POTENTIAL OF ENTOMOPATHOGENIC NEMATODES FOR CONTROL OF THE ERVA-MATE PEST Hedypathes betulinus (KLUG, 1825) (COLEOPTERA: CERAMBYCIDAE)

The intensive exploitation of “erva mate” (Ilex paraguariensis St. Hil.) (Aquifoliaceae) has favored the emergence of pests, including Hedypathes betulinus. We evaluated 18 isolates of entomopathogenic nematodes Steinernema and Heterorhabditis genera for control of H. betulinus adults, and tested nematode in-vivo replication capacity in Galleria mellonella L. (Lepidoptera: Pyralidae) larvae. We also evaluated the effect of adjuvants for foliar application, and the efficiency of nematode isolates selected in erva mate seedlings in the greenhouse via application to soil and plant shoots. The PI and CB40 isolates showed the hgh virulence (92.5% insect mortality by both). The PI isolate showed low productivity in G. mellonella larvae, thus only CB40 was used in subsequent tests. Only emulsified vegetable oil adjuvant was compatible with nematodes. Soil application of nematodes proved ineffective; however, shoot application on erva mate seedlings showed significant mortality (82.5%) after exposure to the CB40 isolate without adjuvant.


INTRODUCTION
Erva mate Ilex paraguariensis Saint Hil.(Aqüifoliaceae) plants are native to southern and midwestern Brazil, Paraguay, and Argentina.Changing in erva mate to monoculture production has increased the productivity, and in Brazil erva mate annual production is estimated approximately in 252,700 tons, concentrated in the states of Brazilian south region (Paraná, Santa Catarina and Rio Grande do Sul states) (IBGE, 2013).Monoculture production in Brazil also favors the occurrence of pests, especially Hedypathes betulinus (Klug, 1825) (Coleoptera: Cerambycidae), named "broca da erva-mate" (PENTEADO et al., 2000;BORGES et al., 2003).Adults live in the shoot, and mate and lay eggs near the base of the plant, on the branches and in bark crevices (CASSANELLO, 1993;GALILEO et al., 1993;D'AVILA et al., 2006).Resident larvae are responsible for the damage, by building galleries on branches and longitudinally along the trunk, preventing sap movement and weakening the plant.In the case of severe infestations or successive pest generations on the plant, death may occur (SOARES, 1998;GUEDES et al., 2000).
Entomopathogenic fungi have been found in natural epizootics association with H. betulinus (SOARES et al., 1995;SOARES and IEDE, 1997;LEITE et al., 2000;LEITE et al., 2006).Also, many studies were carried out with isolates selection in the laboratory and in field evaluation (SOARES;IEDE, 1997MILK et al., 2000BORGES, 2007;GOMM, 2010, FANTI;ALVES, 2013).These studies resulted in a product based on the fungus Beauveria bassiana (Bovemax ® EC) that has since been registered for commercial use in integrated pest management (AGROFIT, 2015).Entomopathogenic nematodes (EPNs) Mermithidae have also been found in association with the pest in the field.However, due to some biological characteristics of this nematode family, there is little potential for their use in controlling H. betulinus.Alves et al. (2009a) in a pioneer study demonstrated the susceptibility of H. betulinus to the nematodes genus Steinernema, in which pests reached 80% mortality within 10 days after treatment.The biology of the pest includes the female behavior of approaching the base of the plant (close to the ground) to oviposit, in which case nematodes present in soil may be control of the pest.Among the advantages of EPNs is their ability to search for hosts, and their symbioses with bacteria, which effectively kills insects.Moreover, in addition to killing the host quickly these agents are not harmful plants or vertebrates, are selective for insects, do not pollute soil or groundwater, are safe for the producers, and can readily establish the ground to maintain long-lasting, small populations to fight pests (FERRAZ, 1998;VOSS et al., 2009).
A successful biological control program depends on the existence of efficient control agents both under laboratory and in the field conditions, and in the case of H. betulinus, such products are rare, highlighting the need to search for new alternatives.This study aimed to evaluate and select isolates of EPNs to control H. betulinus, specifically those showing virulence and high productivity, as well as efficiency in semi-field conditions after soil and leaf application.

Collection of nematodes and insects:
H. betulinus adults were obtained from a commercial erva mate plantation located in Ivaí, Paraná, Brazil.The insects were collected and kept in cages (60 cm × 40 cm wide × 40 cm long) containing erva mate branches and under controlled conditions (26 ± 1°C, 70% relative humidity, 14h photophase) until experiments began, and not exceeding seven days (ALVES, 2009a).Prior to beginning experiments intact insects were selected with movement and normal responses to stimuli (SOARES, 1998).We evaluated 18 isolates of Steinernema and Heterorhabditis nematodes from different research institutions in Brazil, reared in G. mellonella larvae as follows.

Selection of nematode isolates:
We uses 40 plastic 500 ml containers divided into four replications, each containing erva mate branches for food and closed with perforated plastic covers and lined with 150g of moist sand (Alves et al., 2009a).
Nematode suspensions were applied to dry sand (100 Infective Juveniles (IJs)/cm 2 ) (surface area).After application of the suspension, sterile distilled water it was added to obtain 20% substrate moisture.The adults insects were then released onto the branches (one insect/container).For controls, the sand was treated with distilled water (20% moisture).
The containers were closed and maintained at 26 ± 1°C with 14h photoperiod.Measurements were made daily for 15 days, and dead insects were transferred into a dry chamber composed of Petri dishes lined with dry filter paper where they remained for five days.After this time, 2 ml of distilled water was applied to the paper surface, and after 24h the cadavers were dissected and observed under stereomicroscope to search for nematodes.Isolates causing at least 80% mortality were selected for the next step.

Estimation of CL99 of isolate CB40 and in vivo production in Galleria mellonella larvae:
Nematode suspensions were inoculated into Petri dishes (9 cm dia.) lined with two sheets of filter paper at concentrations of 0, 500, 1000, 1500, 2000 and 2500 IJs/plate; 10 G. mellonella larvae were placed on each dish.Dished were incubated at 26 ± 1ºC, with 70 ± 10% RH and a 14 h photophase.Control dishes were treated only with distilled water.After 72 hours post-treatment, dead larvae were transferred into a dry chamber and maintained for five days to confirm nematode infection as described above.Each plate was considered one replicate, with a total of five replicates carried out per nematode concentration.The results were analyzed by linear regression using Excel 2007.After that, in vivo production it was evaluated.Therefore, 2 ml of a nematode suspension were inoculated onto Petri dishes lined by filter paper (630 IJs/plate).Ten previously weighed (lab reared) G. mellonella larvae were placed on the dish.Dishes were covered and sealed with plastic film, and incubated at 24 ± 1°C in the dark for 72 hours.After confirmation of death, cadavers were transferred into a dry chamber and maintained for five days at 24 ± 1°C after which nematode infection was confirmed (WOODRING; KAYA, 1988).After five days, the larvae were transferred to White traps for emergence of the IJs, which were collected and quantified daily until production ended.The most productive strain was selected for the next step.

Effect of adjuvants with CB40 isolates:
We evaluated the effect of foliar fertilizer TEK-F ® and Aureo ® vegetable oil emulsifiable on the CB40 isolate for future application under field conditions.We used the IOBC protocol adapted by Negrisoli et al. (2008), in which a solution of the product is prepared with distilled water at double the dose recommended by the manufacturers.One mL of solution is then transferred to a glass tube, followed by 1 mL of the CB40 nematode suspension (2000 IJs/mL), yielding the product concentration indicated by the manufacturer.For controls, 1 mL of nematode suspension was added to 1 mL of distilled water.Five replicate tubes were prepared for each treatment.The tubes were kept at 22 ± 1 °C with 14 h photoperiod for 48h, after which the viability and infectivity of the isolate was evaluated.Data were subjected to ANOVA (the experiment having a completely randomized design) and means were compared by Tukey test (p <0.05) using the statistical program Sisvar (FERREIRA, 2011).

Greenhouse experiment with erva-mate seedlings:
The CB40 isolate was applied to the aerial parts of the plants (foliar application) and to the soil, as follows.

Foliar application:
CB40 isolate was multiplied in vivo (see in vivo methods above) and stored in aqueous suspension, maintained with aeration to prevent sedimentation and death of IJs.Erva-mate seedlings (~50 cm height) were kept in pots and individually placed in wooden cages (55 × 40 × 40 cm) lined with nylon mesh.Ten adult insects were released in each cage, and the nematode suspension was applied throughout the shoot with manual spray to the point of run-off (approximately 14 mL of suspension/plant containing 5.0 × 10 -3 IJs/mL).Treatments consisted of CB40 nematode suspension in distilled water, CB40 + Aureo ® , nematode suspension, and Aureo ® solution (both treatments with Aureo ® following manufacturer recommendations).Control seedlings were treated with distilled water.Treatments were replicated four times.Five days after application, dead insects were counted and transferred to a dry chamber for confirmation of mortality, as described above.Data were subjected to analysis of variance (F test) and means compared by Tukey test (p <0.05) according to a completely randomized experimental design using the statistical program Sisvar (FERREIRA, 2011).

Soil application:
Application to soil followed the same experimental procedures and statistical analysis described above, except that nematodes were applied directly to soil at a concentration of 100 IJs/cm 2 using a 10 mL pipette, and five H. betulinus adults were released onto the shoot.After five days, dead insects were counted and placed in a dry chamber for 5 days, then dissected to confirm mortality.

Selection of nematode isolates:
There was strong variation in results, with only two out of 18 isolates causing mortality above 90% (CB40 and PI, both belonging to the genus Heterorhabditis).The CH3 isolated, SA, NEPET 11 and CB02 caused between 55 and 75% mortality, while the others caused less than 50% mortality (Table 1).
It's known both among and within-species differences in insect susceptibility to nematode genus ( VAN TOL et al., 2004;VAN TOL;RAUPP, 2006;ALVES et al, 2009b).For example the CB40 and PI isolates both belonging to the genus Heterorhabditis, yet it is also impossible to affirm that this genus best suited to H. betulinus since other isolates in the genus showed very little activity.
Variation in cerambycid susceptibility to different nematode species and isolates has been previously reported.Machado et al. (2003) by selected EPNs for control of Migdolus fryanus (Westwood, 1863) (Coleoptera: Vesperidae) larvae and observed high susceptibility to genus Heterorhabditis, with up to 100% mortality.On the other hand, for Plectrodera scalator (Fabr., 1792) (Coleoptera: Cerambycidae) larvae with only S. carpocapsae All and S. feltiae SN isolates it was obtained mortality higher than 50%, and high concentrations were required to cause significant insect mortality (FALLON et al., 2006).Harvey et al. (2012) evaluated the potential of S. carpocapsae and H. downesi for control of Rhagium bifasciatum Fabr., 1775 (Coleoptera: Cerambycidae) and found that both nematodes were capable of infecting larvae, pupae and adults both in laboratory and field conditions.Table 1.Mortality in Hedypathes betulinus adults caused by entomopathogenic nematodes under laboratory conditions.Tabela 1. Mortalidade de adultos de Hedypathes betulinus causada por nematoides entomopatogênicos em condições de laboratório.These data reinforce the need to carefully select entomopathogenic nematode isolates, because biological features and adaptations to the environment can cause great variation among species, and even among isolates of the same species (GAUGLER et al., 1997).Our results demonstrate this variability, and the importance of conducting tests in order to choose the most appropriate isolate for insect control.So, CB40 and PI isolates were selected to continue the research.

Estimation of CL99 of isolate CB40 and in vivo production in Galleria mellonella larvae:
The PI isolate showed low productivity, requiring a large number of hosts to obtain a small number of individuals, and thus, it was discarded at this stage of the study.The CL 99 of isolate CB40 in G. mellonella larvae was estimated at 9 IJs/cm 2 , which is considered low and indicative of high virulence (Figure 1).Barbosa (2005) show an estimated CL 99 of 43 IJs/cm 2 for H. bacteriophora in G. mellonella larvae.G. mellonella has high susceptibility to most known entomopathogenic nematodes, and it can also be assumed that smaller nematodes can penetrate hosts with greater ease, which may have been a factor in the virulence of isolate CB40.Further, CB40 Heterorhabditis are able to penetrate not only natural openings of hosts, but also the cuticle (GIOMETTI, 2011); this is another feature that may be associated with virulence of this isolate.
About in vivo production, CB40 IJs emerged from G. mellonella larvae for up to 10 days, with peak emergence between the third and fourth days.The total accumulated production in the host was 6.5 × 10 4 IJs/larva or 4.2 × 10 4 IJs/g larva (Figure 2).
The results obtained are lower than previously reported in studies with different isolates in G. mellonella, including Barbosa (2005), which report 4.9 × 10 5 IJs/g larva for H. bacteriophora.
Saenz and Lopes (2011) also evaluated the production of Heterorhabditis sp.isolate SL0708 and obtained between 1.5 and 2.8 × 10 5 IJs/larva.According to the authors, the differences in the production of IJs in G. mellonella are due to various factors involved in the infection process, including the host penetration site (mouth, anus, spiracles or cuticle), inoculum concentration, the number of IJs infecting the host, and the local release of symbiotic bacteria in the hemolymph, all of which can interfere with J1 development within the host.The CL 99 of our CB40 isolate is quite low by comparison, which may have been a factor affecting productivity.This indicates the need to evaluate other inoculum concentrations for improved productivity.

Effect of formulation adjuvants isolated on CB40:
The TEK-F ® product significantly reduced the viability and infectivity (by 84 and 82%, respectively) and is classified as incompatible with CB40 according to the IOBC protocol.The Aureo ® product vegetable oil base did not affect the viability or infectivity of nematodes and results did not differ from the controls, highlighting the possibility of joint use of this product and the nematode (Table 2).

Greenhouse experimentsoil application:
The isolate showed promising results in the laboratory, causing over 90% mortality in the insects, and presented ease of in vivo production in G. mellonella larvae, however it did not show activity against H.