Soil is a non-renewable resource that is under intense pressure by human activities, such as agriculture and husbandry (Olsson et al., 2019). Soil is a highly complex and heterogeneous matrix that supports species assemblage. It is well known that 1 g soil contains up to 10⁹ bacterial cells of 10⁴ taxa, up to 200 m fungal hyphae, and a wide range of different mites, nematodes, earthworms, and arthropods (Bardgett and van der Putten, 2014). This diversity contributes substantially to the total terrestrial biomass and to the ecosystems functioning. Soil fungi play fundamental roles in nutrient cycling and soil health as organic matter decomposers (Lindahl and Tunlid, 2015; Shah et al., 2016), mediators of plant and other soil organisms nutrition (Schneider et al., 2012; Baldrian, 2017), regulating species composition as pathogens (Chapelle et al., 2016), and by providing mutualistic benefits as symbiotrophs (Baldrian, 2017).

Land-use change is a major threat to biodiversity, including belowground diversity, and the ecosystem services that they provide (Bhattacharya et al., 2016). For instance, the soil organic carbon (C) stocks decreased globally 42% when native forests were converted to arable soil (Guo and Gifford, 2002; Bhattacharya et al., 2016). Although deforestation in temperate regions for agriculture or grazing has been reduced mostly or even reversed, it continues unabated in some parts of the world (Currie and Bergen, 2008; Millar and Stephenson, 2015; Food and Agriculture Organization of the United Nations [FAO], 2020) with dire consequences for soil quality. In high-altitude temperate forests in the central highlands in Mexico, conversion of forest to arable land decreased the soil C stock 85% (Fusaro et al., 2019). High soil C stocks reduces the risk of soil erosion and loss of soil fertility, while it facilitates water retention, improves soil structure, and increases soil nutrient content (Lal, 2004). As a consequence, the productivity of maize crops is low in these original forest soils (Montesillo-Cedillo, 2016) and the costs, due to the loss of ecosystems services, are high. Temperate high mountain forests in the tropics are very vulnerable to climate change. If the global temperatures continue to rise, these islands of high biodiversity and endemism would disappear (Halffter and Morrone, 2017; Mastretta-Yanes et al., 2018).

Land-use change induces a series of changes in the abiotic and biotic characteristics of a forest. For instance, the plant community composition (Celentano et al., 2017), edaphic characteristics (Miller and Jastrow, 2000; Lauber et al., 2008; Celentano et al., 2017), availability of plant nutrients (Zhu and Miller, 2003; Wang et al., 2018; Soltangheisi et al., 2019), and the amount and composition of organic matter (Solomon et al., 2002; Raiesi and Beheshti, 2015) are altered by land-use change. All these characteristics are known to affect the soil fungal communities and a number of studies have documented the changes in soil fungi as a result of land-use change in different biomes (Mueller et al., 2016; Marín et al., 2017; Tian et al., 2017). The response of soil fungal communities to land-use change differs across biomes and land-use intensity (Oehl et al., 2010; de la Cruz-Ortiz et al., 2020), depends on the time elapsed since land-use change (Mueller et al., 2016; Zheng et al., 2020), and on the different components of diversity, i.e., species diversity, community composition, and functional guild composition (Pereira et al., 2014; Purahong et al., 2014; Mueller et al., 2016). For instance, fungal species richness was reduced, fungal community composition altered, and generalist fungi enriched when a tropical forest was converted to pasture (Mueller et al., 2016) while it only affected fungal composition in the semiarid Loess Plateau (Tian et al., 2017). In temperate forests, fungal diversity was more linked with vegetation diversity than in any other biome as the ectomycorrhizal host density had a strong effect on the ectomycorrhizal fungal richness (Tedersoo et al., 2014). However, most of these studies have used classic species diversity measures, such as Shannon entropy (Shannon, 1948) and the Gini-Simpson index (Simpson, 1949), that do not measure diversity but entropy. For instance, the Simpson index measures the probability that two randomly chosen DNA sequences truly belong to different operational taxonomic units (OTUs) (Chao et al., 2014). This makes them prone to misuse and misinterpretation (Jost, 2006; Alberdi and Gilbert, 2019). True diversity, on the other side, is based on a mathematical framework easier to interpret and compare between studies (Jost, 2006). As such, it can be hypothesized that conversion of a high-altitude temperate forest to arable land will decrease the soil fungal diversity and specifically ectomycorrhizal fungi in the long-term. Gómez-Acata et al. (2016) found that soil organic carbon content was reduced by deforestation and conversion to arable land and soil characteristics showed a greater variation. As such, forest soils might harbor more heterogeneous microbial communities and thus a higher beta diversity than an arable soil or a pasture soil.

Microorganisms in soil interact with each other within complex food webs. Changes in abundance, diversity and composition of one or more taxa, functional guild or trophic group may affect another group and ecosystems functioning (Wagg et al., 2014). In a forest ecosystem, ectomycorrhizal fungi participate in the decomposition of organic material and become themselves a C substrate and a source of N for bacteria and other fungi (Brabcová et al., 2016).

True diversity indices, or Hill numbers, integrate the most common indices to measure diversity into a single statistical framework that quantify diversity in units of equivalent numbers or equally abundant species or OTUs (Hill, 1973). The weight assigned to the abundant and rare OTUs can be controled using the parameter q, known as the “order of diversity” (Jost, 2006). The larger the q-value, the higher the importance attributed to abundant OTUs. Consequently, the interpretation of diversity concept is more intuitive and its measurement unit is always the same (Tuomisto, 2010a; Chao et al., 2014). Hill numbers integrate measurement, but also estimation, partitioning, comparison of diversities, and components such as phylogeny and function (Alberdi and Gilbert, 2019). Phylogenetic diversity incorporates evolutionary or phylogenetic differences among species into diversity analysis. In phylogenetic Hill numbers, the units of diversity are branch segments or lineages as defined by a phylogenetic tree instead of OTUs or species, and incorporates the tree’s branching pattern, the relative branch lengths and the relative abundances of each node/branch (Chao et al., 2010). The unit of measurement or the effective total attribute value is the effective total branch length. Similarly, in diversity partitioning using Hill numbers, beta diversity is an actual diversity value that measures the effective number of equally large and completely distinct subsystems in a system (Tuomisto, 2010a,b; Chao et al., 2014).

In this study, the effect on the soil fungal communities of converting a high-altitude temperate coniferous forest to a pasture or agricultural land for the cultivation of maize (Zea mays L.) was studied in the natural protected zone “Área de Protección de Flora y Fauna Nevado de Toluca” in Mexico using 18S rRNA metabarcoding. Changes in fungal communities resulting from land-use change can provide important information for soil management and the assessment of the environmental impact of deforestation and land-use change. The soil samples used in this study were the same as those used by Gómez-Acata et al. (2016) to study the effect of land-use change on the bacterial community. As such, the effect of land-use change on the soil fungal communities could be compared with that of bacteria.

Land-use change is considered a strong driver of biodiversity change (Sala et al., 2000). In this study, we confirmed our hypothesis that the conversion of a temperate forest to arable land significantly decreased alpha, beta, and phylogenetic diversity of soil fungal communities, but only the beta diversity decreased when the forest was converted to pasture. Although the effect of deforestation on species richness and diversity can vary, they decrease mostly when a forest is converted to a cultivated system, such as a plantation, perennial cropland, or cropland (e.g., Balami et al., 2020). In this study, the decremental effect of land-use change on fungal diversity can be mainly attributed to changes in vegetation as none of the tested soil physicochemical properties correlated with changes in the fungal taxonomic and phylogenetic alpha diversity. Although pH decreased in the arable soil compared with the forest soil, pH did not correlate with the decrease in fungal diversity. Species richness of ectomycorrhizal fungi, which was higher in the forest soil in this study, was strongly predicted by the richness of their host plants and pH at larger scales (Rousk et al., 2010; Tedersoo et al., 2014). Experimental studies have shown that acidic soils reduce ectomycorrhizal colonization (Thomson et al., 1996).

The variation in the physicochemical properties and soil fungal communities was smaller in the arable soils compared to the pasture and forest soils (Gómez-Acata et al., 2016). The arable soil from the three locations (Dilatada, Capulin, and Rosa Morada) was cultivated in the same traditional way, i.e., maize monoculture, tillage, and removal of crop residue for fodder or fuel, or it was burned (Arriaga-Cabrera et al., 2009). Similar traditional practices reduced the variation in soil organic carbon stocks in arable soil compared to pasture and forest soils in the high mountain ecosystem of the National Park “La Malinche” (Fusaro et al., 2019). The high variation in soil microbial communities in the forest soil might be the result of the high variation in its physicochemical characteristics, e.g., soil organic matter. However, our hypothesis was only partially confirmed as the variation in the edaphic properties did not correlate with the beta diversity of the fungal communities. Mueller et al. (2014) found that changes in the fungal community composition in tropical forest was more closely correlated to plant community composition than to changes in soil properties and Brinkmann et al. (2019) found less dissimilarities between fungal communities at increased intensity in agricultural management. The decline in diversity and the homogenization of fungal communities in the arable soils of the natural protected zone “Área de Protección de Flora y Fauna Nevado de Toluca” might be due to the intensive agricultural practices.

Ascomycota and Basidiomycota were the most abundant fungal phyla in the soils studied and are often dominant in soil of different biomes, such as temperate forest (Buée et al., 2009), tropical rainforest (Mueller et al., 2016), arable soil (Klaubauf et al., 2010; Tardy et al., 2015), grassland (Tardy et al., 2015), and arid and semi-arid soils (Tian et al., 2017). The forests soil in this study had a higher abundance of ectomycorrhizal fungi than the other soils, as could be expected as they establish symbiotic relationships with the root system of coniferous plants. Ectomycorrhizal fungi belong mostly to Basidiomycota and to a lesser extent to Ascomycota, which explained why the relative abundance of Basidiomycota was lower in the arable than in the forest soil (Buée et al., 2009).

Glomeromycota establish important mutualisms with plants, i.e., arbuscular mycorrhiza, that allow them to inhabit a wide range of environments and their proportion and diversity is larger in grassland than in forest or arable soils as found in this study (Oehl et al., 2010; Xiang et al., 2014; Belay et al., 2015; Xu et al., 2017). On the one hand, arbuscular mycorrhizae were enriched in the pasture soil compared to the forest soil. Conifers are usually associated with ectomycorrhizal fungi, which also are capable to degrade resistant organic material abundant in a forest soil. On the other hand, the higher plant diversity in grassland provides more niches for arbuscular mycorrhizal fungi than in cultivated soil. Additionally, intensive agricultural practices, such as high fertilizer application rates, pesticide use and plowing, are detrimental for arbuscular mycorrhizal colonization (Hodge, 2014).

Converting a temperate forest to a pasture and arable land also increased the abundance of animal pathogens. In this study, the relative abundance of the fungal genera Mortierella, Phialophora, Torula, or Acremonium, which contain pathogenic or opportunistic pathogenic animal species (Davies et al., 2010), was higher in the arable and pasture soils. This increase in animal pathogens in some agroecosystems, often as a result of land-use change, has been associated with the emergence and transmission of infectious diseases (Gottdenker et al., 2014; Mendoza et al., 2020). Metarhizium is an entomopathogenic fungus, whose relative abundance decreased in the order arable > pasture > forest. This can be explained by the large increase in the number of insects, such as grasshoppers (Sphenarium purpurascens Charpentier), in ecosystems with a limited plant diversity, such as monoculture of maize (Bullen, 1966). Although the relative abundance of plant pathogens did not increase significantly as a result of land-use change, the plant pathogen Coniochaeta (described by Damm et al., 2010) was enriched in the arable soil. Similarly, Marín et al. (2017) reported that the intensive management of Chilean temperate forests increased the abundance of fungal plant pathogens. The authors speculate that reduction in the complexity of the vegetation facilitates the dispersion of pathogens.

Fungi that can grow as macromycetes are mostly wood decomposers (saprotrophs) and include several ectomycorrhizal species of forest trees (symbiotrophs). The relative abundance of these functional groups of fungi decreased when the forest was converted to pasture or arable land. Total organic C explained 10% of the variation found in the abundance of fungal trophic groups. Saprotrophs and symbiotrophs play significant roles in soil organic matter formation (Wardle and Lindahl, 2014). Thus, land-use change of a high-altitude temperate forest changes the abundance of saprotrophs and symbiotrophs, which might have a great impact on the ecosystem C exchange and sequestration, and this may, in part, explain the great decline in C stocks in this type of ecosystem (Fusaro et al., 2019).

In a previous study, we found that deforestation had also a profound effect on the bacterial community structure (Gómez-Acata et al., 2016). Bacterial and fungal communities in the soil of the high-altitude temperate forest of the “Área de Protección de Flora y Fauna Nevado de Toluca” had a similar sensitivity and similar response to land-use change. Shifts in microbial community composition following land-use change have been associated also with changes in functional genes profiles in high-altitude temperate forest (Muñoz-Arenas et al., 2020). Therefore, land-use change might be strongly and negatively affecting soil functions such as the C and N cycle, which are driven largely by bacterial and fungal communities (e.g., Bardgett and van der Putten, 2014).

Clearing the temperate high-altitude forest in the “Área de Protección de Flora y Fauna Nevado de Toluca” for maize cultivation decreased significantly the effective number of fungal OTUs and their total branch length, but not when converted to grassland. Beta diversity decreased in the order forest > pasture > arable soil. However, none of the measured edaphic characteristics correlated with the changes in alpha, beta, and phylogenetic diversity. The ordination analysis showed a clear effect of land-use intensity as forest soils clustered closer to the pasture soils than to the arable soils. The most abundant fungal phyla in the studied soils were Ascomycota, Basidiomycota and Mucoromycota. Conversion of coniferous forests more than halved the abundance of Basidiomycota, mostly Agaricomycetes such as Lactarius and Inocybe. The number of OTUs belonging to Glomeromycota decreased in the order pasture > forest > arable soil. Symbiotrophs, especially ectomycorrhizal fungi, were negatively affected by deforestation while pathotrophs, especially animal pathogens, were enriched in the pasture and arable soil compared to the forest soil. The total soil organic carbon significantly explained 10% of the variation of fungal trophic groups. Shifts in the soil fungal communities in these soils as a result of land-use change resembled those of the bacterial communities.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, SRP043580; https://www.ncbi.nlm.nih.gov/bioproject/, PRJNA253630.

LD and YN-N designed the study. NM-C, LM-A, and YN-N did the laboratory work. SH-P and YN-N did the bioinformatic analyses. AE-T and YN-N did the statistical analysis. YN-N drafted the initial manuscript. All authors commented on the manuscript.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.