VIRULENCE OF ENTOMOPATHOGENIC NEMATODES ( RHABDITIDA : STEINERNEMATIDAE AND HETERORHABDITIDAE ) FOR THE CONTROL OF Diabrotica speciosa GERMAR ( COLEOPTERA : CHRYSOMELIDAE )

Entomopathogenic nematodes (EPNs) are used in biological control of soil insects and show promise in the control of D. speciosa. The objective of this work was to evaluate the potential of native and exotic entomopathogenic nematode isolates in the control of D. speciosa under laboratory and greenhouse conditions. Results showed that all of EPNs caused larval mortality. The most virulent were Heterorhabditis sp. RSC01 (94%), Steinernema glaseri (84%), Heterorhabditis sp. JPM04 (82%) and Heterorhabditis amazonensis RSC05 (78%). There was no effect of the Heterorhabditis sp. RSC01 and S. glaseri isolates on eggs. The maximum mortality of D. speciosa larvae by Heterorhabditis sp. RSC01 was observed at a concentration of 300 IJ/ insect, while by S. glaseri observed the highest mortality at the concentration of 200 IJ/ insect. The Heterorhabditis sp. RSC01 isolate caused over 80% pupal mortality at a concentration of 250 IJ/insect. The virulence of Heterorhabditis sp. RSC01 and S. glaseri was affected by temperature. The Heterorhabditis sp. RSC01 isolate caused reduction in larva survival under greenhouse conditions at all of the tested concentrations and there was no difference in mortality among different concentrations of infectid juveniles.


INTRODUCTION
The chrysomelid Diabrotica speciosa Germar (Coleoptera: Chrysomelidae) has been causing great concern to farmers, due to its wide occurrence and polyphagous habit.The adults feed preferentially on leaves, shoots, fruits and pollen of cultivated and wild plants, while the larvae prefer the roots.In early attacks on maize, the larvae of D. speciosa can bore into the caulicle of the seedlings, causing drying and death of central leaves.In more developed plants however, they prefer to feed on adventitious roots (GASSEN, 1989).There are few works looking at the use of biological control agents to suppress populations of D. speciosa at this stage; so control has is done exclusively through the use of chemical products.
Entomopathogenic nematodes (EPNs) of the Steinernematidae and Heterorhabditidae families (Rhabditida) are promising biological control agents for the control of soil pests (GREWAL et al., 2001).Studies using EPNs have been carried out in some countries with other species of the genus Diabrotica and have presented good results under both laboratory and field conditons (JACKSON;BROOKS, 1989;WRIGHT et al., 1993;JOURNEY;OSTLIE, 2000;KURTZ et al., 2007).
The objective of this work was to evaluate the virulence of native and exotic entomopathogenic nematode isolates against of D. speciosa under laboratory and greenhouse conditions.

MATERIAL AND METHODS
The research was conducted at the Insect Pathology Laboratory and greenhouse of the Entomology Department (END) of the Federal University of Lavras (UFLA), located in the county of Lavras, MG, from December, 2007 to December, 2008.

Rearing of Diabrotica speciosa and multiplication of entomopathogenic nematodes
For the maintenance of adults and the obtaining of eggs and larvae of D. speciosa, the modified methodology described by Ávila and Milanez (2004) was used.Adults of D. speciosa were collected from bean crops in an experimental field of UFLA and from the Palmital Farm, in the municipal district of Ijaci, MG.
Adults were kept in cylindrical glass jars 17cm high and 10cm in diameter, with a perforated plastic cover, in an acclimatized chamber (temperature 25±1º C, humidity 70±10% and 14 hour photophase).
The jars were lined with a disk of paper towel to avoid excess humidity.Bean leaves were placed within the jars to feed the insects, as well as a 5 cm diameter Petri dish containing a piece of black colored moistened gauze, to provide a surface for oviposition.On the surface of the gauze seminal rootlets of corn were placed to stimulate the oviposition of the females.
The eggs were removed from the oviposition substrate, by washing the gauze in running water over a fine fabric (voile), where they were retained.To avoid the contamination by fungi during the incubation period, the eggs were treated with a 1% copper sulfate (CuSO4) solution for two minutes and, soon afterwards, transferred to 9 cm diameter Petri dishes lined with humid filter paper, which were maintained in an acclimatized chamber (25±1º C, 70±10% humidity and 14 hour photophase) until the larvae hatched.
The larvae were maintained in plastic pots containing vermiculite and corn seedlings according to the methodology described by Ávila et al. (2000) until the beginning of the experiments.
Multiplication of EPNs was conducted using five larvae of Galleria mellonella Linnaeus (Lepidoptera: Pyralidae) raised in the laboratory, according to the methodology described by Dutky et al. (1964) and fed with a modified artificial diet (C.Dolinski, personal communication, March, 2008).The nematodes were multiplied according to the methodology described by Molina and López (2001).

Virulence test of entomopathogenic nematodes to third instar larvae of Diabrotica speciosa
The virulence against D. speciosa larvae of the seventeen isolates of EPNs, belonging to the genera Heterorhabditis and Steinernema was evaluated.The bioassay consisted of five repetitions, each made up of a 9 cm Petri dish containing 15 g of vermiculite, 1.5 g of corn roots and 9 mL of water, to which were transferred ten third instar D. speciosa larvae.
The experimental design was entirely random, with eighteen treatments (17 EPNs isolates and the control).1 mL of aqueous suspension of EPNs was added at a concentration of 1500 infective juveniles (IJ)/mL, corresponding to 150 IJ/insect, while 1 mL of distilled water was added to the control.The Petri dishes were maintained under controlled conditions at 25±1º C, 70±10% humidity and with a 14 hour photophase.
The evaluations were conducted three days after the application of the nematodes, by counting the number of dead larvae.The symptomatology of the dead larvae was verified by the confirmation of the causa mortis, according to the typical characteristics of death caused by nematodes.The larvae killed by nematodes of the genus Heterorhabditis presented a red color, while for Steinernema spp.they were of a whitish color.
The mortality data was submitted to the variance analysis and compared, by the Scott-Knott test of averages, to 5% probability using the SISVAR (FERREIRA, 2002) statistical program.

Test with eggs
The bioassay consisted of five repetitions, each made up of a 5 cm diameter Petri dish lined with a sheet of filter paper, to which ten five-day-old eggs of D. speciosa were transferred.
The experimental design was fully randomized.An aqueous suspension of 0.3 mL of the EPN isolates Heterorhabditis sp.RSC01 and S. glaseri were added to the Petri dish, at concentrations of 25, 50, 100, 150, 200 and 250 IJ/ insect.0.3 mL of distilled water was added to the control and every other day, 0.25 mL of distilled water was added to each of the Petri dishes, to maintain humidity levels.The Petri dishes were maintained under controlled conditions at a temperature of 25±1º C, 70±10% humidity and with a 14 hour photophase.Daily evaluations were made by counting the larvae that had emerged from the eggs in each repetition.
The data on egg mortality, obtained at different concentrations were submitted to the regression analysis at 5% of probability, using the SISVAR statistical program, with the mortality of the insect as the dependent variable and the concentrations of the nematode as the independent variable.

Test with third instar larvae
The test consisted of five replicates, each made up of a 9 cm diameter Petri dish, containing 15 g of vermiculite, 1.5 g of corn roots and 9 mL of distilled water, to which ten third instar D. speciosa larvae were transferred.The experimental design was fully randomised. 1 mL of aqueous suspension of EPN Heterorhabditis sp.isolate RSC01 and S. glaseri were added at the concentrations of 50, 100, 150, 200, 250 and 300 IJ per insect and, 1mL of distilled water was added to the control.The Petri dishes were maintained under controlled conditions at 25±1º C, 70±10% humidity and with a 14 hour photophase.The evaluations were made three days after the start of the experiment, by counting the dead larvae in each repetition.

Test with pupae
The bioassay consisted of four repetitions, each made up of a 9 cm diameter Petri dish containing 15 g of vermiculite and 9 mL of distilled water, to which eight D. speciosa pupae were transferred.The experimental design was fully randomized. 1 mL of aqueous suspension of EPNs/ Petri dish was added at the concentrations of 100, 150, 200, 250 and 300 IJ/insect of the Heterorhabditis sp.RSC01 isolates.1mL of distilled water/Petri dish was added to the control.The Petri dishes were maintained under controlled conditions at a temperature of 25±1º C, 70±10% humidity and with a 14 hour photophase.The evaluations were conducted three days after the start of the experiment, by counting the pupae killed by nematodes in each repetition.The pupae mortality caused by nematodes was confirmed by observing the symptomatology, since the pupae killed by Heterorhabditis sp.RSC01 presented a red coloration.

Effect of temperature on the mortality of Diabrotica speciosa larvae by entomopathogenic nematodes
A bioassay was conducted to study the effect of temperature on the infectivity of IJ, using the Heterorhabditis sp.RSC01 and S. glaseri nematodes.The bioassays consisted of five repetitions/ treatment, each made up of a 9 cm diameter Petri dish containing 15 g of sterilized vermiculite, 1.5g of corn roots and 9 mL of distilled water, to which eight third instar D. speciosa larvae were transferred.The treatments were run at the constant temperatures of 15, 20, 22, 25 and 28º C, in five acclimatized chambers, in which the Petri dishes remained for three days, with 70±10% RH and a 12 hour photophase. 1 mL of standardized aqueous suspension was added at the concentration of 200 IJ/insect, which was the concentration that caused the highest mortality of D. speciosa larvae, according to tests conducted under laboratory conditions.The experimental design was entirely random.
Evaluations were made three days after the start of the experiments, by counting the larvae killed EPNs in each repetition.
Mortality data (larvae and pupae) at different tested concentrations of the infective juveniles or different temperatures were compared using generalized linear models (GLM) with logit link function and binary distribution, using the Statistical Analysis System (SAS) (SAS INSTITUTE, 2003).

Susceptibility of D. speciosa larvae to entomopathogenic nematodes under greenhouse conditions
The bioassay consisted of six repetitions, each made up of a plastic vase with 2.5 liter capacity, containing sterilized soil and sterilized vermiculite medium (1/10) and 500 mL of distilled water.Eight corn seeds were sowed in each vase, which were thinned after seven days, leaving only five plants per vase.Twenty third instar D. speciosa larvae were transferred to each vase.
The larvae were placed at a depth of five centimeters below the soil level, close to the plant roots.Twenty-four hours after the larvae were transferred, 50 mL of EPNs aqueous suspension were added, at the concentrations of 6,500, 13,000, 26,000 and 52,000 IJ of Heterorhabditis sp.RSC01, per repetition (vase).The vases corresponding to the control received 50 mL of distilled water.The concentration of 6,500 IJ/vase was established according to laboratory test results that determined the maxim lethal concentration of Heterorhabditis sp.RSC01 on third instar D. speciosa larvae.The experimental design was entirely random.
The average temperature registered during the experiment was 27.4º C and the average relative humidity 62%.The evaluations were conducted five days after the start of the experiment, by counting the live larvae per each repetition.The survival data was submitted to the variance analysis and the averages were compared by the Tukey test at 5% probability, using the SISVAR statistical program.

Virulence test
All of the tested nematodes caused mortality to third instar D. speciosa larvae, differing from the control, in which mortality was not observed.However, the virulence varied among the different isolates (Table 1).
These results are similar to those observed by Toepfer et al. ( 2005) who, while working with the isolate S. glaseri NC (USA origin), saw that this was pathogenic to third instar larvae of Diabrotica virgifera virgifera LeConte, 1868 (Coleoptera: Chrysomelidae) under laboratory conditions.Converse and Grewal (1998), seeking to select EPNs for the control of Cyclocephala hirta LeConte, 1861 (Coleoptera: Scarabaeidae), verified that isolate NJ65 of S. glaseri was the most virulent among the 22 isolates tested, causing 76.5% larval mortality under laboratory conditions, after three days of exposure to IJ.
It may be noticed that the most virulent nematodes are from Brazil (Table 1).In addition, native isolates have better chances of adaptation to the Brazilian environmental conditions, as well as to the local insect fauna (DOLINSKI; MOINO Jr., 2006).
In this research, several native isolates were used and they showed potential for the control of D. speciosa larvae.These should also be investigated regarding their specificity, to commercially exploit only those that present a high selectivity to the natural enemies (DOLINSKI; MOINO JR., 2006).

Test with eggs
There was no effect of the nematodes Heterorhabditis sp.RSC01 and S. glaseri on the mortality of D. speciosa eggs at any of the tested concentrations.
These results are similar to those related by Jackson and Brooks (1995) who, working with D. v. virgifera, did not verify an effect of the nematode S. carpocapsae on the mortality of eggs of this insect, under laboratory conditions.Journey and Ostlie (2000), testing S. carpocapsae for the control of D. v. virgifera in the field, also verified that this isolate did not present an egg viability reduction.
According to Jackson and Brooks (1995), the absence of infectivity of EPNs in crisomelid beetle eggs can be related to the impermeability of the eggs of those organisms.Toepfer and Kuhlmann (2004) related the natural occurrence of native EPNs in D. v. virgifera eggs in Central Europe, however, the identification of the isolates was not conducted.

Tests with third instar larvae
The mortality of larvae caused by Heterorhabditis sp.RSC01 varied between 64 and 96% (Table 2), at the concentrations of 50 and 300 IJ/insect and between 34 and 90%, at the concentrations of 50 and 200 IJ/insect, respectively, when applying S. glaseri.

Tests with pupae
The Heterorhabditis sp.RSC01 isolate was pathogenic to pupae of D. speciosa, causing mortality at all of the studied concentrations (Table 3).Concentrations of 200, 250 and 300 IJ/ insect did not differ among themselves and caused high mortality of D. speciosa pupae.The maximum mortality of D. speciosa larvae by Heterorhabditis sp.RSC01 was observed at a concentration of 300 IJ/ insect, corresponding to 47.16 IJ/ cm 2 .In the bioassay of S. glaseri the highest mortality was observed at the concentration of 200 IJ/ insect (31.44/ cm 2 ).
Toepfer et al. ( 2005), testing the virulence of S. glaseri on third instar larvae of D. v. virgifera at different concentrations, verified that there was an increase in the mortality of larvae with the increase of the concentration, observing a mortality superior to 77% at the concentration of 15.9 IJ/ cm 2 .These results are similar to those observed by Kurtz et al. (2009) who reported that when testing H. bacteriophora, H. megidis and S. feltiae at a concentration of 16 IJ/ cm 2 in trays with sand, they observed the maximum mortality of the third instar larvae (87.8%) in the treatment with H. bacteriophora.Yang et al. ( 2003), working with S. feltiae for the control Luperomorpha suturalis Chen, 1938 (Coleoptera: Chrysomelidae), verified pupae mortality of 97.1 and 83%, under laboratory conditions, at a temperatures of 25 and 15º C, respectively.
The effect of the nematodes on the D. speciosa pupae can increase the long term action of these entomopathogens in the reduction of field populations of this insect, interfering in population growth and, consequently, avoiding re-infestations by the pest.Furthermore, the reduction of the adult population in the field can contribute to the reduction of the damage that these insects cause to the aerial part of the plants.

Effect of Temperature on the Mortality of Larvae of D. speciosa by Entomopathogenic Nematodes
At 20º C we observed a low mortality of larvae in both treatments (S. glaseri and Heterorhabditis sp.RSC01) (Table 4).To Heterorhabditis sp.RSC01 at the temperatures of 25 and 28° C we verified high mortality of larvae, and no mortality was observed at 15° C.
The bioassay of S. glaseri, at temperatures of 22 and 25° C caused high mortality of larvae, whereas at 28° C there was a reduction in mortality, suggesting that this nematode may be more sensitive to high temperatures.
These results are in accordance with those observed by Boivin and Belair (1989) who, working with S. feltiae for the control of Listronatus oregonensis LeConte, 1857 (Coleoptera: Curculionidae), verified a decrease in TL 50 with the increase of the temperature.feltiae nematodes against four species of corn pests, among them crisomelid beetle larvae, under laboratory and greenhouse conditions, these authors verified that both tested species presented an effect on the mortality of all of the pest species, resulting in lower damage incidence in the corn plants in the parcels that received the EPN treatments in relation to the control.Temperature is the main factor that interferes in the mobility of the nematodes and, consequently, in the expense of their nutritional reserves and in the survival of IJ (MOLYNEUX, 1985).The reduction in the mortality of D. speciosa larvae at the lowest temperatures, for both isolates tested, is a factor that should be taken into account when employing this entomopathogen in the field, keeping in mind its wide territorial distribution and the different climatic conditions in the areas in which this insect is found.In colder environments, insect mortality over a short period of time could be affected.

Susceptibility of D. speciosa Larvae to Entomopathogenic Nematodes under Greenhouse conditions
The Heterorhabditis sp.RSC01 nematode reduced the survival of D. speciosa larvae in the soil at all of the tested concentrations, when compared to the control, without, however, differing among themselves (Figure 1).
The non-verification of mortality differences of D. speciosa larvae among the four tested concentrations demonstrates evidence that the concentration of 6,500 IJ was sufficient to cause the maximum mortality in the greenhouse, the use of higher concentrations being unnecessary under this condition.
These results are similar to those observed by Riga et al. (2001).Testing the efficiency of the S. glaseri and S. The concentration of Heterorhabditis sp.RSC01 that caused the maximum mortality of D. speciosa larvae under laboratory and greenhouse conditions (31.41 IJ/ cm 2 , equivalent to 3.14 x 10 9 IJ/ha) is similar to the concentration used by Yang et al. (2003) that studied S. feltiae for the control of L. suturalis in the field and verified 77.8% larval reduction at the concentration of 30 IJ/cm 2 , thirty-eight days after the treatment and 92.4% larval reduction a hundred days after the treatment.Wright et al. (1993), working with S. carpocapsae applied in the field for the control of D. v. virgifera, under center pivot irrigation in the corn culture, observed reduction in the damage rate of the plants.They also noticed a reduction in adult emergence, with nematode application at concentrations of 1.2 and 2.5 x 10 9 IJ/ha, lower that mentioned in this research.Likewise, Thurston and Yule (1990), testing the nematodes S. feltiae All and S. bibionis Sn against D. barberi larvae in the field, in corn culture, observed that both species of EPNs tested reduced the number of larvae in the soil compared to the control, at concentrations of 1.3 x 10 8 Survival of larval (%) and 1.3 x 10 9 IJ/ha, not verifying a difference among the species or the tested concentrations.Toepfer et al. (2008), working with EPNs for the control of D. v. virgifera larvae under field conditions in the corn culture, verified that all of the species of nematodes (H.bacteriophora, S. feltiae and H. megidis) tested caused reduction of the insect population in the soil.However, the most virulent species was H. bacteriophora, that caused an 81% reduction.The same authors also verified that the tested species reduced the percentage of damage to the plants, proving the potential of EPNs for the control of crisomelid beetle larvae in corn culture.Kurtz et al. (2007) also demonstrated that the H. bacteriophora, H. megidis and S. feltiae species present a capacity to establish themselves in corn for a period of up to five months, being promising for the control of D. v. virgifera in this culture, with massive applications of those EPNs in the field.

CONCLUSION
The high potential that EPNs demonstrate in laboratory, greenhouse and field studies in the control of crisomelid beetle larvae, as well as the results obtained in this study from laboratory and greenhouse tests, confirms the great potential of Heterorhabditis sp.RSC01 for the control of D. speciosa larvae.Their use can be viable in field studies seeking the control of this pest in such cultures as irrigated corn and, in particular potato as it is cultivated in smaller areas, requiring high moisture levels for its development, a characteristic that favors the survival and behavior of EPNs.

Figure 1 -
Figure 1 -Average percentage survival of Diabrotica speciosa larvae at different concentrations of infective juveniles of Heterorhabditis sp.RSC01 per pot, under greenhouse conditions.
1 Means followed by the same letter do not differ at 5% probability when compared to generalized linear models.

Table 3 -
Percentage of confirmed mean mortality of pupae of Diabrotica speciosa at different concentrations of infective juveniles of Heterorhabditis sp.RSC01 isolate (25 ± 1º C, RH 70 ± 10% and a photophase of 14 hours).
1 Means followed by the same letter do not differ at 5% probability when compared to generalized linear models.

Table 4 -
Percentage of confirmed mean mortality of larvae of Diabrotica speciosa by Heterorhabditis sp.RSC01 and Steinernema glaseri isolates at different temperatures (RH 70 ± 10% and of 14 hour photophase).Means followed by the same letter do not differ at 5% probability when compared to generalized linear models. 1