CONTROL OF Phytophthora palmivora ON POSTHARVEST PAPAYA WITH Trichoderma asperellum, T. virens, T. harzianum AND T. longibrachiatum CONTROLE DE Phytophthora palmivora EM MAMÃO NA PÓS-COLHEITA POR Trichoderma asperellum, T. virens, T. harzianum E T. longibrachiatum

Papaya (Carica papaya) is one of the most cultivated and consumed tropical fruit worldwide. Its production might be limited by preharvest and postharvest diseases. The fruit rot caused by Phytophthora palmivora is one of the most important postharvest diseases of papaya in Brazil. The control of these diseases is usually made with fungicide applications. Therefore, studies concerning biocontrol of postharvest diseases might generate data that may reduce the environmental impacts caused by pesticides. Thus, the biological control by Trichoderma in postharvest diseases is an alternative to the use of fungicides for the postharvest control of P. palmivora in the papaya fruit. Four antagonists [T. asperellum (SF04), T. virens (255C1), T. harzianum (THP) and T. longibrachiatum (4088)] were tested, as follow: 1) Trichoderma spp. applied 1 hour after inoculation of P. palmivora and; 2) Trichoderma spp. applied 24 hours after inoculation of P. palmivora; 3) Trichoderma spp. applied 1 hour before inoculation of P. palmivora, and 4) Trichoderma spp. applied 24 hours before inoculation of P. palmivora. All Trichoderma significantly (P≤0,05) reduced the incidence and severity of disease. The 4088 (T. longibrachiatum) isolate was the best controller agent of P. palmivora in postharvest.


INTRODUCTION
Papaya (Carica papaya) has several phytosanitary problems such as pests, fungal and bacterial diseases mainly in post-harvest. Among the fungal diseases feature the root rot and fruit that has caused 60% loss of production . Fruit decays account for significant levels of postharvest losses. It is estimated that about 20% of the harvested fruits are decayed by pathogens during postharvest handling (EL-GHAOUTH et al., 2001;DROBY, 2009).
The causal agent of papaya stem, root and fruit rot is Phytophthora palmivora (Butler) Butler. This disease occurs in almost all producing regions of Brazil, and a reason for it the absence of resistant commercial papaya cultivars (KADER, 2002). The control of P. palmivora is basically done with fungicide and one of those is a combination of metalaxyl and mancozeb. Fungicide application may leave toxic residues on fruits and may select pathogen isolates with resistance to it. In addition, fungicides when not applied properly, might cause environmental contamination and damage to human health (ROBERTS; KUCHAREK, 2005).
The presence of agricultural products in pesticide residues and the accumulation of these substances in the environment have stimulated research of alternative methods of disease control in postharvest, especially by public agencies concerned with health issues and trade relations (OLIVEIRA et al., 2006). Therefore, the biological control applicability is emphasized and studied for the management of various diseases even if informally. Due to the lack of studies related to postharvest especially in producing regions, such as the Southern Bahia one of the leading producers and exporters of papaya whose research and biological control application are still incipient. In addition, the biocontrol has emerged as viable technology in postharvest fruit being favored by specific targets and the fruit storage in controlled conditions, which favors the antagonist establishment (JANISIEWICZ; KORSTEN, 2002).
Trichoderma is well documented as a biological control agent (BCA), with potential use in fruit diseases caused by Phytophthora spp. (ALEXANDER; STEWART, 2001). Trichoderma species are potential antagonists of various pathogenic fungi and have several action mechanisms, the most effective being: the metabolites production and enzymes with antifungal properties, hyperparasitism and nutrients competition (HARMAN et al., 2000). As an additional advantage, these microorganisms are referred to as non-toxic to man and animals and as symbiotic plants associated with avirulent (MERTZ et al., 2009).
In several fruits such as apple, strawberry, citrus and banana the Trichoderma, has been used to effectively control postharvest diseases (PRATELLA; MARI, 1993;PAJMON et al., 1995;BATTA, 2004;NALLATHAMBI et al., 2009). Therefore, the potential in the postharvest diseases' management using biocontrol are relevant to diverse cultures in postharvest, subject to environmental and agribusinesses (JANISIEWICZ; KORSTEN, 2002). Although the Bahia south and Espírito Santo northern are responsible for most of Brazilian production and export of papaya, few studies refer to the biocontrol application in the postharvest fruit and this has compromised the commercial competitiveness. Therefore, the aim of this study was to evaluate the efficacy of Trichoderma spp. for the biocontrol of P. palmivora in postharvest papaya.

MATERIAL AND METHODS
The isolate 356 P. palmivora was obtained from the Collection of Phytophthora 'Arnaldo Medeiros' at the section of Plant Pathology in the Research Center of Cocoa Crop (CEPLAC), Bahia, Brazil, and, where the experiments were conducted.
Phytophthora palmivora (356) was transferred to Petri dishes containing selective medium PARPH (KANNWISCHER; MITCHELL, 1978), grown for five days in the dark (25 ±2ºC), and, then transferred to carrot agar medium (CA). The dishes were kept in incubator (BOD -Biochemistry Oxygen Demand) under continuous light (25 ± 2 ºC), for nine days. Then the pathogen was inoculated into healthy fruit for observation of the P. palmivora characteristic symptoms and again isolated following Koch's postulates. Finally, the cultures were maintained in test tubes containing culture medium CA and preserved for further studies. The four isolates of Trichoderma spp. used in this study were provided by the Collection of fungi antagonists at the Biocontrol Laboratory of CEPLAC. These antagonists were selected as effective biocontrol agents in other studies: SF04 (T. asperellum) (TOCAFUNDO et al., 2010), 255C1 (T. virens) e THP (T. harzianum) (TAVARES, 2009); and, 4088 (T. longibrachiatum) (OLIVEIRA et al., 2009). Trichoderma spp. isolate were transferred to Petri dishes with medium Potato Dextrose Agar (PDA) and maintained for 10 days at 25 ± 2 ºC with 12h light.

Preparation of P. palmivora and Trichoderma spp. suspensions
Suspensions of P. palmivora were obtained from 20 dishes containing the pathogen zoosporangia. To each plate was added 8 mL of sterile distilled water (SDW). For liberation of zoospores the dishes were subjected to thermal shock [5 (± 2 ºC) / 20 min followed by 25 (± 2 ºC) / 25 min]. Afterwards, the zoospores concentration was determined and standardized in 5 x 10 5 zoospores mL -1 suspension using Neubauer chamber, adding 2 drops of fixative solution FAA (formaldehyde, alcohol and acetic acid) for standstill of zoospores.
For Trichoderma spp. suspension was obtained from 20 dishes containing the antagonist candidate. Eight mL of SDW were added on the colonies surface and conidia were removed by friction with a brush. This suspension was filtered through sterile double gauze and conidial concentration (10 8 mL -1 ) was determined and adjusted in hemocytometer.

Evaluation of Trichoderma spp. for biocontrol of postharvest papaya fruit rot
The experiments were performed in the 'Phytophthora' Laboratory, Plant Pathology Section of the Crop Research Center Cocoa (CEPLAC). Two experiments were conducted identically, one in March / 2010 and another in April / 2010. The second one was performed 30 d after the first (Experiment 1). For the evaluation of potential biocontrol were used fruit variety Sunrise Solo, maturation stage 2 (RITZINGER; SOUZA, 2000), from Farm 'Alegria', located at Vera Cruz City, Bahia, Brazil.
The fruits were washed with mild soap and water and dried at room temperature. The Trichoderma were then applied as follows: Trichoderma spp. applied 1 h after inoculation of P. palmivora; Trichoderma spp. applied 24h after inoculation of P. palmivora; Trichoderma spp. applied 1h before inoculation of P. palmivora; Trichoderma spp. applied 24h before inoculation of P. palmivora. Two P. palmivora inoculation methods were also tested: a) 5x10 5 de zoospores mL -1 (20 µL), applied over three 2-mm injured holes at the equatorial region of fruit, and; b) zoospores suspension spray. Each Trichoderma isolate was sprayed as conidial suspension (10 8 conidia mL -1 ) in both types of pathogen inoculation. For comparison purposes with the Trichoderma spp., a fungicide (mancozeb + metalaxyl) treatment was applied on fruits. In addition, a group of fruits was pathogen inoculated and treated with SDW. All treatments were submitted to incubation in humidity chamber for 72h at 25 ± 2 ºC.

Experimental design, disease assessment and statistical analysis
The experiments were conducted in a completely randomized design. The 24 experimental treatments, that were composed by a combination of six Trichoderma isolates with four types of Trichoderma application. All treatments had 10 replications of one fruit for disease severity evaluation and three replications of six fruits each, for disease incidence enumeration. The severity was measured for six days after fruit removal from humidity chamber, by determining injured considering the diameter (mm) of the lesion in two diametrically opposite directions, using the following formula: . Where: S = colony area; D1 = diameter1; D2 = diameter2. The data obtained in the experiments underwent hypothesis t Student test (P ≤ 0.05) and then analysis of variance, the average cluster made by the Scott-Knott test (P ≤ 0.05) [SISVAR 5.3 (FERREIRA, 2011)].

RESULTS AND DISCUSSION
When comparing the two experiments, there was no statistical difference by Student's t test (P ≤ 0.05), between them. Therefore, it was possible to use combined experiment data analysis. Once the pathogen was inoculated 1h or 24 h before the antagonists, it was observed that the isolated 225C1 (T. virens) induced a disease lesion area reduction, compared to the control without any treatment (Table 1). This area reduction on disease severity corresponds to around 49 to 61% of control (Table  2). In addition, it was observed that 225C1 was better than the fungicide treatment, in reduction of the diseased area when the pathogen was applied 24 h before Trichoderma (Tables 1 and 2). The most efficient treatment was the fungicide when the pathogen was applied 1 h before Trichoderma (Tables 1 and 2).
For treatments where the biocontrol agent was applied 1h and 24h before the pathogen, it was observed that isolate 4088 (T. longibrachiatum) caused a decrease on lesion area (Table 1). The isolate 4088 when applied 1h before the pathogen, showed a disease control of over 74% ( Table 2). The isolates 225C1, SF04 and THP also significantly reduced disease severity and increased disease control, compared to the untreated control (Tables 1 and 2). Treatment with isolated 4088 inoculated 24 h before the pathogen, was statistically as effective as the fungicide (Table 1), with a control of around 88% (Table 2) and statistically different from the others. Once the Trichoderma spp. require period of conidial humectation in order to germinate and penetrate into the plant pathogenic fungi (BATTA, 2004a;2004b). The beneficial action of T. virens for pre-treatment of cotton seedlings has been reported. It was demonstrated that the induction of plant defense system and the suppression of pathogen germination by antagonistic compounds produced by germinating cotton seedlings were the dominant biocontrol mechanisms (HOWELL; PUCKHABER, 2005).
Although Trichoderma is well reported as a biocontrol agent (BCA) against various species of Phytophthora (AMORIM; ITAMAR, 1999;COSTA et al., 2000;ALEXANDER;STEWART, 2001; 2) the specific effect on P. palmivora in papaya has not been studied and the results obtained so far are few significant (BUENO; DIANESE et al., 2012;TOCAFUNDO et al., 2010). The control effectiveness of the isolates 225C1 (T. virens), SF04 (T. asperellum) and 4088 (T. longibrachiatum), indicated that these isolates are important for management of the papaya fruit rot (P. palmivora).  Regarding the incidence of P. palmivora in papaya fruit, where the pathogen was inoculated 1h before the BCA and evaluated the 3d following the removal of humid chamber, the isolates 4088, 225C1 and THP induced a delay on disease ( Figure  1). Even as evaluated six days after removal of the humid chamber, the isolate 4088 and 225C1 obtained 50% disease control, statistical differing from the other isolates and check control.
In the treatment where the pathogen was inoculated 24 h before antagonists none of isolates has demonstrated the ability to control or slow the growth of P. palmivora on papaya. Therefore, 24 h of pathogen incubating in fruit was enough for its establishment and consequently reduce the effectiveness of the BCA. The BCA 4088 and 225C1 when inoculated 1 h before the pathogen and evaluated 6 d after incubation, resulted in a significant disease reduction. However, when BCA were applied 24 h before the pathogen resulted in 100% control. The 225C1 (T. longibrachiatum) and 4088 (T. virens) isolates reduced the incidence of disease to under 30% and improved the disease control to over 60%. (Figure 1). This implies that it is necessary an establishment period of the antagonist before having contact with the pathogen, to enable effective disease control. Rapid colonization of fruit wound by the antagonist is critical for decay control, and manipulations leading to improved colonization enhance biocontrol (MERCIER; WILSON, 1994). Thus, microbial antagonists should have the ability to grow more rapidly than the pathogen. Similarly, it should have the ability to survive even under conditions that are unfavorable to the pathogen (DROBY et al., 1992). (1) 4088 -Trichoderma longibrachiatum; 225C1 -T. virens; SF04 -T. asperellum; THP -T. harzianum; Fungicide -mancozeb+metalaxyl; Check control -Sterile distilled water Values followed by the same lowercase letter in each period (3 or 6) and the same capital letter in the same treatment, did not differ statistically (Scott-Knott Test, P≤0,05).
In addition, competition for nutrient and space between the pathogen and the antagonist is considered as the major modes of action by which microbial agents control pathogens causing postharvest decay JIJAKLI et al., 2001). To compete successfully with pathogen at the wound site, the microbial antagonist should be adapted to various environmental and nutritional conditions than the pathogen (BARKAI- GOLAN, 2001;EL-GHAOUTH et al., 2004). In addition, production of antibiotics (antibiosis), direct parasitism, and possibly induced resistance are other modes of action of the microbial antagonists by which they suppress the activity of postharvest pathogens on fruits and vegetables (JANISIEWICZ et al., 2000;BARKAI-GOLAN, 2001;EL-GHAOUTH et al., 2004). Tavares (2009) reported 50% control of papaya seedlings root rot (P. palmivora) using T. virens. In addition, Tocafundo et al. (2010) used T. asperellum (SF04) obtained similar value of control (~57%). The T. longibrachiatum (4088) isolate was successfully used by Oliveira et al. (2009) observing the inhibition of mycelial growth of Fusarium subglutinans f.sp. ananas, indicating how efficient this antagonist was. Adedeji et al. (2008) demonstrated a reduction of the incidence of cocoa black pod (P. megakarya) of 85% when treated with a strain of T. harzianum. Dianese et al. (2012) worked with Trichoderma spp. on papaya and found significant inhibition of P. palmivora by some isolates. Parasitism by Trichoderma spp. on P. palmivora is one common action mechanism (HARMAN, 2000;KUBICEK et al., 2001;BENÍTEZ et al., 2004). Thus, it is evident that the mechanisms of parasitism, antibiosis, and competition for substrates are the possible modes of action of Trichoderma against fungal plant pathogens (VERMA et al., 2007). It is postulated that Trichoderma initially succeed in antagonism via hyphal interactions, probable primal step in antagonism. Later, the BCA fungi kill the pathogenic fungi by means of toxins and consume them using a combination of lysozymes (WHIPPS; LUMSDEN, 2001).
The BCA 4088 could parasitize the pathogen as well inhibited its growth on fruit. It is known that of a BCA success as an antagonist depends of its parasitic capacity, and course of their specificity for the pathogen, requiring some characteristics which antagonists its ability to compete for nutrients, and, its production of pathogen toxic metabolites. The biological control is applied to reduce the amount of inoculum or activity of the pathogen. Biological control rarely eliminates the pathogen site, but reduce the seedlings population, and hence its ability to produce disease. The use of biological control agents, such as fungi of the genus Trichoderma, cannot be the only solution to control the papaya fruit rot, but based on these results, it is a viable tool that can be used in the management of this disease (SHARMA et al., 2009).
As conclusion, the data of this work showed that T. longibrachiatum (4088) was the best BCA of P. palmivora in postharvest, when it was applied 24h before the pathogen.

ACKNOWLEDGMENTS
The authors thank the National Council of Scientific and Technological Development (CNPq -Brazil) and to the Coordination for the Improvement of Higher Education Personnel (CAPES -Brazil). Also, the owner of the 'Alegria' Farm for providing the papayas used in this study and the Commission of the Cacao Agriculture Plan (CEPLAC) for technical support.