Preliminary screening of the larvicidal eff ect of Brevibacillus laterosporus strains against the blowfl y Chrysomya megacephala ( Fabricius , 1794 ) ( Diptera : Calliphoridae )

Introduction: This study evaluated whether different strains of Brevibacillus laterosporus could be used to control larvae of the blowfl y Chrysomya megacephala, a pest that affects both human and animal health. Methods: Mortality rates were recorded after 1-mL suspensions of sporulated cells of 14 different strains of B. laterosporus were added to 2.5g of premixed diet consisting of rotting ground beef fed to fi rst instar larvae of C. megacephala. All bioassays were performed using 10 larvae per strain, with a minimum of three replicates for each bioassay. Larval mortality was recorded daily up to seven days. Results: Strains Bon 707, IGM 16-92, and Shi 3 showed the highest toxicity toward the larvae producing 70.5%, 64.5%, and 51.6% of larval mortality, respectively, which was signifi cantly higher than that in the control group (p < 0.05). In contrast, strains NRS 1642, NRS 661, NRS 590 BL 856, NRS 342, ATCC 6457, Bon 712, and NRS 1247 showed limited or no pathogenic activity against the target larvae. Conclusions: Our preliminary data indicated that B. laterosporus could be used to develop bioinsecticides against C. megacephala.

Brevibacillus laterosporus is known to be a potential biocontrol agent against several orders of invertebrate species, including Diptera, Coleoptera, Lepidoptera, Nematodes, and mollusks (1) (2) (3) ; further, various formulations of this species as a bioinsecticide have been patented (4) (5) .In addition to its established entomopathogenic properties, B. laterosporus has remarkable potential in the biotechnology industry (3) .For example, some strains are known to produce antibiotics such as laterosporamine (6) that act against both gram-positive and gram-negative bacteria and chitinases that have antifungal and insecticidal activities (7) .However, unlike the entomopathogenic strains of Bacillus thuringiensis (8) , the factors responsible for the toxicity of the different strains of B. laterosporus have not yet been determined.
The blow fl y Chrysomya megacephala (Fabricius, 1794) (Diptera: Calliphoridae) is a pest that has signifi cant impacts on human and animal health.It is native to Australian and Oriental regions; it was subsequently introduced into Africa and South America and then rapidly dispersed to the United States of America (9) .This species was fi rst recorded in Brazil in 1976; it was detected in the garbage from a metropolitan region of São Paulo.The presence of these fl ies has been associated with the risk of transmission of fecal pathogens and other disease-causing agents (10) ; however, at present, there are no programs to control this species in Brazil.
In addition to the vectorial capacity of adults, larvae of C. megacephala can cause rare but serious cases of myiasis in domestic animals and humans (11) .In Brazil, there have been reports of human myiasis caused by C. megacephala (12) ; however, studies investigating the direct impact of large populations of this fl y species on public health are lacking.Ecological studies have shown that this species is most frequently observed in open rubbish dumps in both São Paulo (13) and Rio de Janeiro, Brazil (data not published).In Thailand, where this species is considered problematic, bacterial loads carried by C. megacephala were found to be greater than those recorded in Musca domestica Linnaeus, 1758 (Diptera: Muscidae) (14) , a species widely regarded as a pest of public health

METHODS
signifi cance.Furthermore, C. megacephala was shown to be able to carry multi-drug-resistant bacteria, including Escherichia coli and Klebsiella pneumoniae, suggesting its involvement in the potential contribution to the dissemination of antibiotic resistance genes (15) (16) .
Approaches for the control of populations of synanthropic Diptera species, including C. megacephala, included improvements in basic sanitation and application of chemical pesticides such as permethrin and deltamethrin (17) , plant extracts (eucalyptol) (18) , plant metabolites (neolignans) (19) , microhymenopteran parasitoids (20) , insect growth regulators (precocenes) (21) and a range of Bacillus thuringiensis strains (22) .Chemical control has been shown to be generally effective (18) , but constant application might lead to the development of resistance and produce environmental contamination.Owing to the public health threat posed by C. megacephala, complementary control methods that can be used in urban settings need to be developed.Therefore, this study evaluated the potential toxicity of the strains of B. laterosporus against laboratory-reared larvae of C. megacephala as a preliminary step toward the development of an alternative/complementary control strategy.

Collection of fl ies and colony maintenance
Adults of C. megacephala were collected from open dumps in the Amorim community (latitude: -22.875707, longitude: -43.250606),Rio de Janeiro, Brazil, and transported to the laboratory; they were identifi ed using a calliphorid dichotomous key (23) .Flies were reared in net cages maintained in an acclimatized chamber at 27 ± 1°C with 70 ± 10% relative humidity (RH) and a 12-h photoperiod; they were fed a solution of 30% saccharose ad libitum.The fi rst instar larvae of F1 (fi rst) generation were placed in Petri dishes and fed 10g ground beef.The larvae were reared up to adult emergence, and the colony was maintained until the fourth generation, when the bioassays were performed.

Bacterial strains and spore suspensions
A panel of 14 strains of B. laterosporus was tested; these strains were selected based on their toxicity against different targets (2) .All strains were obtained from the Coleção de Culturas do Gênero Bacillus e Gêneros Correlatos (CCGB).Stock cultures were maintained at -20°C in brain heart infusion broth supplemented with glycerol (20% v/v).Working cultures were prepared on nutrient agar (NA) slants that were incubated at 30°C for 72h or until complete sporulation occurred, as assessed by light microscopy.Sporulated cultures were collected from the solid medium by scrapping the entire slope with an inoculating loop and transferred to tubes containing 9mL of autoclaved distilled water.Suspensions were heated at 80°C for 15 min to inactivate the vegetative cells and retain the spores.Serial, tenfold dilutions of spore suspensions were prepared and used to inoculate NA plates for retrospective determination of the colony forming units (CFU•mL -1 ) and the doses that needed to be used in the bioassays.The strategy used in this preliminary screening was to analyze the maximum number of spores to reduce the possibility of not detecting larvicidal activity of a given strain.

Feeding of larvae with experimental diet
All bioassays were conducted using 10 fi rst instar larvae per strain; the larvae were fed 2.5g of premixed diet, consisting of rotting ground beef, along with 1mL of each bacterial suspension.Three replicates were used for each bioassay.The larval rearing diet, which was mixed with 1mL of water, served as the negative control in the bioassays.Assays were conducted in 50-mL plastic pots placed within larger 250-mL pots containing vermiculite for pupation of the larvae and were covered with a mesh; the pots were placed in an acclimatized chamber at 27 ± 1°C with RH of 70 ± 10% and photoperiod of 12h, as reported previously (19) .Pots were examined daily from on the third day of the start of the experiment, and those larvae that did not abandon the diet, or which failed to pupate, were recorded as dead.Surviving larvae were transferred to test tubes (17 × 160mm) containing vermiculite and sealed with hydrophobic cotton wool.Larval mortality was recorded daily until the end of the experiment on day 7. Since the present study aimed to conduct a preliminary screening to identify B. laterosporus strains that could be used for the biological control of C. megacephala, LC 50 values were not calculated.

Statistical analysis
Mortality values in the treated group were corrected using the formula reported by Abbott (24) as follows: Cm (%) = (%Tm -%Com)/(100 -%Com) × 100, where Cm is the corrected mortality, Tm is the mortality in the treated group, and Com is the mortality in the control group.The corrected numbers of dead larvae were subjected to a binomial test of two independent samples by using Bioestat 5.0 program.The signifi cance level was set at 0.05.When the value of the corrected mortality of the treated group was less than that of the control group, no statistical tests were used.
Although standardized growth conditions were used, the number of spores per gram of diet varied between 4.8 × 10 6 and 2.72 × 10 8 .The results of the larval bioassays in decreasing order of mortality are shown in Table 1.The strains Bon 707, IGM 16-92, and Shi 3 showed the highest larvicidal activity (70.5%, 64.5%, and 51.6% corrected mortality, respectively), with significant differences relative to the control group (p < 0.05).Interestingly, the Bon 707 isolate exhibited the highest toxicity, although an intermediate concentration of spores (1.46 × 10 7 ) per gram of diet was used.The mortality caused by strains Shi2 and NRS 1648 was also signifi cantly higher than that of the control group (p < 0.05), but was less than 50%.In contrast, strains NRS 1642, NRS 661, NRS 590, BL 856, NRS 342, ATCC 6457, Bon 712, and NRS 1247 showed limited or no pathogenic activity against the larvae (Table 1).Chrysomya megacephala densities are known to be high in urban areas that have fairs and street markets selling fresh products (9) (10) (13) and open dumpsters, which could serve as their breeding grounds.This situation is very concerning in Brazil owing to poor sanitary conditions in some urban areas.Hence, developing suitable control programs for C. megacephala is necessary in these areas.The drastic measures implemented for the eradication of the myiasis-causing fl y species Cochliomyia hominivorax (Coquerel, 1858) in North America are extremely costly and might have caused negative ecological impacts.Therefore, they are not considered viable or practical for implementation in Brazil (25) .Thus, studies need to focus on the development of integrated pest management strategies, including biological control methods, against this pest.

DISCUSSION
In this study, 14 different strains of B. laterosporus were screened for their potential to be used in the biological control of C. megacephala.In the bioassays, fi rst instar larvae were used since this life stage has been shown to be highly vulnerable to potential biological agents owing to the fact that fi rst instar larvae ingest greater amounts of treated diet over a long time (3) .
As reported previously (2) , the toxicity of the panel of test strains used in this study varied remarkably, indicating the heterogeneity of the pathogenic mechanisms of these strains against the same target species.Only fi ve tested strains induced signifi cant levels of mortality compared to that of the control group.In a previous study (2) , the three strains that exhibited the greatest larvicidal properties in the present study showed negligible or medium toxicity against the larvae of Aedes aegypti (Linnaeus, 1762) (Diptera, Culicidae), Culex quinquefasciatus Say (Diptera, Culicidae), Anticarsia gemmatalis Hübner, 1818 (Lepidoptera: Noctuidae), and Anthonomus grandis Boheman, 1843 (Coleoptera: Curculionidae).Interestingly, similar fi ndings were obtained in a study (26) , in which a strain of B. laterosporus showed elevated toxicity toward Musca domestica.In particular, the bacterial strain was not toxic toward Muscidifurax raptor: Girault & Sanders (Hymenoptera: Pteromalidae), a natural parasitoid of M. domestica (26) .With respect to biocontrol, the narrow spectrum of action of the larvicidal strains identifi ed in this study is both relevant and desirable since their application might not affect non-target fauna (27) .
To our knowledge, this is the fi rst study reporting the use of B. laterosporus for the biological control of Chrysomya spp.Although the entomopathogenic activity of Bacillus thuringiensis against the larvae of Chrysomya spp.has been investigated (22) (28) , the elevated mortality levels in those studies were thought to be associated with the germination of the spores of strain LFB-FIOCRUZ 907 in the hemolymph of both C. megacephala and Chrysomya putoria larvae.
The toxicity of B. laterosporus strains has been speculated to be associated with the activation of spore-associated proteins, or the secretion of binary toxins produced by vegetative cells, or because of the insecticidal protein crystals contained in the parasporal bodies (3) .Strains IGM 16-92 and Shi 3 are known to produce protein crystals; however, the presence of crystals should not be considered to be the only factor linked to toxicity, since strains Bon 707, NRS 1648, and Shi 2 did not produce protein crystals but were effective against C. megacephala larvae.Therefore, further studies are warranted to elucidate the reasons of mortality caused by B. laterosporus in blowfl ies.
In our bioassays, an initial spore suspension was fed to the larvae; however, we did not evaluate whether the spores germinated within the diet and produced vegetative cells, or determined the possible contribution of such cells, if produced, to the observed levels of larval mortality.Further, we did not determine whether the ingested spores germinated within the larvae or if the toxicity was a result of the direct interaction between the spores and the larval gut epithelium.As discussed above, the mechanisms of toxicity of B. laterosporus are complex and, in some cases, multi-factorial.Such analyses would have been impractical and expensive considering that this was a preliminary study.The above-mentioned factors need to be considered in future studies performed using fewer strains with confi rmed toxicity toward C. megacephala.
Previous studies have confi rmed the insecticidal potential of B. laterosporus strains against insects belonging to different orders (2) , including the fl y species M. domestica (29) and Lucilia cuprina (Wiedemann, 1830) (Diptera: Calliphoridae) (30) .This study examined the toxicity of B. laterosporus against other invertebrate species.The mode of action of the protein crystals of B. laterosporus was shown to be comparable to that of B. thuringiensis strains; these crystals caused the deterioration of the larval gut epithelium (31) .Similar to other invertebrate targets, the larval mortality rates caused by B. laterosporus treatment were variable, but signifi cant, depending on the specifi c strain of the bacterium used.In particular, this study identifi ed the strains of B. laterosporus that could be used for the control of C. megacephala.However, further studies are warranted to evaluate whether this bacterium can be used as a component of the alternative/complementary control strategy for C. megacephala.

TABLE 1 -Toxicity of the 14 strains of Brevibacillus laterosporus against the larvae of Chrysomya megacephala (Diptera: Calliphoridae).
a Corrected mortality determined using Abbott's formula(Abbott, 1925); b P value was obtained after binomial test of two independent samples by using Bioestat 5.0 program.*All the strains are deposited in CCGBC.CCGBC: Coleção de Culturas do Gênero Bacillus e Gêneros correlatos.