An enrichment culture technique was used to isolate bacteria responsible for the enhanced biodegradation of ethoprophos in a soil from Northern Greece. Restriction fragment length polymorphism patterns of the 16S rRNA gene, partial 16S rRNA sequence analysis, and sodium dodecylsulfate–polyacrylamide gel electrophoresis total protein profile analysis were used to characterise the isolated bacteria. Two of the three ethoprophos-degrading cultures were pure and both isolates were classified as strains of Pseudomonas putida (epI and epII). The third culture comprised three distinct components, a strain identical to P. putida epI and two strains with 16S rRNA sequence similarity to Enterobacter strains. Isolate epI effectively removed a fresh ethoprophos addition from both fumigated and non-fumigated soil when introduced at high inoculum density, but removed it only from fumigated soil at low inoculum density. Isolates epI and epII degraded cadusafos, isazofos, isofenphos and fenamiphos, but only at a slow rate. This high substrate specificity was attributed to minor (cadusafos), or major (isazofos, isofenphos, fenamiphos) structural differences from ethoprophos. Studies with 14C-labelled ethoprophos indicated that isolates epI and epII degraded the nematicide by removing the S-propyl moiety.
Ethoprophos (O-ethyl S,S-dipropyl phosphorodithioate) is an organophosphorus insecticide–nematicide, used for the control of potato cyst nematode Globodera rostochiensis (Wollemn) and Globodera pallida Stone in potato cultivation. It is applied as granules prior to planting, and incorporated into the first 10–15 cm of the surface soil. Degradation half-lives in soil ranging from 3 to 30 days have been reported, depending on soil and environmental conditions . As with other organophosphorus pesticides, microbial degradation is the primary mechanism of ethoprophos dissipation in soil . Since the early 1980s, there have been several reports correlating reduction in the biological efficacy of some organophosphorus nematicides with enhanced biodegradation following their repeated application at the same site. Nematicides which are reported to be susceptible to enhanced biodegradation included fenamiphos [3,4], fonofos  and isofenphos . Ethoprophos also appears to be susceptible to enhanced biodegradation, and several reports have noted a significant reduction in its biological efficacy in fields with a previous history of ethoprophos exposure [7,8]. In our previous studies, enhanced biodegradation of ethoprophos in soil from potato fields in Northern Greece was demonstrated in studies involving chemical residue analysis and bioassays with nematodes .
Several microorganisms, either in pure or mixed culture, have been isolated from soils exhibiting enhanced biodegradation of organophosphorus nematicides. Ou and Thomas  isolated a consortium of six bacteria, which was able to rapidly metabolise and inactivate fenamiphos. Similarly, Racke and Coats  reported the isolation of a Pseudomonas strain which was able to rapidly degrade isofenphos and use it as a sole carbon source, although it was unable to degrade and grow on other organophosphates, like fonofos, ethoprophos, terbufos or chlorpyrifos. However, to date, there have been no reports of the isolation of microorganisms able to degrade ethoprophos.
This paper reports the isolation and characterisation of ethoprophos-degrading microorganisms using molecular based methods, their ability to degrade ethoprophos when inoculated in sterilised or natural soils, their substrate specificity, and their metabolic activity towards ethoprophos.
2Materials and methods
The soil used for isolation of bacteria was collected from a field site in Northern Greece, which had been exposed to a standard ethoprophos dose of 7–10 kg ai ha−1 almost annually for the last 30 years. Enhanced biodegradation of ethoprophos was evident in this soil in previous studies [9,12]. It had a pH of 6.1, organic matter content of 2.72% (w/w), and moisture content of 21% (ca. −33 kPa). The soil used in the studies of degradation by the isolates was collected in February 1999 from Sheep Pens field at HRI-Wellesbourne, UK. This site had no history of ethoprophos applications. The soil had a pH of 6.8, organic matter content of 2.3% and moisture content of 15% (ca. −33 kPa).
Analytical grade ethoprophos (97% purity, Promochem, Welwyn Garden City, UK) was used throughout this study. An aqueous solution of ethoprophos (500 mg ai l−1 sterile distilled water) was used for the preparation of all ethoprophos-containing media. Two different liquid media were used: a mineral salts medium supplemented with nitrogen (MSMN); and a soil extract medium (SEM). MSMN was prepared as described previously by Cullington and Walker  and SEM was prepared using soil from the field site in Northern Greece. Soil and distilled water were mixed in a 1:1 (w/v) ratio and sterilised for 30 min at 121°C. The sample was centrifuged to remove soil particles and the supernatant was re-autoclaved (30 min at 121°C). After cooling, 20 ml of the aqueous solution of ethoprophos (500 mg l−1) were added to 980 ml of soil extract to give a final concentration of 10 mg l−1. The sterility of SEM was verified by streaking a sample of the medium on Nutrient Agar (NA) and incubating for 4 days at 25°C. The pH of the soil extract medium was 5.9. Mineral salts agar and soil extract agar containing ethoprophos (10 mg l−1) were prepared in a similar way to the MSMN and SEM, except that Difco Bacto agar (15 g l−1) was added. Nutrient Agar (Difco, Appleton Woods, Birmingham, UK) was prepared according to the manufacturer's instructions.
2.3Isolation by enrichment culture
The enrichment culture technique used to isolate pesticide-degrading bacteria from pesticide-treated soils was as described in detail previously [13–15]. At the point of 50% ethoprophos degradation in the second enrichment cycle in either MSMN or SEM, a 10-fold dilution series was prepared on triplicate ethoprophos-containing (10 mg l−1) agar plates of MSMN and SEM, which were incubated at 25°C for 48 h. Twenty well-separated colonies from each medium were assayed for their degrading ability in universal bottles (25 ml) containing MSMN or SEM (3 ml) plus ethoprophos, as appropriate, which were incubated on a shaking platform at 150 rpm at 25°C. Immediately after inoculation and 3, 5, 8, 14 and 21 days later, 0.5 ml of the liquid culture were removed and mixed with an equal volume of hexane in an HPLC vial (2 ml). The contents were vortexed for 10 s and 3 μl of the hexane supernatant was analysed for ethoprophos residues by GLC with a nitrogen/phosphorus detector . Ethoprophos concentrations were also measured in uninoculated samples. Cultures positive for ethoprophos degradation were spread onto agar plates of the corresponding ethoprophos-containing media (MSMN or SEM). Replicate samples (0.7 ml) were stored at −76°C following addition of 15% v/v glycerol. Isolates were routinely cultured on NA plates to check their purity. All plates were incubated at 25°C for 48 h. Pure cultures were used to inoculate MSMN or SEM containing ethoprophos to assay their degradative ability. Isolates positive for ethoprophos degradation were streaked again onto fresh ethoprophos-containing agar plates and NA plates to ensure purity. Cell morphology, motility and Gram-reaction of the purified cultures were examined under a light microscope. Catalase and oxidase activities were determined according to Gerhardt et al. .
To aid PCR amplification of chromosomal DNA, the cells were grown in Luria–Bertrani broth (LB) at 30°C for 18 h and centrifuged at 4000×g for 5 min and the DNA purified. The cell pellet was re-suspended in 1 ml Bacterial Lysis Buffer B1 (Qiagen, Dorking, UK: QIAwell 8 plasmid kit). The manufacturer's instructions were used except for cell lysis. An aliquot of 20 μl of lysozyme (100 mg ml−1, Sigma, Poole, UK) was added and the samples were incubated at 37°C for 15 min. Subsequently, 0.35 ml of Buffer B2 was added and the tubes were incubated at 50°C for 30 min. After incubation, glass beads (0.1 mm diameter) were added and the samples were vortexed for 15 min. Samples were centrifuged at 13 000×g for 5 min and loaded on QIAwell columns (Qiagen, Dorking, UK). The DNA was adsorbed to the column, and eluted under vacuum with 2 ml of buffer QF. The DNA was isopropanol precipitated, washed with 70% ethanol and air-dried. The pellet was suspended in 100 μl of 1 mM Tris pH 8.0.
2.5PCR amplification of the 16S rRNA gene
The 16S rRNA genes of the bacterial strains were amplified with two different sets of universal primers (Life Technologies, Paisley, UK) . The first primer set was CACGGATCCAGACTTTGATYMTGGCTCAG and GTGAAGCTTACGGYTAGCTTGTTACGACTT which amplify most of the 16S rRNA gene (ca. 1504 bp). The second set of primers was AACGCGAAGAACCTTAC and CGGTGTGTACAAGACCC, which amplify an internal fragment of the 16S rRNA gene (ca. 433 bp). Amplification was carried out in 50-μl reactions containing 2 U of DyNAzyme II DNA polymerase (Finnzymes, Espoo, Finland), 100 pmol of each primer, 1× buffer (DyNAzyme buffer), 10 pmol of each nucleotide (dATP, dCTP, dGTP, dTTP). The reaction mixtures were overlaid with 50 μl mineral oil and were amplified as follows: 95°C for 30 s, followed by 30 cycles of 95°C for 1 min, 55°C for 1 min and 72°C for 2 min, with a final extension of 72°C for 5 min. PCR products were purified using a QIAquick PCR purification kit (Qiagen, Dorking, UK) according to the manufacturer's instructions. Samples (7–10 μl; ca. 250 ng template DNA) of the purified 1.5-kb PCR product were digested in 15 μl reaction mixtures with the restriction enzyme CfoI (1 μl). Digestion was performed using the conditions recommended by the supplier (Life Technologies, Paisley, UK), and the digests were subjected to electrophoresis on 2% (w/v) Metaphor agarose (FMC BioProducts, Rockland, USA). For DNA sequencing, purified PCR product (ca. 433 bp) was used (90 ng DNA). Cycle sequencing was performed by using a Taq DyeDeoxy terminator cycle sequencing kit (Applied Biosystems, North Warrington, UK) and samples were analysed on an Applied Biosystems 733 sequencer. Sequences were compared to those on the EMBL database and to each other by using the programs FASTA and GAP, respectively (The Genetics Computer Group, University of Wisconsin, USA).
Cells were grown in LB (4 ml) for 18 h. The cells were pelleted by centrifugation at 4000×g for 5 min and suspended in 100 μl phosphate buffer saline, 0.01 M, pH 7.4 at 25°C (Sigma, Poole, UK). Samples were diluted 1:3 with 4×NuPAGE SDS buffer (Novex Experimental Technology, Frankfurt, Germany) and SDS–PAGE was performed using pre-Cast 3–8% polyacrylamide gels in Tris–acetate buffer (1.0 mm×12 wells). Samples were run at 150 V in 1×NuPAGE Tris–acetate SDS running buffer until the tracker dye had reached the bottom of the gel. Gels were stained with 0.25% (w/v) Coomassie brilliant blue (Sigma, Poole, UK), in 50% v/v methanol, 10% v/v acetic acid for approximately 1 h, and destained with methanol, acetic acid solution (50:10%).
Inoculum was prepared from a culture of isolate epI grown on SEM agar+ethoprophos (10 mg l−1) at 25°C for 48 h. Growth was washed from the plates with 1 ml of SEM and 0.2 ml aliquots of the bacterial suspension were used to inoculate fresh soil extract liquid medium (10 ml) containing ethoprophos (10 mg l−1). Cultures were incubated and ethoprophos degradation was monitored as before. When more than 50% degradation had occurred, an aliquot of the culture (0.2 ml) was transferred into fresh liquid medium containing ethoprophos (10 ml). Liquid cultures of isolate epI produced as described above were used as inoculum when 50% of the initial amount of ethoprophos had degraded. Inoculum densities were measured by preparing a 10-fold dilution series in SEM (or MSMN, as appropriate) and spread on triplicate NA plates, which were incubated overnight at 25°C.
2.8Degradation of fresh ethoprophos residues in soil by isolate epI
Triplicate samples of 300 g wet wt sieved soil were fumigated under vacuum with chloroform for 7 days at 30°C in a glass desiccator as described before . Further triplicate soil samples (300 g wet wt) were stored at 4°C in a sealed polyethylene bag. Residual chloroform was removed from the fumigated soil by repeat evacuations. Fumigated and non-fumigated soils were divided into three 100-g subsamples, which were placed into sterile polypropylene bottles. All samples were treated with 5 ml of a filter-sterilised solution of 200 mg l−1 ethoprophos in methanol to give a final concentration of 10 mg kg−1. Soils were left for 3–4 h for the solvent to evaporate and then mixed with sterile plastic spoons to ensure uniform distribution of the added pesticide. Triplicate fumigated and non-fumigated samples (100 g) were inoculated with 2.5 ml of a fresh SEM culture of isolate epI. The soil was inoculated with a final density of 2×106 cfu g−1 soil. A lower density inoculum was also prepared and 2.5 ml of this culture was added to triplicate fumigated and non-fumigated samples (100 g) to give an inoculum density of 2×103 cfu g−1. Triplicate samples (100 g) of fumigated and non-fumigated soil received 2.5 ml of SEM in place of the bacterial suspension to serve as controls. The inoculum was thoroughly mixed into the soil under aseptic conditions, together with the addition of appropriate amounts of sterile distilled water required for moisture content adjustments (ca. −33 kPa). Samples were incubated overnight at 4°C to allow re-distribution of the applied inoculum and water throughout the soil, before being transferred to a 20°C incubator. Samples (10 g wet wt) were removed from the pots immediately before transfer to the 20°C incubator and at regular intervals over 21 days. Ethoprophos residues were measured by GLC as described previously  and initial recovery of the nematicide was ca. 96%. The moisture content of the soils was maintained constant by daily additions of sterile distilled water if needed.
2.9Utilisation of other organophosphorus nematicides
Analytical grade cadusafos (99% purity, FMC, Princeton, NJ), fenamiphos (97.7%), isofenphos (98%) and isazofos (95%) (Promochem, Welwyn Garden City, UK) were used in this study. Analytical grade standards of fenamiphos sulfoxide and sulfone (Bayer, Leverkusen, Germany) were also used for analytical purposes. The chemical structures of the organophosphorus compounds included in this study are shown in Fig. 1. Aqueous solutions of cadusafos and fenamiphos (200 mg l−1) were prepared as described before for ethoprophos (Section 2.2). In contrast, isazofos and isofenphos, which have lower water solubility than cadusafos, fenamiphos and ethoprophos, were added to sterile, dry bottles as 0.5 ml of a 1000 mg l−1 filter-sterilised methanol solution. The bottles were left uncapped on a laminar flow bench until the methanol had evaporated. MSMN (50 ml) was then added aseptically to the bottles, which were then shaken for 1 h on a wrist-action shaker, to ensure dissolution of the pesticide. Duplicate universal bottles, with 5 ml of MSMN containing the test compound (10 mg l−1), were inoculated with 0.2 ml of a liquid culture of the ethoprophos-degrading isolates as described earlier, except that MSMN was used for inoculum preparation in place of SEM. For comparison, duplicate bottles with 5 ml of MSMN+ethoprophos were also included. Controls receiving 0.2 ml of MSMN in place of the bacterial culture, were prepared for each compound. Viable cell counts were prepared as previously and gave an initial inoculum density of 2×108 cfu ml−1. Universal bottles were incubated for 24 days at 25°C on an orbital shaker at 150 rpm. The degradation of ethoprophos, cadusafos, isazofos and isofenphos was measured by regularly removing and analyzing subsamples of 0.5 ml medium for pesticide residues by GLC as described previously . Isofenphos residues were determined using the same GLC conditions as for ethoprophos with the exception that the oven temperature was 220°C and its retention time was approximately 4.6 min. The limit of detection for pesticides analysed by GLC was approximately 0.005 mg l−1 and recoveries of ethoprophos, cadusafos, isofenphos and isazofos were 102, 101, 93 and 89% respectively. Residues of fenamiphos and its oxidation products were determined by HPLC in 0.7 ml aliquots of media as described before . The limit of detection for fenamiphos and its oxidation products was 0.01 mg l−1 and its recovery was ca. 92%.
2.10Radio-respirometry studies in liquid culture
[Ethyl-1-14C] (98% purity, specific activity 606.8 MBq mmol−1) and [S-propyl-1-14C] (98.6% purity, specific activity 1113.7 MBq mmol−1) ethoprophos were kindly donated by Rhone-Poulenc (Research Triangle, USA). Nine 0.5-ml aliquots from a solution of [ethyl-1-14C] (21.2 kBq ml−1) and nine aliquots of 0.2 ml from a solution of [S-propyl-1-14C] (57.4 kBq ml−1), both in methanol, were placed in sterile bottles (100 ml). The bottles were left uncapped on a laminar flow bench until the methanol had evaporated. MSMN (20 ml) containing (10 mg l−1) unlabelled ethoprophos was added aseptically to the bottles and shaken for 1 h to ensure complete dissolution of the pesticide. This experimental procedure resulted in the preparation of nine cultures of MSMN (20 ml) containing unlabelled (10 mg l−1) and either [ethyl-1-14C] (530 Bq ml−1) or [S-propyl-1-14C] (574 Bq ml−1) labelled ethoprophos. Triplicate cultures containing either [ethyl-1-14C] or [S-propyl-1-14C] ethoprophos were amended with 0.5-ml aliquots from cultures of isolates epI and epII in MSMN as before. Viable cell counts gave an initial inoculum density of 8×106 cfu ml−1 for both isolates. The remaining triplicate bottles containing each of the radio-labelled compounds were amended with the same volume of MSMN (0.5 ml) without bacterial cells to serve as uninoculated controls. Incubation conditions were the same as before and degradation of unlabelled ethoprophos was measured in 0.5-ml aliquots of the media by GLC. The amount of radioactivity remaining in the cultures was measured by mixing aliquots (0.2 ml) of the liquid culture with scintillation fluid (10 ml, Ecoscint A, National Diagnostics, Georgia, USA) and counting in a Rackbeta 1215 liquid scintillation counter. 14CO2 was trapped in a sodium hydroxide solution (1 ml, 1 M) in a sterile glass HPLC vial (2 ml) suspended inside the bottle above the level of the culture liquid using a sterile cotton thread. The vials were removed and the samples were transferred into a scintillation vial and mixed with Soluscint A (10 ml, National Diagnostics, Georgia, USA) before counting in a liquid scintillation counter. A fresh vial was placed in the original sample to take the proceeding measurements. Initial recoveries of unlabelled and labelled ethoprophos were 101 and 93%, respectively.
2.11Nucleotide sequence accession numbers
Nucleotide sequences have been deposited in the EMBL database under the accession numbers AJ276648 (epI), AJ276649 (epII), AJ276650 (epIII), AJ276651 (epIV), and AJ276652 (epV).
3Results and discussion
3.1Isolation and characterisation of ethoprophos-degrading bacteria
Three cultures from the 40 picked initially from the dilution plates appeared to degrade ethoprophos. Two of them were isolated from SEM (cultures epA and epC) and the other from MSMN (culture epB). Cultures epA and epC completely degraded 10 mg l−1 of ethoprophos in 5 days, while culture epB needed 8 days to give complete degradation (Table 1).
Table 1. Ethoprophos (mg l−1) remaining in SEM at different times following inoculation with cultures epA and epC, and in MSMN inoculated with epB
All three cultures when plated on NA gave rise to colonies with variable morphology. While cultures epA and epB appeared to be a mixture of two morphologically similar colony types, culture epC was composed of three distinct colony types. Further molecular characterisation (restriction fragment length polymorphism (RFLP) and sequencing of the internal fragment of the 16S rRNA gene) confirmed that cultures epA and epB were composed of a single bacterial strain with two colony morphologies (epI and epII, respectively). In contrast, culture epC was found to consist of three different bacterial strains (epIII, epIV and epV). The three components of culture epC, when separated, could not individually degrade ethoprophos, which may indicate that all three components were necessary for degradation to occur.
Isolates epI, epII and epIII grew well overnight on NA incubated at 25 and 30°C, but no growth was observed at 37 and 42°C. In contrast, isolates epIV and epV grew well at 25, 30 and also at 37°C, but no growth occurred at 42°C. Cells of all isolates were motile appearing as short rods and classed as Gram-negative. Isolates epI, epII and epIII gave positive catalase and oxidase reactions compared with isolates epIV and epV which gave negative oxidase and positive catalase reactions. Restriction digestion of the 1.5-kb PCR product of the 16S rRNA gene with the enzyme CfoI showed that isolates epI, epII and epIII had identical RFLP patterns, which were different from the patterns obtained for the other two components of culture epC; isolates epIV and epV (Fig. 2). When the partial 16S rRNA gene sequences of isolates epI, epII and epIII were compared to previously published sequences on the EMBL database, greatest homology was seen to Pseudomonas putida 16S rRNA sequences (100% sequence identity). Isolates epI and epIII showed 100% sequence identity to P. putida mt-2 (L28676) which contains the plasmid pWWO and forms part of the BTEX degrading bacteria , P. putida F1 (L37365) which has been shown to degrade toluene, and P. putida MTB6 (AF131103) which shows tolerance to organic solvents. This is in addition to 100% sequence identity to other P. putida strains encompassed in more general Pseudomonas taxonomic studies (D85996, D85997, D37924) [20,21]. When the sequence from epII was compared to those on the EMBL database, 100% sequence identity was shown to P. putida strains used in a phylogenetic study of pseudomonads species (D85995, D8600, D85993) , and to one uncharacterised Pseudomonas strain 35L (AB003628) which was capable of degrading aliphatic polycarbonates . Isolate epII had similar sequence to epI; however, 7 bp of the 390-bp sequence were different, indicating that epI and epII were different strains. In contrast, isolates epIII and epI had identical 16S rRNA sequence in 390-bp sequence. Total protein profiles confirmed these results (Fig. 3), where isolates epI and epIII had similar if not identical protein profile and different from epII. Finally, isolates epIV and epV showed higher homology to the 16S rRNA sequences of Enterobacter cancerogenus (98.4%) and Enterobacter amnigenus (99.2%). The finding that one of the components of culture epC, isolate epIII, was identical to isolate epI in all tests suggests that this component of the consortium was responsible for ethoprophos degradation although it could not degrade ethoprophos when purified. Its loss of degrading ability after purification may be attributed to an instability of its degrading phenotype during subculturing as reported in similar studies with carbofuran-degrading bacteria, where the cultures isolated initially lost their degradative ability after repeated subculturing to fresh media . Sub-cultures of isolate epI have also lost degradative ability on sub-culture which strengthens this possibility. This suggestion is further supported by the observation that no degradation of ethoprophos occurred when the individual isolates separated from culture epC were mixed again and used as a combined inoculum.
The general microbial tests combined with the 16S rRNA sequence results indicate that the isolates capable of ethoprophos degradation belong to the ribosomal rRNA group I pseudomonads, within the P. putida lineage and are probably P. putida or close relatives of this species. Pseudomonads are well known for their metabolic diversity and members of this genus have been reported with the ability to degrade many pesticides including carbofuran , atrazine [25,26], and carbaryl .
3.2Degradation of fresh ethoprophos residues in soil by isolate epI
Degradation of ethoprophos in fumigated and non-fumigated soil inoculated with P. putida epI is shown in Fig. 4. No ethoprophos residues were detected in the fumigated and the non-fumigated samples, inoculated with 2×106 cells g−1 5 and 4 days after inoculation, respectively. Ethoprophos degradation was somewhat slower in the fumigated samples, which were inoculated with the lower cell density (2×103 cells g−1), and its residues were below the limit of detection within 11 days. In contrast, ethoprophos degradation proceeded at a slow rate in the non-fumigated samples, which were inoculated with the low inoculum density and in the corresponding uninoculated samples. About 30% of the initial ethoprophos was recovered in both samples at the end of the study. P. putida epI was therefore equally effective in degrading a fresh ethoprophos addition in both fumigated and non-fumigated soils when added at a cell density of 2×106 cells g−1. However, when a lower inoculum level was used (2×103 cells g−1) rapid degradation was observed only in the fumigated samples. This may be attributed to a growth-promoting effect resulting from the liberation of nutrients from dead cells, or to a reduction in competition for available nutrients and occupation of biological spaces due to the eradication of the indigenous soil microflora in the fumigated samples . Similar results were reported by Ramadan et al.  who found that a Pseudomonas sp. could not mineralise 1 μg of p-nitrophenol ml−1 when introduced in natural lake water at densities of 330 cells ml−1, whereas the compound was mineralised when higher cell densities (3.3×104–105 cells ml−1) were added. The authors suggested that when a low inoculum level is used, the population is not capable of surviving an initial competition from the indigenous microorganisms resulting in a population decline. In contrast, a higher initial inoculum level can compensate for the initial population decline, and the survivors can multiply and perform degradation thereafter.
3.3Utilisation of other organophosphorus nematicides
The degradation patterns of ethoprophos, cadusafos, isazofos, isofenphos and fenamiphos in MSMN inoculated with 2×108 cells of P. putida epI or epII ml−1 are presented in Fig. 5. Isolates epI and epII were able to completely degrade ethoprophos in 3 and 5 days, respectively. Of the other organophosphorus nematicides tested, cadusafos residues declined slowly to about 55% of the amount recovered initially in cultures of P. putida epI at the end of the study, compared with about 70% for isazofos and isofenphos, and 80% for fenamiphos at the same time. When P. putida epII was used, the patterns of degradation of isazofos and isofenphos were similar to those obtained with P. putida epI. However, higher amounts of cadusafos (65%) and lower amounts of fenamiphos (70%) were recovered from the cultures of epII compared with epI at the end of the study. About 8% of the initial amount of fenamiphos were recovered as fenamiphos sulfoxide in the P. putida epI cultures, but only 2% in the P. putida epII cultures at the end of the study. Amounts of fenamiphos sulfone were always below 1% of initial fenamiphos for both isolates. Degradation of the organophosphorus nematicides tested was less than 10% in all uninoculated samples throughout the study (data not shown). Ethoprophos-degrading isolates were able to rapidly degrade ethoprophos, but only slowly degraded the other organophosphorus nematicides tested. The high substrate specificity of the ethoprophos-degrading isolates observed in this study has also been reported for other organophosphorus-degrading bacteria. Racke and Coats  isolated an isofenphos-degrading Arthrobacter sp., which was unable to degrade any of the other organophosphates tested. In addition, Sethunathan and Yoshida  reported the isolation of a Flavobacterium sp., which was able to degrade parathion, but not malathion. This was attributed to the specificity of the enzyme, which was able to break a P–O–C bond in parathion but not the P–S–C bond in malathion. Fenamiphos, isofenphos and isazofos molecules contain aromatic substituents compared with ethoprophos, which contains no ring moieties in its molecule (Fig. 1). However, the similar chemical structures shared by cadusafos and ethoprophos did not lead to rapid degradation of cadusafos by the ethoprophos-degrading bacteria, although the former was degraded relatively faster than any of the other organophosphates tested. It is possible that the existence of an additional methyl group in the –P–S moiety of cadusafos inhibited the ability of the isolated bacteria to degrade it. Cullington and Walker  suggested that chemical similarities could play a significant role in development of cross-adaptation within the substituted-urea group of herbicides. They showed that a single change to substituents of the phenyl ring greatly influenced the rate of degradation by an isolated bacterium. Substitution of a methoxy-group (metoxuron) for a 4-chlorine (diuron) resulted in considerably slower degradation.
Studies with [ethyl-1-14C] and [S-propyl-1-14C] ethoprophos were made to examine the metabolism of ethoprophos by the isolated bacteria. The degradation patterns of unlabelled and radio-labelled ethoprophos, and the evolution of 14CO2 in MSMN inoculated with 8×106 cells of P. putida epI ml−1 are shown in Fig. 6. Similar patterns of degradation of both unlabelled and radio-labelled ethoprophos as well as production of 14CO2 were observed for both isolates (the results from isolate epI only are shown). Degradation of unlabelled ethoprophos was complete in less than 10 days, compared with less than 20% degradation in the uninoculated samples at the same time. When [ethyl-1-14C] ethoprophos was used, the loss of radioactivity and the evolution of 14CO2 was negligible and not significantly different (P=0.05) in inoculated and uninoculated samples (Fig. 6a). In contrast, when [S-propyl-1-14C] ethoprophos was utilised, a significant reduction (P<0.05) of radioactivity in liquid cultures of isolate epI was coincident with the evolution of 14CO2 (Fig. 6b). For example, more than 50% of the initial radioactivity had disappeared from the liquid cultures of isolate epI and about 30% was recovered as 14CO2 10 days after inoculation. Evolution of 14CO2 was negligible in uninoculated samples. The apparent discrepancy in the carbon-14 balance sheet indicating only 80% recovery after about 6 days is not readily explained, although inefficiency in collection and hence quantification of 14CO2 is a possibility. The loss of [S-propyl-1-14C] and the evolution of 14CO2 indicates that degradation proceeds by removal of the –S-propyl moiety of the ethoprophos molecule. This suggestion is further supported by the minimal loss of radioactivity and no 14CO2 production when [ethyl-1-14C] ethoprophos was used and indicates that the O-ethyl group of the ethoprophos molecule is not attacked by the bacteria. Although confirmation of metabolites was not attempted in the present studies, the suggested degradation pathway has been demonstrated to occur in various soils . Ethoprophos degradation in soil proceeded via removal of one of the S-propyl moieties producing either O-ethyl-S-propylphosphorothioic acid or O-ethyl-S-phosphorodithiolic acid, which did not accumulate in the soil, but were further metabolised . These results are also consistent with those from similar studies in the soil from Northern Greece from which the ethoprophos-degrading bacteria were obtained. In this soil [S-propyl-1-14C] ethoprophos degradation was coupled with a significant evolution of 14CO2, compared with significantly lower amounts of 14CO2 when [ethyl-1-14C] ethoprophos was used .
Other experiments have shown that isolates epI and epII are able to degrade ethoprophos in MSMN in less than 50 h with concurrent population growth , thus indicating utilisation of the pesticide as sole carbon source. In the same experiments, it was shown that ethoprophos did not provide an adequate source of phosphorus. In studies involving re-inoculation of soils with isolate epI , inoculum densities as low as 104 g−1 were sufficient to degrade a fresh addition of ethoprophos. Degradation was achieved at temperatures in the range from 5 to 35°C and at soil water potentials from −1500 to −10 kPa, although rates of degradation were restricted in the cooler and drier soils. Isolate epI was also active when inoculated into soils with pH 6.8, and 8.3, but not at pH 5.4, and was less active in soils with higher organic matter content in the range from 0.3 to 8.5%. Ethoprophos-degrading ability was therefore apparent in a wide range of ecological situations .
In conclusion, two pure bacterial cultures and a microbial consortium with the ability to degrade ethoprophos were obtained from a previously treated soil from Northern Greece. This is the first report of the isolation and characterisation of microorganisms able to rapidly degrade ethoprophos. Ongoing research in our laboratory will characterise the catabolic sequences and enzymes produced by these bacteria for ethoprophos degradation.
We would like to acknowledge the State Scholarship Foundation of Greece and the UK Biotechnology and Biological Sciences Research Council for financial support for this work. Special thanks are given to Mrs E. Shaw and Mrs M. Ousley for their technical assistance. Thanks are also expressed to Rhone-Poulenc and Ms J. Williams for supplying the radio-labelled ethoprophos.