Introduction

Diuron, 3-(3,4-dichlorophenyl)-1,1-dimethylurea, is a widely used herbicide belonging to the phenylamide herbicide family. It is used on many agricultural crops, such as wheat, alfalfa, fruit and vineyards along with roads, garden paths and railway lines. Long term dispersion of this compound has contaminated soils, sediments and water. Due to its great persistence (Tomlin 1997), diuron is considered a priority hazardous substance. Two major biotic degradative pathways have been described, i.e., N-demethylation reactions or transformation to anilines (Häggblom 1992). 3,4-Dichloroaniline (3,4-DCA) is the main metabolite (Ellis and Camper 1982; Tixier et al. 2000, 2001). It is highly toxic [median effective concentration (EC50) of diuron, 68 mg l−1; EC50 of 3,4-DCA, 0.49 mg l−1 (Tixier et al. 2000)]. Previous studies showed that various microorganisms were able to degrade diuron. Widehem et al. (2002) isolated from a soil previously treated with diuron, a strain of Arthrobacter sp. N2 able to transform diuron (40 mg l−1) completely in 24 h. Unfortunately, Tixier et al. (2000, 2001, 2002) showed that the degradation was only partial and 3,4-DCA accumulated in the medium. At the same time, several strains were reported to mineralize 3,4-DCA (You and Bartha, 1982; Tixier et al. 2000, 2002; Dejonghe et al. 2002, 2003) without showing the capability of degrading diuron to 3,4-DCA. Comamonas testosteroni W7 and Delftia acidovorans W34 were responsible for the degradation of 3,4-DCA (30 mg l−1) at 0.63 mg l−1 h−1 and 0.54 mg l−1 h−1, respectively (Dejonghe et al. 2002, 2003).

For the first time, a consortium of two bacteria able to mineralize diuron has been tested. Arthrobacter sp. N4 similar to Arthrobacter sp. N2 able to degrade diuron to 3,4-DCA, and D. acidovorans W34, chosen for its ability to mineralize 3,4-DCA, were co-cultivated as free or immobilized cells in a sediment extract medium. In experiments with immobilized cells, Ca-alginate beads were used to immobilize either Arthrobacter sp. N4 or D. acidovorans (co-culture with one immobilized strain and the other one cultivated as free cells), or to co-immobilize both strains.

Material and methods

Microbial strains and preculture conditions

Arthrobacter sp. N4 (Martine Sancelme, personal communication), close to Arthrobacter sp. N2 (Widehem et al. 2002), and Delftia acidovorans W34 (Dejonghe et al. 2003) were pre-cultivated with liquid LB medium for 24 h at 28°C and 200 rpm. Cell density (dry cell wet g l−1) of bacterial suspensions was determined by measuring samples OD at 600 nm and relating the value to a calibration curve.

Cell immobilization

A sterile solution (100 ml) of sodium alginate (30 g l−1) was mixed with bacterial suspension from the preculture (0.097 mg of Arthrobacter sp. N4 cells and 0.078 mg of D. acidovorans W34 cells ml−1 of alginate respectively). Ca-alginate beads of about 3 mm diam. were obtained by dropping the alginate cell mixture into CaCl2 (30 g l−1) using a peristaltic pump supplied with a calibrated needle (Jézéquel et al. 2005).

Mineralization of diuron and 3,4-DCA

100 ml Erlenmeyer flasks were filled with 75 ml Sediment Extract (SE) medium and incubated at 28°C and 200 rpm. This medium was made using sediment from vineyard plots (Rouffach, Alsace, France). It was mixed with tap water (1:1 w/v) and autoclaved (130°C; 1 h). The filtrate was autoclaved (115°C; 0.5 h). A stock solution of diuron (99.5% purity, Sigma-Aldrich) was sterilized by filtration (0.2 μm, Millipore Millex PTFE, Japan) after dissolution in dimethylsulfoxyde. SE medium with diuron (20 mg l−1) was inoculated with 0.0081 mg of Arthrobacter sp. N4 and 0.0064 mg of D. acidovorans W34 free or immobilized cells ml−1 medium, respectively. For immobilized cells, 5 g of alginate beads were inoculated in each flask.

Analysis

Immobilized cells growth was determined by measuring OD600 of the contents of 10 beads from each flask after dissolving them in 3 ml trisodium citrate (50 mM).

Diuron and 3,4-DCA were measured after filtration (0.20 μm) and 20 μl were injected in a HPLC system equipped with a reverse phase column (Lichrosorb RP18, 250 × 4 mm) at room temperature. The mobile phase was acetonitrile/ water (45:55, v/v), at 1 ml min−1. Diuron and 3,4-DCA were detected at 254 nm.

All the experiments were performed in triplicate. Results are shown as mean ± confidence interval. Statistical significance was determined at P = 0.05.

Results and discussion

Cell growth and free cells degradation of diuron and 3,4-dichloroaniline

Bacteria were cultivated in the SE medium, closer to the sediment contents—used for later bioaugmentation studies—than synthetic media. Cell growth of Arthrobacter sp. N4 (Fig. 1a) was delayed by diuron, due to the toxicity of this compound. D. acidovorans W34 was less affected (Fig. 2a).

Fig. 1
figure 1

Free cell growth and degradation of diuron (20 mg l−1) by Arthrobacter sp. N4 cultivated with SE medium. A, medium without diuron (○), medium supplied with diuron (●); B, diuron (♦) and 3,4-DCA (■) in the culture medium. Vertical bars indicated confidence interval (n = 3)

Full size image
Fig. 2
figure 2

Free cell growth and degradation of 3,4-DCA (20 mg l−1) by D. acidovorans W34 cultivated with SE medium. A, medium without diuron (○), medium supplied with diuron (●); B, 3,4-DCA (■) in the culture medium. Vertical bars indicated confidence interval (n = 3)

Full size image

As previously described by Widehem et al. (2002) and Tixier et al. (2002), Arthrobacter sp. N4 degraded diuron within 72 h (Fig. 1b) as the cells grew; simultaneously 3,4-DCA accumulated in the medium, as already shown by several authors (Ellis and Camper, 1982; Tixier et al. 2002). Conversely D. acidovorans W34 was unable to degrade diuron; yet it mineralized 3,4-DCA that was supplied to the culture medium (Fig. 2b) as already shown by Dejonghe et al. (2003). The mineralization rate (i.e., 0.83 mg l−1 h−1) was higher than that reported by You and Bartha (1982) for Pseudomonas putida (0.38 mg l−1 h−1) but lower than Pseudomonas fluorescens 26 K degradation rate (2.36 mg l−1 h−1, Travkin et al. 2003).

Co-culture with free cells of Arthrobacter sp. N4 and Delftia acidovorans W34

The growth kinetic of Arthrobacter sp. N4 co-cultivated with D. acidovorans W34 (Fig. 3a) was very similar to that of D. acidovorans W34 (Fig. 2a). Degradation rate of diuron (Fig. 3b) was low (0.07 mg l−1 h−1 vs. 0.28 mg l−1 h−1 with pure culture of Arthrobacter sp. N4) and 3,4-DCA was only detected after 288 h incubation, revealing a competition between the two bacteria. D. acidovorans W34 was favoured (lower sensitivity to diuron than Arthrobacter sp. N4 (data not shown)). In order to explain 3,4-DCA accumulation in the medium after 288 h, we hypothesized that the growth rate of Arthrobacter sp. N4 was higher when diuron concentration decreased; hence a higher diuron degradation rate compared to that of 3,4-DCA. Although co-culture of these two bacteria allowed diuron mineralization, optimal conditions were not reached. Therefore, cell immobilization was tested as a suitable method to reduce competition between bacteria growing in the same media (O’Reilly and Scott 1995). This technique has been tested in the case of bioaugmentation for the cleaning up of soils contaminated by various molecules among which pesticides and heavy metals (Cassidy et al. 1996; Jézéquel et al. 2005). Alginate is recognized to enhance the degradation rate of some toxic molecules (Bettmann and Rehm 1984; Cassidy et al. 1997).

Fig. 3
figure 3

Free cell growth and degradation of diuron (20 mg l−1) by a co-culture of Arthrobacter sp. N4 and D. acidovorans W34 cultivated with SE medium. A, medium without diuron (○), medium supplied with diuron (●); B, diuron (♦) and 3,4-DCA (■) in the culture medium. Vertical bars indicated confidence interval (n = 3)

Full size image

Co-culture with immobilized cells of Arthrobacter sp. N4 and Delftia acidovorans W34

It has previously been checked that beads used to immobilize strains did not adsorb diuron during incubation (data not shown).

Free- and immobilized-cell growth rates (data not shown) were similar between pure cultures (Fig. 1a, 2a) and co-cultures (Fig. 4a, c, e). Conversely, immobilized cells of one or both bacteria exhibited a lower maximal biomass than free cells, for both strains. It might have been the consequence of, e.g., lower diffusivity of nutrients in beads which modified the nutritional equilibrium and/or the saturation of the bead porosity by cells.

Fig. 4
figure 4

Cell growth and degradation of diuron (20 mg l−1) by co-cultures of free (FC) or immobilized (IC) cells of Arthrobacter sp. N4 (N4) and D. acidovorans W34 (W34) cultivated with SE medium supplied (filled symbol) or not (void symbol) with diuron. A, co-culture of N4CL (●, ○) and W34CI (▴, △); C, co-culture of W34CL (●, ○) and N4CI (▴, △); E, co-immobilization of N4 and W34 (●, ○). B, D, F, diuron (♦) and 3,4-DCA (■) in the culture medium. Vertical bars indicated confidence interval (n = 3)

Full size image

Diuron was mineralized (Fig. 4b, d, f) at a higher rate than that with the free cell co-culture (0.17 mg diuron l−1 h−1 vs. 0.06 mg diuron l−1 h−1, Table 1). This was most probably due to the creation of favourable micro-environments within beads. These micro-environments allowed optimal expression of both bacteria thanks to a better level of oxygen, substrates, etc. In case of co-immobilization, we hypothesized that the partitioning of both strains within beads was the result of the sensitivity of each strain towards diuron, as mentioned above. The gradient of diuron within beads might have favoured D. acidovorans W34 at the surface of alginate beads whereas Arthrobacter sp. N4 colonized the core of the beads. Such favourable micro-environments have already been reported. For instance, water denitrification was successfully achieved by co-immobilizing aerobic Nitrosomas europaea for the oxidation of nitrate to nitrite which colonized the surface of alginate beads, whereas an anaerobic denitrifying Pseudomonas sp. colonized the core of the oxygen deprived beads, allowing to reduce nitrite to N2 (concept of the “magic bead”, Santos et al. 1996). Likewise simultaneous fermentation of a mixture of glucose and xylose was shown to be feasible thanks to the co-immobilization of Saccharomyces cerevisiae and Candida shehatae (Lebeau et al. 1997), although these two yeasts did not require the same oxygen level and that xylose was not consumed as long as glucose was not exhausted with a free cell co-culture. The most efficient mineralization rate was observed with a co-culture of immobilized D. acidovorans W34 and free Arthrobacter sp. N4 cells (0.21 mg diuron l−1 h−1, Table 1). Accumulation of 3,4-DCA in the medium (9.45 mg l−1 at the maximum) between 24 h and 48 h incubation indicated, as above mentioned, a higher degradation rate of diuron into 3,4-DCA compared to the mineralization of 3,4-DCA. We explain this result by considering two simultaneous phenomena: (i) the ratio of Arthrobacter sp. N4 maximal biomass to D. acidovorans W34 maximal biomass was higher than co-immobilization or co-culture (immobilized Arthrobacter sp. N4 cells and free D. acidovorans W34 cells) ratios, and (ii) diuron mineralization was closely related to cell growth. Therefore the limiting step of the diuron mineralization was temporarily the mineralization of 3,4-DCA created by diuron degradation. Diffusion constraints in beads delayed the contact between 3,4-DCA and D. acidovorans W34 which most probably enhanced 3,4-DCA accumulation.

Table 1 Diuron mineralization by co-cultures of D. acidovorans W34 and Arthrobacter sp. N4
Full size table