Environ Sci Pollut Res DOI 10.1007/s11356-014-2512-9
RESEARCH ARTICLE
Effects of a sulfonylurea herbicide on the soil bacterial community Dallel Arabet & Sébastien Tempel & Michel Fons & Yann Denis & Cécile Jourlin-Castelli & Joshua Armitano & David Redelberger & Chantal Iobbi-Nivol & Abderrahmane Boulahrouf & Vincent Méjean
Received: 11 October 2013 / Accepted: 1 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Sulfonylurea herbicides are widely used on a wide range of crops to control weeds. Chevalier® OnePass herbicide is a sulfonylurea herbicide intensively used on cereal crops in Algeria. No information is yet available about the biodegradation of this herbicide or about its effect on the bacterial community of the soil. In this study, we collected an untreated soil sample, and another sample was collected 1 month after treatment with the herbicide. Using a highresolution melting DNA technique, we have shown that treatment with Chevalier® OnePass herbicide only slightly changed the composition of the whole bacterial community. Two hundred fifty-nine macroscopically different clones were isolated from the untreated and treated soil under both aerobic and microaerobic conditions. The strains were identified by sequencing a conserved fragment of the 16S rRNA gene. The phylogenetic trees constructed using the sequencing results confirmed that the bacterial populations were similar in the two soil samples. Species belonging to the Lysinibacillus, Responsible editor: Robert Duran D. Arabet : A. Boulahrouf Laboratoire Génie Microbiologique et Applications, Faculté des Sciences de la Nature et de la Vie, Université Constantine 1, Constantine, Algeria D. Arabet : S. Tempel : M. Fons : C. Jourlin-Castelli : J. Armitano : D. Redelberger : C. Iobbi-Nivol : V. Méjean Institut de Microbiologie de la Méditerranée (IMM), Laboratoire de Chimie Bactérienne UMR7283, Aix-Marseille Université, CNRS, 13402 Marseille, France Y. Denis Institut de Microbiologie de la Méditerranée (IMM), Plate-forme Transcriptome, FR 3479, Marseille, Cedex 20, France V. Méjean (*) Laboratoire de Chimie Bactérienne, Institut de Microbiologie de la Méditerranée, Centre National de la Recherche Scientifique, 31 chemin Joseph Aiguier, 13402 Marseille, Cedex 20, France e-mail: [email protected]
Bacillus, Pseudomonas, and Paenibacillus genera were the most abundant species found. Surprisingly, we found that among ten strains isolated from the treated soil, only six were resistant to the herbicide. Furthermore, bacterial overlay experiments showed that only one resistant strain (related to Stenotrophomonas maltophilia) allowed all the sensitive strains tested to grow in the presence of the herbicide. The other resistant strains allowed only certain sensitive strains to grow. On the basis of these results, we propose that there must be several biodegradation pathways for this sulfonylurea herbicide. Keywords Sulfonylurea . Chevalier® OnePass herbicide . Soil bacterial community . 16S rDNA . Herbicide resistance . High-resolution melting DNA . Bacterial overlay cultures
Introduction Sulfonylurea herbicides were developed in the late 1970s (Lee et al. 2013). Due to their high effectiveness on a wide range of crops at low application rate (10 to 40 g ha−1) and their low mammalian toxicity, they are now used worldwide (Brown 1990; Berger et al. 1998; Sarmah and Sabadie 2002; Hang et al. 2012). More than 50 different products are now available on the market (Lee et al. 2013). The herbicidal activity of the sulfonylureas is based on the inhibition of acetolactate synthase (ALS), the first enzyme implicated in the biosynthesis pathway of the branched chain amino acids (valine, leucine, and isoleucine) in higher plants, bacteria, and fungi, but which is absent in humans and animals (Umbarger and Brown 1958; Lu et al. 2011; Hang et al. 2012). In the absence of ALS, cell division is blocked, resulting in the arrest of the growth process (Beyer et al. 1988). The increasing use of the sulfonylurea herbicides has led to investigations of their environmental impact, especially on the
Environ Sci Pollut Res
soil. Sulfonylurea herbicides have low volatility and do not display remarkable photo-degradability (Sondhia et al. 2013). However, they do display high mobility in soil (Hemmamda et al. 1994; Sondhia 2009a), and some of them (chlorsulfuron and metsulfuron-methyl) could persist for a long time in soil (Nicholls and Evans 1987). Consequently, their trace-level residues could significantly affect both rotation crops and the soil microbiota (Brown 1990; Ismail et al. 1996; Sondhia 2009b). It has been demonstrated that the biological degradation of sulfonylurea by microorganisms is the main route by which they are detoxified in the soil (Sarmah and Sabadie 2002). For instance, chlorsulfuron, metsulfuron-methyl, pyrazosulfuronethyl, and bensulfuron-methyl, which are all sulfonylureas, have been reported to be degraded by fungi, such as Aspergillus niger and Penicillium chrysogenum, and by bacteria, especially strains belonging to the Phyllobacterium, Acinetobacter, and Rhodopseudomonas genera (Zanardini et al. 2002; Valle et al. 2006; Xu et al. 2009; Yin et al. 2011; Sondhia et al. 2013). Species belonging to the Pseudomonas genus have been reported to be able to degrade some sulfonylurea molecules (Manickam et al. 2008; He et al. 2012). Pseudomonas species are known for their ability to survive despite stressful conditions and to resist many pollutants (Madigan and Martinko 2006). Many studies have reported three different biodegradation pathways for the sulfonylureas: cleavage of the sulfonylurea bridge, oxidation, or de-esterification (Zanardini et al. 2002; Lu et al. 2011; Hang et al. 2012). In Algeria, Chevalier® OnePass herbicide (a mixture of two sulfonylureas: iodosulfuron-methyl-sodium and mesosulfuron-methyl plus a third molecule known as mefenpyr-diethyl) is widely used for weed control in cereal agriculture. Hitherto, the microflora in Algerian soil, has received little investigation, and no information is so far available about the biodegradation of iodosulfuron-methyl sodium and the mesosulfuron-methyl. The main objectives of this study were firstly to determine the composition of the bacterial community in agricultural soil that had never been treated with Chevalier® OnePass herbicide and then to evaluate the changes after treatment with the herbicide. We also isolated resistant strains present in the treated soil and tested their ability to degrade the herbicide and thus to enable sensitive bacteria to survive despite the presence of the herbicide.
Materials and methods Collection of soil samples Two soil samples were collected from a wheatfield (Reguada plot, Guettar el Aïech, E 6° 34′ 4″, 36° 11′ 1″ N, Constantine,
Algeria). The first sample was collected in March 2012, a few days before the soil was treated with Chevalier® OnePass herbicide. The second sample was collected 1 month after the treatment. The soil was collected from the top 5 to 20 cm layer in ten flat and distinct areas in the wheatfield in order to get representative samples and to target the microflora of the rhizosphere. The samples were then bagged in sterile packaging to be transported to the laboratory, where roots and stones were removed and the soil samples were thoroughly homogenized by sieving through a 2-mm mesh. The soil was then kept at 4 °C until use. The soil samples were handled under sterile conditions to avoid any contamination. Chemicals Chevalier® OnePass herbicide was purchased from Bayer CropScienceAlgérie (Constantine, Algeria). It is composed of two sulfonylurea molecules: iodosulfuronmethyl sodium [sodium ({[5-iodo-2-(methoxycarbonyl) phenyl] sulfonyl} carbamoyl) (4-methoxy-6-methyl-1, 3, 5-triazin-2-yl) azanide] (30 g kg−1) and mesosulfuronmethyl [methyl 2-[(4, 6-d imethoxypyrimidin-2ylcarbamoyl)sulfamoyl]-α-(methanesulfonamido)-ptoluate] also known as Mesomaxx (30 g kg−1) plus a third molecule, mefenpyr-diethyl also known as safener (90 g kg−1). This latter acts as a catalyst and is highly selective in particular for wheat. The application rate of Chevalier® OnePass herbicide for agricultural purposes is 0.33 kg ha−1. Media The bacterial strains were isolated and grown in Luria Broth (LB) medium composed of yeast extract (5 g L−1), tryptone (10 g L−1), and NaCl (5 g L−1) (Miller 1972). The corresponding solid medium contained 15 g L−1 of agar. In the modified LB used to test the biodegradation capacity of the isolated strains and in the overlay cultures, the 5-g L−1 of yeast extract was replaced by 5 g L−1 of the herbicide before the medium was sterilized at 120 °C. This herbicide concentration allowed growth of the resistant strains. The fusion temperature of the active molecules of the herbicide is greater than 150 °C. Culture conditions Isolation of bacterial clones and growth under aerobic conditions One gram of either untreated or treated soil was mixed with 10 ml of LB in a 50-ml Erlenmeyer flask and then incubated in a rotary shaker at 30 °C for 6 h. This relatively short incubation time was chosen to reduce dominance by the most
Environ Sci Pollut Res
abundant bacterial species. A series of dilutions (10−1 to 10−6) was carried out from the two sample enrichments, and then 100 μl of each dilution was spread on LB agar plate in three replicates and incubated at 30 °C for 24 h. Distinct and separated clones were observed from the 10−3 to the 10−6 dilutions. Bacterial colonies were then distinguished on the basis of their macroscopic appearance (shape, color, dimensions, viscosity, etc.). Each of the clones was grown on a separate agar plate under the conditions described above. Isolation of bacterial clones and growth under microaerobic conditions One gram of either untreated or treated soil was mixed with 10 ml of LB in 10 ml tubes and incubated without shaking at 30 °C for 6 h. A series of dilutions (10−1 to 10−6) of the two sample enrichments was then performed; 100 μl of each dilution was then spread on the LB agar plate in three replicates and placed in an anaerobic jar with gas pack (Anaerocult® A, Merck, Germany) to provide the anaerobic conditions. The jars were then incubated at 30 °C for 48 h. Distinct and separated clones were observed from the 10−3 to the 10−6 dilutions. Bacterial colonies were recovered on the basis of their macroscopic appearance (shape, color, dimensions, viscosity, etc.). Each of the clones was grown on a separate LB agar plate under the conditions described above. Culture on medium containing the herbicide The ability of the isolated strains to use or to resist the herbicide was tested by growing them on a modified LB medium in which the 5-g L−1 of yeast extract had been replaced by 5 g L−1 of the herbicide. The growth of each aerobic and microaerobic isolated strain was determined. The conditions for each type of culture were as described above. As a control, the strains were grown on the usual LB.
Inductively coupled plasma optical emission spectrometry experiment One gram of each of the untreated or treated soil samples was mixed with 10 ml of sterile water and 100 μl of the solution was used. The samples were wet washed with 32.5 % nitric acid (Suprapur, Merck) for 12 h at 100 °C (Neumann et al. 2009) and were then filled to a tenfold volume with water prior to inductively coupled plasma optical emission spectrometry (ICP-OES) analysis. Three replicates for each sample were carried out, and the average sulfur concentration values were calculated. Sulfur analysis was performed using a ThermoFisher ICAP 6000 ICP-OES. Multielement standard solution XVI (Merck) was used as a reference. Metagenome extraction, 16S DNA amplification, and sequencing Soil DNA was extracted using the Power Soil DNA extraction Kit (MO-BIO Laboratories, CA, USA) following the manufacturer’s instructions. PCR was conducted using universal primers FD1 (5′-AGAGTTTGATCCTGGCTCAG-3′) and S6 (5′-GTATTACCGCGGCTGCTG-3′) (Chanal et al. 2006) in order to amplify a conserved fragment (approximately 500 bp) of the bacterial 16S rRNA gene. The 50-μl PCR reactions contained 10 ng of template DNA, the four dNTPs at 100 μM, the Go-Taq Polymerase buffer (Promega), the two convergent primers at 10 μM, one unit of the Go-Taq polymerase (Promega), and distilled water in a final volume of 50 μl. The reaction was carried out under the following conditions: an initial denaturing step at 94 °C for 1 min and 30 s followed by 30 cycles at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 45 s. The amplification of the 16S rDNA of the strains isolated was carried out from bacterial colonies using the same mix and under the same conditions. PCR products were purified using the GenElute PCR Clean-Up Kit (SIGMA-ALDRICH, USA) as indicated by the manufacturer, and the purified DNA was then sequenced.
Overlay cultures High-resolution melting DNA Each of the strains that was able to grow on the medium containing the herbicide (resistant strain) was grown as patches on the modified LB agar medium and incubated at 30 °C for 24 h. In parallel, each sensitive strain isolated from the untreated sample was grown on LB medium and incubated in a rotary shaker at 30 °C for 24 h. A soft layer of the modified LB medium was then laid over the patches of resistant strains, and a 3-μl spot of a culture of one of the sensitive strains was added over each patch. The effect of each of the resistant strains isolated was tested over each of the five sensitive strains. The overlay cultures were then incubated at 30 °C for 24 h.
The high-resolution melting DNA (HRM) runs were done on a CFX96 Real-Time System (Bio-Rad). The PCR reactions were carried out in a final reaction volume of 30 μl, and the mix was composed of 1× SoFast EvaGreen Supermix (BioRad), 500 nM of each primer (FD1 and S6), and 4 μl of each DNA tested (the amplified 500 bp fragment of the two soil samples and the oceanic control were purified and then diluted 100-fold). The oceanic metagenome was a kind gift of Valérie Michotey (Mediterranean Institute of Oceanology). It was extracted from water of the gulf of Fos (the Mediterranean sea, France). Cycling parameters of the real-time PCR
Environ Sci Pollut Res
consisted of an initial step at 95 °C for 2 min followed by 40 cycles at 95 °C for 10 s, 60 °C for 20 s, and 72 °C for 10 s. This program was followed by a high-resolution melting curve consisting of 95 °C for 1 min, 60 °C for 1 min, and ramping at 0.2 °C for 10 s from 79 to 95 °C during which the fluorescence was read at the end of each increment. The results were analyzed using Bio-Rad Precision Melt Analysis Software 1.1 (Bio-Rad). Phylogenic tree construction 16S prokaryotic sequences were downloaded from the Ribosomal Database Project (Cole et al. 2009). The uncultured sequences and the archaea sequences were eliminated from the data set. A first study using NCBI Blast (Johnson et al. 2008) selected the genus of the amplified 16S rDNA sequences (http://blast.ncbi.nlm.nih.gov/Blast.cgi). We extracted the 16S sequences from the Ribosomal Database Project that corresponded to the genus selected. A web server, known as “Phylogeny.fr” (Dereeper et al. 2008), was used for the phylogenetic analysis (http://www.phylogeny.fr/version2_ cgi/alacarte.cgi). The three main steps were used for the phylogenetic analysis. It was first performed on the basis of a multiple alignment of 16S rDNA sequences obtained using MUSCLE version 3.7 (Edgar 2004). The non-similar regions of the aligned sequences were removed by GBlocks version 0. 91b (Talavera and Castresana 2007). The phylogenetic trees were constructed using the BIONJ method (Gascuel 1997) with 100 bootstrap replicates. The numbers shown on each branch denote the percentages of bootstrap support. All methods were run with the default parameters suggested by their respective authors. After the first phylogenetic analysis, we manually removed the unnecessary sequences (we conserved two or three sequences per genus; these sequences were the sequences closest to each of the amplified 16S rDNA sequences in the phylogenic tree) and a second phylogenic analysis was then performed. This is shown in Fig. 3. Nucleotide sequence accession numbers The sequences determined during the present study have been deposited in the GenBank database. The accession numbers are available in Fig. 3. The strains isolated before treatment are indicated by an “N” followed by the number of the clone and then the GenBank accession number in parentheses. The strains isolated after treatment are indicated by “T,” followed by the number of the clone and the accession number in parentheses. Asterisks indicate clones obtained under microaerobic culture conditions. We chose one strain to represent each phylogenetic group of strains isolated in the phylogenetic trees.
Results and discussion Confirmation of the presence of the Chevalier® OnePass herbicide in treated soil using an ICP technique Both the active molecules of the Chevalier® OnePass herbicide contain sulfur. We used this fact to test for the persistence of herbicide residues in the second collection sample. The half-life of sulfonylurea in soil is known to range from a few days to a few months (Kamrin 1997). We collected the second sample 1 month after treatment to give the bacterial community enough time to evolve and to get used to the presence of the herbicide. Three determinations of the sulfur concentration in each soil type (untreated and treated) were carried out using an ICP technique and the average was calculated. In the untreated soil, the average sulfur concentration was 787.8± 18.7 ng ml−1, while in the treated soil, the average value was 1,561±17 ng ml−1, indicating that a significant fraction of the herbicide was still present in the sample. This means that the bacterial community in the soil was still exposed to the effects of the herbicide. Analysis of the amplified 16S rDNA in the untreated and treated soil using the HRM technique In order to obtain an overview of the total bacterial community in each of the soil samples, we used the HRM technique. Whole DNA was extracted and purified from soil samples as indicated in “Materials and Methods.” Conserved fragments of about 500 bp from the 16S rRNA genes were amplified using the FD1 and S6 universal convergent primers (Chanal et al. 2006), and similar fragments of 500 bp 16S rDNA were amplified from an oceanic metagenome for use as a control. The rDNA fragments were then tested using the HRM technique, and the fine analysis of the rDNA melting temperature curves was used to compare the overall bacterial populations present in the samples. Strikingly, the profiles of the rDNA fusion temperature curves were similar in the two soil samples, whereas the curve profile of the oceanic rDNA was clearly distinct (Fig. 1). The minor difference observed between the curve profiles of the untreated and treated soil could indicate limited differences in the bacterial diversity. This result is consistent with previous studies where denaturing gradient gel electrophoresis analysis was used to compare bacterial communities (Valle et al. 2006; Lin et al. 2012). Isolation and analysis of the cultivable strains recovered from the untreated and treated soil According to the HRM results, we expected to find that the culturable bacterial fraction in the untreated and treated soil was similar overall, with the possible appearance of some new
Environ Sci Pollut Res Fig. 1 Comparison of the bacterial communities in the untreated and treated soil using the HRM technique. The metagenomes of the two soil samples were extracted, as explained in “Materials and methods,” and then a conserved 16S rDNA fragment was amplified. The green curve represents the DNA melting profile of the untreated soil and the blue curve that of the treated soil. An oceanic metagenome 16S rDNA (represented by the red curve) was used as a control
species in the treated soil. To confirm this hypothesis, we isolated the cultivable bacterial fraction from both the untreated and treated soil samples. We isolated 259 distinct macroscopic clones under two different culture conditions (aerobic and microaerobic). We were able to distinguish 77 distinct macroscopic clones in the untreated sample and 64 in the treated one under aerobic conditions. Under microaerobic conditions, 66 clones were isolated in the untreated sample, and 52 were found in the treated one (Fig. 2). In order to identify the phylogenetic relationship between the isolated clones and the known bacterial species, a conserved fragment of about 500 bp from the 16S rRNA genes was amplified from each isolated clone. The PCR products were then sequenced,
Fig. 2 Number of strains isolated from the treated and untreated soil under both aerobic and microaerobic conditions. Aerobic cultures were incubated at 30 °C for 24 h. The microaerobic cultures were incubated at 30 °C for 48 h, and the agar plates were placed in anaerobic jars with an Anaerocult A
and the 259 sequences were compared to those of known bacteria to construct phylogenetic trees. Figure 3 strongly suggests that some bacterial groups disappear after treatment under both aerobic and microaerobic conditions. This was true for the group affiliated to the Lysinibacillus genus (N21, N15, and N13 on the phylogenetic tree) and the microaerobic strain related to Clostridium genus (N45* on the phylogenetic tree). The reason could be that the application of the Chevalier® OnePass herbicide has harmful effects on certain bacterial strains and prevents their growth. These strains could therefore be, thus, highly sensitive to this herbicide. On the other hand, some groups persisted after treatment under both culture conditions. This was true for the clones related to Bacillus megaterium (N49 before treatment and T5 after treatment), Achromobacter marplatensis (N5 before treatment and T50 and T1 after treatment), Stenotrophomonas maltophilia (N9 and N1 before treatment and T4 after treatment), Paenibacillus ginsengisoli (N17* before treatment and T9* after treatment), Bacillus cereus (N2* before treatment and T6 and T3* after treatment), and Bacillus circulans (N6* before treatment and T8* after treatment). This indicates that these strains were able to resist the toxicity of the herbicide and thus to spearhead a rapid recovery of a bacterial community. Furthermore, we observed that strains related to Sphingobacterium (T10), Streptomyces (T41), and Pseudomonas (T27 and T32) genera appeared after treatment with the herbicide. Finding new species after treatment with the herbicide could mean that they are better adapted to resist the herbicide, and that they could even degrade it. These results are consistent with the HRM analysis and with previous studies (Valle et al. 2006; Lin et al. 2012).
Environ Sci Pollut Res
Fig. 3 Phylogenetic trees based on comparing 500 bp long sequences of the 16S rRNA gene. The GenBank accession number for each strain isolated is shown in parentheses after the strain name. N indicates the strains isolated from the untreated soil, and T stands for those isolated
from treated soil. An asterisk indicates that the strain was isolated under microaerobic conditions. The scale bar indicates 0.02 substitutions per nucleotide position. Bootstrap values obtained with 100 resamplings were indicated as percentages on all branches
On the basis of the phylogenetic trees, we noted that the presence of the herbicide did not change the global composition of the bacterial community, but that it did affect the growth of certain groups. For instance, the strains related to the Lysinibacillus genus disappeared after treatment even though they were abundant before treatment under aerobic culture conditions (21 % of the isolated strains). On the other hand, some other strains increased after treatment. This was the case of the strains related to Paenibacillus sp. isolated under microaerobic conditions and, as expected, Pseudomonas became the dominant group in the whole cultured fraction isolated from the treated soil under aerobic conditions (27 % of the isolated strains). Indeed, many strains related to the Pseudomonas genus have been described as being able to degrade some sulfonylurea molecules (Li-feng et al. 2007; Manickam et al. 2008; He et al. 2012). This was true for Pseudomonas mediterranea, as we found in this study (see below). Taken as a whole, our results confirm the HRM analysis and appear to be consistent with previous studies. It reveals that the treatment with Chevalier® OnePass herbicide has a drastic effect on the bacterial community, changing the relative abundance of certain strains and probably eliminating others. The resistant strains then recovered and other bacterial groups appeared.
Culture of the isolated strains in the presence of the herbicide and bacterial overlay cultures The analysis of the strains isolated led us to whether they were able to make use of the herbicide or, at least, one of its components. We hypothesized that the strains that disappeared after treatment were highly sensitive to the herbicide, whereas those found in both the soil samples, as well as the new strains found after treatment, were able to resist and probably to degrade the herbicide. If this is true, then they could detoxify the medium and therefore allow the sensitive strains to grow subsequently. To test this hypothesis, we made a modified LB medium in which the herbicide could be used as nutritional resource, as explained in “Materials and Methods.” We first tested the growth of strains representative of each group isolated from the untreated soil. All the strains (isolated under the two culture conditions) were unable to grow in the presence of the herbicide, except the strains related to S. maltophilia (N9 and N1) (data not shown). This could explain why a strain related to this species was found after treatment (T4). Surprisingly, of the ten representative strains isolated aerobically from the treated soil, four were unable to grow on the herbicide medium (T1, T50, T41, and T5). These strains therefore appeared to be sensitive to the herbicide, but,
Environ Sci Pollut Res
nevertheless, we found that they persisted in the treated sample. None of the strains isolated under microaerobic conditions was able to grow in the presence of the herbicide (data not shown). We identified a total of eight resistant strains. As explained in the “Introduction,” the target of the active compounds in Chevalier® OnePass herbicide is the ALS enzyme. We hypothesize that the resistant strains either possess a modified form of ALS that is insensitive to the components of the herbicide or are able to degrade the herbicide to protect the ALS enzyme and thus detoxify the medium. This latter hypothesis could explain the presence of sensitive strains in the treated soil sample. To test this hypothesis, the six resistant strains isolated from the treated soil were grown on the modified LB medium, and the effect of the resistant strain was then tested on the sensitive strains isolated before treatment, as explained in “Material and Methods.” Unexpectedly, among the six resistant strains, only strain T4, which is related to S. maltophilia, allowed all the sensitive strains to grow (Fig. 4a, b). When T4 was grown on the medium containing the herbicide, a dark precipitate formed on the bottom of the agar plate. This effect was not observed for the other resistant strains (Fig. 4a, data not shown). We suppose that the precipitate was a sulfur precipitate resulting from degradation of the sulfur-containing compounds (iodosulfuron-methyl sodium and mesosulfuronmethyl) in the herbicide (Labrenz and Banfield 2004). As indicated in Table 1, the other resistant strains only enabled some of the sensitive strains to grow on the herbicide. For instance, resistant strain T27 (affiliated to P. mediterranea) allowed N13 to grow but without producing the dark precipitate observed with T4 (Fig. 4b). On the basis of these experiments, we suggest that Chevalier® OnePass herbicide could be entirely degraded by the strain related to S. maltophilia.
Table 1 Overlay cultures of resistant and sensitive strains on a medium containing Chevalier® OnePass herbicide
Fig. 4 Analysis of the resistant strains and overlay cultures on the modified LB. a Growth of resistant strains T4 (left) and T27 (right) on a medium containing Chevalier® OnePass herbicide. A dark precipitate appeared during the growth of the T4 strain. b Overlay cultures of
resistant strains T4 and T27 with sensitive strains (only N5, N9, N13, and N15 are shown here) in a medium containing the herbicide. The negative control corresponds to resistant strain T27 that did not allow sensitive strain N1 to grow
RS SS
T4
T20
T10
T32
T27
T6
N13 N15 N21 N49 N5
+ + + + +
− − − + −
− + − − −
+ + − − −
+ − − − −
+ + − − −
(+) indicates that the resistant strain allowed the sensitive strain to grow; (−) indicates that the resistant strain did not allow the sensitive strain to grow SS sensitive strains isolated from untreated soil, RS resistant strains isolated from treated soil
Despite the fact that it is underrepresented in the treated sample, strain T4 therefore seems to be a good candidate for studying the molecular mechanism involved in the biodegradation of Chevalier® OnePass herbicide. This experiment also showed that strains related to P. mediterranea were able to degrade the herbicide, but that they did not allow all the sensitive strains to grow. This could be why they are abundant in the culturable bacterial community after treatment with the herbicide. Since they did not produce the dark precipitate, we suggest that they degrade only partially the sulfonylurea components of the herbicide. In conclusion, the behavior of the bacteria consortium in a treated soil is not as simple as previously thought. In particular, we have shown that both resistant and sensitive strains coexisted in soil containing herbicide. The complete degradation of the Chevalier® OnePass herbicide appears to be achieved only by the strain related to S. maltophilia, a resistant
Environ Sci Pollut Res
strain that allows all the sensitive strains tested to grow. The other resistant strains enabled only certain sensitive strains to grow on the herbicide. This could be related to their herbicidebiodegradation pathway. It would be interesting to decipher the molecular mechanisms involved in these alternative degradation pathways and to explain why strains related to S. maltophilia are capable of completely detoxifying the medium even though they constitute only a small proportion of the total bacterial community.
Conclusion Our results reveal that the presence of the sulfonylurea Chevalier® OnePass herbicide modified the balance between relative species abundances, but did not significantly modify the overall composition of the bacterial community in the soil. Indeed, the soil bacterial community was able to recover, and exposure probably enriched its diversity. We have also shown that the novel HRM technique is an efficient and rapid alternative method that could provide an overview of the entire bacterial composition. This technique is in its infancy, but it will be certainly improved in the future to establish a correspondence between DNA fusion temperatures and bacterial species. New generation sequencing methods are powerful tools for analyzing microbial populations, but they are expensive and more time consuming. Interestingly, our work showed that treated soil contains both resistant and sensitive strains. Furthermore, the resistant strains could use different pathways to degrade the herbicide. Strain T4 could be a good model strain for investigating the biodegradation pathways of iodosulfuron methyl sodium and mesosulfuron methyl, which were studied for the first time here. It will be interesting to clarify the degradation mechanisms used by this strain and to investigate, in detail, the possible use of this strain to detoxify agricultural fields that have been contaminated by Chevalier® OnePass herbicide. Acknowledgments We would like to thank A. Rouabah for his help in the soil sample collection and V. Michotey for supplying the oceanic metagenome. We are grateful to M. Ilbert, C. Aussignargues, and S. Bouillet for the valuable suggestions and discussions. Monika Gosh is acknowledged for improving the English version of the manuscript. This work is funded by the Faculté des Sciences de la Nature et de la Vie, Université Constantine 1 (Algeria), the Centre National de la Recherche Scientifique (CNRS), and the Aix-Marseille Université (France).
References Berger BM, Janowitz K, Menne HJ, Hoppe H-H (1998) Comparative study on microbial and chemical transformation of eleven sulfonylurea herbicides in soil. Z Für Pflanzenkrankh Pflanzenschutz 105: 611–623
Beyer EM, Duffy MF, Hay JV, Schlueter DD (1988) Sulfonylurea. In: Kearney PC, Kaufman DD (eds) Herbic. Chem. Degrad. Mode Action, Dekker. New York, pp 117–183 Brown HM (1990) Mode of action, crop selectivity, and soil relations of the sulfonylurea herbicides. Pestic Sci 29:263–281. doi:10.1002/ps. 2780290304 Chanal A, Chapon V, Benzerara K et al (2006) The desert of Tataouine: an extreme environment that hosts a wide diversity of microorganisms and radiotolerant bacteria. Environ Microbiol 8:514–525. doi: 10.1111/j.1462-2920.2005.00921.x Cole JR, Wang Q, Cardenas E et al (2009) The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res 37:D141–D145. doi:10.1093/nar/gkn879 Dereeper A, Guignon V, Blanc G et al (2008) Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res 36: W465–W469. doi:10.1093/nar/gkn180 Edgar RC (2004) MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinforma 5:113. doi:10. 1186/1471-2105-5-113 Gascuel O (1997) BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data. Mol Biol Evol 14:685–695 Hang B-J, Hong Q, Xie X-T et al (2012) SulE, a sulfonylurea herbicide de-esterification esterase from Hansschlegelia zhihuaiae S113. Appl Environ Microbiol 78:1962–1968. doi:10.1128/AEM. 07440-11 He W-H, Wang Y-N, Du X et al (2012) Pseudomonas linyingensis sp. nov.: a novel bacterium isolated from wheat soil subjected to longterm herbicides application. Curr Microbiol 65:595–600. doi:10. 1007/s00284-012-0187-3 Hemmamda S, Calmon M, Calmon JP (1994) Kinetics and hydrolysis mechanism of chlorsulfuron and metsulfuron-methyl. Pestic Sci 40: 71–76. doi:10.1002/ps.2780400112 Ismail BS, Goh KM, Kader J (1996) Effects of metsulfuronmethyl on microbial biomass and populations in soils. J Environ Sci Health Part B 31:987–999. doi:10.1080/03601239609373049 Johnson M, Zaretskaya I, Raytselis Y et al (2008) NCBI BLAST: a better web interface. Nucleic Acids Res 36:W5–W9. doi:10.1093/nar/ gkn201 Kamrin MA (1997) Pesticide profiles: toxicity, environmental impact, and fate. CRC Press Labrenz M, Banfield JF (2004) Sulfate-reducing bacteria-dominated biofilms that precipitate ZnS in a subsurface circumneutral-pH mine drainage system. Microb Ecol 47:205–217. doi:10.1007/s00248003-1025-8 Lee Y-T, Cui C-J, Chow EWL et al (2013) Sulfonylureas have antifungal activity and are potent inhibitors of Candida albicans acetohydroxyacid synthase. J Med Chem 56:210–219. doi:10. 1021/jm301501k Li-feng G, Jian-dong J, Xiao-hui L et al (2007) Biodegradation of ethametsulfuron-methyl by Pseudomonas sp. SW4 isolated from contaminated soil. Curr Microbiol 55:420–426. doi:10.1007/ s00284-007-9011-x Lin X-Y, Yang Y-Y, Zhao Y-H, Fu Q-L (2012) Biodegradation of bensulfuron-methyl and its effect on bacterial community in paddy soils. Ecotoxicol Lond Engl 21:1281–1290. doi:10.1007/s10646012-0882-7 Lu P, Jin L, Liang B et al (2011) Study of biochemical pathway and enzyme involved in metsulfuron-methyl degradation by Ancylobacter sp. XJ-412-1 isolated from soil. Curr Microbiol 62: 1718–1725. doi:10.1007/s00284-011-9919-z Madigan MT, Martinko JM (2006) Brock Biology of Microorganisms. Pearson Prentice Hall. Manickam N, Ghosh A, Jain RK, Mayilraj S (2008) Description of a novel indole-oxidizing bacterium Pseudomonas indoloxydans sp. nov., isolated from a pesticide-contaminated site. Syst Appl Microbiol 31:101–107. doi:10.1016/j.syapm.2008.02.002
Environ Sci Pollut Res Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory Neumann M, Mittelstädt G, Seduk F et al (2009) MocA is a specific cytidylyltransferase involved in molybdopterin cytosine dinucleotide biosynthesis in Escherichia coli. J Biol Chem 284:21891– 21898. doi:10.1074/jbc.M109.008565 Nicholls PH, Evans A. (1987) The behavior of chlorsulfuron and metsulfuron in soils in relation to incidents of injury to sugar beet. Proc. Br. Crop Prot. Weeds Conf. BCPC Publications, pp 549–556 Sarmah AK, Sabadie J (2002) Hydrolysis of sulfonylurea herbicides in soils and aqueous solutions: a review. J Agric Food Chem 50:6253– 6265 Sondhia S (2009a) Persistence of metsulfuron-methyl in paddy field and detection of its residues in crop produce. Bull Environ Contam Toxicol 83:799–802. doi:10.1007/s00128-009-9822-5 Sondhia S (2009b) Leaching behaviour of metsulfuron in two texturally different soils. Environ Monit Assess 154:111–115. doi:10.1007/ s10661-008-0381-8 Sondhia S, Waseem U, Varma RK (2013) Fungal degradation of an acetolactate synthase (ALS) inhibitor pyrazosulfuron-ethyl in soil. Chemosphere. doi:10.1016/j.chemosphere.2013.07.066
Talavera G, Castresana J (2007) Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol 56:564–577. doi:10.1080/ 10635150701472164 Umbarger HE, Brown B (1958) Isoleucine and valine metabolism in Escherichia coli. VIII. The formation of acetolactate. J Biol Chem 233:1156–1160 Valle A, Boschin G, Negri M et al (2006) The microbial degradation of azimsulfuron and its effect on the soil bacterial community. J Appl Microbiol 101:443–452. doi:10.1111/j.1365-2672.2006.02937.x Xu J, Li X, Xu Y et al (2009) Biodegradation of pyrazosulfuron-ethyl by three strains of bacteria isolated from contaminated soils. Chemosphere 74:682–687. doi:10.1016/j.chemosphere.2008.09. 078 Yin LB, Liu Y, Zhang DY, Zhang SB (2011) Isolation and characterization of Rhodopseudomonas sp. S9-1 capable of degradating pyrazosulfuron-ethyl. Adv Mater Res 356–360:1152–1163. doi:10. 4028/www.scientific.net/AMR.356-360.1152 Zanardini E, Negri M, Boschin G et al (2002) Biodegradation of chlorsulfuron and metsulfuron-methyl by Aspergillus niger. Sci World J 2:1501–1506. doi:10.1100/tsw.2002.281
RESEARCH ARTICLE
Effects of a sulfonylurea herbicide on the soil bacterial community Dallel Arabet & Sébastien Tempel & Michel Fons & Yann Denis & Cécile Jourlin-Castelli & Joshua Armitano & David Redelberger & Chantal Iobbi-Nivol & Abderrahmane Boulahrouf & Vincent Méjean
Received: 11 October 2013 / Accepted: 1 January 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Sulfonylurea herbicides are widely used on a wide range of crops to control weeds. Chevalier® OnePass herbicide is a sulfonylurea herbicide intensively used on cereal crops in Algeria. No information is yet available about the biodegradation of this herbicide or about its effect on the bacterial community of the soil. In this study, we collected an untreated soil sample, and another sample was collected 1 month after treatment with the herbicide. Using a highresolution melting DNA technique, we have shown that treatment with Chevalier® OnePass herbicide only slightly changed the composition of the whole bacterial community. Two hundred fifty-nine macroscopically different clones were isolated from the untreated and treated soil under both aerobic and microaerobic conditions. The strains were identified by sequencing a conserved fragment of the 16S rRNA gene. The phylogenetic trees constructed using the sequencing results confirmed that the bacterial populations were similar in the two soil samples. Species belonging to the Lysinibacillus, Responsible editor: Robert Duran D. Arabet : A. Boulahrouf Laboratoire Génie Microbiologique et Applications, Faculté des Sciences de la Nature et de la Vie, Université Constantine 1, Constantine, Algeria D. Arabet : S. Tempel : M. Fons : C. Jourlin-Castelli : J. Armitano : D. Redelberger : C. Iobbi-Nivol : V. Méjean Institut de Microbiologie de la Méditerranée (IMM), Laboratoire de Chimie Bactérienne UMR7283, Aix-Marseille Université, CNRS, 13402 Marseille, France Y. Denis Institut de Microbiologie de la Méditerranée (IMM), Plate-forme Transcriptome, FR 3479, Marseille, Cedex 20, France V. Méjean (*) Laboratoire de Chimie Bactérienne, Institut de Microbiologie de la Méditerranée, Centre National de la Recherche Scientifique, 31 chemin Joseph Aiguier, 13402 Marseille, Cedex 20, France e-mail: [email protected]
Bacillus, Pseudomonas, and Paenibacillus genera were the most abundant species found. Surprisingly, we found that among ten strains isolated from the treated soil, only six were resistant to the herbicide. Furthermore, bacterial overlay experiments showed that only one resistant strain (related to Stenotrophomonas maltophilia) allowed all the sensitive strains tested to grow in the presence of the herbicide. The other resistant strains allowed only certain sensitive strains to grow. On the basis of these results, we propose that there must be several biodegradation pathways for this sulfonylurea herbicide. Keywords Sulfonylurea . Chevalier® OnePass herbicide . Soil bacterial community . 16S rDNA . Herbicide resistance . High-resolution melting DNA . Bacterial overlay cultures
Introduction Sulfonylurea herbicides were developed in the late 1970s (Lee et al. 2013). Due to their high effectiveness on a wide range of crops at low application rate (10 to 40 g ha−1) and their low mammalian toxicity, they are now used worldwide (Brown 1990; Berger et al. 1998; Sarmah and Sabadie 2002; Hang et al. 2012). More than 50 different products are now available on the market (Lee et al. 2013). The herbicidal activity of the sulfonylureas is based on the inhibition of acetolactate synthase (ALS), the first enzyme implicated in the biosynthesis pathway of the branched chain amino acids (valine, leucine, and isoleucine) in higher plants, bacteria, and fungi, but which is absent in humans and animals (Umbarger and Brown 1958; Lu et al. 2011; Hang et al. 2012). In the absence of ALS, cell division is blocked, resulting in the arrest of the growth process (Beyer et al. 1988). The increasing use of the sulfonylurea herbicides has led to investigations of their environmental impact, especially on the
Environ Sci Pollut Res
soil. Sulfonylurea herbicides have low volatility and do not display remarkable photo-degradability (Sondhia et al. 2013). However, they do display high mobility in soil (Hemmamda et al. 1994; Sondhia 2009a), and some of them (chlorsulfuron and metsulfuron-methyl) could persist for a long time in soil (Nicholls and Evans 1987). Consequently, their trace-level residues could significantly affect both rotation crops and the soil microbiota (Brown 1990; Ismail et al. 1996; Sondhia 2009b). It has been demonstrated that the biological degradation of sulfonylurea by microorganisms is the main route by which they are detoxified in the soil (Sarmah and Sabadie 2002). For instance, chlorsulfuron, metsulfuron-methyl, pyrazosulfuronethyl, and bensulfuron-methyl, which are all sulfonylureas, have been reported to be degraded by fungi, such as Aspergillus niger and Penicillium chrysogenum, and by bacteria, especially strains belonging to the Phyllobacterium, Acinetobacter, and Rhodopseudomonas genera (Zanardini et al. 2002; Valle et al. 2006; Xu et al. 2009; Yin et al. 2011; Sondhia et al. 2013). Species belonging to the Pseudomonas genus have been reported to be able to degrade some sulfonylurea molecules (Manickam et al. 2008; He et al. 2012). Pseudomonas species are known for their ability to survive despite stressful conditions and to resist many pollutants (Madigan and Martinko 2006). Many studies have reported three different biodegradation pathways for the sulfonylureas: cleavage of the sulfonylurea bridge, oxidation, or de-esterification (Zanardini et al. 2002; Lu et al. 2011; Hang et al. 2012). In Algeria, Chevalier® OnePass herbicide (a mixture of two sulfonylureas: iodosulfuron-methyl-sodium and mesosulfuron-methyl plus a third molecule known as mefenpyr-diethyl) is widely used for weed control in cereal agriculture. Hitherto, the microflora in Algerian soil, has received little investigation, and no information is so far available about the biodegradation of iodosulfuron-methyl sodium and the mesosulfuron-methyl. The main objectives of this study were firstly to determine the composition of the bacterial community in agricultural soil that had never been treated with Chevalier® OnePass herbicide and then to evaluate the changes after treatment with the herbicide. We also isolated resistant strains present in the treated soil and tested their ability to degrade the herbicide and thus to enable sensitive bacteria to survive despite the presence of the herbicide.
Materials and methods Collection of soil samples Two soil samples were collected from a wheatfield (Reguada plot, Guettar el Aïech, E 6° 34′ 4″, 36° 11′ 1″ N, Constantine,
Algeria). The first sample was collected in March 2012, a few days before the soil was treated with Chevalier® OnePass herbicide. The second sample was collected 1 month after the treatment. The soil was collected from the top 5 to 20 cm layer in ten flat and distinct areas in the wheatfield in order to get representative samples and to target the microflora of the rhizosphere. The samples were then bagged in sterile packaging to be transported to the laboratory, where roots and stones were removed and the soil samples were thoroughly homogenized by sieving through a 2-mm mesh. The soil was then kept at 4 °C until use. The soil samples were handled under sterile conditions to avoid any contamination. Chemicals Chevalier® OnePass herbicide was purchased from Bayer CropScienceAlgérie (Constantine, Algeria). It is composed of two sulfonylurea molecules: iodosulfuronmethyl sodium [sodium ({[5-iodo-2-(methoxycarbonyl) phenyl] sulfonyl} carbamoyl) (4-methoxy-6-methyl-1, 3, 5-triazin-2-yl) azanide] (30 g kg−1) and mesosulfuronmethyl [methyl 2-[(4, 6-d imethoxypyrimidin-2ylcarbamoyl)sulfamoyl]-α-(methanesulfonamido)-ptoluate] also known as Mesomaxx (30 g kg−1) plus a third molecule, mefenpyr-diethyl also known as safener (90 g kg−1). This latter acts as a catalyst and is highly selective in particular for wheat. The application rate of Chevalier® OnePass herbicide for agricultural purposes is 0.33 kg ha−1. Media The bacterial strains were isolated and grown in Luria Broth (LB) medium composed of yeast extract (5 g L−1), tryptone (10 g L−1), and NaCl (5 g L−1) (Miller 1972). The corresponding solid medium contained 15 g L−1 of agar. In the modified LB used to test the biodegradation capacity of the isolated strains and in the overlay cultures, the 5-g L−1 of yeast extract was replaced by 5 g L−1 of the herbicide before the medium was sterilized at 120 °C. This herbicide concentration allowed growth of the resistant strains. The fusion temperature of the active molecules of the herbicide is greater than 150 °C. Culture conditions Isolation of bacterial clones and growth under aerobic conditions One gram of either untreated or treated soil was mixed with 10 ml of LB in a 50-ml Erlenmeyer flask and then incubated in a rotary shaker at 30 °C for 6 h. This relatively short incubation time was chosen to reduce dominance by the most
Environ Sci Pollut Res
abundant bacterial species. A series of dilutions (10−1 to 10−6) was carried out from the two sample enrichments, and then 100 μl of each dilution was spread on LB agar plate in three replicates and incubated at 30 °C for 24 h. Distinct and separated clones were observed from the 10−3 to the 10−6 dilutions. Bacterial colonies were then distinguished on the basis of their macroscopic appearance (shape, color, dimensions, viscosity, etc.). Each of the clones was grown on a separate agar plate under the conditions described above. Isolation of bacterial clones and growth under microaerobic conditions One gram of either untreated or treated soil was mixed with 10 ml of LB in 10 ml tubes and incubated without shaking at 30 °C for 6 h. A series of dilutions (10−1 to 10−6) of the two sample enrichments was then performed; 100 μl of each dilution was then spread on the LB agar plate in three replicates and placed in an anaerobic jar with gas pack (Anaerocult® A, Merck, Germany) to provide the anaerobic conditions. The jars were then incubated at 30 °C for 48 h. Distinct and separated clones were observed from the 10−3 to the 10−6 dilutions. Bacterial colonies were recovered on the basis of their macroscopic appearance (shape, color, dimensions, viscosity, etc.). Each of the clones was grown on a separate LB agar plate under the conditions described above. Culture on medium containing the herbicide The ability of the isolated strains to use or to resist the herbicide was tested by growing them on a modified LB medium in which the 5-g L−1 of yeast extract had been replaced by 5 g L−1 of the herbicide. The growth of each aerobic and microaerobic isolated strain was determined. The conditions for each type of culture were as described above. As a control, the strains were grown on the usual LB.
Inductively coupled plasma optical emission spectrometry experiment One gram of each of the untreated or treated soil samples was mixed with 10 ml of sterile water and 100 μl of the solution was used. The samples were wet washed with 32.5 % nitric acid (Suprapur, Merck) for 12 h at 100 °C (Neumann et al. 2009) and were then filled to a tenfold volume with water prior to inductively coupled plasma optical emission spectrometry (ICP-OES) analysis. Three replicates for each sample were carried out, and the average sulfur concentration values were calculated. Sulfur analysis was performed using a ThermoFisher ICAP 6000 ICP-OES. Multielement standard solution XVI (Merck) was used as a reference. Metagenome extraction, 16S DNA amplification, and sequencing Soil DNA was extracted using the Power Soil DNA extraction Kit (MO-BIO Laboratories, CA, USA) following the manufacturer’s instructions. PCR was conducted using universal primers FD1 (5′-AGAGTTTGATCCTGGCTCAG-3′) and S6 (5′-GTATTACCGCGGCTGCTG-3′) (Chanal et al. 2006) in order to amplify a conserved fragment (approximately 500 bp) of the bacterial 16S rRNA gene. The 50-μl PCR reactions contained 10 ng of template DNA, the four dNTPs at 100 μM, the Go-Taq Polymerase buffer (Promega), the two convergent primers at 10 μM, one unit of the Go-Taq polymerase (Promega), and distilled water in a final volume of 50 μl. The reaction was carried out under the following conditions: an initial denaturing step at 94 °C for 1 min and 30 s followed by 30 cycles at 94 °C for 30 s, 60 °C for 30 s, and 72 °C for 45 s. The amplification of the 16S rDNA of the strains isolated was carried out from bacterial colonies using the same mix and under the same conditions. PCR products were purified using the GenElute PCR Clean-Up Kit (SIGMA-ALDRICH, USA) as indicated by the manufacturer, and the purified DNA was then sequenced.
Overlay cultures High-resolution melting DNA Each of the strains that was able to grow on the medium containing the herbicide (resistant strain) was grown as patches on the modified LB agar medium and incubated at 30 °C for 24 h. In parallel, each sensitive strain isolated from the untreated sample was grown on LB medium and incubated in a rotary shaker at 30 °C for 24 h. A soft layer of the modified LB medium was then laid over the patches of resistant strains, and a 3-μl spot of a culture of one of the sensitive strains was added over each patch. The effect of each of the resistant strains isolated was tested over each of the five sensitive strains. The overlay cultures were then incubated at 30 °C for 24 h.
The high-resolution melting DNA (HRM) runs were done on a CFX96 Real-Time System (Bio-Rad). The PCR reactions were carried out in a final reaction volume of 30 μl, and the mix was composed of 1× SoFast EvaGreen Supermix (BioRad), 500 nM of each primer (FD1 and S6), and 4 μl of each DNA tested (the amplified 500 bp fragment of the two soil samples and the oceanic control were purified and then diluted 100-fold). The oceanic metagenome was a kind gift of Valérie Michotey (Mediterranean Institute of Oceanology). It was extracted from water of the gulf of Fos (the Mediterranean sea, France). Cycling parameters of the real-time PCR
Environ Sci Pollut Res
consisted of an initial step at 95 °C for 2 min followed by 40 cycles at 95 °C for 10 s, 60 °C for 20 s, and 72 °C for 10 s. This program was followed by a high-resolution melting curve consisting of 95 °C for 1 min, 60 °C for 1 min, and ramping at 0.2 °C for 10 s from 79 to 95 °C during which the fluorescence was read at the end of each increment. The results were analyzed using Bio-Rad Precision Melt Analysis Software 1.1 (Bio-Rad). Phylogenic tree construction 16S prokaryotic sequences were downloaded from the Ribosomal Database Project (Cole et al. 2009). The uncultured sequences and the archaea sequences were eliminated from the data set. A first study using NCBI Blast (Johnson et al. 2008) selected the genus of the amplified 16S rDNA sequences (http://blast.ncbi.nlm.nih.gov/Blast.cgi). We extracted the 16S sequences from the Ribosomal Database Project that corresponded to the genus selected. A web server, known as “Phylogeny.fr” (Dereeper et al. 2008), was used for the phylogenetic analysis (http://www.phylogeny.fr/version2_ cgi/alacarte.cgi). The three main steps were used for the phylogenetic analysis. It was first performed on the basis of a multiple alignment of 16S rDNA sequences obtained using MUSCLE version 3.7 (Edgar 2004). The non-similar regions of the aligned sequences were removed by GBlocks version 0. 91b (Talavera and Castresana 2007). The phylogenetic trees were constructed using the BIONJ method (Gascuel 1997) with 100 bootstrap replicates. The numbers shown on each branch denote the percentages of bootstrap support. All methods were run with the default parameters suggested by their respective authors. After the first phylogenetic analysis, we manually removed the unnecessary sequences (we conserved two or three sequences per genus; these sequences were the sequences closest to each of the amplified 16S rDNA sequences in the phylogenic tree) and a second phylogenic analysis was then performed. This is shown in Fig. 3. Nucleotide sequence accession numbers The sequences determined during the present study have been deposited in the GenBank database. The accession numbers are available in Fig. 3. The strains isolated before treatment are indicated by an “N” followed by the number of the clone and then the GenBank accession number in parentheses. The strains isolated after treatment are indicated by “T,” followed by the number of the clone and the accession number in parentheses. Asterisks indicate clones obtained under microaerobic culture conditions. We chose one strain to represent each phylogenetic group of strains isolated in the phylogenetic trees.
Results and discussion Confirmation of the presence of the Chevalier® OnePass herbicide in treated soil using an ICP technique Both the active molecules of the Chevalier® OnePass herbicide contain sulfur. We used this fact to test for the persistence of herbicide residues in the second collection sample. The half-life of sulfonylurea in soil is known to range from a few days to a few months (Kamrin 1997). We collected the second sample 1 month after treatment to give the bacterial community enough time to evolve and to get used to the presence of the herbicide. Three determinations of the sulfur concentration in each soil type (untreated and treated) were carried out using an ICP technique and the average was calculated. In the untreated soil, the average sulfur concentration was 787.8± 18.7 ng ml−1, while in the treated soil, the average value was 1,561±17 ng ml−1, indicating that a significant fraction of the herbicide was still present in the sample. This means that the bacterial community in the soil was still exposed to the effects of the herbicide. Analysis of the amplified 16S rDNA in the untreated and treated soil using the HRM technique In order to obtain an overview of the total bacterial community in each of the soil samples, we used the HRM technique. Whole DNA was extracted and purified from soil samples as indicated in “Materials and Methods.” Conserved fragments of about 500 bp from the 16S rRNA genes were amplified using the FD1 and S6 universal convergent primers (Chanal et al. 2006), and similar fragments of 500 bp 16S rDNA were amplified from an oceanic metagenome for use as a control. The rDNA fragments were then tested using the HRM technique, and the fine analysis of the rDNA melting temperature curves was used to compare the overall bacterial populations present in the samples. Strikingly, the profiles of the rDNA fusion temperature curves were similar in the two soil samples, whereas the curve profile of the oceanic rDNA was clearly distinct (Fig. 1). The minor difference observed between the curve profiles of the untreated and treated soil could indicate limited differences in the bacterial diversity. This result is consistent with previous studies where denaturing gradient gel electrophoresis analysis was used to compare bacterial communities (Valle et al. 2006; Lin et al. 2012). Isolation and analysis of the cultivable strains recovered from the untreated and treated soil According to the HRM results, we expected to find that the culturable bacterial fraction in the untreated and treated soil was similar overall, with the possible appearance of some new
Environ Sci Pollut Res Fig. 1 Comparison of the bacterial communities in the untreated and treated soil using the HRM technique. The metagenomes of the two soil samples were extracted, as explained in “Materials and methods,” and then a conserved 16S rDNA fragment was amplified. The green curve represents the DNA melting profile of the untreated soil and the blue curve that of the treated soil. An oceanic metagenome 16S rDNA (represented by the red curve) was used as a control
species in the treated soil. To confirm this hypothesis, we isolated the cultivable bacterial fraction from both the untreated and treated soil samples. We isolated 259 distinct macroscopic clones under two different culture conditions (aerobic and microaerobic). We were able to distinguish 77 distinct macroscopic clones in the untreated sample and 64 in the treated one under aerobic conditions. Under microaerobic conditions, 66 clones were isolated in the untreated sample, and 52 were found in the treated one (Fig. 2). In order to identify the phylogenetic relationship between the isolated clones and the known bacterial species, a conserved fragment of about 500 bp from the 16S rRNA genes was amplified from each isolated clone. The PCR products were then sequenced,
Fig. 2 Number of strains isolated from the treated and untreated soil under both aerobic and microaerobic conditions. Aerobic cultures were incubated at 30 °C for 24 h. The microaerobic cultures were incubated at 30 °C for 48 h, and the agar plates were placed in anaerobic jars with an Anaerocult A
and the 259 sequences were compared to those of known bacteria to construct phylogenetic trees. Figure 3 strongly suggests that some bacterial groups disappear after treatment under both aerobic and microaerobic conditions. This was true for the group affiliated to the Lysinibacillus genus (N21, N15, and N13 on the phylogenetic tree) and the microaerobic strain related to Clostridium genus (N45* on the phylogenetic tree). The reason could be that the application of the Chevalier® OnePass herbicide has harmful effects on certain bacterial strains and prevents their growth. These strains could therefore be, thus, highly sensitive to this herbicide. On the other hand, some groups persisted after treatment under both culture conditions. This was true for the clones related to Bacillus megaterium (N49 before treatment and T5 after treatment), Achromobacter marplatensis (N5 before treatment and T50 and T1 after treatment), Stenotrophomonas maltophilia (N9 and N1 before treatment and T4 after treatment), Paenibacillus ginsengisoli (N17* before treatment and T9* after treatment), Bacillus cereus (N2* before treatment and T6 and T3* after treatment), and Bacillus circulans (N6* before treatment and T8* after treatment). This indicates that these strains were able to resist the toxicity of the herbicide and thus to spearhead a rapid recovery of a bacterial community. Furthermore, we observed that strains related to Sphingobacterium (T10), Streptomyces (T41), and Pseudomonas (T27 and T32) genera appeared after treatment with the herbicide. Finding new species after treatment with the herbicide could mean that they are better adapted to resist the herbicide, and that they could even degrade it. These results are consistent with the HRM analysis and with previous studies (Valle et al. 2006; Lin et al. 2012).
Environ Sci Pollut Res
Fig. 3 Phylogenetic trees based on comparing 500 bp long sequences of the 16S rRNA gene. The GenBank accession number for each strain isolated is shown in parentheses after the strain name. N indicates the strains isolated from the untreated soil, and T stands for those isolated
from treated soil. An asterisk indicates that the strain was isolated under microaerobic conditions. The scale bar indicates 0.02 substitutions per nucleotide position. Bootstrap values obtained with 100 resamplings were indicated as percentages on all branches
On the basis of the phylogenetic trees, we noted that the presence of the herbicide did not change the global composition of the bacterial community, but that it did affect the growth of certain groups. For instance, the strains related to the Lysinibacillus genus disappeared after treatment even though they were abundant before treatment under aerobic culture conditions (21 % of the isolated strains). On the other hand, some other strains increased after treatment. This was the case of the strains related to Paenibacillus sp. isolated under microaerobic conditions and, as expected, Pseudomonas became the dominant group in the whole cultured fraction isolated from the treated soil under aerobic conditions (27 % of the isolated strains). Indeed, many strains related to the Pseudomonas genus have been described as being able to degrade some sulfonylurea molecules (Li-feng et al. 2007; Manickam et al. 2008; He et al. 2012). This was true for Pseudomonas mediterranea, as we found in this study (see below). Taken as a whole, our results confirm the HRM analysis and appear to be consistent with previous studies. It reveals that the treatment with Chevalier® OnePass herbicide has a drastic effect on the bacterial community, changing the relative abundance of certain strains and probably eliminating others. The resistant strains then recovered and other bacterial groups appeared.
Culture of the isolated strains in the presence of the herbicide and bacterial overlay cultures The analysis of the strains isolated led us to whether they were able to make use of the herbicide or, at least, one of its components. We hypothesized that the strains that disappeared after treatment were highly sensitive to the herbicide, whereas those found in both the soil samples, as well as the new strains found after treatment, were able to resist and probably to degrade the herbicide. If this is true, then they could detoxify the medium and therefore allow the sensitive strains to grow subsequently. To test this hypothesis, we made a modified LB medium in which the herbicide could be used as nutritional resource, as explained in “Materials and Methods.” We first tested the growth of strains representative of each group isolated from the untreated soil. All the strains (isolated under the two culture conditions) were unable to grow in the presence of the herbicide, except the strains related to S. maltophilia (N9 and N1) (data not shown). This could explain why a strain related to this species was found after treatment (T4). Surprisingly, of the ten representative strains isolated aerobically from the treated soil, four were unable to grow on the herbicide medium (T1, T50, T41, and T5). These strains therefore appeared to be sensitive to the herbicide, but,
Environ Sci Pollut Res
nevertheless, we found that they persisted in the treated sample. None of the strains isolated under microaerobic conditions was able to grow in the presence of the herbicide (data not shown). We identified a total of eight resistant strains. As explained in the “Introduction,” the target of the active compounds in Chevalier® OnePass herbicide is the ALS enzyme. We hypothesize that the resistant strains either possess a modified form of ALS that is insensitive to the components of the herbicide or are able to degrade the herbicide to protect the ALS enzyme and thus detoxify the medium. This latter hypothesis could explain the presence of sensitive strains in the treated soil sample. To test this hypothesis, the six resistant strains isolated from the treated soil were grown on the modified LB medium, and the effect of the resistant strain was then tested on the sensitive strains isolated before treatment, as explained in “Material and Methods.” Unexpectedly, among the six resistant strains, only strain T4, which is related to S. maltophilia, allowed all the sensitive strains to grow (Fig. 4a, b). When T4 was grown on the medium containing the herbicide, a dark precipitate formed on the bottom of the agar plate. This effect was not observed for the other resistant strains (Fig. 4a, data not shown). We suppose that the precipitate was a sulfur precipitate resulting from degradation of the sulfur-containing compounds (iodosulfuron-methyl sodium and mesosulfuronmethyl) in the herbicide (Labrenz and Banfield 2004). As indicated in Table 1, the other resistant strains only enabled some of the sensitive strains to grow on the herbicide. For instance, resistant strain T27 (affiliated to P. mediterranea) allowed N13 to grow but without producing the dark precipitate observed with T4 (Fig. 4b). On the basis of these experiments, we suggest that Chevalier® OnePass herbicide could be entirely degraded by the strain related to S. maltophilia.
Table 1 Overlay cultures of resistant and sensitive strains on a medium containing Chevalier® OnePass herbicide
Fig. 4 Analysis of the resistant strains and overlay cultures on the modified LB. a Growth of resistant strains T4 (left) and T27 (right) on a medium containing Chevalier® OnePass herbicide. A dark precipitate appeared during the growth of the T4 strain. b Overlay cultures of
resistant strains T4 and T27 with sensitive strains (only N5, N9, N13, and N15 are shown here) in a medium containing the herbicide. The negative control corresponds to resistant strain T27 that did not allow sensitive strain N1 to grow
RS SS
T4
T20
T10
T32
T27
T6
N13 N15 N21 N49 N5
+ + + + +
− − − + −
− + − − −
+ + − − −
+ − − − −
+ + − − −
(+) indicates that the resistant strain allowed the sensitive strain to grow; (−) indicates that the resistant strain did not allow the sensitive strain to grow SS sensitive strains isolated from untreated soil, RS resistant strains isolated from treated soil
Despite the fact that it is underrepresented in the treated sample, strain T4 therefore seems to be a good candidate for studying the molecular mechanism involved in the biodegradation of Chevalier® OnePass herbicide. This experiment also showed that strains related to P. mediterranea were able to degrade the herbicide, but that they did not allow all the sensitive strains to grow. This could be why they are abundant in the culturable bacterial community after treatment with the herbicide. Since they did not produce the dark precipitate, we suggest that they degrade only partially the sulfonylurea components of the herbicide. In conclusion, the behavior of the bacteria consortium in a treated soil is not as simple as previously thought. In particular, we have shown that both resistant and sensitive strains coexisted in soil containing herbicide. The complete degradation of the Chevalier® OnePass herbicide appears to be achieved only by the strain related to S. maltophilia, a resistant
Environ Sci Pollut Res
strain that allows all the sensitive strains tested to grow. The other resistant strains enabled only certain sensitive strains to grow on the herbicide. This could be related to their herbicidebiodegradation pathway. It would be interesting to decipher the molecular mechanisms involved in these alternative degradation pathways and to explain why strains related to S. maltophilia are capable of completely detoxifying the medium even though they constitute only a small proportion of the total bacterial community.
Conclusion Our results reveal that the presence of the sulfonylurea Chevalier® OnePass herbicide modified the balance between relative species abundances, but did not significantly modify the overall composition of the bacterial community in the soil. Indeed, the soil bacterial community was able to recover, and exposure probably enriched its diversity. We have also shown that the novel HRM technique is an efficient and rapid alternative method that could provide an overview of the entire bacterial composition. This technique is in its infancy, but it will be certainly improved in the future to establish a correspondence between DNA fusion temperatures and bacterial species. New generation sequencing methods are powerful tools for analyzing microbial populations, but they are expensive and more time consuming. Interestingly, our work showed that treated soil contains both resistant and sensitive strains. Furthermore, the resistant strains could use different pathways to degrade the herbicide. Strain T4 could be a good model strain for investigating the biodegradation pathways of iodosulfuron methyl sodium and mesosulfuron methyl, which were studied for the first time here. It will be interesting to clarify the degradation mechanisms used by this strain and to investigate, in detail, the possible use of this strain to detoxify agricultural fields that have been contaminated by Chevalier® OnePass herbicide. Acknowledgments We would like to thank A. Rouabah for his help in the soil sample collection and V. Michotey for supplying the oceanic metagenome. We are grateful to M. Ilbert, C. Aussignargues, and S. Bouillet for the valuable suggestions and discussions. Monika Gosh is acknowledged for improving the English version of the manuscript. This work is funded by the Faculté des Sciences de la Nature et de la Vie, Université Constantine 1 (Algeria), the Centre National de la Recherche Scientifique (CNRS), and the Aix-Marseille Université (France).
References Berger BM, Janowitz K, Menne HJ, Hoppe H-H (1998) Comparative study on microbial and chemical transformation of eleven sulfonylurea herbicides in soil. Z Für Pflanzenkrankh Pflanzenschutz 105: 611–623
Beyer EM, Duffy MF, Hay JV, Schlueter DD (1988) Sulfonylurea. In: Kearney PC, Kaufman DD (eds) Herbic. Chem. Degrad. Mode Action, Dekker. New York, pp 117–183 Brown HM (1990) Mode of action, crop selectivity, and soil relations of the sulfonylurea herbicides. Pestic Sci 29:263–281. doi:10.1002/ps. 2780290304 Chanal A, Chapon V, Benzerara K et al (2006) The desert of Tataouine: an extreme environment that hosts a wide diversity of microorganisms and radiotolerant bacteria. Environ Microbiol 8:514–525. doi: 10.1111/j.1462-2920.2005.00921.x Cole JR, Wang Q, Cardenas E et al (2009) The Ribosomal Database Project: improved alignments and new tools for rRNA analysis. Nucleic Acids Res 37:D141–D145. doi:10.1093/nar/gkn879 Dereeper A, Guignon V, Blanc G et al (2008) Phylogeny.fr: robust phylogenetic analysis for the non-specialist. Nucleic Acids Res 36: W465–W469. doi:10.1093/nar/gkn180 Edgar RC (2004) MUSCLE: a multiple sequence alignment method with reduced time and space complexity. BMC Bioinforma 5:113. doi:10. 1186/1471-2105-5-113 Gascuel O (1997) BIONJ: an improved version of the NJ algorithm based on a simple model of sequence data. Mol Biol Evol 14:685–695 Hang B-J, Hong Q, Xie X-T et al (2012) SulE, a sulfonylurea herbicide de-esterification esterase from Hansschlegelia zhihuaiae S113. Appl Environ Microbiol 78:1962–1968. doi:10.1128/AEM. 07440-11 He W-H, Wang Y-N, Du X et al (2012) Pseudomonas linyingensis sp. nov.: a novel bacterium isolated from wheat soil subjected to longterm herbicides application. Curr Microbiol 65:595–600. doi:10. 1007/s00284-012-0187-3 Hemmamda S, Calmon M, Calmon JP (1994) Kinetics and hydrolysis mechanism of chlorsulfuron and metsulfuron-methyl. Pestic Sci 40: 71–76. doi:10.1002/ps.2780400112 Ismail BS, Goh KM, Kader J (1996) Effects of metsulfuronmethyl on microbial biomass and populations in soils. J Environ Sci Health Part B 31:987–999. doi:10.1080/03601239609373049 Johnson M, Zaretskaya I, Raytselis Y et al (2008) NCBI BLAST: a better web interface. Nucleic Acids Res 36:W5–W9. doi:10.1093/nar/ gkn201 Kamrin MA (1997) Pesticide profiles: toxicity, environmental impact, and fate. CRC Press Labrenz M, Banfield JF (2004) Sulfate-reducing bacteria-dominated biofilms that precipitate ZnS in a subsurface circumneutral-pH mine drainage system. Microb Ecol 47:205–217. doi:10.1007/s00248003-1025-8 Lee Y-T, Cui C-J, Chow EWL et al (2013) Sulfonylureas have antifungal activity and are potent inhibitors of Candida albicans acetohydroxyacid synthase. J Med Chem 56:210–219. doi:10. 1021/jm301501k Li-feng G, Jian-dong J, Xiao-hui L et al (2007) Biodegradation of ethametsulfuron-methyl by Pseudomonas sp. SW4 isolated from contaminated soil. Curr Microbiol 55:420–426. doi:10.1007/ s00284-007-9011-x Lin X-Y, Yang Y-Y, Zhao Y-H, Fu Q-L (2012) Biodegradation of bensulfuron-methyl and its effect on bacterial community in paddy soils. Ecotoxicol Lond Engl 21:1281–1290. doi:10.1007/s10646012-0882-7 Lu P, Jin L, Liang B et al (2011) Study of biochemical pathway and enzyme involved in metsulfuron-methyl degradation by Ancylobacter sp. XJ-412-1 isolated from soil. Curr Microbiol 62: 1718–1725. doi:10.1007/s00284-011-9919-z Madigan MT, Martinko JM (2006) Brock Biology of Microorganisms. Pearson Prentice Hall. Manickam N, Ghosh A, Jain RK, Mayilraj S (2008) Description of a novel indole-oxidizing bacterium Pseudomonas indoloxydans sp. nov., isolated from a pesticide-contaminated site. Syst Appl Microbiol 31:101–107. doi:10.1016/j.syapm.2008.02.002
Environ Sci Pollut Res Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory Neumann M, Mittelstädt G, Seduk F et al (2009) MocA is a specific cytidylyltransferase involved in molybdopterin cytosine dinucleotide biosynthesis in Escherichia coli. J Biol Chem 284:21891– 21898. doi:10.1074/jbc.M109.008565 Nicholls PH, Evans A. (1987) The behavior of chlorsulfuron and metsulfuron in soils in relation to incidents of injury to sugar beet. Proc. Br. Crop Prot. Weeds Conf. BCPC Publications, pp 549–556 Sarmah AK, Sabadie J (2002) Hydrolysis of sulfonylurea herbicides in soils and aqueous solutions: a review. J Agric Food Chem 50:6253– 6265 Sondhia S (2009a) Persistence of metsulfuron-methyl in paddy field and detection of its residues in crop produce. Bull Environ Contam Toxicol 83:799–802. doi:10.1007/s00128-009-9822-5 Sondhia S (2009b) Leaching behaviour of metsulfuron in two texturally different soils. Environ Monit Assess 154:111–115. doi:10.1007/ s10661-008-0381-8 Sondhia S, Waseem U, Varma RK (2013) Fungal degradation of an acetolactate synthase (ALS) inhibitor pyrazosulfuron-ethyl in soil. Chemosphere. doi:10.1016/j.chemosphere.2013.07.066
Talavera G, Castresana J (2007) Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. Syst Biol 56:564–577. doi:10.1080/ 10635150701472164 Umbarger HE, Brown B (1958) Isoleucine and valine metabolism in Escherichia coli. VIII. The formation of acetolactate. J Biol Chem 233:1156–1160 Valle A, Boschin G, Negri M et al (2006) The microbial degradation of azimsulfuron and its effect on the soil bacterial community. J Appl Microbiol 101:443–452. doi:10.1111/j.1365-2672.2006.02937.x Xu J, Li X, Xu Y et al (2009) Biodegradation of pyrazosulfuron-ethyl by three strains of bacteria isolated from contaminated soils. Chemosphere 74:682–687. doi:10.1016/j.chemosphere.2008.09. 078 Yin LB, Liu Y, Zhang DY, Zhang SB (2011) Isolation and characterization of Rhodopseudomonas sp. S9-1 capable of degradating pyrazosulfuron-ethyl. Adv Mater Res 356–360:1152–1163. doi:10. 4028/www.scientific.net/AMR.356-360.1152 Zanardini E, Negri M, Boschin G et al (2002) Biodegradation of chlorsulfuron and metsulfuron-methyl by Aspergillus niger. Sci World J 2:1501–1506. doi:10.1100/tsw.2002.281
