Soil samples were collected in Nova Esperança, a rural farming community located within the limits of Manaus, AM, Brazil, in the east area of the city, following ISO/FDIS 10381^-1: 2001 guidelines, which consists of a sampling in X shape design, in the following geographical coordinates: sample 1: S3º00'56".5 W59º55'14.0"; sample 2: S3º00'56.9" W59º55'13.7"; sample 3: S3º00'57.7" W59º55'13.5"; sample 4: S3º00'57.7" W59º55'14.2"; sample 5: S3º00'57.3" W59º55'14.1".

Portions (6.6 g) from 5 samples of soil were homogenized totalizing 33 g of a composite soil sample, which was inoculated in a 2 L Erlenmeyer flask containing 500 mL of minimal salts medium - MMS (K₂HPO₄ (2.78 g L-1), (NH₄) 2SO4 (1.0 g L^-1), MgCl₂.6H₂O (0.2 g L^-1), NaCl (0.1 g L^-1), FeCl₃.6H₂O (0.02 g L^-1), CaCl2 (0.01 g L^-1) - to 1 L) and carbendazim - MBC (250 µg mL^-1). A negative control was prepared containing only MMS (500 mL) and carbendazim (250 µg mL^-1), without the soil sample. The flasks were incubated at 28°C under agitation (125 rpm) for seven days for rigorous selection process. After that period, aliquots of 100 mL were removed from each flask and inoculated again in 400 mL of MMS and MBC for more seven days, under the same conditions. This procedure was repeated for two more times. After the last incubation period, aliquots of 1 mL from each flask were submitted to serial dilution (10^-1, 10^-2, 10^-3, 10^-4), using 0.85% saline solution. Aliquots (100 µL) from each dilution were plated on Petri dishes containing MMS agar + MBC (250 µg mL^-1). The experiments were carried out in triplicate. With the aid of a Drigalski handle, the cells were uniformly distributed on the plates which were incubated in BOD at 28°C for ten days to isolate bacteria and fungi. The bacteria which grew were transferred to test tubes containing 5 mL Luria Bertani - LB agar (Himedia) culture medium.

The bacteria were identified following method described by Mota, Back-Brito, and Nóbrega (2009). Amplification of 16S rDNA was performed by PCR using 095F (5' - GTG CCA GCM GCN GGC G - 3') and 094R (5' - GGY TAC CTT GTT ACG ACT T - 3') primers. The sequencing was carried out on an ABI 3500xL automated sequencer (Applied Biosystems^®). Sequences alignments were edited using MEGA6 software (Tamura, Stecher, Peterson, Filipski, & Kumar, 2013) and combined clustering/similarity methods (BLASTn/neighbor joining) were used for identification and comparison with GenBank database sequences.

The bacterial inoculum was prepared from stock cultures LB kept at -20°C, which were transferred to plates containing LB and incubated at 30°C for 24 hours. With the aid of a platinum handle, an aliquot containing approximately 10⁶ CFU mL^-1 bacterial suspension was transferred to 15 mL Falcon type tubes containing 3 mL of LB broth and were incubated at 30°C and under agitation (250 rpm) for 4 hours. The resultant culture was centrifuged (1792 rcf) for 10 minutes. The supernatant was discarded and the cell pellet suspended with MMS. Bacterial biomass was diluted in MMS to obtain O.D. equal to ^-1 absorbance at 610 nm, which corresponds to a population of 10⁹ CFU mL^-1. In the biodegradation tests were used aliquots (1 mL) of bacterial inoculum suspension. In

Degradation tests were carried out with the two finest strains selected after preliminary tests for evaluation of bacterial growth measured by optical density after 72 hours of culture. Concurrently, the degradation was quantified by high-performance liquid chromatography - HPLC. The biodegradation experiments with each isolate or with a consortium of both bacteria were performed in triplicate. Aliquots (1 mL) of the bacterial suspension (D.O. 1) were inoculated in 125 mL Erlenmeyer flask containing MMS (50 mL) and 250 mg L^-1 MBC (97% purity, Sigma-Aldrich) diluted in dimethylsulfoxide (DMSO) and incubated at 30°C under agitation (125 rpm) for 21 days. The positive control group (LB and bacteria) was used to monitor the bacterium viability.

The extraction of MBC was performed using ethyl acetate. To each Erlenmeyer flask, 50 mL of ethyl acetate was added. The material was transferred to a 125 mL separatory funnel and subjected to vigorous agitation for 1 minute. The organic phase was collected and the aqueous phase was extracted twice more, by the addition of 50 mL of ethyl acetate. The end of the extraction was monitored by thin layer chromatography (TLC). In the collected organic phase, anhydrous sodium sulfate was added and after filtration the samples were subjected to vacuum rotary evaporator at 40°C until reduction of the volume of the sample to approximately 10 mL. The sample was transferred to 10 mL volumetric flask containing 10 μg mL^-1 of an internal standard solution. The volume of 1 μL of the sample was injected into HPLC.

The internal standard and surrogate solutions were added with phenanthrene-d10 (phe-d10) and naphthalene d8 (nap-d8) with a purity ≥ 98% (Sigma-Aldrich, 98%), solubilized in HPLC grade hexane. The internal standard (nap-d8) was inserted in the samples diluted in a volumetric flask before injected into HPLC. The surrogate (phe-d10) was inserted in Erlenmeyer flasks before starting the extraction process. The use of these patterns is needed to recognize possible losses during the extraction process. The chromatographic area of the analyte was corrected by extraction recovery rate, which was obtained by the formula: [(CAs/CAis] x 100, where CAis corresponds to the chromatographic area of the internal standard and the CAs for the chromatographic area of the surrogate. The MBC concentration was obtained by a calibration curve using different MBC concentrations: 100, 200, 300, 500, 700, 900 and 1200 μg mL^-1. The percentage of MBC degradation was calculated by the formula: [(CC - CE)/CC] x 100, where CC is the MBC concentration in the control group and CE the remaining concentration of MBC from experiment. Finally, it was calculated the average and the standard deviation of the biodegradation rate in order to obtain greater accuracy of MBC concentration.

HPLC analyses of MBC were performed in a Shimadzu (CBM–20A) Prominence^® UFLC system, coupled with high-pressure binary pump (LC-20AD); auto injector (SIL-20AHT), detector UV-Vis (SPD-20A); column oven heating; CTO-20A. The column used was Phenomenex Luna C18 (250 mm x 4.6 mm, 5u). The Solvent A: ultrapure water/0.1% formic acid and the Solvent B: acetonitrile grade HPLC/0.1% formic acid. Flow: 1.0 mL min^-1. Gradient: 0-2 min, linear gradient from 60 to 85% B; 2-8 min, isocratic at 85% B; 8-9 min, linear gradient from 85 to 60% B. Wavelength set at 285 nm. Retention times at 1.9; 5.9 and 7.9 min for MBC, naphthalene and phenantrene, respectively.

Following the biodegradation assays, a toxicity test was carried out with Artemia salina in triplicates for both strains and the consortium as described by Atayde, Carneiro, Martins, and Palheta (2011). Determination of lethal concentration (LC₅₀) values was performed as described by Harwig & Scott (1971) in triplicate only for extracts that showed toxicity.

Coriandrum sativum rhizosphere soil inoculated in selective medium containing carbendazim as a sole carbon source enabled the isolation of 80 bacteria. It was possible to identify 60 strains and 52 different possible species. Table 1 shows the 20 most common bacteria identified by rDNA sequencing. The predominant genera were Bacillus sp., Ochrobactrum sp., Enterobacter sp. and Stenotrophomonas sp. Among the identified isolates, the genus Bacillus was the most representative, with 28% frequency. From an area adjacent to the area investigated in this study, which had been treated with the fungicide Mancozeb, Rebière (2015) isolated 23 bacteria from rhizosphere soil of Allium fistulosum L. (Welsh onion) crops of which 81.81% were identified as Bacillus sp. Among 23 bacteria isolated from rhizosphere soil collected in China, Hu, Zhao, Liu, Song, and You (2013) identified the genus Ochrobactrum which is able to grow in MMS medium added with the insecticide Imidacloprid (50 µg mL^-1).

After 72 hours of culture in medium containing carbendazim as sole carbon source, two bacterial isolates were selected (Stenotrophomonas sp. and Ochrobactrum sp.) for qualitative and quantitative biodegradation tests. The two selected genera had their sequence annotation submitted to the GenBank-NCBI and were deposited as Stenotrophomonas sp. (accession number KX376306) and Ochrobactrum sp. (accession number KX376307).

The genus Stenotrophomonas sp. is found in harsh environments such as Antarctica (Vazquez, Ruberto, & Mac Cormack, 2005) and can produce bioactive substances, such as laccase enzyme (Galai, Limam, & Marzouki, 2009). Some bacterial strains can degrade PAHs (Zafra, Absalón, Cuevas, & Cortés-Espinosa, 2014), as well as the fungicide Chlorothalonil (Zhang et al., 2014). Stenotrophomonas sp. and Ochrobactrum sp. can degrade Chlorpyrifos (Talwar, Mulla, & Ninnekar, 2014; Deng et al., 2015.), but there is no report on the efficiency of Stenotrophomonas sp. and Ochrobactrum sp. to degrade carbendazim. Therefore, it is important further studies to validate the potential of those microorganisms for degrading carbendazim in Amazonian soils.

Table 1.
The twenty major bacterial isolates able to grow in selective medium containing carbendazim as sole carbon source.

The growth of the bacterial isolates, cultured alone or in consortium, for 21 days in culture medium containing carbendazim, was determined by optical density and chromatographic methods.

The Stenotrophomonas sp. and the consortium of both Stenotrophomonas sp. and Ochrobactrum sp. when cultured at a concentration of 250 µg mL^-1 carbendazim showed the best growth rates (Figure 1). Zhang et al. (2014) evaluated the ability of Stenotrophomonas sp. strain to degrade the fungicide chlorothalonil cultured in minimal salt medium added with chlorothalonil (20 µg mL^-1) for 7 days. Stenotrophomonas sp. optical density ranged 0.17.

The obtained results indicate that both Stenotrophomonas sp. and Ochrobactrum sp. can degrade carbendazim without supplementary sources of carbon. Fang et al. (2010), Zhang et al. (2013), and Salunkhe et al. (2014) showed that bacterial strains used in biodegradation tests can be isolated using carbendazim as the only source of carbon. According to Fang et al. (2010), such characteristics are required when isolating microorganisms to be used for bioremediation purposes.

Figure 1.
Optical density of culture medium containing Ochrobactrum sp. or Stenotrophomonas sp. bacterial isolates from Coriandrum sativum L. rhizosphere cultured for 21 days at 30°C under agitation (125 rpm) separately or in consortium. Error bars represent standard errors of the means.

Carbendazim biodegradation analyzed by HPLC

HPLC analysis of extracts from Stenotrophomonas sp. cultures indicated the higher potential (68.9%) of this bacterial isolate to degrade carbendazim, if compared to the isolate Ochrobactrum sp. cultured separately or in consortium with Stenotrophomonas sp. (Figure 2).

Figure 2.
Biodegradation of carbendazim in 21 days by bacteria isolated from Coriandrum sativum L. rhizosphere. Error bars represent standard errors of the means.

Several bacteria efficient in degrading MBC have been cited in the literature, such as Pseudomonas sp. (Xiao et al., 2013), Bacillus subtilis TL-171 (Salunkhe et al., 2014), and Brevibacillus borstelensis (Arya & Sharma, 2015). Salunkhe et al. (2014) reported that four strains of B. subtilis individually degraded from 75.7 to 95.2% MBC in a concentration of 10 µg mL^-1 after 15 days of incubation. The concentration of MBC used to validate the degradation potential by the bacterial isolates Stenotrophomonas sp. and Ochrobactrum sp. was 25% higher than that used by Salunkhe et al. (2014).

The lowest degradation percentage showed by the consortium of both bacterial isolates may have occurred due to competition for nutrients caused by the co-cultivation of isolates of different genera. However, Arya & Sharma (2015) investigating the degradation ability of bacteria alone or in consortium found a significant reduction of MBC caused by bacteria in consortium. Luo et al. (2009) isolated three marine bacterial lineages: Ochrobactrum sp., Stenotrophomonas sp., and Pseudomonas sp. to degrade benzopyrene an aromatic polycyclic hydrocarbon of high molecular weight. Faster degradation was observed when the isolates were co-cultured in consortium.

The obtained results indicate that the isolate Stenotrophomonas sp. cultured alone shows greater potential to degrade carbendazim if compared to Ochrobactrum sp. MBC degradation potential, no matter if cultured alone or in consortium with Stenotrophomonas sp.

Toxicity tests using Artemia salina

Values obtained in toxicity tests carried out with A. salina (Table 2) ranged 0-0.7% using Stenotrophomonas sp. extract and the consortium of both isolates. Harwig & Scott (1971) established criteria for the extract toxicity classification using only the number of microcrustaceans and considered as non-toxic (NT), mortality less than 0.9%. However, Ochrobactrum sp. extracts showed slight toxicity (LT).

Table 2.
Toxicity tests using Artemia salina

M% - mortality percentage

From the result with the Ochrobactrum sp., culture extracts of this lineage were tested in A. salina at concentrations of 50, 25, 13, 7 and 3% to determine the lethal concentration (LC50), median lethal dose values indicated that extracts from this isolate are not toxic in all tested concentrations.

jucileuzasantos@hotmail.com