Evaluation of Cuban Bacillus thuringiensis (Berliner, 1911) (Bacillales: Bacillacea)isolates with larvicidal activity against Aedes aegypti (Linnaeus, 1762) (Diptera:Culicidae)

Aileen González Rizo, Camilo E Casta?et Martinez, Celeste Ramírez Cardentey, Ariamys Companioni Iba?ez,Zulema Menéndez Díaz, Lianet Monzote Fidalgo, Hilda M Hernandez álvarez

1Vector Control Department, Institute of Tropical Medicine ¨Pedro Kourí¨, AutopistaNovia del Mediodía km 6?, La Habana 11400, Cuba

2Virology Department, Institute of Tropical Medicine ¨Pedro Kourí¨, Autopista Novia del Mediodía km 6?, La Habana 11400, Cuba

3Parasitology Department, Institute of Tropical Medicine ¨Pedro Kourí¨, Autopista Novia del Mediodía km 6?, La Habana 11400, Cuba

ABSTRACT

Objective: To evaluate 11 Cuban native Bacillus (B.) thuringiensis isolates in order to select one with the best larvicidal activity against Aedes (Ae.) aegypti and low cytotoxicity.

Methods: The cry and cyt genes of the isolates (A21, A51, L95,L910, M29, R84, R85, R87, R89, U81 and X48) were amplified by PCR. The influence of organic matter and NaCl on the larvicidal activity was tested by bioassays. Cytotoxicity was assayed on peritoneal macrophages of BALB/c mice.

Results: The cyt1 (Aa, Ab, Ba), cyt2, cry4aA, cry4Ba, cry11 (Aa, Ba,Bb) and cry10 genes were identified in all native Cuban isolates.The larvicidal activity (LC90) of seven isolates was affected by the presence of organic matter in the water, while A21, A51, L910, R84,U81 and X48 had better LC50, LC90, LC95 than the 266/2 9-Ⅶ-98 control strain. The LC50 of two isolates was affected by the presence of NaCl and A21, A51, R85 isolate had better larvicidal activity than the 266/2 9-Ⅶ-98 control strain. In terms of toxicity against macrophages, the extracts of nine isolates were less cytotoxic than the control strains.

Conclusions: Native isolate A21 had the main virulence factors against Ae. aegypti larvae, displayed a good larvicidal activity in presence of different factors related with Ae. aegypti breeding sites, and had low citotoxicity against macrophages. These results can contribute to the improvement of existing biological control strategies and the development of new biolarvicides.

KEYWORDS: Mosquitoes; Biological control agent; Bacillus thuringiensis; Bioassays; Aedes aegypti; cry and cyt genes

Significance

Vector control strategies should be adapted to the local conditions, mainly in low- and middle-income countries which are most affected by dengue and climate change. This study provides an integral, objective and practical evaluation of Bacillus thuringiensis Cuban native isolates in order to select the best isolates for biolarvicide development. This kind of evaluation (based not only on the cry and cyt genes)emphasizes the importance of obtaining non-toxic isolates that maintain their high larvicidal activity against Aedes aegypti in presence of different factors associated with the breeding sites. It is a valuable tool for the development of new and safe biolarvicides, more adapted to Aedes aegypti breeding sites conditions.

1. Introduction

Climate change, global warming, human activities, among other factors increase the abundance and worldwide geographical distribution of Aedes (Ae.) aegypti (Linnaeus, 1762) (Diptera:Culicidae)[1]. This mosquito is considered the principal vector that transmits Zika, dengue, chikungunya and yellow fever in the Americas; therefore, its control is of paramount importance to interrupt the transmission of these diseases[2].

In this sense, the most effective method to reduce Ae. aegypti populations is the use of chemical insecticides aimed to control immature or adult insects[2]. However, the increase in insecticide resistance[3] requires alternative methods of control such as microbial insecticides[4]. The most widely used microbial biopesticides are derived from Bacillus (B.) thuringiensis (Berliner,1911) (Bacillales: Bacillacea)[5].

Biolarvicides based on B. thuringiensis are specific to a limited number of insect species with no toxicity against humans or other organisms, and an effective tool for Ae. aegypti larval control[6].The principal virulence factors of this bacterium (cry and cyt toxins) have a more distinct mode of action on mosquito larvae than chemical insecticide[6]. Nevertheless, the larvicidal activity of B. thuringiensis in field has a low persistence owing to the low stability of its toxins under field conditions[6,7]. In particular, the larvicidal activity of B. thuringiensisis is conditioned by several factors, namely organic enriched habitats, exposition to UV light,temperature increase, changes in pH, chlorination or bacterial degradation[8-10]. Thus, the continuous search of native isolates is a current need in order to generate biolarvicide formulations more adapted to the conditions of each region and provide a highly effective and low-cost product[11-13].

In Cuba, previous studies reported native isolates of B. thuringiensis with a high larvicidal activity against Ae. aegypti[14,15], as well as the influence of temperature and water chlorination on this activity[16].In this context, the present study carry out the final evaluation of Cuban native isolates in order to select the better isolates for biolarvicide development based on: 1) the presence of cry and cyt genes; 2) the influence of organic matter and water salination on the larvicidal activity, and 3) the cytotoxicity on macrophage.

2. Materials and methods

2.1. Bacterial control strains, isolates and mosquitos

B. thuringiensis serotype H-14, IPS 82 from the International Entomopathogenic Bacillus Centre, Institute Pasteur; Paris, France and B. thuringiensis var. israelensis serotype H-14 266/2 9-Ⅶ-98(strain isolated from the most extensive biolarvicides used in Cuba:Bactivec? Labiofam, Cuba) were used as control strains.

Native B. thuringiensis isolates: A21, A51, L95, L910, M29, R84,R85, R87, R89, U81, and X48 were isolated from soil samples of the Cuban archipelago[14,15]. These isolates belong to the entomopathogenic bacteria collection from the Biological Control Laboratory of the Tropical Medicine Institute “Pedro Kourí”, IPK,Cuba.

Ae. aegypti (Rockefeller strain), a laboratory susceptible strain of Caribbean origin colonized after the 1930s, was provided by the Center for Disease Control and Prevention (CDC) Laboratory in San Juan, Puerto Rico.

Mosquitoes were maintained on 10% sucrose solution at(26.0±0.5) ℃, 80%-85% relative humidity with a 12 h light/dark cycle. Female mosquitoes were given access to an anesthetized mouse and allowed to blood feed for 30 min weekly. The larvae were fed with finely powdered fish food (CENPALAB, Cuba)[17].

2.2. Detection of cry and cyt genes

To detect the cry and cyt genes a 12 h of B. thuringiensis culture(control strains and isolates) in a nutrient medium plate was used.A loopful of cells was transferred to 0.1 mL of H2O and treated with lysozyme for 2 h at 37°℃ to obtain DNA using the procedure described by Maxwell? 16 Tissue DNA Purification Kit (Promega,USA). The PCR mix consisted of 1× green buffer (Promega, USA),2 mM MgCl2; 0.2 mMdNTP; 0.5 μM each primer (forward and reverse, Table 1); 2.5 U Go taq Flexi DNA polymerase (Promega,USA); and 2 μL of template DNA for a final volume of 50 μL. and PCR was carried out in a Mastercycler personal Eppendorf AG,Germany, as follows: 2 min at 95°℃; 30 cycles of 1 min at 95 ℃, 1 min annealing at 46 ℃ to 54 ℃ ( according to each primer combination,Table 1), and 1 min at 72 ℃; and 5 min at 72 ℃. Fifteen μL of PCR product was electrophoresed on 2% agarose gel and run 250 V during 45 min.

Table 1. Primers used in the cry and cyt gene detection.

2.3. Influence of organic matter and NaCl on the toxicity of B. thuringiensis native isolates

Bacterial isolates and control strains (B. thuringiensis IPS-82 and B. thuringiensis 266/2 9-Ⅶ-98) were grown in a fermentation medium consisting of sucrose (2 g/L), bacteriological peptone(2 g/L), yeast extract (1 g/L), and inorganic salts (12.5 mmol/L MgSO4; 0.05 mmol/L MnSO4; 1.2 mmol/L FeSO4; 1.2 mM ZnSO4;25 mmol/L CaCl2); and incubated at 30°℃ and 150 rpm shaking,until sporulation was completed (48-72 h). Concentrations were expressed in mg/mL (dry weight).

Quantitative bioassays were conducted following the World Health Organization (WHO) protocol[21]. Twenty-five larvae (Ⅲ-Ⅳinstar) were placed into 120 mL cups with 100 mL of dechlorinate water. Five concentrations of bacterial formulation that cause mortalities between 10% and 90% were accepted for validating the bioassay in order to calculate the lethal concentrations (LC).Four replicates were performed for each concentration tested per bioassay. Each bioassay was repeated four times in independent assays. Larval mortality was recorded 24 h after treatment.

To detect the effect of organic matter and NaCl on larvicidal activity the biosassays were performed with: 300 mg of non contaminated powdered leaf litter in 100 mL of dechlorinate water and dechlorinated water with a NaCl concentration of 5 g/L, respectively.The biosassays performed only in dechlorinated water were used as control. Finally, the influence of organic matter and NaCl versus dechlorinate water on the larvicidal activity of the Cuban isolates was tested and compared.

2.4. Macrophage cytotoxicity assay

The spore-crystal mixtures of native isolates and control strains (B.thuringiensis IPS- 82 and B. thuringiensis 266/2 9-Ⅶ-98) were resuspended in 50 mM Na2CO3for 1 h at 37°℃. After that, the supernatants were centrifuged at 13 000 × g during 10 min at 4 ℃. Then the clarified supernatants were passed through a 0.45 μm membrane filter, and the pH was adjusted to 8.0.

The filtered supernatant was used directly (aqueous extract)or diluted in alcohol at 80% (hydroalcoholic extract). Both solutions were kept standing for 7 days at 4°℃ with occasional manual shaking (3 times a day for 1 minute). Subsequently, the solvent from the samples was evaporated in a Concentrator Plus(Eppendorf, Germany) during 4 h. The supernatant was removed and the pellet was re-suspended in dimethylsulfoxide (DMSO;BDH, England), until a final concentration of 20 mg/mL was obtained. In parallel, a control with culture medium was included.

Peritoneal macrophages for cytotoxic assays were collected from healthy female BALB/c mice as follows: twelve animals were euthanized by cervical dislocation and macrophages were obtained by lavage with 5 mL of RPMI-1640 medium (Sigma, USA) into the peritoneal cavity.

The median cytotoxic concentration (CC50) of the extracts on macrophages was determined. Peritoneal macrophages in RPMI-1640 medium supplemented with antibiotics (penicillin 200 UI,streptomycin 200 μg/mL) were seeded in 96-well V-bottom plates at a concentration of 3×105cells/well and incubated for 2 h at 37 ℃ in 5% CO2to obtain a monolayer culture. The non-adherent cells were removed by washes with phosphate-buffered saline solution (PBS).

Then, in each well, 50 μL of medium with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich, USA) and antibiotics (penicillin 200 UI, streptomycin 200 μg/mL) were added, into the wells of column 2 and 7, additional 48 μL of medium were dispensed and 2 μL of tested extracts and two-fold serial dilutions down each lane were carried out to give final concentrations from 12.5 to 200μg/mL. Thereafter, the treated macrophages were incubated at 37℃ in an atmosphere of 5% CO2. After 72 h of incubation 15 μL of a solution of 3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide (MTT, Sigma, USA) was added to each well. After incubating for 4 h, at the same conditions, the formazan crystals were dissolved in DMSO (100 μL per well). Absorbance was measured at 560 and 630 nm as the reference wave length[22] and lineal concentration response curves were constructed. Evaluations were performed in triplicate in independent assays.

The extracts from native isolates with CC50higher than the CC50obtained with the controls strain used in the study (IPS-82 and 266/2 9-Ⅶ-98) were considered non-cytotoxic.

2.5. Statistical analysis

In all bioassays Ae. aegypti larval mortality data were used to calculate the lethal concentrations for 50%, 90% and 95% of exposed individuals (LC50, LC90and LC95respectively) through log probit analysis[23] using the program SPSS 21. The means of larval mortality caused by each isolate and the control strains against Ae.aegypti were calculated. Once the lethal doses were calculated, the LC95/LC50ratio was performed to determine how many times it is necessary to increase the LC50in order to obtain higher mortality.A lower ratio is indicative of better formulation efficiency[24].

To detect the effect of organic matter and NaCl on larvicidal activity data analysis was performed by t-Student test using the statistical package SPSS 21. In all cases, statistically significant differences were identified at P＜0.05 level.

In macrophage cytotoxicity assay the medium cytotoxic concentration (CC50) was obtained from linear dose-response.Results are expressed as median and 95% confidence intervals (CI)of three independent replicates. The statistical differences between CC50of the control and isolates extracts were determined using Kruskal-Wallis with Statistica for Windows Program (Version 13.1,StatSoft, Inc 2016), considering statistical differences as P＜0.05.

2.6. Ethical approval

All the experimental procedures involving animals were conducted in accordance with the Guide for the Care and Use of Laboratory Animals, Eighth Edition, which was approved by the Ethics Committee (CEI-IPK 21-16), Havana, Cuba.

3. Results

3.1. Detection of cry and cyt genes

The specific cry and cyt type primers were used to detect cry and cyt genes in the isolates by PCR analyses, cry11-type, cry4-type,cyt1-type, and cyt2-type genes were found in all native isolates(Table 2). The presence of cyt1 (Aa, Ab, Ba), cry11 (Aa, Ba, Bb) and cry10 were detected in all isolates. On the other hand, we could detected other cyt1 genes (Aa, Ab) in 10 isolates (90.1%). The presence of cry10Aa gene was only detected in two isolates, L910 and M29 (18.2%). In 10 isolates, a band of 305 bp was obtained as a result of the amplification with cb-11 primer (cry11 A, B).

3.2. Influence of organic matter on the toxicity of B.thuringiensisnative isolates

In the performed bioassays, the control mortality was lower than 5.0%. The LC90of A21, A51, L95, L910, M29, R85 and X48 isolates were affected (P＜0.05) (Table 3) by the presence of organic matter in the water comparing with those exposed to declorinated water. A21, A51, L910, R84, U81 and X48 isolates exhibited better larvicidal activity (LC50, LC90and LC95) than the 266/2 9-Ⅶ-98 strain in presence of organic matter as shown in Table 3. A51 isolate had lower LC90than IPS-82 control strain.Efficiency for R85 in presence of organic matter was 9.0, which was the most affected isolate (Table 3).

3.3. Influence of NaCl on the toxicity of B. thuringiensisnative isolates

The LC50of R84 and X48 were affected by the presence of NaCl.(Table 3). The larvicidal activity (LC50, LC90, LC95) of A21, A51,R85 and U81 isolates were significant better (all P＜0.05). A51 and U81 isolates had lower LC90than IPS-82 strain with presence of NaCL (both P＜0.05) (Table 3).

In summary, A21, A51 and U81 isolates exhibited better larvicidal activity than 266/2 9-Ⅶ-98 strain in presence of organic matter and NaCl.

3.4. Macrophage cytotoxicity assay

The aqueous extracts of: A21, L95, L910, M29, R84, R85 and U81 isolates, as well as the hydroalcoholics of: A21, L95, L910,M29, R84, R87, R89 and U81 isolates did not show cytotoxicity given at 200 μg/mL (Table 4). On the contrary, both aqueous and hydroalcoholic extracts of X48 isolate showed CC50values significantly lower (P＜0.05) than the strains used as control and therefore they were considered cytotoxic.

Table 2. Detection of cry and cyt genes in Bacillus thuringiensis native isolates and control strains.

Table 3. Lethal concentration (LC) of Bacillus thuringiensis isolates and control strains against Aedes aegypti larvae after 24 h exposure obtained from probit analysis (mg/L).

Table 4. Cytotoxicity of aqueous and hydroalcoholic extracts of Bacillus thuringiensis native isolates and control strains on peritoneal macrophages.

4. Discussion

B. thuringiensis exhibits high toxicity for diptera larvae[12,13].The breeding sites treated with this bacterium attract Ae. aegypti female and act as lethal ovitraps[25,26]. However, the efficacy of the products based on this bacterium is affected by environmental conditions[8-10,27]. For this reason, the evaluation of native strains is an important step for developing biolarvicides adapted to our natural conditions. Cuban B. thuringiensis isolates, collected from soils samples of different environments[14,15] and evaluated in this and others studies[16,28], exhibited some differences that permitted a correct selection.

The genetic studies of native isolates allowed corroborating the presence of the main virulence factors against Ae. aegypti detected in B. thuringiensis[6,18,29,30]. The cry10 gene (primer cb-7) was identified in 11 isolates, while the amplification of cry10Aa was only obtained in the control strain IPS-82 and in M29 and R84 isolates. According to the literature reviewed, the cry10Aa genetic variant active against Diptera has been described for the cry10 gene[6,29-31]. However, the differences in cry10 gene amplification with different primers, suggest the possibility of other genetic variants. The cry and cyt genes detected confirm the proteins patterns previously reports for these isolates[14,15].

The high larvicidal activity of the B. thuringiensis deltaendotoxins against mosquitos is attributed to complex interactions between their proteins[32]. The combinations: cry4Aa and cry4Ba[32], cry4Aa and cry11Aa, cry4Ba and cry11Aa, cry10Aa and cyt2Ba[32], cry10Aa with cyt1Aa[6,29,32], cyt2Ba with cry4Aa,cyt1Aa and cry11Aa[32]; are synergistic against Ae. aegypti larvae.The detection of different cry and cyt genes in all isolates allowed us to suggest the presence of these protein combinations, which would justify the high larvicidal activity previously reported[14,15].

The use of isolates with cry and cyt active proteins against Diptera would delay the development of resistance, taking into account that cyt proteins act as additional receptors for cry proteins and potentiate their activity[32,33]. Field and laboratory resistance to B. thuringiensis were reported in Culex quinquesfasciatus and Culex pipiens larvae[34,35]. However, only insignificant levels of resistance were attained against Ae. aegypti in laboratory conditions. In both genera of Diptera, resistance behaves unstable, and in absence of selection pressure it reverts to 50% after three generations[32,34,35].Therefore, the detection of cyt1A, B and cyt2 genes in all native isolates can predict low resistance in the field to future products based on these isolates.

It is good for us to have native B. thuringiensis isolates with excellent combinations of cry and cyt genes. However, the influence of different factors, like temperature increase, presence of chlorine,salt and organic matter, over B. thuringiensis larvicidal activity is another highlight to be considered.

According to a study carried out in 2019[16], some Cuban B.thuringiensis native isolates (A21, A51, L910, R85, and X48)maintained a good larvicidal activity against Ae. aegypti in presence of temperature increase (25-35°℃) and chlorine. Nevertheless, the correct selection of native isolates implies the evaluation of other factors, such as salt and organic matter, to determine their influence on the larvicidal activity of these isolates.

Biolarvicides based on B. thuringiensis var. israelensis show low activity in organically enriched habitats[8-10]. Rydzanicz et al demonstrated that the optimum larvicidal effect of B. thuringiensis can be achieved in breeding habitats with limited organic content[9].In this study, the larvicidal activity was significantly affected by organic matter in seven of 11 native isolates. This decrease may be associated to diversification of the food source of the Ae. aegypti larvae by the organic matter and consequently, they ingest a lower concentration of toxins, spores and vegetative cells. Additionally,the lamellar envelope of the toxic crystal of our isolates may interact powerfully with organic matter particles, leading to a major decrease of larvicidal activity[8]. On the other hand, the cyt proteins detected in these isolates may be bind irreversibly to the organic matter present in the medium and thus preventing their synergistic effect with cry proteins. This inhibitory effect was previously reported by Tetreau et al in 2012[8]. Notwithstanding, the larvicidal activity of six isolates was better than 266/2 9-Ⅶ-98 control strain.

The larvicidal activity of four native isolates increased significantly in presence of NaCl, which could be associated to the specific characteristics of each isolate. In some B. thuringiensis strains the NaCl increases the sporulation process and delta-endotoxins production[36], leading to a major larvicidal activity.On the other hand, water salinity may lead to osmotic stress in Ae. aegypti larvae, which will increase the feeding needs and to compensate it, they will consume more B. thuringiensis toxins.Dawson et al, in 2019, did not obtain a reduction in the larvicidal activity of B. thuringiensis in presence of Na+and Cl-[10]. However,other study, such as Jude et al, reported a significant reduction in the larvicidal activity of B. thuringiensis against Ae. aegypti in the presence of NaCl[37].

A high larvicidal activity against Ae. aegypti is very important in the selection of native isolates but a low cytotoxicity is essential in order to obtain safe candidates for biolarvicide development.Macrophages are essential effectors of the immune system response against microorganisms. The ability of some species of the Bacillus genus such as B. cereus (a species phylogenetically close to B. thuringiensis) to eliminate macrophage cells explains the persistence and dissemination of virulent strains in mammals[38].The lower cytotoxicity against macrophages obtained with the extracts of 10 isolates is the first step that suggests safety in their use in future formulations. In this and previous studies, the X48 isolate showed a high larvicidal against Ae. aegypti[14,16]. This isolate has a principal virulence factors against Ae. aegypti larvae,but it was more cytotoxic against peritoneal macrophages than 266/2 9-Ⅶ-98 and IPS- 82 strains. This result allows us to preclude it for biolarvicide development.

According to our results, the U81 isolate kept a high larvicidal activity in presence of organic matter and NaCl, and it was less cytotoxic against peritoneal macrophages than 266/2 9-Ⅶ-98 and IPS-82 strains. Nevertheless, its activity was significantly affected by temperature increase and chlorine presence[16]. Taking into account the average of temperature increase in Cuba[39] and that chlorine is one of the most commonly used domestic water disinfectants in the world[40], we analyzed the larvicidal activity obtained with others isolates.

In this sense, A51 isolate had a better larvicidal activity based on the results obtained in this and in preceding studies[16], although the presence of beta exotoxins[28] excluded it as a candidate for biolarvicide development.

On the other hand, the results obtained in this and others studies[14-16,28] allow us to recommend A21 isolate as an active ingredient of biolarvicides. Its high larvicidal activity in presence of different factors related with Ae. aegypti breeding sites, their mains virulence factors against Ae. aegypti larvae and its low citotoxicity against macrophages are important points for this selection.

The results obtained from the evaluation and selection of native strains more adapted to Ae. aegypti breeding sites conditions can contribute to the improvement of existing biological control strategies and the development of new biolarvicides. Further investigations should be done with Cuban native isolates aiming to sequence the complete genome, to evaluate its larvicidal residual activity, and to carry out metabolomic studies; in order to clarify or improve the high larvicidal activity described.

Conflict of interest statement

The authors declare that there is no conflict of interest.

Authors’ contribution

AGR: Conceptualization, methodology data curation, formal analysis, investigation, writing-original draft, writing-review &editing, final approval of the version to be published; CECM,CRC: Formal analysis, investigation, writing-review & editing,final approval of the version to be published; ACI, ZMD: Formal analysis, investigation, writing- review & editing. final approval of the version to be published; LMF: Methodology data curation,formal analysis, investigation, writing-review & editing, final approval of the version to be published; HMHA: Resources,supervision, formal analysis, investigation, writing-review &editing, final approval of the version to be published.