INTRODUCTION

Many studies have identified alternative con trol methods for plant parasitic nematodes. Biological control of plant parasitic nematodes is an important alternative due to the negative impact caused by the use of nematicides on the environ ment and on human health (^{Ferraz et al., 2010}).

One of the most studied biological control agents is the fungus Pochonia chlamydosporia Zare & Gams (sin. Verticillium chlamydosporium God-dard), which is a facultative parasite of eggs and females of both cyst and root-knot nema-todes (^{Kerry and Bourne, 2002}). This antagonist showed potential use as a biocontrol agent, both in greenhouse and field trials (^{Hernández and Hidalgo-Díaz, 2008}). Isolates of P. chlamydosporia var. chlamydosporia, Pc-3, Pc-10, and Pc-28, select ed in Brazil, when individually applied to the soil, were able to decrease the number of M. javanica eggs up to 80% (^{Lopes et al., 2007}; Dallemole-Gi-aretta et al., 2008; ^{Dallemole-Giaretta et al., 2012}). This fungus can also be used in combination with other management techniques for plant parasitic nematodes, or applied with another antagonist to maximize the control of nematodes (^{Siddiqui and Ehteshamul, 2000}; ^{Cannayane and Rajendran, 2001}). However, there are no reports on applica tions of two or more isolates of P. chlamydosporia to potentiate nematode control.

This fungus has desirable characteristics for a biocontrol agent, such as rhizosphere compe tition and colonization of roots of a wide range of plants. However, the colonization of the root system of plants differs according to the isolate of P. chlamydosporia (^{Bourne et al., 1994}). Otherwise, it is known that roots infected by nematodes are more heavily colonized by the fungus than the roots of healthy plants (Bourne et al., 1994; Bourne et al., 1996).

The objective of this study was to evaluate the mixing of isolates of P. chlamydosporia var. chlam-ydosporia for the control of Meloidogyne javanica (Treub) Chitwood in tomato plants.

MATERIALS AND METHODS

Three isolates of P. chlamydosporia var. chlamydosporia (Pc-3, Pc-10, and Pc-28) were used in this study. These fungal isolates had already presented potential in controlling M. javanica in previous studies, and they belong to the mycology collection of the Laboratory of Biological Control of Plant Parasitic Nematodes (BIONEMA) of the Federal University of Viçosa (UFV - Universidade Federal de Viçosa), Brazil.

The isolates of P. chlamydosporia were cultivat ed on sterilized rice grains at 26°C for 21 days in the dark. After 21 days, the number of chlamydo-spores per gram of substrate was determined ac cording to methodology described by ^{Kerry and Bourne (2002}). The chlamydospores production of isolates of P. chlamydosporia Pc-3, Pc-10, and Pc-28 were 1.75 x 105, 3.12 x 106 and 1.32 x 106 per gram of substrate, respectively.

The experiment was conducted in the green house of the UFV Department of Phytopathology (20.7603° S, 42.8678° W). The treatments evaluat ed were the fungal isolates Pc-3, Pc-10, and Pc-28 of P. chlamydosporia applied individually and combinations of isolates Pc-10 + Pc-28 + Pc-3; Pc-10 + Pc-28; Pc-28 + Pc-3, and Pc-10 + Pc-3. The fungus was not applied to the control treatments. The control without nematodes was used to con firm that there were neither nematodes nor fun gus in the soil. The containers were arranged in a complete randomized block design with eight replications per treatment.

Pots of 1,000 mL of capacity were filled with a methyl bromide-fumigated mixture of soil and sand (1:1). Soil of each pot was placed into a plas tic bag of 5,000 mL of capacity with 3 g of sub strate colonized with the respective fungal iso late, homogenized manually and replaced in the pot. Then, the soil was infested with 3,000 eggs of M. javanica per pot. The nematode inoculum was put in four holes of approximately 5 cm of depth, with spacing of 5 cm in the center of the pot.

A single 21-day-old seedling of 'Santa Clara' tomato was transplanted into each pot 15 days after inoculation with both organisms. The pots were maintained at 60% field capacity for mois ture and fertilized once a week. The average tem perature in the greenhouse was 25°C and ranged from 18°C to 32°C.

Evaluations of plant height, weight of fresh aboveground parts and root systems, and num bers of galls and eggs were evaluated 45 days af ter transplanting. Data were analyzed by analysis of variance (ANOVA) and means separated using Duncan's multiple range procedures, with 5% of probability.

RESULTS AND DISCUSSION

All of the treatments had significantly lower aboveground weights compared to the control treatment without fungus and nematode. It was found that the weight of aboveground parts of to mato plants cultivated in pots without nematodes was lower than those infested with M. javanica (Table 1). The application of fungal isolates Pc-3, Pc-10, and Pc-28 of P. chlamydosporia var. chlam-ydosporia to the soil, individually or combined did not reduce the number of galls of M. javanica when compared to the nematode alone treatment (Table 1). There was a significant difference in the number of M. javanica eggs per root when the fungal isolate Pc-10 was incorporated individual ly into the soil compared to the nematode alone (307.7 and 597.2 eggs per root, respectively).

Previous experiments (^{Lopes et al., 2007}; ^{Dallemole-Giaretta et al., 2012}) with the same fungal isolates used in this experiment demonstrated that they are able to control the root-knot nematode. However, the results obtained in this study show that isolates Pc-3 and Pc-28 did not significantly re duce the number of eggs compared to the control untreated with the fungus.

Although many studies have been conducted with P. chlamydosporia, complex interactions be tween biotic and abiotic factors may influence the antagonistic potential of the fungus. Therefore, there is a need to identify the factors that affect virulence and growth of saprophyte isolates of P. chlamydosporia (^{Esteves et al., 2009}).

One of the factors that may have affected the low effect of P. chlamydosporia isolates may be the quantity of inoculum used in this study. The quan tity of inoculum applied to the soil was 5.25 x 102, 9.36 x 103, and 3.96 x 103 chlamydospores per gram of soil, of P. chlamydosporia var. chlamydosporia, iso lates Pc-3, Pc-10, and Pc-28, respectively. In case of isolate Pc-3, the quantity of inoculum used was probably low. In general, the fungus is effective at 5 x 103 chlamydospores per gram of soil (^{Kerry, 2001}; ^{Dallemole-Giaretta et al., 2012}). However, the possible reduction of chlamydospore avail ability may have affected the establishment of isolates Pc-3 and Pc-28 in the soil. ^{Lopes at al. (2007}) demonstrated the potential of isolate Pc-28 as a biological control agent against M. javanica, apply ing 300 chlamydospores of this isolate per gram of soil. However, they also reported that isolate Pc-10 allowed the establishment and reproduction of the fungus on root systems when applied at a rate of 9.36 x 103 chlamydospores per gram of soil. The number of M. javanica eggs decreased by 48.47% (Table 1) when compared to the control treatment, confirming the potential of this isolate as a biocon-trol agent of M. javanica (Dallemole-Giaretta et al., 2008).

The combined application of P. chlamydosporia var. chlamydosporia isolates was expected to decrease M. javanica populations. Isolates used in this study were selected because of the biocontrol action demonstrated in previous studies. There fore, the mixture of antagonists must be compat ible in order to control phytopathogens (^{Akrami et al., 2009}; ^{Lucon et al., 2009}). An example of incompatibility of biological control agents was reported by Lucon et al. (2009). In this case, it was found that when applying five isolates of Trichoderma spp. to control Rhizoctonia solani only two combinations of isolates of the antagonist result ed in the control of R. solani.

In the present study, combinations of P. chlamydosporia isolates did not result in the control of M. javanica when compared to the application of the isolate Pc-10 individually. Further studies are required in the field to verify whether adverse conditions affect the synergism between isolates, or whether isolates of P. chlamydosporia tested will be incompatible.

Table 1 Effect of different combinations of Pochonia chlamydosporia isolates (Pc-3, Pc-10, Pc-28) on the growth of tomato plants and on the number of galls and eggs of Meloidogyne javanicaz. 

CONCLUSION

This experiment confirms the ability of P. chlamydosporia var. chlamydosporia isolates to re duce the number of M. javanica eggs. A range of variability was observed among the different isolates and their combinations. However, the combination of isolates was not able to increase nematode control. Studies about compatibility of isolates should help addressing the complex interaction among isolates, and host and patho gens.