Technical grade carbofuran with 99.3 per cent purity obtained from FMC India Private Limited, Bengaluru, was used for this study. Carbofuran stock solution of 1000 ppm (0.101g dissolved in 100 ml of acetone) was prepared for supplementing the MSM. Another carbofuran stock solution of 1000 ppm (0.101g dissolved in acetonitrile and diluted to 100 ml) was prepared, and five working standards at concentration levels of 0.1, 0.25, 0.5, 1, and 2.5 μg/ml were prepared by appropriate dilution of the 1000 μg/ml stock standard solution with acetonitrile for calibrating the HPLC for residue analysis.
Five bacterial isolates, isolated and purified from the carbofuran enrichment cultures (Nafeesa. M, 2009), were selected for the study i.e. CF-1, CF-2, CF-3, CF-4, and CF-7. Mineral Salt Medium (MSM) (K₂HPO₄ – 4.8g, KH₂PO₄ - 1.2g, NH₄NO₃ – 1.0g, MgSO₄ – 0.2g, Ca(NO₃)₂ – 0.4g, and Fe₂(SO₄)₃ – 0.001g per litre) provided with carbofuran as the sole source of carbon (50μg ml⁻¹) was used. Before supplementing the media with carbofuran, the media was sterilized by autoclaving it at 1.2 Kg cm⁻² pressure for 20 min. Two and a half ml of the stock solution with acetone was added into the sterilized conical flasks. The acetone portion was allowed to evaporate fully at room temperature to ensure the growth of the microorganisms in the media. Fifty ml of the MSM was aseptically transferred into the conical flasks, which were already supplemented with carbofuran.
The inoculums were prepared by growing the five isolates separately in 50 ml of MSM supplemented with carbofuran. After inoculating the media with appropriate cultures, the flasks were kept overnight on a rotary shaker at 30°C and at 200 rpm. All inoculations were made at a one per cent level, i.e. the equivalent of one ml of inoculum to 100 ml of liquid media (0.5 ml for 50 ml of the media) (Singh et al., 2004). The flasks were incubated at room temperature (30±2°C).
Five isolates were inoculated into 250 ml conical flasks containing 50 ml of the MSM supplemented with carbofuran as a carbon source. A control (without the bacterial inoculum) was also maintained in the experiment. There were six treatments viz., CF-1, CF-2, CF-3, CF-4, CF-7, and control. Each treatment was replicated thrice. In each replicate, six duplicate flasks were maintained to study the growth of the isolate and to estimate the carbofuran residue level in the media.
The growth of the isolates in the medium was estimated by measuring the increase in turbidity of the medium using a UV/VIS spectrometer (Lambda Bio®, PERKIN ELMER GmbH, Germany) at 600 nm. Five ml suspension from the incubated flasks (duplicate flasks) was removed for microbial analysis. The readings were taken at four-hour intervals for four days from the day of inoculation. Control (MSM without inoculum) was kept as blank while taking the readings.
Sampling and Residue Estimation in the Media
Sampling and residue estimation in the media were done by combining and following the methods adopted by Ramanad et al. (1988), Yan et al. (2007), and Talebi and Walker (1993). Samples were taken from the duplicate flasks at desired intervals i.e. 0, 1, 2, 4, 8, and 15 DAI after homogenization of the media by placing the flasks in a shaker (Orbitech, BVN, Instrument Pvt. Ltd., Chennai) for one hour at 150 rpm. Five ml of homogenized media from each replicated flask was removed aseptically and pooled. From the pooled suspension, five ml was taken as a representative sample for the extraction of carbofuran residue.
Fifty ml of dichloromethane was added to the five ml portion of MSM containing carbofuran taken in a separating funnel. The mixture was vigorously shaken for 10 min and left to stand until phase separation took place. The dichloromethane layer (bottom layer) was filtered through anhydrous sodium sulfate in a funnel fluted with Whatman No.1 filter paper. The samples were extracted twice. The extracts were combined and evaporated on a rotary evaporator at 40°C under vacuum near to dryness. The residue was dissolved in five ml of acetonitrile.
The prepared samples were diluted to 50 times (200 μl of the sample to 10 ml with acetonitrile) before injection. Dilutions of the samples were done to get a lower concentration to protect the HPLC column.
Recovery Studies and Residue Analysis
Recovery studies and residue analysis were done using HPLC - UV (Shimadzu LC 20 AT, Shimadzu India Pvt. Ltd., Mumbai). The operational parameters (Column - RP 18(10 μm) (25 ×3.2 i.d), Mobile phase - Acetonitrile:Water (60:40 v/v), Flow rate - 1ml/min, Sample Injection Volume - 20 μl) were fixed by referring to and modifying the parameters used by Talebi and Walker (1993) and Farahani et al. (2007). The HPLC was calibrated using five working standard solutions of carbofuran at concentration levels of 0.1, 0.25, 0.5, 1, and 2.5 μg/ml.
For recovery studies, fortifications were done in three concentration levels (0.25, 0.5, and 1 ppm). Fifty ml of the culture medium was fortified with carbofuran in the three above-mentioned levels. After the additions of the chemical, the media was homogenized by placing it in a shaker for one hour. Five ml portion of the culture media was taken as a representative sample for extraction. The standard solutions were loaded after conditioning, and recoveries were calculated from the chromatogram of the standard solution.
Recovery studies were conducted in triplicate at three concentration levels (0.25, 0.5, and 1 ppm). The amount of insecticide residue recovered was measured by comparing the sample response with the response of the standard in the HPLC chromatogram using the formula (Kohout, 1974).
Statistical Analysis
Statistical scrutinies of the results were made as per the method described by Panse and Sukhatmi (1985). Dissipation curves of carbofuran in different treatments were fit as suggested by Nigg and Stamper (1980). Other statistical parameters like confidence intervals for intercept (a), slope of regression line (b), half-life (T₀.₅), and coefficient of determination (r²) were obtained following the procedures described by Timme et al. (1986). The different functions were worked out for the residue data to fit the dissipation curves as per Regupathy and Dhamu (2001). The best-fit model for control and for each isolate was selected by comparing the correlation coefficient values for different functions and modified r².
