SEARCH

SEARCH BY CITATION

Keywords:

  • Difloxacin;
  • resistance;
  • DGGE ;
  • soil;
  • manure;
  • rhizosphere

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

Difloxacin (DIF) belongs to the class of fluoroquinolone antibiotics that have been intensively used for the treatment of bacterial infections in veterinary and human medicine. The aim of this field study was to compare the effect of manure from DIF-treated pigs and untreated pigs on the bacterial community structure and resistance gene abundance in bulk soil and rhizosphere of maize. A significant effect of DIF manure on the bacterial community composition in bulk soil was revealed by denaturing gradient gel electrophoresis (DGGE) of bacterial 16S rRNA gene fragments amplified from total community DNA. In few samples, quinolone resistance genes qnrB and qnrS1/qnrS2 were detected by PCR and subsequent hybridization, while qnrA was not detected. Quantitative PCR revealed an increased abundance of the integrase gene intI1 of class I integrons and sulfonamide resistance genes sul1 and sul2 in DIF manure-treated bulk soil and rhizosphere, relative to 16S rRNA genes, while traN genes specific for LowGC-type plasmids were increased only in bulk soil. Principal component analysis of DGGE profiles suggested a manure effect in soil until day 28, but samples of days 71 and 140 were found close to untreated soil, indicating resilience of soil community compositions from disturbances by manure.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

Difloxacin (DIF) belongs to the class of fluoroquinolones, a group of synthetic antibiotics commonly used in human and veterinary medicine for the treatment of both Gram-negative and Gram-positive infections (Aarestrup, 2005; Cabello, 2006; Adriaenssens et al., 2011). They inhibit the essential bacterial enzymes DNA gyrase and DNA topoisomerase IV (Drlica & Zhao, 1997). In recent years, the resistance against fluoroquinolones has increased and limited the treatment of bacterial infections (Dalhoff, 2012). The use of fluoroquinolones in animal husbandry is assumed to contribute to the emergence and spread of antimicrobial resistance (Collignon et al., 2009), and recently, a potential transfer of plasmid-mediated quinolone resistance genes from wastewater of swine feedlots to adjacent fields was described (Li et al., 2012). After oral application to the pigs, only a minor part of DIF is metabolized by the animal, and the main part of the parent compound is excreted in bioactive concentrations (Sukul et al., 2009). In manure, DIF is very stable during storage (Lamshöft et al., 2010) and reaches agricultural fields where it may affect the structure and function of the microbial community and may lead to an increased abundance and transfer of antibiotic resistance genes. Recently, in the field experiment reported here and in other trials, it was shown that DIF was hardly (bio) accessible and rapidly formed nonextractable residues, which might limit its effect on the microbial community (Rosendahl et al., 2012). In another experiment, 14C-labeled DIF was applied with manure or water to soil, and the mineralization of 14C-DIF was shown to be low (below 0.2%), while nonextractable residues of 14C-DIF increased to 70–74% after 56 and 120 days (Junge et al., 2012). Nevertheless, in a microcosm experiment, effects of DIF on the microbial biomass, respiration, potential denitrification, and ratio of bacteria/fungi were observed up to eight days after application by manure (Kotzerke et al., 2011). To date, five different transferable mechanisms of quinolone resistance (TMQR) are described in the literature including quinolone target protection (qnr genes), enzymatic inactivation (ACC(6′)-Ib-cr), plasmid-mediated effects on slow growth, plasmid-encoded efflux pumps, and exogenous exchange of DNA (reviewed by Ruiz et al., 2012b). Qnr genes have been found in a large variety of plasmids (Strahilevitz et al., 2009). However, DNA sequences in the vicinity of qnr genes were reported to be rather similar and usually integrated in complex sul1-type integrons and associated with insertion sequence common region 1 (ISCR1) (Nordmann & Poirel, 2005; Garnier et al., 2006). Integrons can be used by Gram-positive and Gram-negative bacteria to stockpile and express different exogenous open reading frames including resistance genes and are considered to play a central role in the worldwide problem of antibiotic resistance (Mazel, 2006; Gillings et al., 2008a). The presence of antibiotics such as fluoroquinolones can lead to the activation of the SOS response in bacteria, which results in integrase overexpression and correspondingly in the raise of recombination events of gene cassettes (Stalder et al., 2012). Therefore, the abundance of class 1 integrons and related sul genes in a microbial community might be used as an estimate of potential coselection of resistance gene cassettes. In addition, an induction of natural competence in response to fluoroquinolone antibiotics was described (Charpentier et al., 2012), which might contribute to the spread of resistance genes in bacterial populations via transformation. Therefore, we hypothesize that despite the reported rapid formation of nonextractable residues of DIF in soil (Rosendahl et al., 2012), the application of manure from DIF-treated pigs will affect the structure of the bacterial communities in field soil as well as the abundance of resistance genes and mobilizing elements such as integrons compared with soil treated with manure from unmedicated pigs, which might contribute to the public health risk of antibiotic resistance.

The aim of this field study was to assess the effects of manure from DIF-treated pigs on the bacterial community structure in bulk soil and rhizosphere of maize by denaturing gradient gel electrophoresis (DGGE) of 16S rRNA gene fragments amplified from total community (TC) DNA. Detection of the quinolone resistance genes qnrA, qnrB, and qnrS1/qnrS2 was performed by PCR from TC DNA with subsequent Southern blot hybridization. Sulfadiazine resistance genes sul1 and sul2 as well as class 1 integron-integrase genes (intI1) and traN specific for LowGC-type plasmids, which were recently described to play an important role in dissemination of resistance against sulfonamides in manure and manure-treated soils (Heuer et al., 2009), were quantified in TC DNA by real-time PCR as marker for a potential coselection and transferability of resistance genes.

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

Field experiment

Manure was produced as described previously (Rosendahl et al., 2012). In brief, to produce DIF manure, the prescribed dose of DIF (5 mg kg−1 body weight, Dicural injectable solution, Fort Dodge, Würselen, Germany) was administered intramuscularly to pigs on four consecutive days, and manure was collected for 10 days (composite sample). Control manure was produced by collecting manure from untreated pigs (composite sample).

The field trial was conducted from May 2010 until February 2011 as described previously (Rosendahl et al., 2012). At the field site Merzenhausen located near Jülich (Germany), which only received inorganic fertilizer during the last 30 years, DIF manure or control manure was applied in an amount of 30 m3 ha−1 on experimental plots (3 × 6 m, n = 4) and incorporated to a depth of 12 cm (day 0), followed by sowing of maize (Zea mays L.). Bulk soil was sampled on days 0, 7, 14, 28, 71, 105, and 140 after manure application in between rows using a soil corer (5.6 cm inner diameter × 12 cm height) by mixing three subsamples. Rhizosphere samples were taken on days 28, 71, 105, and 140 by digging out ≥ 3 plants, shaking, and cutting roots plus remaining soil into small pieces and thorough mixing. Five grams of root material with attached soil was treated in a stomacher blender three times with 15 mL 0.85% sodium chloride solution for 30 s at high speed (Stomacher 400 laboratory blender; Seward, UK), and the suspensions were decanted into 50-mL Falcon tubes. Rhizosphere soil pellets were generated by centrifugation of the rhizosphere solutions for 15 min at 3780 g at 4 °C, and the pellets were homogenized. Additionally, soil was sampled before the application of manure (composite sample).

DNA extraction, PCR, and DGGE

TC DNA from manure, soil samples, and rhizosphere soil pellets was extracted using the FastDNA® Spin Kit for soil (MP Biomedicals, Heidelberg, Germany) and purified by the GeneClean® Spin Kit (MP Biomedicals), following the instructions of the manufacturer. Amplification of 16S rRNA gene fragments by PCR and separation of PCR products by DGGE were performed as described previously (Weinert et al., 2009). PCR products were arranged randomly on the DGGE gels to avoid clustering effects based on the gel position in the following community analysis. After the alignment, a Pearson correlation was performed with the program GelCompar II® (version 6.5, Applied Maths, Austin). The similarity matrix was used to test for significant treatment effects and to determine the adjusted difference between bacterial communities as described previously (Kropf et al., 2004). In short, permutation tests (104 random permutations) for the comparison of groups of lanes based on pairwise similarity measures were applied. The difference in community structure between treatments was measured using the average of correlation coefficients within treatment minus the average of correlation coefficients between treatments.

Principal component analysis (PCA)

Ordination was performed using canoco 4.5 (Microcomputer Power, Ithaca, NY). Densitometric curves were produced with GelCompar II® from DGGE lanes and digitalized, and the resulting signals at the positions of the gel were used as densitometric scores data. Sampling days and treatment (control or DIF treatment) have been binary-encoded and can hence be shown as environmental variables in the biplot. The scores of these environmental variables were calculated based on the densitometric scores composition of all samples at the respective sampling times or in all DIF-treated or control samples, respectively. Indirect gradient analysis was used for the ordination; hence, while working with only the densitometric curves data (densitometric scores), environmental data were included in the biplots only for facilitating interpretation. Manure and soil samples that were not treated with manure were not used for the ordination, but are shown as supplementary samples. Sample scores for these supplementary samples can be calculated based on their densitometric scores composition. Detrended correspondence analysis was used first to check whether linear or unimodal response of the data should be considered. The length of the gradient for the first canonical axis was 1068 for the bulk soil samples and 1456 for the rhizosphere samples using the ‘detrending by segments’ option, hence indicating a linear response of the densitometric data. Subsequently, ordination was performed with PCA. Focus scaling was set to ‘intersample distances’, and densitometric scores were posttransformed by standard deviation. The explained variance being associated with the PCA axis resulted from a transformation of the original data and is reported in terms of the quotient between the explained and the overall variance in the data (%). From the percentage of the explained variance, only the part that is explained by the first and second axes, hence not 100% of the explained variability, is shown in the biplots. We report this as percentage of the explained variability that is shown in the biplots (%).

Principal response curve (PRC)

PRCs were calculated based on redundancy analysis (RDA) of the densitometric curves for the rhizosphere and bulk soil samples. The same design matrix as for the PCA was used for RDA. The principle of PRC is to consider time points explicitly as cofactors for the statistical analysis and to separate between time and treatment influence by assuming that an effect is related to the product between time point and treatment. In this way, the contribution of time and treatment can be separated quantitatively from each other (Van den Brink & Ter Braak, 1999). The RDA was performed on nontransformed densitometric curve data with focus scaling on intersample distances and scaling of densitometric scores by standard deviations. Time points were classified as covariables, and products between time points and DIF treatments were considered as interactions of explanatory variables. The Cdt values on the y-axis of the PRC diagrams (Fig. 2) can be interpreted as the aggregated community response, in this case the aggregated response of the densitometric curves, to the treatment over time. This information is completed by the densitometric scores, in this case the scores of the single DGGE gel positions, that aggregate the response of the DGGE bands to the treatment (Figs 3 and 4).

PCR and hybridization of resistance genes

The occurrence of qnrA, qnrB, and qnrS genes was analyzed by PCR amplification using specific primers as previously described (Robicsek et al., 2006), while for qnrS2, 5′-GCGACATTTATTAACTGCAA-3′ was used as forward primer. Subsequent Southern blot hybridization of PCR products was performed as described previously (Sambrook et al., 1989). Probes were generated from PCR products obtained from plasmids pQC for qnrA; pMG298, pMG307, and pMG305 for qnrB; and pMG306, pGNB2, and pMG308 for qnrS, which were excised from the agarose gel after electrophoresis and labeled with digoxigenin according to the instructions of the manufacturer (Roche, Applied Science, Mannheim, Germany).

Relative abundance of sul1, sul2, traN, and intI1

The copy numbers of target genes were quantified by real-time PCR 5′-nuclease assays in a CFX96™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA) as described previously for sul1, sul2, bacterial 16S rRNA genes (rrn) (Heuer & Smalla, 2007; Heuer et al., 2009), traN of LowGC-type plasmids (Heuer et al., 2009), and intI1 of class 1 integrons (Barraud et al., 2010). The relative gene abundance was calculated by dividing the respective gene abundance by rrn copy number to compensate for differences in DNA extraction and amplification efficiency between samples.

The effect of DIF manure on the relative abundance of sul1, sul2, traN, and intI1 in bulk soil and rhizosphere was analyzed by a linear mixed model using proc mixed of the sas 9.2 (SAS Institute Inc., Cary, NC). For a conservative estimation of relative abundance of intI1 and traN, values below detection limit of the qPCR method were replaced by the calculated value of the detection limit threshold. To compare the effect of DIF manure between bulk soil and rhizosphere, the means and differences between the respective treatments were calculated for days 28, 71, 105, and 140 and analyzed by a linear mixed model using proc mixed. Effects were declared as significant at < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

Pearson correlation and statistical evaluation of DGGE profiles

DGGE analysis of 16S rRNA gene fragments amplified from TC DNA revealed that the bacterial community composition of the bulk soil treated with manure from DIF-medicated pigs differed significantly from the composition of the soil that was treated with control manure (permutation test according to Kropf et al. (2004); P < 0.05). The difference between the treatments was smaller than the difference between time points 0, 7, 28, and 140 with 1.5% for treatment differences and 11.1% for time point differences, respectively. For the rhizosphere, the difference between time points 28, 71, and 140 was significant with 17.2%. The difference between treatments was smaller than the difference between replicates. DGGE gels with randomized sample loading, which were used for the analysis, as well as DGGE gels with blockwise sample loading are shown in the Supporting Information (Figs S1–S4).

PCA

PCA were performed using densitometric curves data of the DGGE profiles from DGGE gels with randomized sample loading. For the bulk soil samples, the PCA explained in total 42.4% of the variability in the densitometric data composition of the samples. The biplot of the sample and environmental data scores (Fig. 1a) showed in total 75.9% of the variance that is explained by the PCA. The samples of the first two time points 0 and 7 were mainly separated from each other and from the other time points, whereas control and DIF samples could not be separated in the biplot. In the biplot, the sampling time points seemed to determine the bacterial community composition as major influencing factor. The distances between DIF and control were smaller as compared to the distances between the sampling times, indicating that the community composition of the DIF and control samples was similar to each other, whereas the community composition on the different time points showed more dissimilarity.

image

Figure 1. Biplots of sample scores and environmental scores for bulk soil (a) and rhizosphere (b) as the result of PCA of DGGE gels. Abbreviations indicate manure from untreated pigs (control manure), manure from pigs treated with DIF (DIF manure), soil samples taken before the start of the experiments from unmanured soil (untreated soil), bulk soil samples from day i mixed with control manure (CBS-i), bulk soil samples from day i mixed with DIF manure (DBS-i), rhizosphere soil samples from day i mixed with control manure (CRH-i), and rhizosphere soil samples from day i mixed with DIF manure (DRH-i). Some labels have been deleted in the biplots for reasons of readability. For bulk soil samples (a), the PCA explains on all axes 42.4% of the variance in the DGGE data. For rhizosphere samples (b), the PCA explains on all axes 55.7% of the variance in the DGGE data.

Download figure to PowerPoint

For the rhizosphere samples, the PCA explained in total 55.7% of the variability in the densitometric data composition of the samples. The biplot of the sample and environmental variable scores (Fig. 1b) showed in total 84.3% of the total variance that is explained by the PCA. The distribution in the biplot was very similar to that from the bulk soil samples. Samples and environmental variables were well separated at the first two time points 0 (bulk soil) and 28 (rhizosphere), whereas the influence of control and DIF was small in comparison. The rhizosphere samples from days 71 and 140 grouped tightly together in the biplot.

Additionally, the densitometric data from manure from DIF-treated pigs (DIF manure) and control pigs (control manure) as well as from samples of unmanured soil (untreated soil) had been included into the biplot by calculating the sample scores based on their densitometric score composition. However, these data were not used to calculate the ordination to avoid that the difference between manure and all other samples would have been the main factor governing the grouping of data points (results not shown). Manure and untreated soil samples were only displayed in the biplots to show similarity to other samples or classes. The manure samples were clustering well with day 0 and day 7 samples, whereas untreated soil samples were clustering with days 71 and 140.

Principal response curve (PRC)

PRC were used to quantify the variance in the densitometric data of DGGE lanes, which can be explained by time and treatment. For the bulk soil (Fig. 2a), time explained 34.5% of the data variance, whereas the DIF treatment explained 18.1% of the variance in densitometric data composition. The missing 47.4% was attributed to the differences between the replicates. The corresponding densitometric scores (right-hand side of Fig. 3) show DGGE gel positions. The DGGE gel positions 200–220, 690–710, and 760–790 showed the highest scores; hence, bands in these regions of the DGGE gel with low to moderate GC content showed the strongest deviations from the main composition of the community in the positive direction. The highest negative scores were found for the gel positions 1760–1780.

image

Figure 2. Principal response curves (PRC) based on RDA of the densitometric curves for the (a) bulk soil and (b) rhizosphere samples, showing the canonical coefficients of the RDA over time. For bulk soil, PRC attributes 34.5% of the variability to time, 47.4% to differences between replicates, and 18.1% to the DIF treatment, while 59.6% of the treatment variance is displayed on the first axis. For rhizosphere, PRC attributes 52.7% of the variability to time, 39.3% to differences between replicates, and 8% to the DIF treatment, while 54.9% of the treatment variance is displayed on the first axis.

Download figure to PowerPoint

image

Figure 3. DGGE profiles of 16S rRNA gene fragments amplified from total genomic DNA extracted from manure and bulk soil samples. Profiles of the randomized DGGE gels were aligned and sorted by treatment and days postapplication. Additionally, profiles of unmanured soil (a), DIF manure (b), and control manure (c) are shown. Densitometric scores for each DGGE coordinate indicate its contribution to the explained variance. Positive values indicate an increase in band intensity in the DIF treatment compared with the control.

Download figure to PowerPoint

For the rhizosphere soil (Fig. 2b), time explained 52.7% of the data variance, whereas the DIF treatment explained 8% of the variance in densitometric data composition. The missing 39.3% was attributed to the differences between the replicates. The corresponding densitometric scores (right-hand side of Fig. 4) showed highest values at DGGE gel positions 720–740, 1070–1100, and 1140–1180 with low to moderate GC content, whereas the positions 1460–1480 deviated with the largest negative scores. Comparing the densitometric scores of the bulk soil and rhizosphere DGGEs, the three peaks in the positive direction mentioned above mark bands with similar melting behavior.

image

Figure 4. DGGE profiles of 16S rRNA gene fragments amplified from total genomic DNA extracted from manure, bulk soil (BS), and rhizosphere samples. Profiles of the randomized DGGE gels were aligned and sorted by treatment and days postapplication. Additionally, profiles of unmanured soil (a), DIF manure (b), and control manure (c) are shown. Densitometric scores for each DGGE coordinate indicate its contribution to the explained variance. Positive values indicate an increase in band intensity in the DIF treatment compared with the control.

Download figure to PowerPoint

Occurrence of quinolone resistance genes qnrA, qnrB, and qnrS

The occurrence of quinolone resistance genes qnrA, qnrB, and qnrS1/qnrS2 was analyzed by PCR and Southern blot hybridization for the DNA extracts of DIF manure and control manure as well as of bulk soil on days 0, 7, 14, 28, 71, 105, and 140 and rhizosphere on days 28, 105, and 140 after manure treatment. The gene qnrB was detected in one replicate of DIF manure-treated bulk soil on days 14 and 105, in two replicates of DIF manure-treated bulk soil on day 7, and in all four replicates of the rhizosphere of maize grown in DIF manure-treated soil on day 28, but not in control manure-treated soil samples.

The genes qnrS1/qnrS2 were detected in only one replicate of control manure-treated bulk soil and rhizosphere on day 71 and in one replicate of DIF manure-treated bulk soil on day 105. The gene qnrA was not detected.

Effect of DIF manure on relative abundance of sul1, sul2, intI1, and traN

Quantitative PCR revealed that in untreated soil, the relative abundance (log target gene copies/rrn copies) of the resistance genes sul1 and sul2, of intI1, and of the traN gene fragment specific for LowGC-type plasmids was in the range of −5.8 log units for sul1 and below detection limit for sul2, intI1, and traN. The relative abundances in control manure from untreated pigs were −2.4 (sul1), −2.1 (sul2), −3.1 (intI1), and −5.1 (traN) log units. In manure from DIF-treated pigs, the relative abundances were except for intI1 higher with −2.3 for sul1, −1.8 for sul2, −3.2 for intI1, and −4.4 for traN. After application of manure from DIF-treated and untreated pigs, the relative abundance (log target gene copies/rrn copies) in the soil increased by about two orders of magnitude for sul1 reaching values of −4 log units and was now detectable with about −3 to −2 log units for sul2, −4 log units for intI1, and about −5 to −3 for traN (Fig. 5). Over the experimental period of 140 days, the relative abundance decreased again reaching values in the range of the untreated soil. The statistical analysis by a linear mixed model revealed that the relative abundances of sul1 and sul2 were significantly increased in the DIF manure-treated soil compared with the control (Table 1). Except for the rhizosphere on day 71, the relative abundance of intI1 genes of class I integrons was increased in DIF manure-treated bulk soil and rhizosphere compared with the control, but decreased over time after manure application. The relative abundance of traN genes of LowGC-type plasmids was significantly increased in the bulk soil until day 71, but not in the rhizosphere.

Table 1. The influence of factors treatment (control and DIF manure), day, and their interactions on the relative abundance of resistance genes sul1 and sul2 as well as of intI1 and traN specific for class 1 integrons and LowGC-type plasmids, respectively, were analyzed for bulk soil and rhizosphere by a mixed linear model
CompartmentClassLog (sul1/rrn)Log (sul2/rrn)Log (traN/rrn)Log (intI1/rrn)
F P F P F P F P
  1. Significance and F-values are shown. Samples from days 7, 14, 28, 71, 105, and 140 were used for the bulk soil analysis. Rhizosphere samples were analyzed starting on day 28.

Bulk soilTreatment42.760.000856.360.000335.06< 0.000137.690.0008
Day21.94< 0.000171.12< 0.000110.87< 0.000119.48<0.0001
Day × treatment0.590.70612.350.07623.970.00811.450.2476
RhizosphereTreatment31.860.000554.61< 0.00012.50.147315.020.008
Day15.4< 0.000132.66< 0.00011.380.2833105.47<0.0001
Day × treatment0.450.72180.750.54061.610.224510.810.0011
image

Figure 5. Time course of the relative abundance of sulfadiazine resistance genes sul1 and sul2 as well as of intI1 and traN specific for class 1 integrons and LowGC-type plasmids, respectively. Values for manure-treated soil (light gray bars) and DIF manure-treated soil (dark gray bars) are shown for bulk soil (B) and rhizosphere (R) of maize. Error bars indicate standard deviations (n = 4).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

The bacterial community composition of bulk soil treated with manure from DIF-medicated pigs differed significantly from the soil that was treated with control manure, suggesting a selective effect of DIF on the soil bacterial community. The concentrations of (bio) accessible DIF in soil were very low, but DIF was constantly detectable (Rosendahl et al., 2012). Previously, it was shown that very low concentrations of antibiotics may select for resistant bacteria (Gullberg et al., 2011). Correspondingly, although none of the tested qnr genes could be detected in the DIF and control manure, the quinolone resistance gene qnrB was detected in some of the replicates of DIF manure-treated bulk soil beginning on day 7 and on day 28 in the DIF manure-treated rhizosphere. This might indicate a low natural abundance of populations carrying quinolone resistance genes in Merzenhausen soil, which had no history of previous manure amendment (Kotzerke et al., 2008), and may suggest a low selection for the tested qnr genes following the application of DIF with manure. These results differ from the findings of Dalkmann et al. (2012) who found no relationship between concentrations of the fluoroquinolone antibiotic ciprofloxacin in soil and the abundance of qnr resistance genes qnrA, qnrB, and qnrS in soils after long-term irrigation with untreated wastewater, but they are in accordance with results of Li et al. (2012) reporting a correlation of plasmid-mediated quinolone resistance and (fluoro) quinolone residues in soil adjacent to swine feedlots. In contrast to the study by Li et al. (2012), qnrS1/qnrS2 genes were only found in two samples in DIF manure- and control manure-treated soil, while qnrA was not detected, indicating a low natural abundance. These results correspond to the findings of Dalkmann et al. (2012) who detected qnrB and qnrS, but not qnrA genes in soil irrigated with wastewater containing ciprofloxacin and sulfamethoxazole among other compounds. However, other quinolone resistance determinants such as efflux pumps-encoding genes (qepA and oqxAB) or the variant aminoglycoside acetyltransferase gene [aac(6′)-Ib-cr] were not tested in this study.

Besides a potential selective pressure exerted by DIF on the soil microbial community, bacteria that were applied to soil by manure might have contributed to the differences in bacterial community structure observed between DIF manure- and control manure-treated soils. Antibiotic resistance is a common phenomenon in swine effluents (Chee-Sanford et al., 2009), and resistant commensal bacteria with other resistance mechanisms than the qnr genes tested in this study might have been selected in the pig gut by the administration of DIF. Accordingly, DGGE results showed clear differences between community compositions of DIF and control manure. Further, the incorporation of manure and manure-derived bacterial communities had a strong influence on the composition of the soil bacterial community, as demonstrated by the PCA biplots of DGGE band patterns from soil samples, which cluster clearly with DIF manure and control manure samples on day 0 (Fig. 1). Additionally, the application of manure to the soil led to an increase in intI1 genes in bulk soil and rhizosphere, which was below the detection limit of the real-time PCR method in untreated soil, although intI1 genes were reported to be common in environmental samples (Gillings et al., 2008b). The significantly higher abundance of intI1 genes in the DIF manure-treated bulk soil and rhizosphere compared with the control soils indicates an increase in populations carrying intI1 genes and might suggest an increased potential of the bacterial community to acquire and exchange antibiotic resistance genes via gene cassettes in the DIF manure-treated soil. This might also include other quinolone resistance genes that were not tested in this study, for example the aac(6′)-Ib-cr gene, which was located in the variable region of a class 1 integron in one Escherichia coli and two Klebsiella pneumonia strains as observed by Ruiz et al. (2012a). Additionally, following the application of manure, the relative abundance of sul1, sul2, and traN increased by several orders of magnitude, confirming the results of previous studies showing the same trend after the application of manure to the same soil (Heuer & Smalla, 2007; Heuer et al., 2009; Jechalke et al., 2013; Kopmann et al., 2013). Further, although the pigs were not treated with sulfonamides, a significant increase in sul1 and sul2 abundance was observed in the DIF manure-treated bulk soil and rhizosphere compared with control manure-treated soil. This increase in about one log unit was in the same range for bulk soil and rhizosphere. Additionally, traN genes specific for LowGC-type plasmids, which in recent years were assumed to play an important role in conferring sulfadiazine resistance in manure-treated soils (Heuer et al., 2009; Jechalke et al., 2013; Kopmann et al., 2013), were increased in DIF manure-treated bulk soil.

Previous studies suggested that manure bacteria might not be well adapted to the soil environment and decrease in abundance after application (Hammesfahr et al., 2008; Heuer et al., 2008), which might depend on different environmental conditions such as temperature, moisture, pH, and the indigenous community present (Chee-Sanford et al., 2009). Correspondingly, significant differences in community composition were observed between DGGE profiles of different time points. Over time, the differences of the fingerprints between the DIF manure-treated soils and control manure-treated soils decreased, and the manured soil samples approached the untreated soil in the PCA plot, indicating resilience of the soil bacterial community composition from the disturbance by manure application, which was potentially promoted by a depletion of nutrients and a decrease in abundance of bacteria that entered the soil with manure. However, differences in sul1, sul2, and intI1 gene abundances were still detectable on day 140, which might indicate horizontal gene transfer of these genes from the manure bacteria to the indigenous soil community.

In the rhizosphere, a trend similar to bulk soil was observed over time. However, the effect of DIF on the bacterial communities in the rhizosphere seemed to be smaller than in bulk soil. A reduced effect of DIF in rhizosphere soil might be explained by an accelerated dissipation of the substance that was indeed observed for this field study in the surrounding of plant roots (Rosendahl et al., 2012). This accelerated dissipation may have caused a reduced selective pressure in the rhizosphere, which in turn might have reduced the effect of DIF in the rhizosphere compared with bulk soil. The PRC analyses corroborate this finding, as the amount of explained variance attributed to the DIF treatment was calculated to be 18.1% for bulk soil, but only 8% for the rhizosphere soil. The rhizosphere can have a distinct effect on the microbial community, which can be explained by nutrients provided by root exudation (reviewed by Berg & Smalla, 2009; Doornbos et al., 2012). Correspondingly, the distance in the biplot displayed in Fig. 1 for bulk soil on day 0 and rhizosphere samples on day 28 indicated that the composition of the bacterial community adjacent to plant roots is very different from the bulk soil community. This rhizosphere effect can be influenced by changes in the quality and quantity of root exudation over the growth period, which can be caused by the age of the plant and external factors such as biotic and abiotic stressors (Badri & Vivanco, 2009) and hence might explain the heterogeneity between replicates.

In conclusion, although DIF was reported to be hardly accessible to the soil bacterial community due to the rapid formation of nonextractable residues (Rosendahl et al., 2012), manure from DIF-treated pigs significantly altered the soil bacterial community and the abundance of resistance genes and mobile genetic elements over a period of 140 days compared with manure from untreated pigs. These findings have to be considered in further assessments of risks associated with the spread of resistance genes in the environment.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

This work was funded by the Deutsche Forschungsgemeinschaft (DFG) in the framework of the Research Unit FOR 566 ‘Veterinary medicines in soil: Basic research for risk analysis’ (SM59/5-3).

Statement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information

Field application of manure from difloxacin-treated pigs altered the soil bacterial community and increased the abundance of resistance genes and mobile genetic elements compared with manure from untreated pigs.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information
  • Aarestrup FM (2005) Veterinary drug usage and antimicrobial resistance in bacteria of animal origin. Basic Clin Pharmacol Toxicol 96: 271281.
  • Adriaenssens N, Coenen S, Versporten A et al. (2011) European Surveillance of Antimicrobial Consumption (ESAC): outpatient antibiotic use in Europe (1997-2009). J Antimicrob Chemother 66: 312.
  • Badri DV & Vivanco JM (2009) Regulation and function of root exudates. Plant, Cell Environ 32: 666681.
  • Barraud O, Baclet MC, Denis F & Ploy MC (2010) Quantitative multiplex real-time PCR for detecting class 1, 2 and 3 integrons. J Antimicrob Chemother 65: 16421645.
  • Berg G & Smalla K (2009) Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere. FEMS Microbiol Ecol 68: 113.
  • Cabello FC (2006) Heavy use of prophylactic antibiotics in aquaculture: a growing problem for human and animal health and for the environment. Environ Microbiol 8: 11371144.
  • Charpentier X, Polard P & Claverys J-P (2012) Induction of competence for genetic transformation by antibiotics: convergent evolution of stress responses in distant bacterial species lacking SOS? Curr Opin Microbiol 15: 570576.
  • Chee-Sanford JC, Mackie RI, Koike S, Krapac IG, Lin YF, Yannarell AC, Maxwell S & Aminov RI (2009) Fate and transport of antibiotic residues and antibiotic resistance genes following land application of manure waste. J Environ Qual 38: 10861108.
  • Collignon P, Powers JH, Chiller TM, Aidara-Kane A & Aarestrup FM (2009) World Health Organization ranking of antimicrobials according to their importance in human medicine: a critical step for developing risk management strategies for the use of antimicrobials in food production animals. Clin Infect Dis 49: 132141.
  • Dalhoff A (2012) Resistance surveillance studies: a multifaceted problem—the fluoroquinolone example. Infection 40: 239262.
  • Dalkmann P, Broszat M, Siebe C, Willaschek E, Sakinc T, Huebner J, Amelung W, Grohmann E & Siemens J (2012) Accumulation of Pharmaceuticals, Enterococcus, and resistance genes in soils irrigated with wastewater for zero to 100 years in central Mexico. PLoS ONE 7: e45397.
  • Doornbos RF, van Loon LC & Bakker P (2012) Impact of root exudates and plant defense signaling on bacterial communities in the rhizosphere. A review. Agron Sustain Dev 32: 227243.
  • Drlica K & Zhao XL (1997) DNA gyrase, topoisomerase IV, and the 4-quinolones. Microbiol Mol Biol Rev 61: 377392.
  • Garnier F, Raked N, Gassama A, Denis F & Ploy MC (2006) Genetic environment of quinolone resistance gene qnrB2 in a complex sul1-type integron in the newly described Salmonella enterica serovar Keurmassar. Antimicrob Agents Chemother 50: 32003202.
  • Gillings M, Boucher Y, Labbate M, Holmes A, Krishnan S, Holley M & Stokes HW (2008a) The evolution of class 1 integrons and the rise of antibiotic resistance. J Bacteriol 190: 50955100.
  • Gillings MR, Krishnan S, Worden PJ & Hardwick SA (2008b) Recovery of diverse genes for class 1 integron-integrases from environmental DNA samples. FEMS Microbiol Lett 287: 5662.
  • Gullberg E, Cao S, Berg OG, Ilbäck C, Sandegren L, Hughes D & Andersson DI (2011) Selection of resistant bacteria at very low antibiotic concentrations. PLoS Pathog 7: e1002158.
  • Hammesfahr U, Heuer H, Manzke B, Smalla K & Thiele-Bruhn S (2008) Impact of the antibiotic sulfadiazine and pig manure on the microbial community structure in agricultural soils. Soil Biol Biochem 40: 15831591.
  • Heuer H & Smalla K (2007) Manure and sulfadiazine synergistically increased bacterial antibiotic resistance in soil over at least two months. Environ Microbiol 9: 657666.
  • Heuer H, Focks A, Lamshöft M, Smalla K, Matthies M & Spiteller M (2008) Fate of sulfadiazine administered to pigs and its quantitative effect on the dynamics of bacterial resistance genes in manure and manured soil. Soil Biol Biochem 40: 18921900.
  • Heuer H, Kopmann C, Binh CTT, Top EM & Smalla K (2009) Spreading antibiotic resistance through spread manure: characteristics of a novel plasmid type with low %G+C content. Environ Microbiol 11: 937949.
  • Jechalke S, Kopmann C, Rosendahl I, et al. (2013) Increased abundance and transferability of resistance genes after field application of manure from sulfadiazine-treated pigs. Appl Environ Microbiol 79: 17041711.
  • Junge T, Claßen N, Schäffer A & Schmidt B (2012) Fate of the veterinary antibiotic 14C-difloxacin in soil including simultaneous amendment of pig manure with the focus on non-extractable residues. J Environ Sci Health, Part B 47: 858868.
  • Kopmann C, Jechalke S, Rosendahl I et al. (2013) Abundance and transferability of antibiotic resistance as related to the fate of sulfadiazine in maize rhizosphere and bulk soil. FEMS Microbiol Ecol 83: 125134.
  • Kotzerke A, Sharma S, Schauss K, Heuer H, Thiele-Bruhn S, Smalla K, Wilke BM & Schloter M (2008) Alterations in soil microbial activity and N-transformation processes due to sulfadiazine loads in pig-manure. Environ Pollut 153: 315322.
  • Kotzerke A, Hammesfahr U, Kleineidam K, Lamshöft M, Thiele-Bruhn S, Schloter M & Wilke B-M (2011) Influence of difloxacin-contaminated manure on microbial community structure and function in soils. Biol Fertil Soils 47: 177186.
  • Kropf S, Heuer H, Grüning M & Smalla K (2004) Significance test for comparing complex microbial community fingerprints using pairwise similarity measures. J Microbiol Methods 57: 187195.
  • Lamshöft M, Sukul P, Zühlke S & Spiteller M (2010) Behaviour of C-14-sulfadiazine and C-14-difloxacin during manure storage. Sci Total Environ 408: 15631568.
  • Li J, Wang T, Shao B, Shen J, Wang S & Wu Y (2012) Plasmid-mediated quinolone resistance genes and antibiotic residues in wastewater and soil adjacent to swine feedlots: potential transfer to agricultural lands. Environ Health Perspect 120: 11441149.
  • Mazel D (2006) Integrons: agents of bacterial evolution. Nat Rev Microbiol 4: 608620.
  • Nordmann P & Poirel L (2005) Emergence of plasmid-mediated resistance to quinolones in Enterobacteriaceae. J Antimicrob Chemother 56: 463469.
  • Robicsek A, Strahilevitz J, Sahm DF, Jacoby GA & Hooper DC (2006) qnr Prevalence in ceftazidime-resistant Enterobacteriaceae isolates from the United States. Antimicrob Agents Chemother 50: 28722874.
  • Rosendahl I, Siemens J, Kindler R, et al. (2012) Persistence of the fluoroquinolone antibiotic difloxacin in soil and lacking effects on nitrogen turnover. J Environ Qual 41: 12751283.
  • Ruiz E, Sáenz Y, Zarazaga M, Rocha-Gracia R, Martínez-Martínez L, Arlet G & Torres C (2012a) Qnr, aac(6)-Ib-cr and qepA genes in Escherichia coli and Klebsiella spp.: genetic environments and plasmid and chromosomal location. J Antimicrob Chemother 67: 886897.
  • Ruiz J, Pons MJ & Gomes C (2012b) Transferable mechanisms of quinolone resistance. Int J Antimicrob Agents 40: 196203.
  • Sambrook J, Fritsch EF & Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  • Stalder T, Barraud O, Casellas M, Dagot C & Ploy M-C (2012) Integron involvement in environmental spread of antibiotic resistance. Front Microbiol 3: 119.
  • Strahilevitz J, Jacoby GA, Hooper DC & Robicsek A (2009) Plasmid-mediated quinolone resistance: a multifaceted threat. Clin Microbiol Rev 22: 664689.
  • Sukul P, Lamshoft M, Kusari S, Zuhlke S & Spiteller M (2009) Metabolism and excretion kinetics of C-14-labeled and non-labeled difloxacin in pigs after oral administration, and antimicrobial activity of manure containing difloxacin and its metabolites. Environ Res 109: 225231.
  • Van den Brink PJ & Ter Braak CJF (1999) Principal response curves: analysis of time-dependent multivariate responses of biological community to stress. Environ Toxicol Chem 18: 138148.
  • Weinert N, Meincke R, Gottwald C, Heuer H, Gomes NCM, Schloter M, Berg G & Smalla K (2009) Rhizosphere communities of genetically modified zeaxanthin-accumulating potato plants and their parent cultivar differ less than those of different potato cultivars. Appl Environ Microbiol 75: 38593865.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Statement
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
fem12191-sup-0001-FigS1-S4.pdfapplication/PDF2688K

Fig. S1. Not randomized denaturing gradient gel electrophoresis (DGGE) gel of 16S rRNA gene fragments amplified from total genomic DNA extracted from DIF manure, control manure and bulk soil treated with manure or not.

Fig. S2. Not randomized denaturing gradient gel electrophoresis (DGGE) gel of 16S rRNA gene fragments amplified from total genomic DNA extracted from DIF manure, control manure, bulk soil from day 0 treated with manure or not as well as rhizosphere treatments.

Fig. S3. Randomized denaturing gradient gel electrophoresis (DGGE) gel of 16S rRNA gene fragments amplified from total genomic DNA extracted from DIF manure, control manure and bulk soil treated with manure or not, which was used for the analysis described in the main manuscript.

Fig. S4. Randomized denaturing gradient gel electrophoresis (DGGE) gel of 16S rRNA gene fragments amplified from total genomic DNA extracted from DIF manure, control manure and bulk soil from day 0 treated with manure or not as well as rhizosphere treatments, which was used for the analysis described in the main manuscript.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.