Evidence for enhanced dissipation of chlorpyrifos in an agricultural soil inoculated with Serratia rubidaea strain ABS 10

The insecticide 14C-chlorpyrifos was found mineralized in a Tunisian soil with repeated exposure to it. From this soil, a bacterial strain was isolated that was able to grow in a minimal salt medium (MSM) supplemented with 25 mg L−1 of chlorpyrifos. It was characterized as Serratia rubidaea strain ABS 10 using morphological and biochemical analyses, as well as 16S rRNA sequencing. In a liquid culture, the S. rubidaea strain ABS 10 was able to dissipate chlorpyrifos almost entirely within 48 h of incubation. Although the S. rubidaea strain ABS 10 was able to grow in an MSM supplemented with chlorpyrifos and dissipate it in a liquid culture, it was not able to mineralize 14C-chlorpyrifos. Therefore, it can be concluded that the dissipation capability of this bacteria might be attributed to its capacity to adsorb CHL. It can also be ascribed to other reasons such as the formation of biogenic non-extractable residues. In both non-sterile and sterile soil inoculated with S. rubidaea strain ABS 10, chlorpyrifos was more rapidly dissipated than in controls with DT50 of 1.38 and 1.05 days, respectively.


Introduction
After the ban on organochlorines (such as DDT, dieldrin and heptachlor) and carbamates (such as carbofuran), organophosphorus insecticides (OPs) have been used extensively in agriculture as substitutes because of their high efficiency, supposedly relatively low persistence and fewer harmful effects on the environment (Yang et al. 2005). However, the widespread use of OPs has led to severe environmental issues because these compounds are often transported away from the sites where they have been applied. For instance, OPs may enter aquatic environments via soil percolation, air drift or surface runoff (Liang et al. 2011). Numerous reports indicate that OPs are toxic to humans, causing metabolic disorders and neuropathy in response to both acute and chronic exposure (Iyer and Makris, 2010). Of all the OPs, chlorpyriphos (CHL) is one of the most frequently used due to its broad-spectrum activity. CHL has relatively low water solubility (2 mg.L −1 ) and is spontaneously hydrolyzed to 3,5,6-trichloro-2-pyridinol (TCP) . Although applied on crop cover, a large proportion of this insecticide reaches the soil, where both abiotic (i.e., sorption chemical degradation) and biotic (biodegradation) processes control the fate and activity of this compound. The half-life (DT 50 ) of chlorpyrifos in soil is usually between 10 and 120 days, depending on soil type, climate and other environmental conditions such as composition of microbial communities (Abraham and Silambarasan 2016). Indeed, microbial activity has been deemed the most influential and significant way to remove OP pesticides (Li et al. 2005). The ability of microorganisms to degrade OPs is viewed as the primary means of removing these agrochemicals from soils (Cycon et al. 2013). A range of species of bacteria capable of degrading OPs especially CHL by co-metabolism or by using pesticides as a source of carbon and phosphorous (Cycon et al. 2013) has been reported by several researchers (Enterobacter strain B-14, Stenotrophomonas sp. Strain YC-, Sphingomonas sp. Strain Dsp-2, Paracoccus sp. strain TRP, Bacillus pumilus strain C2A1, Cupriavidus sp. DT-1, Alcaligenes faecalis, Flavobacterium sp, Klebsiella sp, Serratia sp, Pseudomonas sp) Chisti et al. 2013). Furthermore, it has been observed that repeated application of OPs led to enhanced biodegradation due to the selection of degrading microbial populations (Singh et al. 2003). More investigation on microbial degradation of the OPs is required in order to not only understand processes involved in their degradation but also to be able to develop bioremediation strategies to clean contaminated soils (Cycon et al. 2013). Indeed, microbe-based remediation relying on biosorption, bioaccumulation, biotransformation or biomineralization processes (Ayangbenro and Babalola 2017) has received increasing attention as it seems an applicable and cost-effective biotechnology to clean up soils polluted with OPs (Singh et al. 2006;Chen et al. 2011). On one hand, biotransformation and biomineralization both contribute to the transformation of the pollutant, the latter viewed as the gold standard as it leads to the complete transformation of the pollutant. Moreover, biosorption and bioaccumulation contribute to the stabilization of the pollutant in the environmental matrix by forming non-extractable residues that are almost not-transferable to other compartments of the environment. From this point of view, these last two processes are of interest in cleaning contaminated water by using microbial biomass as a low-cost biobased adsorbent (Khadivinia et al. 2014) or in stabilizing the pollutant in a given contaminated matrix in order to avoid further dispersion in the environment.
Hence, the present study aims not only to estimate the adaptation of the microbial community of an agricultural soil regularly exposed to CHL to its enhanced mineralization, but also to isolate and characterize bacterial isolates able to grow on CHL as the sole carbon source. The ability of one bacterial isolate to dissipate CHL was estimated in a liquid culture and in soil microcosms incubated under controlled conditions in the laboratory in order to estimate its interest for bioremediation purposes.

Chemicals and culture medium
The tested compounds CHL (99.5% purity) was purchased from Sigma-Aldrich (St. Louis, MO, USA). All other reagents (dimethyl sulfoxide DMSO and dichloromethane) used in this study were of high purity and analytical grade. Organic free water was prepared with a Milli-Q/Milli-Ro system (Millipore Corp., Bedford, MA, USA). Stock solutions of chlorpyrifos were prepared at 5000 mg L −1 in dimethyl sulfoxide (DMSO). For each experiment, 1 mL of CHL was added in the medium. For the microbial assays, mineral salt medium (MSM) and Luria-liquid broth (LB) were used. The medium (MSM) contained 20 mL of K 2 HPO 4 , 10 mL of MgSO 4 7H 2 O, 10 mL of NaCl, 1 mL of CaCl 2 , 2 g of boric acid, 0,2 g of ZnSO 4 , 0,8 g of CuSO 4 , 0,25 g of NaNO, and 0,05 g of CO(NO 2 ) 2 , 1 mL of FeSO 4 6H 2 O was added per liter of distilled water. The medium (LB) contained 10 g of peptone, 5 g of yeast extract and 10 g of NaCl per 1 L of distilled water (pH 7.0). After autoclaving (121 °C, 20 min) and cooling the medium was supplemented with a suitable CHL solution prepared as described above.

Sampling
The soil sample used in this study was collected from field site located in Mornag approximately 20 km away from Tunis (36° 67′ 70.39 N, 10° 27′ 53.78 E), Tunisia. The sampling site has been in use for intensive agricultural practices since long and this soil has received leachates after application of chlorpyrifos for several years. Soil sample was collected from a depth of 20 cm. Soil was mixed thoroughly and plant debris were removed. Then soil was sieved at 2 mm and kept at 4 °C until its use (for less than 3 weeks).

Mineralization of 14 C-CHL in soil microcosms
The potential of the soil microbial community to mineralize chlorpyriphos (CHL) was evaluated using 14 C-labelled CHL, the molecule used in this study was marked on the cycle, (Izotop, specific activity = 118 µCi/mg) as described by El-Sebai et al. (2007). Four individual replicates of 20 g soil microcosms were studied. For each replicate 1 mL of a methanol solution comprising 60 mg.L −1 of 14 C-CHL and 0.068 µ Ci of 14 C-CHL was applied on an aliquot of 1 g of dried soil. After evaporation CHL contaminated soil was thoroughly mixed with the rest of soil. Soil humidity was then adjusted to 40% of the water holding capacity (WHC) and kept constant all along the 70 days of incubation in the radiorespirometer. 14 CO 2 evolved from 14 C-CHL was quantified by liquid scintillation counting (LS 6500 Multi-Purpose Scintillation Counter, Beckman) using ACSII scintillation fluid (Amersham) (Storck et al. 2017).

Enrichment procedure, isolation, and characterization of chlorpyrifos degrading strain
Preliminary screening experiments were performed to obtain strains that were tolerant to CHL. The study was conducted as described previously by Ben Salem et al. 2016.
Only one soil with three replicates was used in this study. Fresh soil sample was divided in six subsamples of 50 g dry weight equivalent. Aqueous CHL solution was prepared at 4.8 g. L −1 (which corresponds to 10 times the recommended dose agronomic purpose) the day of its application starting from Robust ® formulated solution. The duration of experiment was 2 weeks.
Soil samples were incubated at room temperature under laboratory conditions. They were watered every 2 days to keep soil humidity at 40% of water holding capacity (WHC). Every 2 days, they were irrigated with 10 mL of commercial chlorpyrifos solutions as 10 × concentrated solutions of pesticides to exert a selective pressure favorable to the emergence of degrading bacterial populations. Three samples were not treated with CHL but with equivalent amount of pure water (control). Two days after CHL application, soil was sampled to immediately carry bacterial isolation. Briefly, one g of soil was added to 10 mL of physiological water (NaCl 9‰) and serially ten-times diluted. 100 µL of 10, 10 -2 , and 10 -4 dilutions were streaked on PCA plates that were incubated for 16 h at 37 °C. Each colony growing on the plate was purified using the Z streak technique.
Primary distinction between all the isolates was based on the size, color, and morphology of their colonies on the PCA plates. Bacterial colonies showing different morphologies were selected and further characterized using API 20E (Biomérieux, Lyon, France) following the recommendations of the Bergey's manual of systematic bacteriology (Krieg and Holt 1984). In addition, 16S rRNA amplicon generated by PCR using the 27f and 1492r universal primer pair (Gürtler and Stanisich 1996) from DNA extracted from the bacterial isolate was sequenced as previously described (Devers et al. 2008). 16S rRNA sequence was compared to other sequences available in GenBank database (http: //www. ncbi.nlm.nih.gov/ genbank) using the BlastN search analysis (http:// www. blast. ncbi. nlm. nih. gov/). Sequence was deposited in the Genbank database (SUB8916609 S10 MW494965).

Bacterial growth kinetics in different liquid media
To investigate the growth of Serratia rubidaea strain ABS 10 with chlorpyrifos, 200 µl of strain (OD 600 = 0.8) were inoculated into 20 mL of either MSM or nutrient broth medium added with CHL at a final concentration of 25 mg.L −1 . The culture was incubated at 30 ± 2° C on a rotary shaker at 120 rpm. The bacterial growth was regularly monitored for 5 days by measuring the turbidity of the culture using a spectrophotometer at 600 nm.

Inoculum preparation
A bacterial colony was inoculated in LB medium and grown to reach the exponential phase. It was then collected by centrifugation at 5000 g for 5 min. The cell pellet was washed twice with 0.9% of sterile NaCl and then re-suspended in NaCl to obtain the bacterial suspension at a concentration of approximately 3.10 6 CFU / mL. The cell density (OD 600 nm) was measured using UV-Visible spectrophotometer (Lu et al. 2013).

Dissipation of chlorpyrifos by S. rubidaea strain ABS 10 in aqueous medium
CHL dissipation studies were performed in 250-mL Erlenmeyer flasks containing 100 mL of sterile MSM supplemented with CHL at 25 mg L 1 and inoculated with 1 mL of bacterial strain (approximately 3 × 10 6 cells.mL −1 ). Uninoculated media comprising with the same concentration of CHL were used as control. All the samples were incubated at 30 ± 2 °C on a rotary shaker at 120 rpm for 5 days. Samples were periodically taken from the culture under aseptically conditions to measure the remaining pesticide concentration the culture medium.

Dissipation of chlorpyrifos by S.rubidaea strain ABS 10 in soil
To study the dissipation of chlorpyrifos in sterilized (SS) or natural (NS) soil inoculated with S. rubidaea strain ABS 10 (B) or not S. rubidaea strain ABS 10. Briefly, 100 g of sterilized soil (SS) or natural soil (NS) was placed in a 250-mL Erlenmeyer flask, inoculated or not with 30 mL of a S. rubidaea strain ABS 10 suspension (B) containing 3 × 10 6 cells.mL −1 and treated with CHL at 25 mg kg −1 . The amount of carbon, nitrogen, and phosphorous were calculated using the relationship C/N/P 100:10:1. The sources of carbon, nitrogen, and phosphorous were glucose, (NH 4 ) 2 SO 4 and K 2 HPO 4 respectively ). Sterilized soil not inoculated with S. rubidaea strain ABS 10 was used to estimate abiotic dissipation of CHL (Pino and Penuela 2011). All flasks were incubated in an incubator at 30 ± 2 • C. Samples were periodically removed aseptically to determine the pesticide concentration. Each treatment was performed in triplicate (n tot = 12).

Analytical methods
At regular intervals, 5-10 mL cultures were withdrawn from aqueous medium and centrifuged at 7200 × g for 10 min to obtain a cell-free medium. CHL was extracted twice from the supernatant with an equal volume of dichloromethane (DCM). Organic layers of DCM were pooled and evaporated at 28 ± 2 °C. For the analysis of CHL in soil, 5 g of soil samples was mixed with 10 mL of dichloromethane. The samples were ultra-sonicated for 30 min at 30 °C. After that, the mixture was centrifuged for 30 min on a rotary shaker at 120 rpm. Then the samples were allowed to stand until the soil had settled, and the clear supernatant was used to determine the pesticide concentration by GC-MS. Levels of CHL were measured by GC-MS using an Agilent 6850 N gas chromatograph (Agilent Technologies, USA), equipped with an Agilent6973 MS detector. A capillary column HP-5MS (30 m, 0.25 mm, 0.50 mm) was used while chromatographic separation was achieved with the following method: the GC oven temperature was initially set at 70 °C, for 2 min, and raised to 270 °C at a rate of 20 °C/min and held for 10 min. The injector and detector were set at 250 and 280 °C, respectively. The carrier gas Helium was used as at a constant flow rate of 1 mL/min. Electron impact (70 eV) mass spectra were recorded from 100 to 550 amu (atomic mass unit).

Data analysis
The four kinetic models proposed by the FOCUS working group on pesticide degradation kinetics (FOCUS 2006) were used to calculate pesticide dissipation kinetic parameters. The four kinetic models proposed by the FOCUS working group on pesticide degradation kinetics (FOCUS 2006) were used to calculate pesticide dissipation kinetic parameters: the single first order kinetic model (SFO), and the biphasic models hockey stick (HS), first order multi-compartment model (FOMC) and double first order in parallel model (DFOP).
Data obtained from the dissipation experiments were fit to the exponential decay model. The first-order model and the DT 50 was calculated as follows: (1) C t = C 0 e −kt where C t is the concentration of pesticide remaining in MSM or soil after t days, C 0 is the initial concentration of pesticide in MSM or soil. k and t are the rate constant (d −1 ) and degradation time in days respectively (Focus 2006).

Statistical analysis
All the experiments were performed in triplicates. The data were statistically analyzed using two-way analysis of variance (ANOVA). When significant differences test (P ≤ 0.05) were observed, the means were separated using Graphpad Prism, v7.00.

Mineralization of 14 C-CHL in soil microcosms
After a 10-day lag-phase, the soil microbial community efficiently mineralized 14 C-CHL reaching up to 70% of mineralization of initial radioactivity applied (Fig. 1). During the exponential phase, the rate of mineralization was 1.3% 14 C-CO 2 per day. The mineralization curve had sigmoid shape characteristics from microbial community adapted to enhanced degradation of pesticides. One could hypothesize that in response to the repeated CHL treatments applied to this arable soil the microbial community adapted to its enhanced biodegradation that provides nutrient and energy sources for the growth CHL-degrading community.
Indeed, the emergence of degrading microorganisms, among which CHL-degraders, has been observed in soils regularly exposed to different pesticides (Singh 2009). The use of pesticides as nutrient and energy sources provides a selective advantage over other microorganisms (Copley 2009). Adaptation to enhanced degradation seems to be a common trait to soil microorganisms in response to repeated pesticide treatments (Crouzet et al. 2010;De Andrea et al. 2003;Hussain et al. 2011;Vischetti et al. 2008;Weaver et al. 2007). This is an environmental-friendly functional trait because it decreases the persistence of pesticide residues in the soil as well as their dispersion in the environment and their ecotoxicological impact to non-target organisms and supported ecosystem services (Topp et al. 2004). Furthermore, it is important to know the position of the marking used because it is known that the cycle of chlorpyrifos is very difficult to cleave. It is thus necessary to specify if the marking was on the cycle which was the case of studied molecule or on the chains. Therefore, in the studied soil there is about 60% of the amount of CHL initially brought in that sees its cycle cleaved and is mineralized.
Other studies also confirmed this hypothesis.The position of the labeled atom in the respective tracer molecule also affects the NER results (Barriuso et al. 2008;Gaultier et al. 2008). For example, if the label is positioned in a molecular moiety of a contaminant, that may easily be cleaved and evolve as 14 CO 2 . For example, in a carboxylic group, the mineralization of the complete molecule will be overestimated, whereas the detected amounts of non-extractable residue (NER) tend to be low (Barriuso et al. 2008).
In addition, the position of the labeled C atom within the parent molecule requires consideration in terms of incorporation into biomass. For instance, due to the oxidation state of the C atoms in the triazine ring of atrazine, they will not be incorporated into microbial biomass (Struthers et al. 1998) but will be released as CO 2 , which then may be assimilated by microorganisms, since CO 2 fixation is also a relevant process in soils contributing to the bioNER formation.
All this could explain the mineralization of CHL in soil microcosms and obtained resutls was in aggreement with Kastner et al. 2014. High mineralization of a compound in a short time course is mostly accompanied by microbial degradation with the formation of metabolites and microbial biomass. This degradation generally results in high probability for biogenic NER 14 C label positions in carboxylic moieties or other highly oxidized positions of the parent compound are an exception from this rule, since the label in these positions is lost as CO 2 even without any further transformation of the compound (Kastner et al. 2014).

Isolation and characterization of bacterial strain growing on CHL medium
In order to isolate bacterial strain able to transform CHL, enrichment cultures on MSM medium added with CHL were conducted. In total we have been able to isolate four bacterial isolates able to grow on MSM-CHL medium. Among these isolates, only one has the capability to grow on liquid MSM-CHL. This isolate is aerobic, none spore forming, Gram negative, straight rods with rounded ends bacteria producing small circular colonies on the nutrient agar plates. Its freshly grown culture showed positive tests for oxidase, catalase, and exhibited the ability of nitrate reduction (Table 1). Its 16S rRNA gene sequence is 99% similar to Serratia rubidaea strain NBRC 103,169 (Ac n° NR_114232), JCM1240 (Ac n° NR_024644), and DSM 4480 (Ac n° NR_114716). Consequently, we proposed to name it Serratia rubidaea strain ABS S10 (S10).The phylogenetic analysis of 16SrRNA sequence of Serratia rubiaea is presented in Fig. 2.
Moreover, the growth of S10 was monitored in MSM-CHL and NB-CHL media (containing 25 mg.L −1 of CHL) (Fig. 3). As expected S10 was unable to grow on MSM not supplemented with CHL. Contrariwise it slightly grew on MSM supplemented with CHL as sole carbon source reaching 0.06 ± 2 a.u. after 72 h. Likewise, the growth of S. rubidaea strain ABS 10 was promoted by approximately a factor two in the NB medium complemented with CHL as compared to the control. In both cases, a sharp increase in growth was observed up to 1 day and the maximum growth was obtained after 3 days. These results suggest that S10 can use CHL for its growth. However, on mineral salt medium the culture poorly grew suggesting some metabolic limitations. This was confirmed by the fact that S. rubidaea strain ABS 10 was unable to mineralize 14 C-CHL labelled on the pyridine ring, suggesting that this strain was not able to get access to C of the ring. Our observation is in accordance with earlier studies reporting that a range of bacterial isolates such as, Serratia sp (Xu et al. 2007) Sphingobacterium sp, Alcaligens sp ) Serratia marcescens (Cycon et al. 2013) Pseudomonas kilonensi (Khalid et al. 2016) able to grow in minimum salt medium supplemented with CHL as sole carbon source. In addition, the supplementation of MSM-CHL medium with simple C source such as glucose or sucrose or more complex one such as NB was shown to promote the growth of Pseudomonas kilonensi SRK1 suggesting that easily degradable C source can fuel CHL degradation (Khalid et al. 2016). In fact, CHL has reported to be degraded by bacteria co-metabolically that required additional carbon sources (Singh et al. 2006;Xu et al. 2008;). Among Serratia species, S. marcescens was shown to be able to use CHL concentration of 50 mg.L −1 as the only carbon source when grown in MSM ). Furthermore, others species of Serratia were characterized for their high potential to grow in MSM supplemented with another OPs such as diazinon (Abo-Amer 2011).

Dissipation of CHL by strain S. rubidaea strain ABS 10 in liquid medium
The dissipation of CHL by S10 culture was assessed in resting cell experiment (Fig. 4). S10 rapidly dissipated CHL with a rate of approximately 2.03 mg CHL per day. Within 24 h, 92% of CHL initially added was dissipated by S. rubidaea strain ABS 10 while only 20% of dissipation was observed in the control, which consists of sterile medium not inoculated with this strain. It is noteworthy that as expected CHL was also dissipated in the control but at slower rate than in the S. rubidaea strain ABS 10 culture, and after 5 days of incubation 50% of the initial dose of chlorpyrifos remains. CHL half-life in S10 culture was significantly lower than in the control (1.15 vs 4.95 days, respectively) thereby confirming its ability to dissipate it. Keeping in mind that S. rubidaea strain ABS 10 was unable to mineralize 14 C-CHL, one could suggest that the rapid CHL dissipation in S. rubidaea strain ABS 10 culture was not due to its mineralization but probably to its adsorption on bacterial cells as it was previously shown for the herbicide 2,4-D (Benoit et al. 1998) or the insecticide chlordecone (Merlin et al. 2014) on fungal biomass. Indeed, giving the fact that CHL is highly hydrophobic, it has strong affinity to phospholipid bilayer constituting microbial cell membrane on which it can sorb. CHL was previously reported to be hydrolyzed co-metabolically by microorganisms (Chisti, et al. 2013) fueled by other C sources than CHL which, in this case, does not constitute a source of C or energy to sustain its growth (Singh et al. 2006;Xu et al. 2008;).

Dissipation of CHL in soil
Dissipation of CHL was monitored in sterile- (Fig. 5, panels A) and native-soil (Fig. 5, panels B) microcosms inoculated with S10 or not (control). Similar CHL dissipation kinetics were observed in non-inoculated sterile and native soils. These two kinetics of dissipation were biphasic and CHL DT 50 was estimated to 2.1 days for both treatments. Despite the fact that the native soil microflora is able to mineralize this insecticides (CHL), no differences in the dissipation kinetics observed in sterile-and native-soils treated CHL). This apparent discrepancy is explained by the fact that the dissipation was monitored only during a 6 days period of time during which mineralization of 14 C-CHL is very low (Fig. 1). Therefore, over this short period of time CHL mineralization does not contribute to observed dissipation which is most likely mainly governed by abiotic processes such as sorption on soil components. The inoculation of S10 in the sterile and native soils resulted in a marked increase in the rate of CHL dissipation as compared to their respective control (Fig. 5). In fact, chlorpyrifos residues were extracted from the supernatant of microbial cultures or from the soils and quantified by GC-MS to monitor its dissipation overtime (Fig. 6).
In the inoculated native and sterile soils, the CHL DT 50 were estimated to 1.1 days and 1.4 days, respectively (Table 2)  . Having in mind that the dissipation of CHL in S10 culture was controlled by its sorption to bacterial cells, the improvement of the dissipation of CHL observed in inoculated soils might be attributed to its biosorption on microbial cells. This hypothesis is supported by the hydrophobic nature of CHL which provides to it a strong lipophilicity compatible with the  (Angelova and Schumauder 1999).
In this short-term experiment, the dissipation of CHL was mainly governed by abiotic processes which combined the sorption to soil components and the biosorption to S. rubidaea strain ABS 10 cells (3.10 6 CFU.mL −1 ).
Several microorganisms including bacteria, fungi, and algae have been already reported as effective biosorbents for removal of dyes, metal and even pesticides due to its low cost, non-toxic approach regeneration capability and high efficiency for pollutant uptake (Pathak and Dikshit 2011). Particularly, bacteria have been used as biosorbents owing to their ubiquity, size, and ability to grow under controlled conditions and resilience to an extensive range of environmental conditions (Ayangbenro and Babalola 2017). Various heavy metals have been tested on bacteria species such as Pseudomonas, Enterobacter, Bacillus and Micrococcus species. Their excellent sorption capacity is due to their high surfaceto-volume ratios and their numerous potential active chemosorption sites, such as the teichoic acid on the cell wall (Mosa et al. 2016). Likewise, Zakeri et al. 2010 reported also that Serratia sp was an efficient radium biosorbent and might be appropriate candidate for designing biosorption remediation system.
On the other hand, the dissipation capability of this bacterium might be ascribed to the formation of biogenic nonextractable residues since major contributions of biogenic residues to NER formation are to be expected, if the respective organic contaminant is degraded under significant formation of CO 2 (Kästner et al. 2014).
In fact, Kästner et al. 2014, reported that the contaminants can be volatilized or leached to the groundwater, sequestered (sorbed or entrapped), or immobilized as nonextractible residues (NER) via binding to soil components, degraded abiotically, taken up by living organisms, and/or biodegraded by microorganisms. In principle, NER may be formed from parent xenobiotics and metabolites (according to the IUPAC definition) or from labeled carbon converted to microbial biomass and even from labeled CO 2 produced during biodegradation To sum up, xenobiotic NER derived from parent pesticides and primary metabolites sorbed or entrapped within the soil organic matter (Type I) or covalently bound (Type II) pose a considerably higher risk than those derived from productive biodegradation. However, biogenic non-extractable residues (bioNER) (Type III) resulting from conversion of carbon (or nitrogen) from the compounds into microbial biomass molecules do not pose any risk for the environnement (Kästner et al. 2014).

Conclusion
This study showed that in response to long-term exposure to CHL, the soil microflora adapted to its enhanced mineralization. S. rubidaea strain ABS 10, a bacterial strain able to grow in a mineral salt medium added with CHL as the sole carbon source, was isolated from this soil. Although this strain was able to rapidly dissipate CHL in a liquid culture, it was not able to mineralize 14 C-CHL labelled on the pyrazine ring. In addition, S10 was shown to be a good biosorbent, able to fully dissipate CHL within 1 day both in a liquid medium and in soil microcosms. The present study offers new insight in the development of a remediation technology of CHL and other hydrophobic pollutants, based on the use of S10 as a biosorbent. Prospectively, the monitoring of chlorpyriphos degradation products such as TCP needs further research.
Funding This work was in part financially supported by the collaborative project PHC-MAGHREB (32618SA, CMCU), by Bayer CropScience and ANGed (National Agency of management of waste).

Data availability
The datasets from which the current study was created are available from the corresponding author on reasonable request.