Selection of fungi with biocontrol potential against the black spot disease of papaya 1

are necessary to control the disease (Barreto et al. 2011). Despite being the main form of control, the intensive use of fungicides may select resistant isolates and negatively affect the human health, environment and fruit commercialization. Depending on the amount of fungicide found in the fruit, its commercialization may be unviable in the country, as well as its exportation, since the quality of the fruit

In view of the problems caused by fungicides in papaya crops, it is necessary to adopt alternative control strategies.Among them are the use of resistance-inducing products (Oliveira & Nishijima 2014, Santos et al. 2017) and genetic resistance for the selection of resistant genotypes (Vivas et al. 2012and 2015a, Poltronieri et al. 2017).In spite of the efforts employed in the development of control methods against the black spot disease of papaya, no product is known to induce resistance in the plant, and neither there are cultivars resistant to the disease.
An alternative strategy that may be adopted is the biological control.This type of control is considered a promising method to reduce the pesticide use and protect crops, thus adding value to the product going to the market, as it is based on procedures that reduce the pathogen population in a sustainable way and may be part of an integrated system for disease control.For papaya, the biological control is still little studied, with some reports on biocontrol agents such as bacteria and yeasts against several post-harvest fungi (Bautista-Bañosa et al. 2013, Lima et al. 2014).However, in the literature, there is no research on the black spot disease of papaya biocontrol.Thus, this is a pioneering study of great importance for the selection of isolates with mycoparasitic potential against A. caricae.
A fundamental step to implement a biological control program is the selection and isolation of antagonists.For this purpose, biocontrol agents can be selected by methods that are based on the evidence that the relevant organism somehow interferes with the pathogen development or reduces the disease.Considering the different forms of action of the antagonist on the pathogen, it is believed that a greater diversity of isolates leads to the use of more virulent antagonists, thus contributing to the identification of promising biological control agents.
Given the above, the present study aimed to obtain fungi with mycoparasitic potential, evaluate the fungal colonization by inoculation in the host and select the fungi that are agent precursors for the biological control of the black spot disease of papaya.

MATERIAL AND METHODS
Fungi with mycoparasitic potential were obtained from black spot lesions on papaya leaves collected in different locations, in 2012.The study also included mycoparasitic fungi obtained from leaves of okra [Abelmoschus esculentus (L.) Moench.] and scarlet eggplant (Solanum aethiopicum L. var.gilo Raddi) showing pathogens with signs of mycoparasitism similar to the black spot observed in papaya, as well as leaves of coffee (Coffea arabica L.) and frangipani (Plumeria rubra L.) showing rust mycoparasites (Table 1).
The mycoparasitic fungi obtained in pure cultures were submitted to monosporic isolation.In this procedure, spores diluted in a drop of sterile water were spread with a Drigalski spatula in a 2 % agar-water medium under a glass slide, and then incubated in a humid chamber at 25 ºC and a 12-hour photoperiod.Under light microscopy, the freshly germinated spores were transferred to a potatodextrose-agar (PDA) culture into Petri dishes.The pure cultures were maintained in tubes containing PDA culture medium tilted at 10 ºC and also preserved using the Castellani's method (Gonçalves et al. 2016).
The morphological structure of the mycoparasites was observed by light microscopy and compared with the genus descriptions from a taxonomic manual of asexual forms of fungi (Seifert et al. 2011).In order to confirm the identification of the isolates, phylogenetic analyses were carried out.To do so, the mycelium of the isolates was macerated for extraction of genomic DNA with the protocol modified by Pinho et al. (2012), using the Wizard Genomic DNA Purification Kit (Promega).The eluted DNA was stored at -20 ºC until use.
Nucleotide sequences were edited with the DNA Dragon software (Hepperle 2011).All sequences were corrected manually and the arrangement of the nucleotides at ambiguous positions was corrected using the primer sequences in the 5 '-3' and 3'-5' orientations (Pinho et al. 2012).After verification, the sequences were compared in the GenBank database using the BLAST tool and aligned in the Muscle ® program (Edgar 2004), by the MEGA v. 7 software (Kumar et al. 2016).Gaps (insertions/deletions) were treated as nonexistent.The phylogenetic tree was processed by Bayesian Inference (BI) analysis, using the Markov Chain Monte Carlo (MCMC) method with 10,000,000 generations sampled per 1,000 and 25 % burn-in.The selected substitution model was GTR + I + G, according to the Akaike Information Criterion (AIC).The tree was root-dip inoculated with Verticillium dahliae DB 08052009-3.
In order to evaluate the isolate colonization in the black spot lesions, two experiments were carried out in a shadehouse covered with 50 % shade cloth.The first experiment took place in July 2013 (with indoor temperature of 17-32 ºC and relative humidity of 57-78 %) and the second one in November 2013 (indoor temperature of 20-30 ºC and relative humidity of 64-98 %) (Figure 1).The experiments were designed in randomized blocks, with 25 treatments (24 mycoparasitic isolates and 1 control -water) and three replications.The experimental plot consisted of one plant per pot.The experiments used the 'Golden' papaya cultivar, considered susceptible to the black spot disease (Vivas et al. 2012).
The conidial suspension used in the experiment was prepared with isolates grown in a PDA culture medium for 13 days, at 25 ºC and a 12-hour photoperiod.After this period, the fungus was suspended in distilled water using a Drigalski spatula, and then the conidia were counted with a Neubauer chamber.The plants, grown in 5-L pots, 6 months old and with symptoms of the disease, had the lower part of their leaves sprayed with a 10 5 conidia mL -1 * A = Acremonium spp.; H = Hansfordia pulvinata; I = Sarocladium implicatum; L = Lecanicillium lecanii; S = Simplicillium lanosoniveum.suspension of each of the mycoparasitic fungi included in the treatment.During the treatments, the plants were kept separated from each other, to avoid genetic drift of the mycoparasitic isolates.
The average time, in days, for the appearance of colonization signs (white mycoparasitic structures covering black spot lesions), the number of black spot lesions and the incidence (%) of white mycoparasitized lesions at 15 days after inoculation were evaluated.For this evaluation, 10 cm of the midrib of the last viable leaf was considered, from tip towards petiole.
The data were submitted to analysis of variance.When a significant effect was observed, the Scott-Knott test (p < 0.05) was conducted.In order to select the most promising isolates with ability to colonize black spot lesions, the isolates were included in a Biplot graph.All analyses were performed using the Genes software package (Cruz 2013).

RESULTS AND DISCUSSION
The phylogenetic analysis, performed with 37 taxa, resulted in the alignment of sequences of 516 characters, of which 229 were parsimony informative, 264 variable and 251 conserved.Confirming the morphological analyses, the 24 isolates were grouped into five clades (Table 1, Figure 2), which contain species previously described as having a biocontrol potential.
The isolates L-620 and L-622, derived from the causal agent of the coffee leaf rust, were identified as Lecanicillium lecanii (Zimm.)Zare & W. Gams.The isolate S-599 from Simplicillium lanossoniveum (J.F. H. Beyma) Zare & W. Gams was obtained from the causal agent of frangipani leaf rust.These genera belong to the same family (Cordyciptaceae) and are studied as entomopathogenic and mycoparasitic fungi in leaf rust (Jackson et al. 2012, Gauthier et al. 2014).
For the first four species aforementioned, the phylogenetic analysis shows that the gene under study has enough phylogenetic signal to separate them into species (Figure 1).On the other hand, these molecular markers do not have enough phylogenetic signal for the different species of Acremonium spp., including the isolates A-598, A-601, A-602, A-603, A-604, A-616, A-617, A-618, A-619 and A-621, mycoparasites of A. caricae in papaya (Table 1, Figure 2).The Acremonium genus may be potentially used in agriculture, acting as endophytic and mycoparasitic fungi (Wicklow et al. 2005, Choi et al. 2008).As previously mentioned, the obtained fungi have a biocontrol potential in other pathosystems.However, in papaya, this is the first time that the potential of such fungi in the biological control of Asperisporium caricae is investigated.
Regarding the number of days elapsed between the inoculation and appearance of mycoparasitism signs, it was not possible to perform the analysis of variance, as some isolates did not show any signs in the evaluated period (Figure 3).A variation between the isolates was observed in both experiments, concerning the appearance of mycoparasitism signs.In the first experiment, the signs appeared between 8 and 11 days after inoculation, while, in the second one, the signs appeared between 6 and 9 days after inoculation.The isolates of H. pulvinata stood out for taking a shorter time to mycoparasitize A. caricae (Figure 3).
Regarding the incidence of leaves with mycoparasitized black spot lesions, a significant variation (p < 0.01) was observed among the isolates.As for the number of black spot lesions on the leaves, no significant effect was noticed between the treatments and, consequently, all the isolates were inoculated in plants with a uniform number of lesions.It was also observed that, during the first experiment, the number of black spot lesions was lower than that observed in the second experiment.This could be explained by differences in the air temperature and humidity between the first and second evaluated periods (Figure 1).In addition, before the inoculation of the mycoparasites, the plants were under the same incubation conditions, under a shadehouse covered with 50 % shade cloth (Sombrite ® ).This way, environmental conditions affected both the manifestation of the disease symptoms and the efficiency of the inoculated mycoparasites.
In the second experiment, the environmental conditions were more favorable for both the pathogen (A.caricae) and the mycoparasites, since there were more lesions produced by the pathogen infested by the mycoparasites.Such conditions led to an increase in the incidence rate for all the mycoparasite isolates.In addition, the signs of mycoparasitism were observed earlier.With reference to the environmental conditions, temperatures of 20-30 ºC and humidity of 64-98 % were described in experiments as favoring both the pathogen and the mycoparasite (Suzuki et al. 2007, Vivas et al. 2005b).Considering the temperature variation in the two experiments, it is believed that the temperature had a smaller influence on the mycoparasitism, when compared to the humidity, since this was higher in the second experiment.However, further studies are needed to elucidate the effect of climatic factors on the occurrence of mycoparasitism on black spot lesions.The results obtained for the pathogen corroborate field observations by Suzuki et al. (2007), in the north of the Espírito Santo state, Brazil, who found that favorable conditions for the black spot disease in the field include temperatures of 25-30 ºC and a high relative humidity of 80-100 %.
Thus, it is possible to identify potentially promising isolates for the biological control of the black spot disease of papaya, being H-611, H-613, H-614 and H-615 (H.pulvinata) the ones with higher mycoparasitic efficiency.These isolates were the ones that stood out in the two inoculation moments (Figure 4), reaching an incidence rate of 75.46 % for the isolate H-611, in the second experiment.
Based on the results herein presented, the isolates L-620 and L-622 (L.lecanii), S-599 (S. lanosoniveum) and I-609 and I-621 (S. implicatum), although already related to the biological control in other studies, have not yet been reported as natural mycoparasites of Asperisporium caricae.Nevertheless, the present study shows that these fungi have the capacity to infect the pathogen, notably in favorable environmental conditions, as observed in the second experiment (Figures 3 and 4).
Isolates of Acremonium spp., although showing a greater potential than the isolates of L. lecanii, S. lanosoniveum and S. implicatum, did not stand out, when compared to isolates of H. pulvinata.A positive aspect concerning the use of Acremonium spp.lies in the fact that they produce chlamydospores.These give them a greater competitive and adaptive advantage to the environment, thus surviving longer under field conditions, for instance, fungi used as biological control agents such as Trichoderma spp., The obtained results confirm the mycoparasitic potential of the A. caricae phytopathogen for use in biological control programs against the black spot disease of papaya.In this study, isolates of the H. pulvinata species were the most promising ones, as they were the first to present signs of mycoparasitism, showing a higher efficiency in colonizing the black spot pustules under the different environmental conditions evaluated, especially the isolates H-611, H-613, H-614 and H-615.Further research should be conducted until a bioproduct is formulated and considered for commercial use.However, the first step was taken toward the selection of isolates potentially useable for biological control of the black spot disease of papaya.

Figure 1 .
Figure 1.Temperature (line) and humidity (bar) measured during the experiments: the first took place in July 2013 (a) and the second one in November 2013 (b).

Figure 2 .
Figure 2. Bayesian inferences based on ITS-rDNA sequences.Subsequent probabilities are shown above the nodes.The isolates obtained in this study are highlighted in bold.The tree was root-dip inoculated with Verticillium dahliae DB 08052009-3.

Figure 3 .
Figure 3. Average and standard deviation of the days after the inoculation of mycoparasites in papaya leaves with black spot symptoms, in a shadehouse, in two periods: July (white bars) and November (gray bars) of 2013.

Table 1 .
Isolates of mycoparasitic fungi obtained from different hosts and Brazilian locations.