1 Introduction

The regulation of mosquitoes by the use of pathogens has long being proposed as a sustainable alternative to the classical chemical controls (Tetreau et al., 2013). While classical chemical control is the method of choice, its indiscriminate use has resulted in development of resistance in mosquitoes and several health and environmental problems. Thus, non-hazardous biocontrol methods would be the best choice to tackle mosquitoes. Several bacterial, fungal and protozoan pathogens are effective biocontrol agents against mosquitoes (Copping and Menn, 2001; Soni and Prakash, 2012; Chandra et al., 2016; Chatterjee et al., 2007) but they have not been exploited significantly probably due to absence of assured source of the products. Nevertheless, limited research has been done to identify and exploit the natural microbial pathogens of mosquitoes in different habitats and regions of the country. Research has been done to control mosquito population with the help of novel botanicals (Ray et al., 2014; Singha and Chandra, 2011; Bhattacharya and Chandra, 2014; Haldar et al., 2011; Panneerselvam et al., 2013). In eastern India, rice is grown mainly under flooded conditions where several mosquitoes breed and it is the exclusive habitat of Cx. vishuni group, vectors of Japanese encephalitis. Luxananil et al., (2001) reported the presence of the bacteria like Bacillus sp. from the guts of mosquito larvae Ae. aegypti and Cx. quinquefasciatus, collected from various natural breeding habitats. Therefore, information on the natural microbial pathogens of mosquitoes of the rice fields would be very useful for vector control (Teng et al., 2005) especially the potent indigenous pathogens which would be best suited to local conditions and could be augmented for the purpose. Bacteria, especially Bacillus thuringiensis (Bt) and B. sphaericus (Bs), are the most potent and successful pathogens for effective control of vectors of diseases (Cooping and Menn, 2011; Luxananil et al., 2001; Tangsongcharoen et al., 2011; Subramaniam et al., 2012). Both of the pathogens produce inclusion bodies i.e. toxins, although susceptibility differs with mosquito species. The larvicidal factor of B. sphaericus is unique which consists of two proteins of 51 and 42 kDa (Tangsongcharoen et al., 2011; Park et al., 2010). Both of which are required for toxicity i.e. potent organisms contain the two protein components. The present study has been designed to isolate and characterize the pathogenic B. sphaericus from dead Cx. quinquefasciatus larvae, to check the crystal protein component besides testing a particular formulation of the bacteria for its larvicidal potential against three mosquito species (Culex quinquefasciatus Say, 1823, Anopheles subpictus Grassi, 1899, Armegeres subalbatus Coquiliett, 1898), which are important vectors and nuisance pests (Hati et al., 1989; Chatterjee and Chandra, 2000; Rudra et al., 2013), both under laboratory and natural field conditions. The results are expected to evaluate the virulence of this strain of the bacteria and provide a basis for its further examination and exploitation as larvicide.

2 Materials and Methods

Dead larvae of Cx. quinquefasciatus were collected from the submerged rice fields of Burdwan University Agricultural Farm, Burdwan, West Bengal, India. Each larva was washed with 70% alcohol followed by sterile distilled water (three times) and the gut contents were inoculated on nutrient agar (NA) plates (g/l: peptone 5, beef extract 3, agar 3, and pH 7). The petriplates were incubated at 31 ± 0.1^°C in the BOD incubator for 24h, the colonies were checked under a phase-contrast microscope and those having spherical spores and crystals were picked up, purified by dilution plating on NA plates and maintained at 4±0.1^oC on NA slants. Morphological, physiological and biochemical characters of the bacteria were studied following standard methods (Collee and Miles, 1989; Lacey, 1997; Pelczar et al., 1957). Antimicrobial tests were done with standard antibiotics discs as follows.

2.1 Antibiogram by disc diffusion method

Antibiotic susceptibility of isolate was tested by Kirby Bauer disc diffusion method (Bauer et al., 1996) following the criteria of the National Committee for Clinical Laboratory Standards (National Committee for Clinical Laboratory Standards, 1993) with an inoculum of approximately 10⁶ cfu/ml spread on Müller Hinton Agar plates. Commonly used antibiotic discs (Hi-media) were placed on the Müller Hinton Agar plates and incubated at specific temperature of 37°C. After incubation, the zone diameters of inhibition (ZDI) around each antibiotic disc were measured to the nearest millimeter.

2.2 Sample preparation for scanning electron microscopy (SEM)

At first freshly died Cx. quinquefasciatus larvae were collected from the test solution of bacterial formulation. Then the larval gut was dissected out using needle and forceps under binocular. The dissected gut was fixed in 2.5 % gluteraldehyde for 2 hrs. It was then subjected to dehydration through upgraded alcohols: 50% alcolhol (5 mins), 70% alcohol (30 mins-with two changes), 90% alcohol (30 mins- in two changes) and absolute alcohol (30 mins-with two changes). After completion of the dehydration procedure the sample was then kept serially into 3 different ratios of mixtures of absolute alcohol: amyl acetate with 3 changes viz. 3:1 (30 mins), 2:2 (30 mins), 1: 3 (30 mins). After that the gut was kept in amyl acetate for 30 mins. Then the sample was subjected to SEM to observe the effect of bacterial infection on the inner gut wall.

2.3 Preparation of crystal protein from Bacillus sphaericus

Crystal protein was isolated (Maniatis et al., 1982), trypsinized and analyzed by sodium dodecyl sulphate-polyacrylamide gel electophoresis (SDS-PAGE) (Janssen, 1994). Proteins were observed as blue bands through a visible light transilluminator and photographed through the gel video photodocumentation system. Data were analyzed through relevant software.

2.4 Larvicidal bioassay

To determine the toxicity of B. sphaericus against different mosquito species namely, An. subpictus, Ar. subalbatus and Cx. quinquefasciatus, the bio-assay tests were carried out at 30±2^oC; using 100 larvae (late third instar) kept in 1000 ml stored water in glass bowls. The larvae were exposed to different doses (5ml/L, 10ml/L, 15ml/L) of B. sphaericus suspension (6.8x10⁶ bacteria/ml). Each test was replicated three times at three different times along with a control and the mortality (%) was determined following Abbott’s formula (Abott, 1925): (% mortality in the experiment) – (% mortality in control) Mortality (%) = x 100, 100 - (% mortality in control). The data obtained on the laboratory trials were subjected to probit analysis to obtain the median dose. The regression equations of dose mortality values were compared with respect to times of exposure and the mosquito species to note the differences in the mortality rate with time and the mosquito species (Zar, 1999).

2.5 Field evaluation of B. sphaericus suspension

To evaluate the mortality of mosquitoes in the field conditions, B. sphaericus was added (at the dose rate of 100 ml/m²) in selected shallow ponds (20-35 m x 15-25 m x 15-20 m) containing An. subpictus and drains (3-4 m x 10-15 m x 4-5 m) containing Ar. subalbatus and Cx. quinquefasciatus, respectively. The study area was located in Burdwan, West Bengal, India. The range of water temperature, pH and dissolved oxygen were 26-31⁰C, 6.34-6.61 and 5.28-6.47 mg/L respectively during the experiment in late summer of 2009. Similar untreated habitats in the adjacent areas were monitored as control. A 250 ml dipper was used to estimate larval densities by taking 25 dips in each breeding habitat. Larval density of An. subpictus, and Cx. quinquefasciatus was determined before and after treatment in control and treated habitats at 24 h intervals upto 10 days.

3 Results

The bacteria formed circular, white, flat and undulate colonies (Table 1). The bacteria were Gram positive and rod shaped having spherical spores and amorphous spherical crystals measuring 2-4 x 1 μm. They did not produce acid and gas from different carbon sources (Table 1). The organisms were positive for catalase, urease, protease and oxidase but negative for H₂S production, indole production, nitrate reduction, Vogues-Proskauer test and lipase (Table 1). The organisms were sensitive to recommended doses of chloramphenicol, kanamycin, eryrthromycin, lomefloxacin, tobramycin, gatifloxacin, amikacin, gentamycin, sparfloxacin, levofloxacin amoxycillin but resistant to penicillin G, tetracycline, ampicillin, norfloxacin, nalidixic acid and cefuroxime (Table 1). The isolate was identified as B.sphaericus on the basis of morphological, physiological and biochemical characters. Analysis of SDS-PAGE profile of crystal protein revealed that the bacteria contained 42 kDa protein which is mosquitocidal (Tangsongcharoen et al., 2011; Park et al., 2010; Silva-Filha et al., 1999; Srisucharitpanit et al., 2013; Hire et al., 2009).

Table 1 Phenotypic characterization of the bacteria (B-1) isolated from Culex quinquefasciatus

Results of larvicidal bioassay in the laboratory have been presented in Figure 1. Application of 15 ml/L (6.8x10⁶ bacteria/ml) B. sphaericus suspension resulted in 70.03% and 59.90% death of An. subpictus and Ar. subalbatus larvae, respectively but 100% mortality of Cx. quinquefasciatus larvae after 3h of exposure (Figure 1). Within 6 h, cent percent mortalities of all the mosquito species tested was recorded at the dose of 10ml/L. The probit analysis revealed a higher LD₅₀ value for Ar. subalbatus (11.29 ml/L) compared to An. subpictus (9.78ml/L) and Cx. quinquefasciatus (2.88ml/L) (Table 2). The regression equation of dose mortality of the three mosquito species are presented in Table 2. Comparison of slopes and elevations for the mosquito species An. subpictus irrespective of hours of exposures indicated that the significant difference exist for the elevation but not for the slopes (For Slopes F = 0.431, df= 2,30 (N.S.); For Elevation F = 22.91, df= 2,32, P<0.05, S.E.= 4.67350) of regression equations. When the regression equations were compared to the time of exposure, significant differences were observed in the mortality rates for the mosquito species (Table 3). The slopes and the elevations differed significantly for Cx. Quinquefasciatus (For Slopes F =3.32, df= 2,30 ; For Elevation F =53.479, df= 2,32, P<0.05 S.E.= 2.574) (Table 4), while the only the elevations differed for An. subpictus and An. subalbatus (An. subpictus: For Slopes F = 0.431, df= 2,30 (N.S.); For Elevation F = 22.91, df= 2,32, P<0.05, S.E.= 4.6735; An. subalbatus : For Slopes F = 0.471, df= 2,30 (N.S.); For Elevation F = 22.11, df= 2,32, P<0.05, S.E.= 4.680). Evaluation of slopes and elevations for the three different mosquito species under 3 h exposures revealed significant difference in elevation but not in slopes ((For Slopes F = 0.878, df= 2,30 (N.S.); For Elevation F = 264.036, df= 2,32, P<0.05, S.E.= 1.62815). The results were consistent with the comparison of six hour’s and nine hours’ data. In case of six hour’s comparison significant difference exist for the elevation but not for the slopes (For Slopes F = 2.2973, df= 2,30 (N.S.); For Elevation F = 21.0865, df= 2,32, P<0.05, S.E.= 3.8562) of regression equations. Assessment of slopes and elevations for the three different mosquito species under 9 h exposures indicated the same ((For Slopes F = 3.089, df= 2,30 (N.S.); For Elevation F = 4.983, df= 2,32, P<0.05, S.E.= 4.983).

Figure 1 Mortality of different species of mosquito larvae after 3, 6  and 9 hours exposures to  B. sphaericus preparation (v/v) under laboratory conditions

Table 2 The values [LFL-UFL, Mean dose ± S.E.] of the bacteria B. sphaericus as a mosquito larvicide agent, under laboratory conditions (n= 9 trials per dose). The χ2 values indicate the significant level of the probit analysis in LFL and UFL and Regression equation on mortality rate and dose of B. sphaericus used to deduce the effective mean dose (ml/l of bacterial spores)

Table 3 Comparison of slopes and elevations for the three mosquito species irrespective of hours of exposures.  (PR = Pooled regression, C = Common, Res – Residual

Table 4 Comparison of slopes and elevations for the three different mosquito species under 3 h, 6 h and 9 h of exposures. (PR = Pooled regression, C = Common, Res – Residual)

The observation revealed that the B. sphaericus was more effective against Cx. quinquefasciatus than An. subpictus and Ar. subalbatus larvae (Figure 1). SEM analysis revealed the invasive nature of the bacteria in the mosquito gut wall (Figure 2; Figure 3) presents the photography of crystal proteins of the isolated strain of B. sphaericus. The number of mosquito immatures was considerably reduced in the field upon application of this strain of B. sphaericus. Average per dip larval density of An. subpictus in shallow ponds was 10.33 before treatment which was reduced by 96.77% within 7d after treatment (Figure 4). Pretreated population of Ar. subalbatus and Cx. quinquefasciatus in drains was 14.67 and 68.67 per dip, respectively which were declined by 70.46% and 100% for A. subalbatus and Cx. quinquefasciatus, respectively by 7d (Figure 4). No larval emergence was observed upto tenth day in case of Cx. Quinquefasciatus.

Figure 2 Scanning electron photograph showing invasion of B. sphaericus in the inner gut wall of Culex quinquefasciatus

Figure 3 Crystal protein of Bacillus sphaericus under scanning electron microscope

Figure 4 Reduction of larval density of three different mosquito immatures on different days of treatment

4 Discussion

The isolated strain of B. sphaericus is considerably effective in killing larvae of all the three species of mosquito though Cx. quinquefasciatus larvae are more susceptible in comparison to other two species. Larval mortality increases with the increment of both exposure and dose. The results corroborate that B. sphaericus is an intestinal toxicant and potent pathogen of mosquitoes (Silva-Filha et al., 2004; Opota et al., 2008). Average per dip larval density of An. subpictus in shallow ponds and Ar. subalbatus and Cx. quinquefasciatus in drains reduced significantly in the field condition. But after 10d, the effect of B. sphaericus decreased, which was evident by the reappearance of mosquito larvae in most of the experimental habitats. Cx. quinquefasciatus was highly susceptible showing 100% reduction in larval density after 10d. The crystals of the organism produced a fraction of protein of about 42 kDa size. The protein is an established mosquitocidal toxin (Tangsongcharoen et al., 2011; Park et al., 2010) which would be effective against mosquitoes tested in our study also. So, the results showed that the use of the isolated strain of B. sphaericus during the present study would be helpful in public health programmes for the management of vector-borne diseases.

5 Conclusion

The bacteria B. sphaericus studied against Culex, Anopheles and Armegeres larvae were found to have toxic effect on larval gut wall. Further detail study need to be carried out for larger scale field application. Results from the field trials showed that only a very low dosage of 100 ml/sq m is required to control larvae successfully. Such low application dosages offer the possibility to keep operational costs low and it is eco-friendly also. Since the strain has a close relationship with the mosquito larvae in the native environment and are capable to recolonize in the guts of mosquito larvae, this strain can be considered as a promising new host for an effective aid of mosquito-larvicidal toxins.

Acknowledgements

We are grateful to UGC-DRS (Grant no. F.3-9/2012 SAP-II) and DST-WB for providingfinancial assistance.

Conflict of interest statement

We pronounce that we don’t have any conflict of interest.

References

Abott W.S., 1925, A method of computing the effectiveness of an insecticide, Journal of Economic Entomology, 18(2): 265-267

https://doi.org/10.1093/jee/18.2.265a

Bauer A.W., Kirby W.M., Sheris J.C., and Turck M., 1996, Antibiotic susceptibility testing by a standardized single disc method, American Journal of Clinical Pathology, 45: 149-15

Bhattacharya K., and Chandra G., 2014, Phagodeterrence, larvicidal and oviposition deterrence activity of Tragia involucrata L.(Euphorbiaceae) root extractives against vector of lymphatic filariasis Culex quinquefasciatus (Diptera: Culicidae), Asian Pacific Journal of Tropical Disease, 4: S226-S232

https://doi.org/10.1016/S2222-1808(14)60444-8

Chandra G., Banerjee S., and Bhattacharya K., and Chatterjee S., 2016, Efficacy of a Locally Isolated Strain of Bacillus cereus as Mosquito Larvicide, Molecular Entomology, 7(05):1-12

Chatterjee S., and Chandra G., 2000, Role of Anopheles subpictus as a primary vector of malaria in an area in India, Japanese Journal of Tropical Medicine and Hygiene, 28(3): 177-181

https://doi.org/10.2149/tmh1973.28.177

Chatterjee S.N., Bhattacharya T., Dangar T.K., and Chandra G., 2007, Ecology and diversity of Bacillus thuringiensis in soil environment, African Journal of Biotechnology, 6 (13): 1587-1591

Collee J.G, and Miles P.S., 1989, Tests for identification of bacteria, In: Practical medical microbiology, Eds., NY, USA, pp.141-160

Copping L.G., and Menn J.J., 2001, Biopesticides: a review of their action, applications and efficacy, Pest Management Science, 56: 651-676

https://doi.org/10.1002/1526-4998(200008)56:8<651::AID-PS201>3.0.CO;2-U

Haldar K.M., Ghosh P., and Chandra G., 2011, Evaluation of target specific larvicidal activity of the leaf extract of Typhonium trilobatum against Culex quinquefasciatus Say, Asian Pacific Journal of Tropical Biomedicine, s199 -s203

https://doi.org/10.1016/S2221-1691(11)60156-1

Hati A.K., Chandra G., Bhattacharyya A., Biswas D., Chatterjee KK., and Dwibedi H.N., 1989, Annual transmission potential of bancroftian filariasis in an urban and a rural area of West Bengal, India, The American Journal of Tropical Medicine and Hygiene, 40(4): 365-367

Hire R.S., Hadapad A.B., Dongre T.K., and Vinay Kumar V., 2009, Purification and characterization of mosquitocidal Bacillus sphaericus BinA protein, Journal of Invertebrate Pathology, 101(2): 106-111

https://doi.org/10.1016/j.jip.2009.03.005

Janssen K., 1994, Current protocols in molecular biology, Greene Publ. Assoc. Inc. and John Wiley Sons Inc., NY, USA

Lacey L.A., 1997, Manual of techniques in Insect pathology. Acad. Press, NY, USA, pp. 409

Luxananil P., Atomi H., Panyim S., and Imanaka T., 2001, Isolation of bacterial strains colonizable in mosquito larval guts as novel host cells for mosquito control, Journal of Bioscience and Bioengineering, 92 (4): 342-345

https://doi.org/10.1016/S1389-1723(01)80237-1

Maniatis T., Fritsch E.F., and Sambrook J., 1982, Molecular cloning. A laboratory manual, Cold Spring Harbor, NY, USA

National Committee for Clinical Laboratory Standards: Methods for antimicrobial susceptibility testing of anaerobic bacteria: approved standard. 3^rd ed. NCCLS document M11-A3, Villanova, Pa, National Committee for Clinical Laboratory Standards, 1993

Opota O., Charles J.F., Warot S., Pauron D., and Isabelle Darboux I., 2008, Identification and characterization of the receptor for the Bacillus sphaericus binary toxin in the malaria vector mosquito, Anopheles gambiae, Comparative Biochemistry and Physiology-Part B: Biochemistry, Molecular Biology, 149(3): 419-427

https://doi.org/10.1016/j.cbpb.2007.11.002

Panneerselvam C., Murugan K., Kovendan K., Kumar P.M., and Subramaniam J., 2013, Mosquito larvicidal and pupicidal activity of Euphorbia hirta Linn. (Family: Euphorbiaceae) and Bacillus sphaericus against Anopheles stephensi Liston. (Diptera: Culicidae), Asian Pacific Journal of Tropical Medicine, 6(2): 102-109

https://doi.org/10.1016/S1995-7645(13)60003-6

Park H.W., Bideshi D.K., and Federici B.A., 2010, Properties and applied use of the mosquitocidal bacterium, Bacillus sphaericus, Journal of Asia Pacific Entomology, 13(3): 159-168

https://doi.org/10.1016/j.aspen.2010.03.002

Pelczar M.J., Bard R.C., Burnett G.W., Conn H.J., Demoss R.D., and Euans E.E., 1957, Manual of microbiological methods, Soc. Amer. Bacteriol, McGraw Hill Book Company, Inc., NY, USA, pp. 315

Ray A.S., Bhattacharya K., Singh A., and Chandra G., 2014, Larvicidal Activity of Nelumbo nucifera Gaertn. (Nymphaeaceae) against Anopheles stephensi (Liston 1901) and its Effect on Non-target Organisms, Journal of Mosquito Research, 4 (1): 1-7

Rudra S.K., Paramanik M., and Chandra G., 2013, Studies on Armigeres subalbatus Mosquitoes in Tribal and Non-tribal Areas of Bankura District, West Bengal, India, Journal of Mosquito Research, 3(3): 14-20

https://doi.org/10.5376/jmr.2013.03.0003

Silva-Filha M., Oliveira C., Regis L., Yuan Z., Rico C.M., and Nielsen-LeRoux C., 2004, Two Bacillus sphaericus binary toxins share the midgut receptor binding site: implications for resistance of Culex pipiens complex (Diptera: Culicidae) larvae, FEMS Microbiology Letters, 241(2): 185-191

https://doi.org/10.1016/j.femsle.2004.10.018

Silva-Filha M.H., Nielsen-LeRoux C., and Charles J.F., 1999, Identification of the receptor for Bacillus sphaericus crystal toxin in the brush border membrane of the mosquito Culex pipiens (Diptera: Culicidae), Insect Biochemistry and Molecular Biology, 29(8): 711-721

https://doi.org/10.1016/S0965-1748(99)00047-8

Singha S., and Chandra G., 2011, Mosquito larvicidal activity of some common spices and vegetable waste on Culex quinquefasciatus and Anopheles stephensi, Asian Pacific Journal of Tropical Medicine, 4: 412-420

https://doi.org/10.1016/S1995-7645(11)60088-6

Soni N., and Prakash S., 2012, Entomopathogenic fungus generated Nanoparticles for enhancement of efficacy in Culex quinquefasciatus and Anopheles stephensi, Asian Pacific Journal of Tropical Disease, 2(1): S356-S361

https://doi.org/10.1016/S2222-1808(12)60181-9

Srisucharitpanit K., Yao M., Chimnaronk S., Promdonkoy B., Tanaka I., and Boonserm p., 2013, Crystallization and preliminary X-ray crystallographic analysis of the functional form of BinB binary toxin from Bacillus sphaericus, Acta Crystallographica, 69: 170-173

Subramaniam J., Kovendan K., Kumar P.M., Murugan K., and Walton W., 2012, Mosquito larvicidal activity of Aloe vera (Family: Liliaceae) leaf extract and Bacillus sphaericus, against Chikungunya vector, Aedes aegypti, Saudi Journal of Biological Sciences, 19(4): 503-509

https://doi.org/10.1016/j.sjbs.2012.07.003

Tangsongcharoen C., Boonserm P., and Promdonkoy B., 2011, Functional characterization of truncated fragments of Bacillus sphaericus binary toxin BinB, Journal of Invertebrate Pathology, 106(2): 230-235

https://doi.org/10.1016/j.jip.2010.10.004

Teng H.J., Lu L.C., Wu Y.L., and Fang J.G., 2005, Evaluation of various control agents against mosquito larvae in rice paddies in Taiwan, Journal of Vector Ecology, 30: 126-132

Tetreau G., Chandor-Proust A., Faucon F., Stalinski R., Akhouayri I., Prud’homme S.M., and Raveton M., 2013, Contrasting patterns of tolerance between chemical and biological insecticides in mosquitoes exposed to UV-A, Aquatic Toxicology, 140–141: 389-397

https://doi.org/10.1016/j.aquatox.2013.07.004

Zar J.H., 1999, Biostatistical Analysis, IV ed. New Delhi, India, Pearson Education Singapore (Indian Branch) Pvt. Ltd., pp 66