Biodegradation of chlorpyrifos using isolates from contaminated agricultural soil, its kinetic studies

Extensive pesticides use is negatively disturbing the environment and humans. Pesticide bioremediation with eco-friendly techniques bears prime importance. This study evaluates the bioremediation of chlorpyrifos in soil using indigenous Bacillus cereus Ct3, isolated from cotton growing soils. Strains were identified through ribotyping (16s rRNA) by Macrogen (Macrogen Inc. Geumchen-gu, South Korea). Bacillus cereus Ct3 was resistant up to 125 mg L−1 of chlorpyrifos and successfully degraded 88% of chlorpyfifos in 8 days at pH 8. Bacillus cereus Ct3 tolerated about 30–40 °C of temperature, this is a good sign for in situ bioremediation. Green compost, farmyard manure and rice husk were tested, where ANOVA (P < 0.05) and Plackett–Burman design, results indicated that the farm yard manure has significant impact on degradation. It reduced the lag phase and brought maximum degradation up to 88%. Inoculum size is a statistically significant (P < 0.05) factor and below 106 (CFU g−1) show lag phase of 4–6 days. Michaelis–Menten model results were as follows; R2 = 0.9919, Vmax = 18.8, Ks = 121.4 and Vmax/Ks = 0.1546. GC–MS study revealed that chlorpyrifos first converted into diethylthiophosphoric acid and 3,5,6-trichloro-2-pyridinol (TCP). Later, TCP ring was broken and it was completely mineralized without any toxic byproduct. Plackett–Burman design was employed to investigate the effect of five factors. The correlation coefficient (R2) between experimental and predicted value is 0.94. Central composite design (CBD) was employed with design matrix of thirty one predicted and experimental values of chlorpyrifos degradation, having “lack of fit P value” of “0.00”. The regression coefficient obtained was R2 = 0.93 which indicate that the experimental vales and the predicted values are closely fitted. The most significant factors highlighted in CBD/ANOVA and surface response plots were chlorpyrifor concentration and inoculum size. Bacillus cereus Ct3 effectively degraded chlorpyrifos and can successfully be used for bioremediation of chlorpyrifos contaminated soils.

Plackett-Burman design (PBD) for screening significant factors. In this study activity of five variables (both nutritional and environmental) were selected for statistical analysis. These variables and their higher (+ 1) and lower (− 1) values are presented in Table S5. Plackett-Burman design was designed using Minitab 16.2.2 software 24 and the comprehensive matrix (standard 12 run) is presented in Table 1. The impact of each variable on biodegradation of chlorpyrifos was calculated through following equation: where Y represents the chlorpyrifos biodegradation (response), A o represents the intercept, B i represents the linear factor coefficient and X i is the level of each variable. Probability value (P value < 0.05) in regression analysis

Central composite design (CCD) of experiments and response surface methodology (RSM).
Four significant factors viz concentration of chlorpyrifos, temperature, carbon source and inoculum size were established as critical determinants for biodegradation of chlorpyrifos as the outcome of PBD. Full factorial central composite design (CCD) (2 4 ) was adopted with 31 experimental runs, seven center points and eight axial points 25 . The significance of model, prediction equation, regression coefficient and case statics were calculated through analysis of variance (ANOVA). A second-order polynomial equation was employed to fit results and analyze interaction among factors, as shown by Eq. (3): where 'Y' represents the chlorpyrifos degradation, 'X' represents the factors that influence degradation, 'A o' ' represents the intercept coefficient, 'B i ' represents the linear coefficient, 'B ii ' represents the quadratic coefficient and 'B ij ' represents the interaction coefficient. Response surface plots were obtained using Minitab (statistical software, version 16.2.2) to identify effect of factors. ANOVA for CCD was generated using Minitab Software.
Ethical approval. This article does not contain any studies with human participants or animals performed by any of the authors.
Informed consent. Informed consent is not-applicable in this study.

Results and discussion
The demand of eco-friendly solution and the use of indigenous species to restore pesticide contaminated soils is growing globally 26 . Biodegradation is considered best option for in situ restoration operations 6 . Soils contaminated with polycyclic aromatic hydrocarbon was successfully restored by Nocardia farcinica, Nocardia asteroids and Nocardia cashijiensis 27 . Chlorpyrifos is widely used on cotton and vegetables to control pests. Excessive use of the pesticide and its accumulation in soil changes the physiochemical properties of soil. Which lead to change in soil micro-flora and fertility. Abigail et al. 7 reported that loss of soil fertility is due to toxicity and mutagenicity in nostoc muscorum and anabaena doliolum. To restore chlorpyrifos contaminated soil we isolated resistant strain Ct3, it was resistant up to 150 mg of chlorpyrifos. The resistance detail in different concentrations of chlorpyrifos is given in Table S1. This strain is not much resistant against other pesticides (Table S2). Its phylogenetic trees identified it as Bacillus cereus (Fig. 1). The use of indigenous species is preferred as they do not pose any negative impact on micro-flora. The degradation rate becomes slow at higher CP concentration, whereas at low concentration with apparently no lag phase the rate of biodegradation becomes fast (Fig. 2a). To maintain minimum number of bacteria the more time is required that becomes the reason of longer lag phase 28 . Results showed that the CP bioremediation is concentration dependent. ANOVA reveled that concentration is a significant (P < 0.05) factor (Tables 1, 2). 4.7% bioremediation was noted with the initial concentration of 75 mg kg −1 and when the concentration was increased up to 125 mg kg −1 the bioremediation decreased to 2% (in 2 days). After 4 days the rapid degradation started when initial concentration of 75 mg kg −1 , whereas after 8 days of incubation, rapid degradation started when concentrations was 100 and 125 mg kg −1 . At 16th day more degradation (66%) was noted with 75 mg kg −1 of CP and less degradation (43) was noted with 125 mg kg −1 of CP. The 2 main reasons which impact the CP degradation are as follows; firstly, as the parent pollutant (CP) concentration decreases the attractive forces between soil particles and CP molecules increases 11 . Secondly, the growth of bacterial population The failure and success of biodegradation is due to temperature variation 32 . Degradation capacity of Bacillus Ct3 was significantly affected temperature (P < 0.05) (Tables 1, 2). After 6 days rapid degradation was started (at 35 °C), whereas, lag phase continued up to 8 days at 30 °C and 40 °C. Excessive degradation of 45%, 60% and 54% was observed at 30 °C, 35 °C and 40 °C, respectively (Fig. 2b). According to the results, the isolated Bacillus sp tolerated wide range of temperature; this is a good sign for in situ bioremediation. Liu et al. 9 stated that at 30 °C Pseudomonas putida worked most effectively and achieved 78%.
In bioremediation process, soil pH plays one of the most effective roles as abiotic factors. The CP bioremediation rate changes with small fluctuation in pH. The maximum degradation (75%) of CP was observed at 8 pH (Fig. 2c). After 6 days, the CP degradation increase sharply and later at day 14 it entered into stationary phase. Maximum degradation rate of 67% was observed at pH 8.5 and the minimum (60%) was observed at pH 7. All the applied treatments of pH showed significant difference from each other at the 16th day. Whereas, non-significant results were observed at 2nd and 4th day. Impact of pH change was non-significant (P < 0.05) on degradation (Table 2). Results also showed that with wide range of pH the Bacillus can effectively degrade CP (Fig. 2c). Other research projects also highlighted the significance of pH in bioremediation 33 . In the rapid changing environment, the strains which had the ability to work at variable pH are more likely to succeed in degradation and are more acceptable. Enterobacter sp. was isolated by Singh et al. 14 , this strain showed optimum degradation at higher pH, whereas low pH had negative effect on this strain. Pseudomonas kilonensis SRK1 was isolated from wastewater which degraded initial concentration of 150 mg L −1 having CFU (3.6 × 10 6 ), pH (8) and glucose as additional carbon source 34 . Samual et al. 35 reported optimum biodegradation of p-nitrophenol at pH 7 using Pseudomonas putida. Gong et al. 36 stated that optimum pH is one of the critical factor in biodegradation and successive cycling of pollutants in forest.
In bioremediation process the organic amendments significantly impact the rate and course of degradation. In this study, rice husk, farmyard manure and green compost was added in the soil as carbon source. In presence of organic amendments Bacillus sp. showed changed behavior towards bioremediation (Fig. 2d). Farmyard manure shows more promising results towards bioremediation process in comparison to non-carbon treatment. ANOVA (P < 0.05) and PBD indicated that the carbon source has significant impact on degradation ( Table 2). It reduced the lag phase and brought maximum degradation up to 88%. During whole experiment the degradation percentage varies significantly. The increase in degradation rate from 1st to 14th day was gradual, but degradation rate becomes slow after day 14th (Fig. 2d). With addition of green compost the degradation was also facilitated and reached to the level of 83% without lag phase. However, addition of rice husk showed different pattern. Initially lag phase was observed, later degradation increased at 6th day and reached at stationary phase at 14th day. Rice husk is 13% less effective than farmyard manure. This result shows that Bacillus sp. can effectively use different organic material to speed up the bioremediation process. Many studies have recommended the effectiveness of amendments especially organic one, common examples includes public green compost, farmyard manure, municipal waste, cow manure 37 , mushroom spent, nut shells 38 and coconut fiber 39 .
In CP bioremediation the inoculum size/density also play a vital role. At different inoculum densities this strain depicts almost the similar pattern (Fig. 2e). With 10 8 CFU g −1 the CP degradation was up to 100% (in www.nature.com/scientificreports/ 10 days). Inoculums density of 10 7 CFU g −1 also led to 100% degradation. However, medium inoculums densities (10 4 and 10 5 CFU g m −1 ) reported slightly less degradation rate initially but later fast degradation starts. Inoculums density of 10 7 (CFU g −1 ) also attained 100% degradation. Cupriavidus sp. DT-1 also degraded 100% of CP and 94% of TCP. Inoculum is a statistically significant (P < 0.05) factor in chlorpyrifos degradation ( Table 2). The whole experiment lasted for 30 days with 10 6 cells g −1 40 . The use of sewage sludge for improving bioaugmentation was tested and found effective 41 . Jariyal et al. 42 has reported that instead of single strain, consortia can be used for better degradation. Use of consortium is also very beneficial in synergistic biodegradation of aromatic-aliphatic copolyester plastic. This consortium was isolated from marine ecosystem and have the tendency to use pollutant as sole carbon source in 15 days 43 . The Hanes plot's straightness is signified by the R 2 (0.9919) and V max (18.7627) (Fig. 2f). Effectiveness of the isolates is usually predicted by the ratio of V max /K s, and in this study this ratio is 0.1546. Maya et al. 22  Samuel et al. 44 reported that pseudo-secound order model to be best fit for Pb(II) removal. The kinetic data was as follows, R 2 = 0.987, q e = 24.64, K 2 (min −1 ) = 0.0342 and q max (mg g −1 ) = 112. 35. In GC-MS chromatogram, 2 peaks appeared at retention time (RT) of 9.14 and 15.13 min (Fig. 3). These peaks were designated "A" and "B". Peak "B" with retention time 15.13 min match with the chlorpyrifos. With the passage of time this peak decreased gradually. Peak "A" with the RT of 9.14 min match with TCP (Fig. 3). Characteristic fragment ion peak and molecular ion (m/z) also supported the identification of TCP. Peak "A" was most prominent around 7-15 days, later this peak started decreasing. The CP first converted into diethylthiophosphoric acid (DETP, m/z = 172) and TCP (m/z = 197) by hydrolysis (Table S3). Figure 4 presents the proposed degradation pathway of chlorpyrifos. The TCP ring was broken and it was completely mineralized without any accumulation of byproduct. The degradation products were non-toxic as they did not hinder the growth of Bacillus cereus Ct3. Very good growth was noted till 14th day of biodegradation experiment (Table S4). Similar pathway and metabolites were reported by other studies 22,45 . Hamzah et al. 10 reported 3,5,6-trichloro-2-pyridinol and CPF-oxon as metabolite of CP degradation by Pseudomonas aeruginosa. The metabolites identified in this study were diethyl phosphate and 3,5,6-trichloro-2-pyridinol, these 2 metabolites were also reported by Liu et al. 9 . Samuel et al. 46 calculated q max = 71.9 and R 2 = 0.99 for biosorption of chromium by Aspergillus niger. Table 3 reports the chlorpyrifos degradation efficiencies of different strains in different conditions. Plackett-Burman design was employed to investigate the effect of five factors. Table 1 represents the experimental design with the actual chlorpyrifos degradation values and the predicted chlorpyrifos degradation values. The correlation coefficient (R 2 ) between experimental and predicted value is 0.94, which is in reasonable agreement. Table 2 represents the ANOVA results for PBD, where P value less than 0.05 represents the significance of the factor for chlorpyrifos biodegradation. The polynomial equation for independent and dependent factors in this study could be written as: Base on the results of PBD following four significant factors were identified; concentration, temperature, carbon, inoculum size. Central composite design (CBD) was employed and its design matrix with thirty one predicted and experimental values of chlorpyrifos degradation is represented in Table 4. The second-degree polynomial equation obtained was as follows: The ANOVA result for CBD is presented in Table 5. It represents the individual main effects and the interaction among different factors. The "F-value" of the model was 1050.7 which suggest that the model was significant. For the present study the "lack of fit P value" was 0.00. Figure 5 represents the experimental values and the predicted values for the biodegradation of chlorpyrifos. The regression coefficient obtained was R 2 = 0.93 which indicate that the experimental vales and the predicted values are closely fitted. Figure 6a-f represents the combined effect of different factors for biodegradation of chlorpyrifos. The degradation increase with the increase in concentration, temperature, carbon source and inoculum but at certain point it reaches the maximum. Further increase in any factor beyond this point will decrease the degradation of chlorpyrifos. The most significant factors highlighted in ANOVA and surface response plots were concentration and inoculum size. Changes in concentration and inoculum size impacts more on degradation then other factors. This is in argument with the work reported by Zhou et al. 58 and Ungureanu et al. 59 , wherein the inoculum size was the important (4) Y = −0.51 − 0.0313 * concentration + 0.09 * temperature + 0.422pH + 1.07 * carbon + 2.737 * Inoculum.

Conclusion
Rapid increase in pesticide is negatively linked with ecological disturbance, decrease in soil fertility and changes in microbial community. Natural and eco-friendly techniques are gaining popularity to protect ecosystems from the side effects of pesticides. To restore chlorpyrifos contaminated soil, we isolated 156 strains from cotton growing Punjab region of Pakistan.    www.nature.com/scientificreports/ be used for bioremediation of chlorpyrifos contaminated soils. Future study will focus on the whole genome sequencing of Bacillus cereus Ct3, identification of genes involved in biodegradation and action of enzymes responsible.