Antifungal and proteolytic activities of endophytic fungi isolated from Piper hispidum Sw

Endophytes are being considered for use in biological control, and the enzymes they secrete might facilitate their initial colonization of internal plant tissues and direct interactions with microbial pathogens. Microbial proteases are also biotechnologically important products employed in bioremediation processes, cosmetics, and the pharmaceutical, photographic and food industries. In the present study, we evaluated antagonism and competitive interactions between 98 fungal endophytes and Alternaria alternata, Colletotrichum sp., Phyllosticta citricarpa and Moniliophthora perniciosa. We also examined the proteolytic activities of endophytes grown in liquid medium and conducted cup plate assays. The results showed that certain strains in the assemblage of P. hispidum endophytes are important sources of antifungal properties, primarily Lasiodiplodia theobromae JF766989, which reduced phytopathogen growth by approximately 54 to 65%. We detected 28 endophytes producing enzymatic halos of up to 16.40 mm in diameter. The results obtained in the present study highlight the proteolytic activity of the endophytes Phoma herbarum JF766995 and Schizophyllum commune JF766994, which presented the highest enzymatic halo diameters under at least one culture condition tested. The increased activities of certain isolates in the presence of rice or soy flour as a substrate (with halos up to 17.67 mm in diameter) suggests that these endophytes have the potential to produce enzymes using agricultural wastes.


Introduction
Biological control through microorganisms that inhibit or antagonize plant pathogens and pests reduces or eliminates the use of chemical products. Fungal endophytes are effective antagonists (Azevedo et al., 2000) and constitute a taxonomically and metabolically diverse group of organisms that colonize internal plant tissues without causing apparent harm to the host plant (Wilson et al., 1991). Indeed, endophyte-mediated biological control has been investigated both in vivo and in vitro through screening experiments to verify the activity of endophytes against phytopathogens and pests (Andreote et al., 2009;Badalyan et al., 2002;Campanile et al., 2007;Flores et al., 2013;Mejía et al., 2008;Rocha et al., 2009;Rubini et al., 2005;Sánchez et al., 2007;Specian et al., 2012).
Endophytic and phytopathogenic fungi compete and interact within the same ecological niche through the action of hydrolytic enzymes such as proteases and chitinases, which degrade the hyphal cell walls of pathogenic microorganisms (Almeida et al., 2007;Guthrie and Castle, 2006;Sánchez et al., 2007). This enzymatic activity is closely associated with the fungus-host specificity: the fungal strains of a given species isolated from the same host plant are remarkably homogeneous with respect to enzymatic production (Leuchtmann et al., 1992;Petrini et al., 1992). To facilitate the entry of endophytes into host tissues through natural or artificial openings, hydrolytic enzymes including pectinases, cellulases and lipases are secreted (Polizeli et al., 1991).
Proteases or proteolytic enzymes have commercial importance (Rao et al., 1998), as these enzymes are used in bioremediation and waste treatment, detergents, cosmetics and leather manufacture, silk degumming, animal cell culture, contact lens cleaning, therapy and diagnosis and the pharmaceutical, photographic and food industries. In addition, proteases are considered as insecticidal agents because these enzymes are required for the complete digestion of complex insect cuticles (Anwar and Saleemuddin, 1998;Gupta et al., 2002;Harrison and Bonning, 2010;Hasan et al., 2013;Kumar and Takagi, 1999;Murthy and Naidu, 2010;Nielsen and Oxenboll, 1998).
The medicinal plant Piper hispidum Sw. (Piperaceae), commonly known as "bayuyo" (Cuba), "cordoncillo" (Mexico), "jaborandi" or "falso-jaborandi" (Brazil), harbors a diverse endophytic fungal community (Orlandelli et al., 2012a), including fungi presenting activity against human pathogenic bacteria (Orlandelli et al., 2012b). Considering the shortage of information concerning the antifungal and enzymatic activities of the endophytes from this plant, the aim of the present study was to evaluate the antagonism and competitive interactions between endophytic and phytopathogenic fungi in dual culture experiments and to detect the proteolytic activity of these endophytes using a cup plate assay and different growth substrates.

Materials and Methods
Endophytic and pathogenic fungi A total of 98 endophytic fungi were isolated from the leaves of P. hispidum plants located in a forest remnant in southern Brazil (Orlandelli et al., 2012a) and belong to the fungal culture collection of the Laboratório de Biotecnologia Microbiana, Universidade Estadual de Maringá, Paraná, Brazil. These fungal strains were molecularly identified as Alternaria sp., Bipolaris sp., Colletotrichum sp., Colletotrichum gloeosporioides, Phyllosticta capitalensis, Lasiodiplodia theobromae, Marasmius cladophyllus, Phlebia sp., Phoma herbarum, Diaporthe sp., Schizophyllum commune and one isolate from the order Diaporthales. Molecular identification was based on sequencing of the ITS1-5.8S-ITS2 region of rDNA (GenBank accession numbers JF766988 to JF767008).
For the experiments, all fungi were previously grown in Petri dishes containing potato dextrose agar (PDA) medium (Smith and Onions, 1983) at 28°C under biochemical oxygen demand (BOD) for seven days.

In vitro antagonism and competitive interactions between endophytic and phytopathogenic fungi in dual culture
A modified version of the dual culture method of Campanile et al. (2007) was used. Briefly, 6-mm endophyte and phytopathogen plugs were combined in triplicate and inoculated onto PDA dishes, with a 4-cm distance between each plug. Filter paper plugs inoculated with 10 mL of fungicide Derosal plus® (with a 10 -1 dilution of methyl benzimidazol-2-ylcabamato + tetramethylthiuram disulfide) or fungicide Tiofanil® (with a 200 mg/mL dilution of chlorothalonil + thiophanate-methyl) were used as positive controls, and autoclaved distilled water was used as a negative control.
The antagonism index (AI) was calculated as previously described (Campanile et al., 2007) using the following formula: AI = (RM -rm)/RM x 100, where rm represents the ray of the colony toward the antagonist, and RM represents the average of the three rays of the colony in the other directions. The competitive interaction (CI) between endophytes and phytopathogens was determined according to the Badalyan rating scale (Badalyan et al., 2002), which considers three main types of interactions (A, B and C) and four interaction sub-types (C A1 , C B1 , C A2 and C B2 ). Types A and B represented deadlock (mutual inhibition) at mycelial contact (A) or at a distance (B), whereas type C was replacement or overgrowth without initial deadlock. The intermediate interaction subtypes scored consisted of partial (C A1 ) or complete (C A2 ) replacement after initial deadlock with mycelial contact and partial (C B1 ) or complete (C B2 ) replacement after initial deadlock at a distance.

Conditions for protease production and cup plate assay
The endophytic fungi were grown as previously described (Sena et al., 2006) in liquid inducer medium (IM) containing powdered skim milk (Nestlé®) as the inducer substrate to stimulate protease secretion. The cultivation conditions were adapted from Sena et al. (2006), and the endophytes were also grown in IM containing two different substrates (carbon sources): rice or soy flour (5 g/L). Liquid medium incubated without fungal inoculation was used as a negative control. The cultures were incubated under stationary conditions (BOD at 28°C for 10 days). Subsequently, the liquid medium was filtered using sterile gauze to separate the fungal mycelia.
For the cup plate assay, the filtered media were inoculated (50 mL) onto Petri dishes (9 cm) containing gelatin milk agar medium (Sena et al., 2006) with the surface perforated with cup plates (6-mm diameter). A commercial protease from Aspergillus oryzae (Sigma®) (³ 500 U/g) was used as a positive control.
The experiment was performed in triplicate, and the dishes were incubated under BOD at 28°C for 24 h. The en-zymatic activity was evaluated as the presence of clear halos on an opalescent background and measured in millimeters (Dingle et al., 1953).

Statistical analyses
All experiments were performed using a completely randomized design (CRD) and analyzed by ANOVA (analysis of variance). The mean values were compared using the Scott-Knott test (p < 0.05) in the statistical program SISVAR 4.3 (Ferreira, 1999).

Results and Discussion
Evaluation of in vitro antagonism (AI) and competitive interaction (CI) between endophytic fungi and phytopathogens The dual culture method has been broadly applied in antagonism studies because this analysis facilitates the in vitro screening of agents that can be used for biological control (Faria et al., 2002;Mariano, 1993). In the present study, ANOVA showed differences among the in vitro antagonistic actions, as varying degrees of phytopathogen mycelial growth inhibition were observed. The results obtained after screening all 98 P. hispidum endophytes are shown in Figure 1-A, and the types of CI observed between the endophytes and A. alternaria, Colletotrichum sp., P. citricarpa and M. perniciosa are shown in Figure 1-B. More details regarding AI and CI between the 21 molecularly identified endophytes tested and phytopathogens are shown in Table 1.
The AI values obtained for the best antagonist (L. theobromae JF766989) varied between 54.16 and 64.79%, and these results were higher than those obtained in a previous study (Campanile et al., 2007), where the best result for antagonism was 28.5%. Badalyan et al. (2002) observed that most of xylotrophic mushrooms and cereal phytopathogens present subtypes of the type C interaction. According to the scale proposed by the same authors, interaction types A and B indicate a deadlock or mutual inhibition in which neither organism overgrows in the presence of the other; in contrast, type C and associated subtype interactions indicate a replacement involving the inhibition of one organism. L. theobromae JF766989 partially overgrew (interactions C A1 and C B1 ) in the presence of all phytopathogens; however, most of the 98 P. hispidum endophytes presented deadlock interactions with mycelial contact (A).
Although these results suggest L. theobromae JF766989 as an antagonist of phytopathogenic fungi, most of the endophytes tested were more effective than fungicides for reducing the growth of the phytopathogens.

Evaluation of the proteolytic activity of endophytic fungi
Screening for new producers of novel and industrially useful enzymes is of great interest for biotechnology research (Kumar and Takagi, 1999). Proteases are physiologically necessary and have been isolated from a wide diversity of sources, such as plants, animals, and microorganisms (Rao et al., 1998). Microbial proteases have several characteristics necessary for biotechnological application and represent a large portion of the total worldwide sale of enzymes, with low production costs compared with animal or plant proteases. Moreover, microorganisms are preferred as a source of proteases due to their rapid growth, limited space requirements for cultivation and ease of genetic manipulation to generate new enzymes with desirable properties (Najafi et al., 2005;Said and Pietro, 2002).
The cup plate assay in the present study showed that 28 of the 98 endophytes (28.57%) presented proteolytic activity when grown in inducer medium. An ANOVA showed differences in the observed enzymatic halos, with means ranging from 1.33 to 16.40 mm in diameter; the highest value was observed for S. commune JF766994 ( Fig. 2 and Table 2).
Approximately 28.57% of the P. hispidum endophytes evaluated presented proteolytic activity under the conditions assayed. In some cases, the enzyme production was significantly higher after the medium was changed. Species of the genus Mucor are protease producers of commercial value (Alves et al., 2002), with 82% of 56 isolates belonging to 11 different species presenting proteolytic activity. Djamel et al. (2009) reported that only 10 (3.9%) of 253 Penicillium strains examined presented significant proteolytic activity, as based on the hydrolysis of milk casein (clear zones around the colony) and the mycelium colony diameter, with clear halos greater than 9 mm.
The 28 P. hispidum endophytic isolates that initially presented enzymatic activity were grown in the presence of rice or soy flour (Table 2). When rice flour was used as the substrate, 14 endophytes produced enzymatic halos, ranging from 7.27 to 15.40 mm, with the best values observed for S. commune JF766994. In addition, two unidentified isolates (G53-83 and G36-112) presented statistically superior enzymatic activity when grown on this substrate. In the presence of soy flour, positive results were obtained for 10 endophytes, with enzymatic halos ranging from 5.0 to 17.67 mm in diameter. The best result was obtained for P. herbarum JF766995, which presented statistically superior enzymatic activity when grown on this substrate; similar results were obtained for isolate G05-05.
Agro-industrial and other wastes can be used as substrates for fermentation, suggesting a cost-effective approach to enhance enzymatic production, as these substrates are cheap and abundant natural carbon sources (Blesson, 2009;Singh et al., 2012). Singh et al. (2012) showed that sugarcane bagasse, wheat bran, corncob, wheat straw and, in particular, rice bran are suitable substrates for the production of amylases and xylanases from thermophilic actinobacteria. In addition, substrates such as soy, wheat and rice bran, mango and banana peel, gelatin and fish flour have been used for the production of microbial proteases (Murthy and Naidu, 2010;Paranthaman et al., 2009;Souza et al., 2008).
Consistent with the results of the present study, Souza et al. (2008) used the cup plate assay to investigate the production of enzymes from Amazonian basidiomycetes cultivated on different substrates, obtaining enzymatic halos of up to 23.80 mm in diameter. Smaller halos (up to 18.07 mm in diameter) were obtained using a medium supplemented with protein sources, and halos of up to 19.11 and 18.64 mm in diameter were obtained on soy bran and fish flour, respectively. Paranthaman et al. (2009) verified protease production under the solid-state fermentation of Aspergillus niger using different varieties of broken rice as substrates, and the results varied between 44.7 and 67.7 U/g.

Conclusions
Endophytes constitute a novel and important new source of active substances that can be employed in different biotechnological industries. Diverse strains, even members of the same endophytic fungal species, can exhibit characteristic metabolite production with enzymatic or antifungal potential. Some positive antifungal phenotypes of endophytes might reflect competition for space or nutrients, as demonstrated through dual culture experiments. The results of the present study suggest that in the assem- 362 Orlandelli et al.

Figure 1
-Antagonism index (AI) and competitive interaction (CI) between 98 P. hispidum endophytic fungi and phytopathogenic fungi in dual culture. A) AI indicates the reduction (%) in phytopathogen mycelial growth. *Means of triplicates. Different letters indicate that the AI intervals are significantly different according to the Scott-Knott test (p < 0.05). B) *Badalyan rating scale (Badalyan et al., 2002): A = deadlock with mycelial contact; B = deadlock at a distance; C A1 = partial replacement after initial deadlock with contact; C B1 = partial replacement after initial deadlock at a distance. **N = no competitive interaction was observed (absence of endophyte antagonism).   (Badalyan et al., 2002): A = deadlock with mycelial contact; B = deadlock at a distance; C A1 = partial replacement after initial deadlock with contact; C B1 = partial replacement after initial deadlock at a distance. ***N = no competitive interaction (absence of endophyte antagonism). (C+1) Positive control (10 -1 in distilled water); (C+2) positive control (200 mg/mL in distilled water); (C-) negative control.
blage of P. hispidum endophytes, certain strains are important sources of antifungal properties, particularly L. theobromae JF766989, which reduced the growth of A. alternaria, Colletotrichum sp., P. citricarpa and M. perniciosa by approximately 54 to 65%.
Investigators in Brazil should further explore the potential to generate new enzymes from microbial sources, as this country has a continental area that includes hundreds of plant species with diverse endophytes. The results of the present study highlight the proteolytic activity of the endo-phytes P. herbarum JF766995 and S. commune JF766994, which presented the highest enzymatic halo diameters under at least one culture condition tested. As some isolates showed increased activity in the presence of rice or soy flour as a substrate, these endophytes have the potential to produce enzymes from agriculture wastes.

Acknowledgments
The authors would like to thank CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior) 364 Orlandelli et al.