Amazonian tropical forest soils are supposed to hold high microbial biodiversity, since they support by litter recycling one of the most luxuriant ecosystems. However anthropogenic practices of slash-and-burn, mainly for pasture establishment, induce deep changes in the biogeochemical cycles, and possibly in the composition and function of microbial species (Cenciani et al., 2009).

While the diversity of microorganisms in the soil is immense, only a very low percentage is cultivable (around 1%) under laboratory conditions. The limited range between the bacteria species, for example, hampers the detection by microscopy techniques. Additionally the methods of obtaining bacteria in culture medium are not very effective for its quantification, due to difficulties in reproducing the conditions that every species or groups require in their natural habitats (Felski & Akkermans, 1998). Estimates of the global diversity of fungi indicate that a small percentage is described in the literature, especially due to limitations found in techniques of cultivation to assess the diversity of fungi. Apart from this the lack of taxonomic knowledge hinders the identification of bacterial and fungal species found in the soil (Kirk et al., 2004).

The study of prokaryote diversity is extremely complex because the definition of species for these organisms is a question still open. Currently a prokaryotic species is regarded as a group of strains including the standard strain, characterized by some degree of phenotypic consistency showing 70% or more DNA-DNA homology and more than 95% similarity between the 16S rRNA gene sequences. In this context we highlight the importance of polyphasic taxonomy, which aims to integrate different datasets and phenotypic, genetic and phylogenetic information about the microorganisms (Gevers et al., 2005).

With the advance of molecular biology it became possible to identify bacteria, fungi and other microorganisms in the soil and plants without need to isolate them. One of cultivation — independent molecular tool that has often been used to analyze the diversity and dynamics of microbial populations in the environment is the polyacrylamide gel electrophoresis in denaturing gradient (DGGE). The DNA is extracted and purified and only a fragment of the rRNA gene is amplified by the polymerase chain reaction (PCR). The amplification products are analyzed by gel electrophoresis, which allows the separation of small PCR products, commonly up to 400 bp according to their contents of guanine plus cytosine (G+C) Consequently, the fingerprinting pattern is distributed along a linear denaturing gradient (Muyzer & Ramsing, 1995; Courtois et al., 2001; Cenciani et al., 2009).

1.1 Fungi diversity assessed by PCR-DGGE

The role of fungi in the soil is complex and fundamental to maintain the functionality of the biome. Fungi play an active role in nutrient cycling and develop pathogenic or symbiotic associations with plants and animals, besides interacting with other microorganisms (Anderson & Cairney, 2004).

Working with soils in the Amazonia, Monteiro et al. (2007) described the changes in the genetic profiling of soil fungal communities caused by different land use systems (LUS): primary forest, secondary forest, agroforestry, agriculture and pasture. The author conducted her study in the following sequence: DNA extraction — total DNA was extracted using the Fast DNA kit (Qbiogene, Irvine, CA, USA), according to the manufacturer’s instructions; PCR — a fragment of the 18S rRNA gene (1700 bp) of fungi was amplified by PCR according to Oros-Sichler et al. 2006; DGGE — amplicons were separated on an acrylamide gel containing bisacrilamide and a linear gradient of urea and formamide (Fig. 2).

Diversity Database program (BioRad) was used to determine the richness of amplicons. The non-metric multidimensional scaling (NMDS) tool was used to determine the effect of land use changes under the fungi communities through the PRIMER 5 program (PRIMER-E Ltd., 2001).

The DGGE of the 18S rRNA gene combined with NMDS statistic tool showed the presence of distinct communities in each of the areas analyzed, with the presence of single bands. Results indicated the dominance of specific fungal groups in every treatment, especially in the area converted to pasture, distant from the other systems of land use (Fig. 2).

Following this pattern the authors asserted that the banding profile generated by DGGE represent fungi communities from different soils, and were shown to be more similar among samples from the same system of land use than among samples of different systems of land use. However the clustering of samples through NMDS showed that there is a tendency for samples from pasture be different of the other sites, which are closest relatives among them ( Fig. 3). Finally the results obtained by the authors show that changes in the land use affected the community structure of soil fungi; as well it is also possible that the type of vegetation covering has a key role in such changes (Monteiro et al., 2007).

Although molecular fingerprinting approaches such as cloning and sequencing are being used increasingly for evaluation of fungal communities, there are scarce studies reaching the diversity of fungi in soils of native forests, and in the same soils but impacted by agricultural management. Within this context changes in the genetic profile of fungi according to each system of land use, and the environmental stress can provide valuable information for the sustainable management of forest soils (Monteiro et al., 2007).