collect fresh plant material as a substrate for the growing fungus, while others, mainly of the Cyphomyrmex, Myrmicocrypta, and Apterostigma genera, collect carcasses, feces, and other materials to feed the fungus.
In contrast, Sericomyrmex and Trachymyrmex have transitional habits (Rojas Fernández 2001;Leal et al. 2011;Sosa-Calvo et al. 2018). It is hypothesized that the development of the metapleural gland played a crucial role in colonization by some ant species in soil (rich in microorganisms). This gland secretes hydroxy and phenylacetic acids that differentially inhibit the growth of microorganisms inside the nests; not surprisingly, species adapted to live in trees have lost this gland (Yek et al. 2012;Vieira et al. 2012;Offenberg and Damgaard 2019;Offenberg et al. 2022).
The impact of ants' activity on soil health has been moderately addressed and principally considers the modification of physicochemical properties. For example, some studies describe the impact of soil, vegetation, and soil fauna transportation on the construction of underground nests, as well as the modification of chemical properties due to mineralization, organic matter accumulation, and decomposition, among other factors (Wagner et al. 1997;Kristiansen and Amelung 2001;Rojas Fernández 2001;Dostál et al. 2005;Pinto-Tomás et al. 2009;Fernandez et al. 2014;Wang et al. 2017;Zhang et al. 2018). 1 3

Associated Mycobiota
Based on the available information and the scope of our current research, this review mainly focuses on fungi isolated from terrestrial ants and their potential as a source of functional secondary metabolites; however, the approaches to study arboreal ants and their associated mycobiota are also summarized in the following paragraph. It is worth mentioning that the chemical composition of ants themselves has already been reported (Mans et al. 2016;Seabrooks and Hu 2017).
Ants with ground-level nests, from which fungi have been isolated, are scarce. Taxonomically (Fig. 2), these individuals are mainly grouped into tribe Attini, which is represented by ants of 7 genera: Apterostigma, Mycocepurus, Myrmicocrypta, Atta, Trachymyrmex, Acromyrmex, and Cyphomyrmex. Other tribes under study with one genus each are Solenopsidini, Tetramoriini, Pheidolini, Formicini, Camponotini, and Tapinomini. The morphology and lifestyle of these species considerably differ depending on their tribe, and the size and structure of their nests. Furthermore, they have a different native distribution, and colonize a great variety of ecosystems. The study of ant-associated mycobiota has been carried out by field sampling of soil from nests in nine countries, although the availability of research on filamentous fungi in North and South America stands out (Fig. 3). However, some research on mycobiota is carried out on colonies maintained in the laboratory under controlled environmental and feeding conditions. In the case of Atta sexdens rubropilosa and Atta cephalotes, for example, such studies have led to a better understanding of the fungal diversity inside the nests, chambers, and the ants themselves (Fisher et al. 1996;Carreiro et al. 1997Carreiro et al. , 2004Rodrigues et al. 2005b).
Fungi associated with wild mature nests have been characterized by sampling mound's soil, taken from the surface to a certain depth (3-40 cm)-often discarding detritus (Brill et al. 1996;Duff et al. 2016;Maksimova et al. 2016)-as well as the walls and contents of chambers (waste, garden, and breeding), by means of a reasonable excavation aside the entrances (Fisher et al. 1996;Ba et al. 2000;Rodrigues et al. 2014). Likewise, fungi away from the nest (1-10 m) are also considered in studies for comparative purposes (Arcuri et al. 2014;Rodrigues et al. 2014;Duff et al. 2016;Maksimova et al. 2016). Conversely, nest mycobiota do not always consider a description of fungi on or inside the ants (Arcuri et al. 2014;Maksimova et al. 2016).
Ant-associated fungi from soil of nest and fragments of the fungal garden are commonly isolated by conventional techniques based on sample dilution in sterile water and direct seeding in Petri dishes containing, yeast-malt extract agar (Fisher et al. 1996), potato dextrose agar , glucose-peptone-yeast agar (Maksimova et al. 2016), and medium for aquatic yeasts (Carreiro et al. 1997). Antibiotics such as chloramphenicol (Arcuri et al. 2014;Rodrigues et al. 2014), amoxicillin (Jiménez-Arreola et al. 2020), oxytetracycline (Fisher et al. 1996), and levomycetin (Maksimova et al. 2016) are frequently added to the medium to prevent bacterial growth. It is worth mentioning that in addition to molecular methods some of the filamentous fungi reported in this compilation were identified using classic microbiological methods, which is limited in the identification of fungal taxonomic groups without properly defined morphological keys of their colonies or spores in the bibliography (Fisher et al. 1996;Rodrigues et al. 2008). Conversely, yeast identification is usually performed by genetic means (Carreiro et al. 1997;Rodrigues et al. 2009;Pagnocca et al. 2010;Arcuri et al. 2014).
Studies on fungal microbiome frequently examine anthills of Acromyrmex and Atta ants from Canada, the USA, and some Central and South American countries, mainly Brazil. Additional research has also been carried out in Nigeria and Russia; however, there is a lack of knowledge of the mycobiota found in this niche in Europe, Asia, and Oceania ( Fig. 3). Research focused on studying non-mutualistic ant-associated fungi from Brazil led to the identification of species from Acremonium, Escovopsis, Fusarium, Mucor, Escovopsioides, Metarhizium, and Mariannaea genera; all isolates related to A. sexdens rubropilosa. Moreover, Trichoderma species have also been isolated from ants of the Acromyrmex genus (Rodrigues et al. 2005b;Augustin et al. 2013;Montoya et al. 2016;Armando et al. 2017). Species from the Leucocoprinus genus have received particular attention since they are cultivated by mycophagous ants (Lugo et al. 2013;Luiso et al. 2020). Yeast species from Candida, Trichosporon, Debaryomyces, Geotrichum, Hannaella, Cryptococcus, and others have also been described in association with Atta, Myrmicocrypta, and Solenopsis ants in Brazil and the USA (Ba et al. 2000;Pagnocca et al. 2010;Arcuri et al. 2014). A review on the diversity of yeasts associated with fungus-growing ants found that the orders 1 3 Saccharomycetales, Tremellales, and Trichosporonales are the most common (Bizarria et al. 2022).
Based on comparative studies, it is known that, in general, the mycobiota differs by ant species and even between the interior and exterior of the nests, tending to isolate more rare taxa in ant colonies. Not surprisingly, fungal microbiota differs even between nests of the same ant species; such variations are mainly due to topographic gradients, temporality of the ecosystem, and plant substrate for leafcutter ants (Kaspari et al. 2010;Rodrigues et al. 2011;Reis et al. 2015;Pereira et al. 2016;Russell et al. 2017;Delgado-Baquerizo et al. 2019;Parmentier et al. 2020;Lindström et al. 2021;Siedlecki et al. 2021). These differences also extend to nematodes. For example, it has been proposed that within the nests, a unique microenvironment is created, in which microbes adapt and diverge from the microbiome of the surrounding soil, phenomena mainly due to temperature stability and the continuous processes of organic material decomposition (Laakso et al. 1998;Lindström et al. 2018Lindström et al. , 2019. Specifically, in the nests of Aphaenogaster rudis, the richness of phytopathogenic fungi was lower, compared to that of the nearby soil (Tarsa et al. 2018). In nests of the arboreal ant Lasius fuliginosus and Azteca trigona, it was also observed that the fungal community is different from that of the surrounding land over time (Lucas et al. 2017;Brinker et al. 2019). Moreover, some patterns have been noted in several ant species, for example, fungi isolated from Pogonomyrmex occidentalis nest in wood-decaying differ from the mycobiota in dead pines or the dominant vegetation around the nest (Carlson and Whitford 1991; Lupala et al. 2019). Another example involves mycorrhizal fungi, which are more copious in the mounds of harvesting ants (Pogonomyrmex spp.) than in the surroundings. Moreover, it has been projected that the foraging activity of Messor spp. influences soil fertility in nests, increasing microbial biomass and its activity (Snyder et al. 2002;Ginzburg et al. 2008). Additionally, it was found that the Trichoderma genus and mycorrhizal fungi are more abundant in plant material rejected by Atta spp. than that deposited in the mound (Van Bael et al. 2009;Rocha et al. 2017).
The diversity of fungi isolated from ants and their nests is summarized in Fig. 4, regardless of whether a comparison was made among the surrounding soil, or the colony was captive or wild (Ba et al. 2000;Arcuri et al. 2014;Fig. 2 The evolutionary analysis was constructed by the maximum likelihood method. The evolutionary history was inferred by using the maximum likelihood method and General Time Reversible model, using the cytochrome oxidase subunit I gene (sequences used to perform the analysis, alignment, as well as the original tree are provided as Supporting Information file). The tree with the highest log likelihood (-2079.03) is shown. Initial trees for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach and then selecting the topology with superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+ G, parameter = 0.1234)). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. This analysis involved eight nucleotide sequences. Andrena xera (GenBank ID: OQ223003.1) was used as an outgroup. Codon positions included were 1st + 2nd. There was a total of 722 positions in the final dataset. Evolutionary analyses were conducted in MEGA11 (Tamura et al. 2021) Rodrigues et al. 2014;Duff et al. 2016;Maksimova et al. 2016;Montoya et al. 2016). The mycobiota at the genus level isolated from ant nests in the field, compared to that of the colony surroundings, as raised in some investigation, is summarized in Fig. 5 Duff et al. 2016;Montoya et al. 2016).

Ant Cuticles
Fungi isolated from ants' cuticles and their interior are scarce with only six reports found in the literature (Fig. 4), all of them are attributable to ants, and just a few reports dealing with fungi isolated from the nest. Fungi isolated from ants are described in this subtopic. Conversely, actinobacteria living in symbiosis with ants are the most studied microorganisms. In this review, fungi carried by ants were considered part of the nest mycobiota, considering they are distributed by these arthropods from outside and between nest spaces, either from their cuticle or, for example, from their gut through feces. Hyaline filamentous fungi were isolated in Brazil from Atta laevigata workers. The most abundant fungal isolates were Alternaria arborescens, Bipolaris sorokiniana, Bipolaris eleusines, Bipolaris zeae, Curvularia trifolii, and Paraphaeosphaeria michotii, species mainly associated with plants (Guedes et al. 2012). A search for fungi in young queen ants of Atta capigura and A. leviagata concluded that the genera Cladophialophora, Cladosporium, Exophiala, Ochroconis, Phaeococcomyces, Phialophora, and Penidiella are the most common filamentous fungi (mainly dematiaceous), whereas Aureobasidium, Candida, and Cryptococcus are the most represented yeasts Miranda Duarte 2013;Duarte et al. 2014). Interestingly, ants play a role as dispersers of mycobiota in nature. In fact, it has been proven that they are vectors of pathogenic species to humans. For example, it was reported that in several hospital areas in Greece, Brazil, France, and Iran, ants carry some opportunistic fungi such as Penicillium sp., Acremonium sp., and Mucor sp. All represent the known causal agents of mycoses (Aquino et al. 2013).
Remarkably, although entomopathogenic fungi are usually isolated from insect bodies, the prevalence of Cordyceps sp. in healthy ants is low, as demonstrated by Hughes in 2004, according to a study in which the cuticle of four different species of foraging ants from a tropical forest in Panama was examined (Hughes et al. 2004). Interestingly, dead ant specimens are a better source of entomopathogenic fungi and have led to the isolation of Aspergillus parasiticus, Aspergillus ochraceus, Beauveria bassiana, Metarhizum anisopliae, Pandora formicae, Ophiocordyceps sp., and Kneallhazia solenopsae species, all of them from dead females of Atta mexicana, workers of Atta bisphaerica, specimens of Formica sp., Camponotus sp., and Solenopsis sp. (Sokolova and Fuxa 2008;Ribeiro et al. 2012;Kobmoo et al. 2015;Małagocka et al. 2017). Furthermore, Aspergillus nomius was isolated from A. sexdens rubropilosa, and it was suggested that its pathogenicity comes from the mycotoxins it produces, such as aflatoxins B1 (2)

Anthills
The interiors of anthills are a series of complex, interlaced tunnels with various chambers to provide some larger areas inside the nest. These chambers have a variety of uses including storage areas for food, nurseries for ant larvae and their caretakers, and even areas for ants to simply rest. According to the literature, several filamentous fungi and yeasts have been isolated from anthills, the former being the most studied. To date, due to the number of species grouped together, Trichoderma spp. stand out among the filamentous ones, while Candida and Cryptoccocus are the most reported among yeasts. The association of this fungal diversity with some ant species is described below, although it is important to consider that most of the available information contemplate ant species of the Attini tribe (Fig. 4). So far, 360 species of fungi associated with ants and their nests have been isolated, 229 are filamentous, and 131 yeast-like, classified into 120 and 53 genera, respectively. This information was generated from the study of 34 species of ants, belonging to 13 genera, mainly Acromyrmex and Atta. The apparent diversity of filamentous fungi in Fig. 4 owes to the fact that most of these studies have focused on such microorganisms. Moreover, it can be noted that in terms of the number of fungal species, those found only in the nests are more copious than those found in the surrounding soil, for both filamentous and yeast-like species (Fig. 5). It cannot be overlooked that the fungal diversity associated with nests is the result of modifying that originally found in the soil, and that the differences observed are in fact, the effect of the establishment of ant's colonies. In general, the diversity of fungi associated with nests is outstanding, although the genus varies from ant species, time, and geographical location.
From the nest mounds of F. aquilonia ants in Novosibirsk, Russia, yeast species belonging to the Debaryomycetaceae family (Debaryomyces hansensii and Schwanniomyces vanrijiae) have been isolated. Notably, fungi isolated from the mounds differed from those species colonizing the surroundings (Trichosporon moniliforme and Cystofilobasidium capitatum) (Maksimova et al. 2016).
Similar differences in the distribution of filamentous fungi were found for Formica ulkei in Alberta and Nova Scotia, Canada. In this case, Aspergillus navahoensis, Aspergillus pseudodeflectus, Aspergillus spp., Paecilomyces sp., and Penicillium sp. were the most copious fungi in the mounds. In contrast, Cladosporium cladosporioides, Geomyces pannorum, Acremonium sp., Fusarium sp., Penicillium sp., and Phoma strains were the dominant species in the surroundings. Temperature and water availability in mounds are hypothesized to differentiate the ant/ nest-associated fungi from species colonizing nearby soil (Duff et al. 2016). It should be noted that nest sediments and neighboring anthills are a great source of entomopathogenic fungi, especially M. anisopliae and B. bassiana (Hughes et al. 2004;Angelone and Bidochka 2018), highlighting the ecological role of those fungal species in preventing anthill from other insects.
From nest interiors, especially fungal gardens, and waste deposits of A. sexdens and Myrmicocrypta sp., yeasts of Candida, Rhodotorula, and Trichosporon genus prevail, with Candida dubliniensis, Candida oleophilia, Cryptococcus haglerorum, Trichosporon chiarellii, and Hanseniaspora uvarum being the most common species (Carreiro et al. 1997;Middelhoven et al. 2003;Pagnocca et al. 2010). In addition, Candida parapsilosis and Candida lipolytica were identified in the brood chambers of Solenopsis invicta or "ant fire" (Ba et al. 2000). Concerning filamentous fungi within ants' nests, Leucoagaricus gongylophorus stands out as a symbiont of foraging ants; in-depth aspects such as its distribution among ant species or their lineage are not included in this review. However, according to the available information, this fungus and its parasite (Escovopsis sp.) are widely distributed throughout various ant species, and considered to have coevolved together with ants (Seifert et al. 1995;Currie et al. 2003;Lugo et al. 2013;Mueller et al. 2017;Bich et al. 2020). It should be added that only certain species of Escovopsis are considered opportunistic parasites and a pair of members have been reclassified into new genera, Sympodiorosea and Luteomyces (Montoya et al. 2021). In addition, E. nivea has also been identified as an antagonist of fungal gardens (Pietrobon et al. 2022). Up to today, the chemical composition of L. gongylophorus remains largely unknown, but given its importance as food for ant colonies, its study may be of interest from a nutraceutical point of view. It has also been suggested that certain fungi collaborate with L. gongylophorus on degrading plant material (Rodrigues et al. 2005b). Moreover, inhibition of the symbiotic fungus is a proposed target for ant control (Della Lucia et al. 2014;Araújo et al. 2022).
Several species of Trichoderma genus are commonly found in the chambers-nest garden of Atta and Acromyrmex genus, for example, A. sexdens rubropilosa, A. lobicornis, and A. lundii (Montoya et al. 2016;Armando et al. 2017). From Acromyrmex, Trichoderma lentiforme, Trichoderma inhamatum, Trichoderma virens, Trichoderma koningiopsis, and Trichoderma aff. Neotropical, among others, have been identified (Armando et al. 2017). Trichoderma spp. have been proposed to be a pathogen of Trachymyrmex isolated from nests and in their vicinity. In the case of nests, sampling points are specified. The genera are presented together with the number of species they include; only those > 1 are specified. In parentheses is the total number of genera, and at the end of each section the species of ants in the nests are listed. This figure was generated with data reported in the Supporting Information file septentrionalis fungal gardens (Kyle et al. 2023). In addition, several new species of Trichoderma, including T. attinorum, T. texanum, and T. longifialidicum spp. Nov., have been discovered (Montoya et al. 2016).
Interestingly, it was found that the foraging ant Atta colombica does not prefer plants with a high abundance of endophytes; the 30-43% increase in the time required to cut, transport, and clean this type of tissue was shown to be an undesirable cost. Furthermore, since this preference is independent of endophyte diversity, it is suggested that the techniques used by ants are general. Further, it was found that the colonization by the symbiont fungus showed the same performance in plant tissues with a high and low abundance of endophytes (Bittleston et al. 2011;Van Bael et al. 2012).
The mycobiota diversity isolated from fungal gardens may reflect the endophytic and epiphytic mycobiota of the plant material collected by ants, which vary over time, as detailed in a report on A. cephalotes from which Aspergillus niger, Epicoccum nigrum, Fusarium culmorum, Fusarium solani, Phomopsis glandicola, Phomopsis illicit, and Phomopsis quercella, among other species, have been identified (Fisher et al. 1996). In addition to the fungal composition in healthy Fig. 6 The evolutionary analysis was constructed by the maximum likelihood method. The evolutionary history was inferred by using the maximum likelihood method and General Time Reversible model, using the 18S rDNA sequence (SSU region). All sequences were obtained from the NCBI or SILVA database (Glöckner et al. 2017) (sequences used to perform the analysis, as well as the original alignment and tree files, are provided as Supporting Information files). The tree with the highest log likelihood (-6451.37) is shown. Initial trees for the heuristic search were obtained automatically by applying Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances estimated using the maximum composite likelihood (MCL) approach and then selecting the topology with superior log likelihood value. A discrete Gamma distribution was used to model evolutionary rate differences among sites (5 categories (+ G, parameter = 0.6421)). The tree is drawn to scale, with branch lengths measured in the number of substitutions per site. This analysis involved 154 nucleotide sequences. Giardia ardeae (SILVA accession number: Z17210) was used as an outgroup (branch length = 2.20). There were a total of 419 positions in the final dataset. Evolutionary analyses were conducted in MEGA11 (Tamura et al. 2021) nests of A. sexdens rubropilosa, there has been an interest on knowing the fungi that mainly establish in the fungal garden of intoxicated or abandoned colonies, noting that the parasite Escovopsis spp. cannot only establish in these weakened nests (Rodrigues et al. 2005a, b;Carlos et al. 2011).
Today, the fungal strains isolated from ants or their nest are all within the Ascomycota, Basidiomycota, Mucoromycota, and Zoopagomycota phyla. Among them, the most represented subphylums are Pezizimycotina with 82 plus genera, Saccharomycotina (19), Agaricomycotina (35), Ustilaginomycotina (6), Mucoromycotina (5), and finally Mortierellomycotina and Zoopagomycotina with one genus each (Fig. 6). This data highlights the broad fungal diversity carried by ants and that enclosed in their nests, and at the same time, projects this ecological niche as a rich source for rare cultivable fungi.

Chemical Compounds
Chemical information available on ant-associated fungi demonstrates the ability of these microorganisms to synthesize complex structures with unknown ecological roles. Notwithstanding, specialized metabolites produced by fungi are incredibly valuable in drug discovery programs aiming to find new molecules to treat human infectious and chronic diseases (O'Brien and Wright 2011). Currently, only 27 specialized metabolites reported in the literature have been isolated from ant/nests-associated fungi, including triterpenoids, indolterpenoids, and linear and polyaromatic polyketides. This chemical information has been partially described in some compilations focused on the chemical ecology of microbial and insect interactions (Beemelmanns et al. 2016;Van Arnam et al. 2018;Worsley et al. 2018;Batey et al. 2020;Menegatti et al. 2021). A brief description of the source and biological activity of all isolated molecules is presented in the following lines.
Sulfated triterpenoids A-108835 (4) and A-108836 (5) were obtained from Fusarium compactum cultivated in tomato and oat medium. This fungal strain was isolated from nest-mound soil harvested in Rantan (Nigeria, Africa). The major compound A-108835 (4) is substituted with an acetyloxy group, whereas the minor triterpene sulfate A-108836 (5) differs from A-108835 in the arrangement of the heteroannular diene with 7,8:9,11 position of rings B and C instead of the 8,9:14,15 positions of the C and D rings (Brill et al. 1996). Also, in A-108836 (5), the hydroxy group at C-11 is missing and a methoxy group at C-12 replaced the acetyloxy subtituent at C-12 in A-108835 (4). The initial report on the structure of compound 4 referred to this molecule as a diastereomer with the sodium 3-sulfate of 4,4,24-trimethylcholesta-8,14,24(28)trien-2α,3β,11α,12β-tetrol-12-acetate, a compound isolated from the plant pathogen Fusarium graminearum. In vitro assays demonstrated that compound 4 inhibited the proteolytic activity of rhinovirus 3C protease (IC 50 48 μg/ml), a molecular target to treat the common cold. However, it does not show antibacterial and antifungal activity against human pathogens (Brill et al. 1996). It was found that these triterpene sulfates and some semi-synthetic derivatives can inhibit HIV-1 integrase enzyme with IC 50 values ranging from 4.8 to 15 μM, essential in HIV-1 virus replication. A charged group, e.g., a sulfate or carboxy, is required for the antiviral activity demonstrated in a viral spread assay. Unfortunately, these molecules show a very small therapeutic window (Singh et al. 2003).
Chemical investigation of the parasitic fungus Escovopsis weberi obtained from the nest of Acromyrmex echinatior dereplicated a series of shearinine-type indoletriterpenes, including shearinine L (6), M (7), D (8), E (9), and A (10), which were subsequently isolated from its axenic culture in soy flour media. In addition, emodin (11) and cycloarthropsone (12) were also obtained (Dhodary et al. 2018). 10 is a part of a family of indole alkaloids isolated from sclerotioid ascostromata of Eupenicillium shearii, reported as antifeedants of Helicoverpa zea larvae plague in topical application trials (Magnus and Mansley 1999). More recently, the related structure 6 also showed insecticidal effect by decreasing A. octospinosus foraging of substrate impregnated with it, adding a negative allelopathic effect of the parasite Escovopsis to the nests (Dhodary et al. 2018). The insecticidal activity of the indole alkaloids has been linked to the ABC ring system, 1,3,9,9a-tetrahydroindeno[2,1-c] pyran, present in other families of compounds with the same effect (Belofsky et al. 1995). Compounds 8 and 9 were isolated from the seafloor fungus Penicillium janthinellum; these molecules have been described in vitro as inducers of apoptosis in human leukemia HL-60 cells. Compound 9 also prevented epidermal growth factor-induced malignant transformation of JB6 P + Cl 41 cell line in agar with an IC 50 of 13 µM (Smetanina et al. 2007). Product 8 was previously dereplicated from Escovopsis sp. by comparison of spectrometric data with databases, together with shearinines F and J. Furthermore, it was determined that aromatic compounds 11 and 12 inhibit the growth of Leucoagaricus gongylophorus (symbiotic fungus) in agar diffusion tests (Dhodary et al. 2018), acting as fungicides against the symbiont fungus, promoting foraging ant control. Moreover, it was previously described that certain coumarins related to 11 inhibit Leucoagaricus sp. in vitro, especially those possessing the 2H-chromen-2-one unit (Godoy et al. 2005).
Molecular network analysis guided the isolation and characterization of several specialized metabolites, including the 22,23-dehydroshearinine A (13), and the sulfur-containing diketopiperazines melinacidin IV (14), melinacidin III (15), chetracin B (16), and chetracin C (17) from the extracts of Escovopsis sp. Compounds 14 and 8 are synthesized in excess during fungal garden infection and inhibit the growth of Pseudonocardia sp. . Additionally, the linear polyketide conocandin B (18) and its derivative, conocandin C (19), were isolated from this fungus. Interestingly, 18 also inhibited the growth of Pseudonocardia sp., when evaluated in vitro, whereas 19 did not show the same activity (Bae et al. 2021) (Fig. 7).

Fig. 7
Symbiotic interactions between ants and the microbiota inside their nests. (A) Under normal conditions, ants of the Attini tribe cultivate the symbiotic fungus to feed their larvae; (B) however, occasionally, an opportunistic fungus settles in the fungal garden. When it occurs, (C) ants detect the problem and apply a chemical treatment, a product of a different symbiotic relationship. (D) It has been found that certain actinomycetes that inhabit ants' cuticles synthesize antifungals that stop the parasite. In addition, (E) ants ensure the contact of compounds with the unhealthy part of the fungal garden by mixing them in their mouths Bionectria sp. was isolated from the fungal garden of Apterostigma dentigerum in Costa Rica. Its fermentation in potato dextrose agar (PDA) revealed the production of bionectriol A (20), a glycosylated polyketide, which did not show antifungal activity when tested in coculture trials against Escovopsis sp. and Candida albicans (Freinkman et al. 2009). Structurally related compounds were isolated from Bionectria chroleuca, of which TMC-151E, TMC-151F, and bionectriol C moderately inhibited Candida sp. biofilm formation (CI 50 815.3, 23.5, and 42.3 μM, respectively) (Wang et al. 2014).
From the fungus Talaromyces sp. IQ-313, isolated from the mound of an anthill located at Huasteca Hidalguense, Hidalgo, Mexico, a series of asymmetric dimeric phenalenones: duclauxin (21), talaromycesone B (22), xenoclauxin (23), and bacillisporin G (24) were found when the fungus was cultivated in oat cereal. Biological evaluation of such compounds as inhibitors of a full-length protein tyrosine phosphatase 1B (hPTP1B 1-400 ), a therapeutic target for developing antidiabetic and anticancer drugs (Jiménez-Arreola et al. 2020), evidenced the potential of these molecules as prototypes for developing hypoglycemic drugs. Other oxyphenalenones were isolated from fungi such as Penicillium duclauxii, Gilmaniella humicola, and Neonectria sp., and have been evaluated mostly as inhibitors of the growth of cancer cell lines (Elsebai et al. 2014).
As for the symbiont fungus, it has been ruled out that its volatile compounds play an ecological role. However, some compounds act as signalers in their interaction with pathogens and maximize their growth in the garden (Folgarait et al. 2011;Bueno de Oliveira 2020;Sousa et al. 2020). Characterization of volatile compounds in culture assays of Escovopsis sp. and L. gongylophorus has allowed their tentative identification by solid-phase microextraction and gas chromatography-mass spectrometry analysis (Masiulionis and Pagnocca 2020).
The limited number of specialized metabolites isolated from ant/nest-associated fungi considerably focuses on the chemical composition of the parasitic fungus Escovopsis sp. and the symbiotic taxa Leucoagaricus sp. These data show the lack of research, not only regarding the mycobiote of the enormous diversity of ant species available, but also the chemical composition of their fungi and how this could favor the discovery of new functional chemical entities with potential applications in human or veterinary medicine, agriculture, and research, to mention a few areas (Scherlach et al. 2013;Meena and Siddhardha 2019). However, the importance of understanding the role that these secondary metabolites of microbial origin play in these biological systems should not go unnoticed (Frey Klett et al. 2011;O'Brien and Wright 2011;Moreau 2020).

Nest Interspecific Relationships
Studies on the ecological interaction of fungi inhabiting ants or their nests have proposed that species of the Attini tribe, mainly those cultivating Leucoagaricus gongylophorus, establish a mutualistic symbiotic relationship, which is complemented with Streptomyces, Pseudonocardia, and Burkholderia actinobacteria (Currie et al. 1999Currie 2001;Santos et al. 2004;Poulsen et al. 2009;Barke et al. 2010;Seipke et al. 2011;Sit et al. 2015;Dângelo et al. 2016;Goldstein and Klassen 2020;Nascimento et al. 2022). Specialized necrotrophic parasitic fungi such as some species of Escovopsis sp. can lead to colony death, especially when invading the fungal gardens of susceptible nests; however, certain actinobacteria are known to prevent the growth of this parasite by producing antifungal compounds (Reynolds and Currie 2004;Sen et al. 2009;Schoenian et al. 2011;Jiménez-Gómez et al. 2021). For example, Trachymyrmex cf. zeteki carries actinobacteria in its cuticle and mixes them with unhealthy parts in an infraoral pouch to clear infections; other species of the Attini group carry them in complex cuticular modifications (Little et al. 2006;Li et al. 2018) (Fig. 7). The putative symbiotic nature of the ant-actinobacteria relationship remains under discussion; nevertheless, the recruitment of effective actinobacteria against Escovopsis sp. by ants has been described . A similar relationship has been described for Allomerus decemarticulatus and Allomerus octoarticulatus ants that inhabit plant cavities constructed with fungi of the order Chaetothyriales, carrying bacteria of Streptomyces and Amycolatopsis genera capable of inhibiting the growth of fungi contaminating their nests, for example, Annulophypoxylon spp. and Chaunopycnis spp. (Seipke et al. 2012). In addition, it has been proposed that black yeasts (Phialophora genus) are an additional component of the ant-actinobacteria relationship, since some isolated species of ants can inhibit the development of actinobacteria and impact their effect on Escovopsis sp. Currie 2007, 2008). The opportunistic microorganisms of the nests can also be bacteria. Amycolatopsis bacterium was isolated from the fungal garden of Trachymyrmex smithi; from this microorganism, nocamycin V (29), a heterotricyclic compound capable of inhibiting the growth of Pseudonocardia sp. strains under laboratory conditions, was obtained from its extract (Hansen et al. 2022).
Some of the compounds that have been reported as inhibitors of fungi Ecovopsis sp. and its derivatives are summarized below. Several dentigerumycins have been found since the discovery of dentigerumycin (1) from a strain of Pseudonocardia associated with the nest of Apterostigma dentigerum. Compound 1 was one of the first molecules identified as a mediator of ant-actinobacteria mutualism by in vitro growth inhibition assays of Escovopsis sp. (IC 50 2.8 µM), but it does not affect the symbiotic fungus ). Another molecule, candicidin C (30), was dereplicated as the major component of the bioactive fractions of Streptomyces sp., isolated of Acromyrmex octospinosus. This molecule is capable of inhibiting the development of E. weberi and Escovosis aspergilloides but not L. gongylophorus (Haeder et al. 2009). Subsequent investigations with Pseudonocardia associated with Attini ants led to the finding of other compounds. GE37468 (31) is a known thiopeptide isolated from bacteria related with the cuticle of T. septentrionalis ant, which antagonize other bacteria of the same genus in the nest (Chang et al. 2020). Other important molecules are dentigerumycin F (32) and anttimicin (33); the former was found in the actinobacteria extract when it was cocultured with the Escovopsis supernatant, while the latter, known as a siderophore compound, inhibited the development of the parasite in its iron-depleted form (IC 50 12.5 µg/ml) (Bae et al. 2021;Fukuda et al. 2021). Dereplication of compounds from antimicrobial extracts of Pseudonocardia sp., Streptomyces sp., and Escovopsis sp. by spectrometric analysis and the identification of multiple biosynthetic pathways of antibiotics have been recurrent (Barke et al. 2010;Seipke et al. 2011), as well as by direct analysis of the ant exoskeleton (Boya et al. 2017;Gemperline et al. 2017).
All these compounds that share chemical and biological features, for example, are made up of highly modified amino acids, the cyclic depsipeptide, or polyene cores, and inhibit the growth of Escovopsis sp. in vitro. Nevertheless, in competition evaluations Pseudonocardia strains also inhibit the growth of the entomopathogenic fungi M. anisopliae (Bruner-Montero et al. 2021). Given the lack of specificity, it is speculated that such molecules also model the ant/nest-associated mycobiota, which would also be expected as an effect of the metapleural glands in some ants and its defensins in general (Schoenian et al. 2011;Yek and Mueller 2011;Vander Meer 2012;Sakolrak et al. 2018). Compound 33 is also synthesized in response to entomopathogenic fungi (Gómez-Díaz et al. 2022), as evidenced by the behavior of ants inside the nests during sanitization of fungal gardens by some species (Goes et al. 2022). In this context, fungi in the nest are expected to have developed adaptation mechanisms based on evolutionary competition, for example, through resistance mechanisms or allelochemical interactions mediated by specialized metabolites. Analogous to the bioprospecting of antifungals, antibacterial, and antiparasitic from ant-associated actinobacteria (Desireé et al. 2013;Holmes et al. 2016;Efimenko et al. 2020;Ortega et al. 2021;Wu et al. 2022), it is proposed that the nest mycobiota constitutes a rich source of antibacterial molecules, especially useful in the search for treatments for species resistant to conventional drugs (OMS 2020).

Perspectives and Future Directions
The information available on fungal diversity associated with ants and their nests is limited; in this review, it can be noted how most of the research is related to Attini ants and given their economic importance it is not expected for such interest to diminish. Likewise, we anticipate that the description of chemicals from micromycetes will not depart from the study of their role in interactions within Attini nests, in the same way that the subject favored the discovery of a variety of metabolites of bacterial origin (Ramadhar et al. 2014;Artavia-León et al. 2018;Chevrette et al. 2019;Van Moll et al. 2021;Selim et al. 2021). Regarding the rest of ant species, certain bioprospecting studies have allowed to unveil the chemical composition of their mycobiota, although exclusively from filamentous fungi and using conventional techniques in organic chemistry. Perhaps this mycobiota, both filamentous and yeast-like, goes unnoticed, likely due to the belief of its similarity with species isolated from soil; however, according with the information presented, it is proposed to abandon this point of view, as ants modify not only the physicochemical characteristics of the soil in which they settle, but also its microbiota. Accordingly, the way in which this difference can be exploited is promising, as well as the micromycetes associated with arboreal and warrior ants. We believe that studies on the variation of the mycobiota in anthills will continue to offer interesting information, even more with the facilities of massive sequencing techniques, and the development of analytical methods to study their chemical composition.

Conclusions
Based on the literature revision, the biological diversity of fungi inhabiting ant nests has not been extensively studied. Pioneering research has focused on analyzing soil samples from mounds and nest chambers, and indirectly from ant cuticles. As a result, only a few hundred of ant-associated filamentous fungi and yeasts have been described, being Trichoderma, Penicillium, and Fusarium the most reported genera within the Pezizomycotina subphylum and Candida and Cryptococcus in the Saccharomycotina clade. The majority of chemical entities have been isolated from 14 ant species, mainly of the Acromyrmex and Atta genera found in the American continent, perhaps because of their economic importance as pests. Since this information was obtained almost exclusively through classical methods in microbiology, the application of modern massive sequencing techniques would help to fill this gap due to many ant species and their cosmopolitan nature. Also, the diversity of the secondary metabolites produced by these microorganisms and their potential application are promising fields. In the background, compounds with in vitro antibacterial, antifungal, antidiabetic, and antiviral activity have been mentioned, as well as interspecific relationships that place the anthills' mycobiota as a source of chemical entities whose bioactivity should be pondered through bioprospecting studies, leading to discover molecules with applications in human or veterinary medicine, agriculture, and research, to mention a few. Finally, the ecological role of fungi in nests is just beginning to take place in relationships that have been investigated for a few years.
Acknowledgements ASAC thanks CONACyT for the scholarship (885837) for postgraduate studies.
Author Contribution ASAC collected and analyzed all data presented and wrote the first draft. JRC analyzed and revised the manuscript. Both the authors read and approved the final manuscript.

Data Availability
The data used and/or analyzed during this review are provided as supporting information files.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.