Molecular characterization and virulence of Beauveria bassiana and Metarhizium anisopliae against Galleria mellonella ( Lepidoptera : Pyralidae ) and Tenebrio molitor ( Coleoptera : Tenebrionidae ) larvae

1 Plant Health and Applied Biotechnology Program Instituto de Ciência Biológica e da Saúde Federal Rural University of Rio de Janeiro BR 465 Km 07 ZIP. 23890-000, Seropédica RJ, Brazil. 2 Health, Food Safety and Environmental in Agribusiness Instituto Biológico ZIP. 04014-002, São Paulo, SP, Brazil. 3 Phytopathology and Biochemistry Laboratory, Instituto Biológico, São Paulo, SP, Brazil. 4 Embrapa Agrobiologia BR 465, Km 07 ZIP. 23890-000, Seropedica RJ, Brazil. 5 Mycology Laboratory Research Center in Mycology and Mycotoxicology Veterinary Institute Federal Rural University of Rio Janeiro BR 465 Km 07 ZIP. 23890-000, Seropédica RJ, Brazil.


INTRODUCTION
It is well known that synthetic insecticides have deleterious effects on the worldwide environment and consequently, attention has addressed to biocontrol agents as suitable alternatives (Safavi et al., 2007).Entomopathogenic fungi are natural common enemies of arthropods, particularly insects, and they might be used in the management of pest populations in agroforestry and urban ecosystems (Lacey and Kaya, 2007).The most important species, Beauveria bassiana (Vuill.)Balsam and Metarhizium anisopliae (Metch) Sorok.are associated with the Phylum Ascomycota, Order Hypocreales, Families Clavicipitaceae and Cordycipitaceae, respectively (Hibbett et al., 2007).Insect pathogenic fungi have to meet several host challenges producing enough new infectious spores in each generation maintaining viable populations.First, successful transmission often requires the release of massive spore numbers and/or sticky spore surfaces or substances maximizing adhesion in other ways (Vega et al., 2012).Second, spores should germinate and initiate penetration of the solid insect exoskeleton relatively quickly (Cole and Hoch, 1991).Third, the fungal cells must proliferate inside the hemocell, muscles, or other host body tissues collapsing the host immune system in order to the host dies shortly after (Vega et al., 2012).Fourth, the fungal pathogen should manage the host cadaver optimizing spore production and dispersal under prevailing environmental conditions (Roy et al., 2006).
Entomopathogenic fungi act by contact and are able to infect different live stages of insects with piercing-sucking mouthparts.They present horizontal and vertical transmissions, viz., inside the population (among individuals) and over time (across the progenies), respectively.Despite these advantages, a number of factors, such as the difficulty for environmental adaptability, resistance to chemicals, and lower readyeffect compared to the chemicals may limit the use of these fungi for the pest management in greenhouse, and especially, in field (Shah and Pell, 2003;Quesada Moraga et al., 2006a;Wraight et al., 2007).Nevertheless, a wide variability within and among species has been detected in pathogenicity, virulence and ecological features in many entomopathogenic fungi.This variability has been considered and analyzed for selecting candidates for myco-insecticides production (Quesada Moraga et al., 2006a;Wraight et al., 2007).
This study aimed at characterizing and evaluating the pathogenicity of B. bassiana and M. anisopliae strains isolated from Atlantic Forest soil, Rio de Janeiro, Brazil, against two insect hosts models: Galleria mellonella and Tenebrio molitor, also comparing the effect of entomopathogenic fungi against Lepidoptera and Coleoptera orders.

Source of insects and fungal isolates
Four B. bassiana (Bb27MI, Bb11MI, Bb79MI and Bb53MI) strains and four M. anisopliae (Ma58MI, Ma10MI, Ma11MI and Ma12MI) from Atlantic forest soils in the Rio de Janeiro state, Brazil, were Mora et al. 663 used in the experiment.The fungi were isolated according to the methodology described by Esparza Mora et al. (2016).The strains were maintained at 4°C on potato dextrose agar (PDA).Last instar larvae of G. mellonella and T. molitor were obtained from the Biological Institute in Campinas -SP.G. mellonella was fed by an artificial diet containing 472 g wax pure bee, 96 g skimmed milk powder, 188 g powdered yeast, 385 g maize flour, 160 g soya, 416 g of glycerin, 300 ml water distilled, while T. molitor was reared on bread.

Molecular identification of strains
After cultivation, DNA from the strains on PDA media for 7 days at 25 ± 1°C, DNA was extracted according to the CTAB method (Doyle and Doyle, 1987).The polymerase chain reaction (PCR) for amplification of fragment Internal Transcribed spacers (ITS) region was performed by a pair of universal primers ITS1 (5'-TCGGTAGGTGAACCTGCGG-3') and ITS4 (5'-TCCTCCGCTTATTGATATGC-3') (White et al., 1990), generating estimated size of 500 bp product.The amplification reaction conditions consisted of 2 min at 94°C followed by 40 cycles of 15 s at 94°C, 30 s at 56°C and 2 min at 72°C with a final extension of 4 min at 72°C.The products were visualized under UV light in 1% agarose gels stained by ethidium bromide.The PCR products were purified by precipitation with polyethylene glycol according to the protocol described by Schmitz and Riesner (2006).The sequencing was performed by chain termination method with 3.1 BigDye (Applied Biosystems) reagent and ABI3500 automatic sequencer (Applied Biosystems).After sequencing, phylogenetic tree with similar sequences of the ITS1 and ITS4 strains regions was completed and sequences of different fungal species were obtained from the GenBank/NCBI, using the Maximum Likelihood method (ML), evaluating the strength of the topology, the tree was assessed by the Jukes-Cantor method with 1000 bootstrap replicates, and phylogenetic analysis was carried out in the Molecular Evolutionary Genetics analysis-MEGA 6.0.

Preparation of inoculum
Each strain was cultivated in Petri dishes containing sterile parboiled rice, maintained at 25 ± 1°C for 8 days.The conidia were collected with by sterile loop and suspended in sterile distilled water containing 0.1% Tween 80, and then agitated in vortex for 3 min.Four M. anisopliae and B. bassiana conidia concentrations: 1×10 9 , 1×10 8 , 1×10 7 and 1×10 6 conidia.ml - were prepared.The treatment control received only sterile distilled water with 0.1% Tween 80.The conidial counts in each suspension with an improved Neubauer Hemacytometer (Marienfeld, Germany) were determined.

Virulence of conidia
Ten (10) G. mellonella and T. molitor larvae were immersed in 10 ml of fungal suspension for 30 s.Thereafter, the treated insects were placed on filter paper (Whatman No. 1) inside a plastic box (18 × 8.5 × 4.5 cm) with ventilation holes (2 cm 2 ) on three sides.Food was introduced into the box and the ventilation holes were covered with a metal screen.The set-up was placed inside a climatic chamber at 25 ± 1°C.Three replicates per treatment with 30 species of each insect per repetition were accomplished.*Corresponding author.E-mail: aleesparza18@hotmail.com.
Author(s) agree that this article remains permanently open access under the terms of the Creative Commons Attribution License 4.0 International License Figure 1.Phylogenetic tree constructed by ML method, Jukes-Cantor showing the relation between the strains of entomopathogenic fungi obtained from Atlantic Forest, and other related species deposited in the GenBank-NCBI.The sequences used are of the ITS regions of rDNA.Built tree in MEGA 6.0 program.The "bootstrap" values for 1,000 repetitions are indicated next to the tree branches.The Puroureocillium lilacinum ITS sequences was designated as the outgroup for rooting the tree.
Mortality was daily recorded for 10 days after inoculation.The fungi were re-isolated from dead insect samples selected randomly from each replicate box.

Molecular characterization
The ITS1 -5.8 -ITS4 rDNA sequences were compared with those available in the GenBank data base for M. anisopliae (accession numbers HM055427.1)and B. bassiana (KC121560.1).The amplification of the ITS region resulted in a single product for all isolates.The size of the product was about 500 bp.
The alignments and phylogenetic analysis confirmed the taxonomic identity of the strains used in our study.M. anisopliae Ma10MI, Ma12MI, Ma58MI and Ma11MI strains grouped from the same species strain sequence from the database, at 100% similarity.Among the B. bassiana species, it was observed that the strains Bb11MI, Bb79MI, Bb53MI and Bb27MI composed a grouping with B. bassiana strain from GenBank, at 100% similarity (Figure 1).

DISCUSSION
The development of PCR amplification from different rDNA regions has greatly facilitated the fungi taxonomic studies.Alignments and phylogenetic analyses confirmed the B. bassiana and M. anisopliae strains taxonomic identity.Analysis of ITS-rDNA sequences have been applied to determine the genetic diversity of M. anisopliae and B. bassiana (Wall et al., 2005;Entz et al., 2005;Becerra et al., 2007;Carneiro et al., 2008;Freed et al., 2011).Thus, Bautista-Galvez et al. ( 2012), made the genetic characterization of M. anisopliae strains obtaining fragments of 600 to 800 bp by PCR amplification from the ITS1-ITS4 rDNA regions.The size of the DNA fragments from M. anisopliae strains collected in sugarcane crops were similar to those ones reported for the M. anisopliae (L0909) reference strain.Comparison of ITS rRNA obtained from fungal isolates in regarding to the GenBank database (NCBI) indicated that the MM0801, AD0702, AD0803, CD0804, FC0805, FC0706, AS0807, GB0808, L0909, BC0710, SF0811, M370 and M374 strains corresponding to M. anisopliae and the JC0816 and AR0814 corresponding to B. bassiana.Mergulhão et al. (2014) used the ITS rDNA region, as well as assessed the genetic variability and phylogeny from five B. bassiana strains (API 145,API 148,API 223,API 225 and API 226).The ITS region was amplified by the ITS1 and ITS4 primers, wherein amplicon was observed with an average size of 800 bp which was sequenced.The results showed that all strains, except for IPA145, presented themselves as monophyletic group.The IPA145 showed a genetic identity with B. bassiana and Cordyceps bassiana 99%.The IPA148 was the most genetically distant in regarding to the other ones, at less than 40% similarity.G. mellonella and T. molitor are known to be susceptible to B. bassiana and M. anisopliae, for this reason they have been used as baits for isolation of entomopathogenic fungi from the soil (Zimmerman, 1986;Bidochka et al., 2002;Montesinos et al., 2011) or, as a probe to pathogenicity of other insect species (Bharadwaj et al., 2011).Mortality and concentrations were considered parameters in the study of the best strains behavior once indicating the pathogen colonization capacity, surpassing all competitor agents present in the insect (Neves, 1998).It was observed that it requires less inoculums amount by Bb79MI B. bassiana and Ma58MI M. anisopliae treatments to kill the G. mellonella and T. molitor larvae.Similar results by Garcia et al. (2011) were obtained, evaluating the insecticidal activity from 8 B. bassiana strains and from 4 M. anisopliae on Spodoptera frugiperda and Epilachna varivestis larvae at six concentrations (10 4 to 10 9 ); the BB18 B. bassiana strain was more virulent for E. varivestis larvae with a 93.3% mortality, LC 50 = 1.20×10 6 conidia.ml - and LT 50 = 5.1 days.Bb42 B. bassiana strain presented the highest mortality on S. frugiperda larvae (96.6%, LC 50 = 5.92×10 3 conidia.ml - and LT 50 = 3.6 days).Differences among lethal times is a tool widely used in selecting strains, because it is interesting that the fungus quickly eliminate its host, as well (Lohmeyer and Miller, 2006).A significant interaction was also observed between the conidial concentration and the time.The highest conidia concentrations require less time causing 100% insects mortality.The amount of conidia used should to attain a certain concentration and thus, achieving an efficacious penetration of the fungus on the insect cuticle and causing host death (Zhioua et al., 1997).All strains were able to cause infection and mortality against G. mellonella and T. molitor by contact.In the T. molitor larvae the shortest time to cause the maximum mortality was 6 days at 1×10 9 conidia.ml - concentration with the Ma12MI, Ma10MI and Ma58MI M. anisopliae strains, while for the Bb79MI B. bassiana strain the time was 7 days.On G. mellonella the shortest time to cause 100% mortality was 6 days with the Bb79MI B. bassiana and Ma58MI M. anisopliae strains.At the 1×10 9 conidia.ml - concentration, 100% mortality for the T. molitor larvae with the Bb11MI, Bb53MI, Bb72MI B. bassiana and Ma11MI of M. anisopliae strains was not obtained, as well as for the G. mellonella larvae with Bb11MI and Bb27MI B. bassiana and Ma12MI and Ma11MI M. anisopliae strains.These results are similar to the ones obtained by Khalid et al. (2012), evaluating the virulence of 90 B. bassiana and 15 M. anisopliae strains on G. mellonella larvae using 10 2 , 10 3 , 10 4 , 10 5 and 10 6 conidia.ml - concentration.The BbaAUMC3076, BbaAUMC3263 and ManAUMC3085 strains caused 100% mortality at concentrations of 5.5×10 6 , 5.86×10 5 and 4.8×10 6 conidia.ml - .The LC 50 values were 1.43×10 3 , 1.04×10 5 and 5.06×10 4 conidia.ml - Bba3263AUMC, Bba3076AUMC and Man3085AUMC, respectively.Khosravi et al. (2015), evaluated pathogenicity from B. bassiana strains against Arge rosae larvae.The bioassay was performed by the immersion method at 2×10 4 , 2×10 5 , 2×10 6 , 2×10 7 and 2×10 8 conidia.ml - concentrations; IRAN403C strain presented the highest mortality being 70% at 2×10 7 conidia.ml - concentration, LC 50 and LT 50 obtained was 5.54×10 5 conidia.ml - and 3.92 days at 2×10 8 conidia.ml - concentration, respectively.However, the findings of this study demonstrated direct proportional relationship between the amount of conidia applied on the insects and mortality from the same ones, so the more diluted suspensions the virulence and pathogenicity were less pronounced, resulting in lower mortality for a longer time than the observed one at higher concentrations.This fact was also observed by other authors studying M. anisopliae and B. bassiana strains on different insect species (Filho et al., 2002;Silva et al., 2003;Cunha et al., 2008).According to Neves and Alves (2000), as more conidia penetrating, more toxins or enzymes are released, increasing the insect mortality.However, the fungus action speed depends, besides the concentration, of the host species involved (Sosa-Gomez and Moscardi, 1992).According Paccola-Meirelles and Azevedo (1990), St. Leger (1991) and Kleespies and Zimmermann (1998), variation in virulence of entomopathogenic strains is a result of differences in the enzymes and toxins production in conidia germination speed, mechanical activity in the cuticle penetration, colonization capacity and cuticle chemical composition.
It might be concluded that B. bassiana and M. anisopliae fungi present different capacity cause mortality of the insects, with the Bb53MI and Bb79MI B. bassiana strains as the most pathogenic for G. mellonella, as well as the Bb79MI B. bassiana strain was the most pathogenic for T. molitor.Ma58MI and Ma10MI M. anisopliae strains presented the highest virulence for G. mellonella and T. molitor, and also presenting the lowest LC 50 and LT 50 values.
Institute in Campinas -SP for supplying G. mellonella and T. molitor larvae.
different letters, small letters on columns and capital letters on lines differ by Scott Knott test at 5%.Mortality average at 8 days.
different letters, small letters on columns and capital letters on lines differ by Scott Knott test at 5%.Mortality average at 8 days.

Table 1 .
Mortality G. mellonella larvae at different conidia concentrations of B. bassiana and M. anisopliae.

Table 2 .
Mortality T. molitor larvae at different conidia concentrations of B. bassiana and M. anisopliae.

Table 3 .
LC 50 and LT 50 values of B. bassiana and M. anisopliae strains against G. mellonella and T. molitor.

Table 4 .
CL 90 and LT 90 values of B. bassiana and M. anisopliae strains against G. mellonella and T. molitor.