Endophytes are microorganisms living in the internal tissues of plants without causing any symptoms. Efforts tending to characterize libraries of fungi isolated as endophytic from plant species have acquired relevance in order to add information to biodiversity and due to their important role as producers of bioactive metabolites. Fungal isolates are usually characterized by morphologic and molecular approaches (DNA barcoding), but a wrong identification or no identification occurs frequently. The advent of omics techniques has increased the available skills to solve those issues, especially proteomics and metabolomics, which have acquired great relevance in the field. The combination of morphological, molecular, proteomic and metabolic information certainly can help more deeply and consistently to solve the identification of those unidentified fungal isolates. The aim of this work was to identify a library of 63 endophytic fungi isolated from Peperomia obtusifolia (L.) A. Dietr. (Piperaceae) by morphological analysis, MALDI-TOF MS, sequence-based identification, and metabolic analysis by means of DI-ESI-QToF-MS and HPTLC-UV techniques.
Fungal endophytes previously isolated from P. obtusifolia were sub-cultured in potato dextrose agar with salts (APDS) for 7 days, and according to their macro and micro morphological aspect, they were grouped into 10 groups. Fungal isolates from two groups were assigned to the genera Fusarium and Alternaria, however in the other 8 groups asexual structures were absent and/or morphological characters were not enough in order to identification.
Fungal genomic DNA was obtained from cultures, and the ITS region from rDNA was amplified and sequenced using the primers ITS1 and ITS4. Sequences were compared with GenBank deposited sequences, and  97 % sequence similarity was considered in order to assign a genus and species name. Phomopsis/Diaporthe (24) was the most frequently identified genus among the fungal library, being Diporthe infecunda the most frequent isolated species. The other isolates belonged to Fusarium (7) and Mycosphaerella (7), Phoma (6), Colletotrichum (4) and Alterinaria (4) among the most represented genera.
In addition, the MALDI-TOF MS spectral profiles of fungal ribosomal proteins (range of m/z 2,000 to 20,000 Da) were obtained by culturing fungi in Sabouraud-Glucose agar And the MS analyses were performed on a MicroFlex LT MALDI-TOF mass spectrometer (Bruker Daltonics).The comparison of the spectral profiles and data processing was carried out by multivariate analysis of the main components (PCA) using Bionumerics software 7.6.
Metabolomic analysis of fungal isolates was performed by analysis of their metabolites by DI-ESI-QToF-MS and HPTLC. Fungal extracts were obtained from 10-days cultures in APDS using ethyl acetate as extracting solvent. PCA performed on the spectral data allowed to group fungal isolates in clusters, including strains which were unidentified by molecular and morphologic techniques.
The data obtained here allowed us to get a comprehensive characterization of all the isolates, including taxonomic identification and metabolic profiling. Phenotyping and genomic techniques demonstrate adequate complementarity in order to obtain a reliable identification of fungal isolates, adding information about their potential as producers of interesting metabolites.