The almonds were fermented in “Fazenda Konagano”, is located in district of the state of Tomé-Açu, Pará. Afterwards, they were submitted to three types of drying, in the Laboratory of Agronomy Engineering of UFRA as shown in Figure 1.

The 10 samples of cocoa beans were collected, in the initial and final stages of each process (drying).

For drying, the samples were collected in their respective dryers at the beginning of drying in ideal conditions, then sent to the Microscopy laboratory of UFRA to identify the fungus and its incidence for each drying method used in the study [7] .

When evaluated the fungi, the colonies were cultivated in culture medium, Sabouraud agar (Laborclin), at a concentration of 65 g per 1000 ml of distilled water, sterilized in an autoclave for 15 min at a temperature of 118˚C [8] . Then, the sample was cultured and incubated 25˚C ± 1˚C/21days, the readings of the plates were taken at the time interval of 7, 14 and 21 days [9] .

The plates with mixed culture were frown with the fungi of interest in Potato Dextrose Agar (concentration of 18 g of dextrose, 200 g of potato and 20 g of agar, autoclaved at 121˚C for 20 min), incubated at 25˚C ± 1˚C/14days. In order to guarantee the purification of the fungus, 2 ml of distilled water was diluted in the plate where the fungus is isolated. The culture was poor in nutrient, water agar, and the fungus solution was striated by “spread plate” technique and incubated for 24 hours, thus obtaining the isolated colony.

To visualize the fungal spores and hyphae, a glass slide was prepared, the dye Methylene Blue was added, then, with the help of a needle, a fungal sample was collected and, on the plate, superimposed on a cover slip and visualized under an Optical Microscope (SP labor, Model: BLUE1000-BIL-BI) [10] .

DNA from bacteria was extracted using the CTAB (cetyl trimethylammonium bromide) protocol described by [11] , modified.

For DNA extraction, the fungi were initially grown on enriched and differential solid medium, Blood-Agar-(AS) with growth for 36 hours. After cultivation, one to three colonies of bacterial growth were added to microtubes containing 567 µL of 1× TE and vortexed. To fisicla break the cell wall of the fungi, expose and extract the DNA, 30 UL of 10% SDS, 3 µl of proteinase K were used and incubated for 1 to 1.5 hours at 37˚C, then 100 µl of 5 M NaCl, 80 IU of CT 80 µL of CTAB/NaCl placed in a 65˚C water bath for 10 minutes, 780 µL of chloroform/isoamyl alcohol and shaken manually for 10 minutes and centrifuged for 5 minutes at 14.000 rpm. Then the supernatant was transferred to new tubes and 360 µl of ice-cold isopropanol added. Then, the microtube was inverted until the DNA became visible and it was centrifuged at 14.000 rpm for 25 minutes. However, immediately afterwards, the supernatant must be discarded, the DNA washed with 1 ml of 70% ethanol and centrifuged for 3 minutes at 14.000 rpm and the DNA was allowed to dry for 20 minutes in a support laminar. The DNA was resuspended in 50 µL of 0.1× TE added with ribonuclease (10 mg∙mL⁻¹) and incubated at laboratory temperature 25˚C for 12 hours.

After the extraction was completed, the samples were quantified in Qubit^TM. The amplicons were analyzed on Agarose Gel (2%). Then, the samples were diluted to concentrations of 10 to in µL⁻¹ and stored at −20˚C.

For amplification of the genetic material (the DNA), DNA concentration was performed using the Biodrop µLite spectrophotometer (Thermo Fisher Scientific, USA). For identification, the 18S rDNA gene was applied with primers 27F (5'-AGAGAGTTGATCMTGGCTCAG-3’) and 1492R (5'-ACCTTGTTTTACGACTT-3’) with approximately 1450 base pairs of amplification.

PCR was performed with a total volume of 25 µl of a reaction mixture consisting of 12.5 µl of 2× Master Mix PCR (Promega GoTaq® Master Mix, Wisconsin, USA), 20 pmol of the primers and 100 ng/µl of bacterial DNA. PCR was performed in a thermocycling apparatus (Mastercycler, Eppendorf, Germany). PCR was performed with the following parameters: initial denaturation 4 minutes at 94˚C, followed by 25 cycles of denaturation 94˚C for 1 minute, annealing 55˚C for 1 minute and extension 72˚C for 1 minute and final extension 72˚C for 7 minutes. PCR amplification was analyzed by agarose gel electrophoresis (2% w/v). The run was performed in 0.5× TBE buffer at 120 V for 40 min. Visualization was performed with a transilluminator device and images were captured using UPV software (BioDoc-it TM).

PCR products were purified using the Exo + SAP enzyme mixture containing two hydrolytic enzymes: recombinant shrimp alkaline phosphatase (rSAP) and Exonuclease I (Exo I), following the manufacturer’s recommendations.

For bacteria, DNA-templates were used the primers 27F (5'-AGAGTTTGATCMTGGCTCAG-3') and 1492R (5'-ACCTTGTTACGACTT-3’). The sequencing of the samples was carried out at the company ACTGene Analyses Moleculares Ltda. (Biotechnology Center, UFRGS, Porto Alegre, RS) using the AB 3500 Genetic Analyzer automatic sequencer equipped with 50 cm capillaries and POP7 polymer (Applied Biosystems). The DNA-templates were purified with the ExoSAP-IT^TM PCR Product Cleanup reagent (Applied Biosystems) and quantified in the Nanodrop 2000 c instrument (Thermo Scientific). Afterwards, they were labeled using 2.5 pmol of specific primer and 0.5 μL of BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems) reagent in a final volume of 10 μL. Labeling reactions were performed in an LGC XP Cycler thermocycler with an initial denaturation step at 96˚C for 3 min followed by 25 cycles of 96˚C for 10 sec, 55˚C for 5 sec and 60˚C for 4 min. Once labeled, samples were purified by precipitation with 75% isopropanol and washing with 60% ethanol. Precipitated products were diluted in 10 μL of Hi-Di^TM formamide (Applied Biosystems), denatured at 95˚C for 5 min, cooled on ice for 5 min and electro injected in the automatic sequencer. Sequence data were collected using Data Collection 3 software (Applied Biosystems) with parameters Dye Set “Z”; Mobility File “KB_3500_POP7_BDTv3.mob”; BioLIMS Project “3500_Project1”; Run1 Module “FastSeq50_POP7_50m_cfv_100”; and Analysis Module 1 “BC-3500SR_Seq_FASTA.saz”.

The resulting Data Collection files (.ab1; electropherograms) were converted to FASTA files (.seq; text) by Sequence Analysis Software v. 6 (Applied Biosystems) under default parameters. These sequences were compared to with 18 species rDNA gene sequences for fungi available in the GenBank database, the blast algorithm (http://www.ncbi.nlm.nih.gov/Genbank/index.html). Multiple sequence alignments were performed using ClustalW (Kyoto University, Bioinformatics Center; http://www.genome.jp/tools/clustalw/ ).

Phylogenetic relationships among the fungi strains were performed by the alignment of sequences using the Clustal X 2.0 software [12] . The Unweighted Pair Group Method with Arithmetic Mean (UPGMA) phylogenetic trees were constructed using the software Mega 11 [13] .

The fungi most found in the drying of yards were Aspergillus spp. and Penicillium spp. Morphological characterization, Figure 2, for being suggestive fungi, for being xerative fungi, for low moisture content to avoid deprivation, must be in accordance with [6] .

Molecular Taxonomic Identification of Fungi Isolated from Cocoa beans.

An identity above 80% was found in all Indians, being considered reliable for molecular identification, Table 1.

Fonte: Os Autores (2021).

The species found in this study: Geotrichum candidum and Galactomyces geotrichum are endophytic fungi. On the other hand, Ascomycota sp. and Cladosporium cladosporioides are environment fungi (Table 1).

The taxonomy result demonstrates the proximity relationship between all analyzed species.

The studied sequences showed similarities with sequences deposited in a database. Most of them shared the same phylogenetic clade. Galactomyces geotricum R1 was similar to Geotrichum candidum MW493646.1. Cladosporidium species was similar in its accession number, Geotrichum candidum R3 sequences was close to Galactomyces geotricum JN903644.1 and Ascomycota sp. KM458803.1 was similar to its sequence in accession number (Figure 3).