Detection of conjugative plasmids and antibiotic resistance genes in anthropogenic soils from Germany and India

Authors


  • Present address: Abdul Malik, Department of Agricultural Microbiology, Faculty of Agricultural Sciences, Aligarh Muslim University, Aligarh-202 002, India.

  • Editor: Elizabeth Baggs

Correspondence: Elisabeth Grohmann, Technische Universität Berlin, Umweltmikrobiologie/Genetik, Franklinstraße 28/29, 10587 Berlin, Germany. Tel.: +49 30 31473187; fax.: +49 30 31473673; e-mail: elisabeth.grohmann@tu-berlin.de

Abstract

PCR typing methods were used to assess the presence of plasmids of the incompatibility (Inc) groups IncP, IncN, IncW, IncQ and rolling circle plasmids of the pMV158 type in total DNA extracts from anthropogenic soils from India and Germany. Ten different soils from two different locations in Germany, the urban park Berlin Tiergarten and the abandoned sewage field Berlin-Buch, and from four different locations in India were analysed. PCR amplification of the total DNA extracts revealed the prevalence of IncP-specific sequences in Berlin Buch and Indian soil samples. The detected IncP plasmids contained at least one transfer function, the origin of transfer, oriT. In contrast to IncP-specific sequences, IncQ, IncN, IncW and pMV158-specific sequences were never detected. The presence of ampC, tet (O), ermB, SHV-5, mecA, and vanA antibiotic resistance genes was also tested. Three Indian soil samples irrigated with wastewater contained the ampC gene, whereas the other resistance genes were not found in any of the samples. Detection of IncP trfA2 and oriT sequences by PCR amplification and hybridization is a clear indication that IncP plasmids are prevalent in these habitats. Exogenous plasmid isolation revealed conjugative plasmids belonging to the IncPβ group encoding resistance to ampicillin.

Introduction

The importance of horizontal gene transfer in the evolution and diversification of bacteria is now well established. Plasmids play a primordial role in these exchanges. The rapid dissemination of antibiotic resistance genes in bacterial populations as a consequence of the intensive use of antibiotics in medicine and agriculture can be partly attributed to plasmid-mediated horizontal transfer (Zechner et al., 2000; Grohmann et al., 2003; Franiczek et al., 2006). Most of the compounds used in medicine are only partially metabolized by patients and are then discharged into the hospital sewage system or directly into the municipal wastewater treatment plant. They may pass through the sewage system and end up in the environment, mainly in soil. Obviously, most of the antibiotics are not fully eliminated during the wastewater purification process. In soil, tetracycline concentrations in the range of several hundred μg kg−1 have been detected some months after manure application (Hamscher et al., 2002). Ampicillin concentrations of 80 ng L−1 have been reported in hospital effluents (Kümmerer, 2004). For this study, six different antibiotic resistance genes were chosen. They were selected due to the abundance of resistant microorganisms in the environment (e.g. Esiobu et al., 2002; Kümmerer, 2004; D'Costa et al., 2006; Guardabassi & Agerso, 2006; Schmitt et al., 2006; Patterson et al., 2007): ampC (ampicillin resistance), mecA (methicillin resistance), SHV (extended β-lactam resistance), ermB (erythromycin resistance), tet(O) (tetracycline resistance), and vanA (vancomycin resistance).

Only a limited fraction of the indigenous soil bacterial population can be cultured in the laboratory. The percentage of culturable bacteria depends on the selected media and growth conditions as well as on the studied species. The large nonculturable fractions of soil bacteria (Ward et al., 1990; Riesenfeld et al., 2004) could represent a potential reservoir of antibiotic resistance genes. Resistance genes could transfer to pathogenic bacteria via crops or contact with the environment causing untreatable infections in animals and humans (Nwosu, 2001).

Plasmids provide their host with a large array of phenotypes: antibiotic (Datta & Hughes, 1983) or heavy metal resistance (Silver, 1996; Malik et al., 2002; Aleem et al., 2003), the ability to degrade recalcitrant compounds and to use them as carbon and energy source (Top et al., 1998; Nawab et al., 2003). Broad host range (BHR) plasmids can transfer or mobilize their genes into taxonomically distant species (Thomas, 2000). Most plasmids belonging to the Inc groups IncP, IncN, IncW are conjugative and many IncQ and rolling circle plasmids are mobilizable (Götz et al., 1996; Haines et al., 2006; Smalla et al., 2006). Plasmids from these Inc groups have been frequently detected in environmental systems (Götz et al., 1996; Smalla et al., 2006). The present study focused on the detection of plasmid specific sequences from prevalent plasmid Inc groups and on the detection of the antibiotic resistance gene pool in wastewater irrigated soils from India and urban soils from Berlin, Germany by PCR and DNA hybridization.

Materials and methods

Soil samples were collected from two different locations in Berlin (Germany) and four different locations in different cities (Aligarh, Kanpur and Ghaziabad) in India.

German soil samples

The first experimental site is located in the urban park Tiergarten in the centre of Berlin, it is located in a sunbathing area, which is marked by anthropogenic influences, like fertilization, sprinkling and the soil is compacted due to mowing. The site consists of sandy soil and is characterized by small scale heterogeneity due to different anthropogenic substances as pollutants (Braun et al., 2006). Three different samples were collected in 3–15 cm depth and designated as NW (nonwettable), W (wettable) and Rz (rhizospheric).

The second sampling site is located at the northern city limit of Berlin, in Berlin-Buch. Since 1890, untreated wastewater was applied on 13 000 ha of sewage farm land around Berlin. Up to 10 000 mm year−1 were applied. In 1985, the wastewater application was stopped, the soil surface was levelled and an effort was made to afforest the fields (Täumer et al., 2005). The soil, a hortic anthrosol, consists of 40–60 cm organic topsoil upon medium sized sand. The top soil shows a high and very heterogenous organic matter content. Three samples were collected from 15 cm depth at this location and designated according to decreasing hydrophobicity as determined by the Water Drop Penetration Time method (Dekker & Jungerius, 1990) as B1 with a water drop penetration time (WDPT) of 40 min, B2 with 2 min and 40 s and B3 with a WDPT <1 s. From each site 1000 g of soil sample was collected, 20 g of each sample was homogenized and three subsamples of 5 g from each site were taken for further analysis. The pH was determined in H2O (DIN ISO 38404). Organic carbon content was measured by drying the sample at 550 °C for 5 h following DIN EN 12879. Water content was defined according to DIN EN 12880.

The Tiergarten test site had water contents of 28±10.2 % in the wettable soil samples, 0±2.1% in the nonwettable samples and 9.6±2.0% in the rhizospheric samples. pH (H2O) for the wettable samples was 4.8±1.3, for the nonwettable samples 4.3±0.4 and for the rhizospheric samples 6.0±0.3. Organic carbon content in the wettable soil samples was 5±2%, in the nonwettable samples 5±1%, and in the rhizospheric samples 4±1%.

In the Buch test site water content in the wettable regions (B2 and B3) ranged from 7% up to 24% and in the dry, nonwettable regions (B1) from 4% to 9%. pH (H2O) in the Buch samples ranged between 4.8 and 6 (Hoffmann, 2002; Täumer et al., 2005). The organic matter content of the topsoil horizon (0–60 cm) ranged between 0.04 and 0.06 g g−1 (Täumer et al., 2005).

Indian soil samples

Composite soil samples were collected from 15 cm depth at four different sites in India from (1) Mathura road, Aligarh, which has been irrigated with wastewater (mainly from lock manufacturing and steel industries) for more than two decades until now (Aleem & Malik, 2003), (2) agricultural field soil from the Jajmau area of Kanpur which receives the partially treated effluents from the tannery industries, (3) agricultural field soil from Ghaziabad which had been irrigated with industrial wastewater of steel and electroplating industries until now and (4) an agricultural field soil which is irrigated constantly with ground water (Aligarh, Bypass).

For each site, 1 kg of soil from four different locations within the site, were collected and homogenized to form a 4 kg composite soil sample. Subsamples of 5 g, three from each site were taken for further physico-chemical analysis. Each measurement was carried out in duplicate. The pH was determined in H2O (DIN ISO 38404). Organic carbon content was measured by drying the sample at 550 °C for 5 h following DIN EN 12879. Water content was defined according to DIN EN 12880. The sandy loamy soil from Mathura road, Aligarh, had a water content of 22.2±1.9%, the Kanpur soil (mainly clay and loam) had a water content of 34.5±3.0%, the loamy Ghaziabad soil had a water content of 20.3±2.2% and the sandy loamy Aligarh Bypass soil had a water content of 22.0±1.7%. pH (H2O) of Mathura road soil was 8.1±0.2, from Kanpur soil 7.7±0.2, from Ghaziabad soil 8.3±0.3, and Aligarh Bypass soil 8.3±0.2. Organic carbon content of Mathura road soil was 0.7±0.2%, of Kanpur soil 1.3±0.1%, of Ghaziabad soil 0.4±0.05% and of Aligarh Bypass soil 0.6±0.3%.

Bacterial strains and plasmids

Bacterial strains used in this study are listed in Table 1. A plasmid containing tet (O) was obtained from Marilyn Smith (University of Kansas, Medical Center, KS). The plasmid pIP501 containing ermB was obtained from the German culture collection (DSMZ No. DSM8629). The Escherichia coli strains harboring the target plasmids were cultured on LB agar plates with the respective antibiotics. Escherichia coli HMS (recA, F, WsdR, (rk mk+), rifampicin (Rif) resistance, courtesy of M. Espinosa, CSIC, Spain) was cultured in LB medium supplemented with 100 μg mL−1 rifampicin and used as recipient in the biparental matings. Tryptic soy agar (TSA) plates for selecting transconjugants contained the following antibiotics: rifampicin (100 μg mL−1) and either ampicillin (Amp) (10 μg mL−1), kanamycin (Kan) (30 μg mL−1), tetracycline (20 μg mL−1) or chloramphenicol (50 μg mL−1). In addition cycloheximide (300 μg mL−1) was added to the selective media to prevent fungal growth. Enterococcus faecalis OG1X was cultured in brain heart infusion medium (Condalab, Madrid, Spain). Before PCR, colonies of the Escherichia coli host strains were picked with a sterile toothpick, resuspended in 50 μL sterile bidistilled water and heated at 95 °C for 15 min. Enterococcus faecalis was submitted to an alkaline lysis before PCR (Sambrook et al., 1989). PCR amplification was carried out with the primers listed in Table 2 according to the amplification profiles described in Götz et al. (1996).

Table 1.   Bacterial strains used in the study
NameInvestigated plasmids/
resistance gene
Plasmid incompatibility
group/replication type
Source
Enterococcusfaecalis OG1XpMV158rolling circleM. Espinosa, Centro de Investigaciones
  Biológicas, Madrid, Spain
Escherichiacoli SCS1RP4IncPE. Lanka, Max-Planck-Institut für
  Molekulare Genetik, Berlin, Germany
Escherichiacoli K12pULB2432IncNM. Couturier, Université Libre de Bruxelles, Faculté
  de Sciences, Genetique des Procaryotes,
  Brussels, Belgium
Escherichiacoli DH5αR388IncWF. de la Cruz, Universidad de Cantábria,
  Departamento de Biologia Molecular,
  Santander, Spain
Escherichiacoli JE723pJE723IncQE. Lanka, Max-Planck-Institut für
  Molekulare Genetik, Berlin, Germany
Enterobacter cloacaeampC DSM46348
Enterococcus faeciumvanA DSM17050
Klebsiella pneumoniaeSHV-5 DSM116609
Staphylococcus aureus ssp. aureusmecA DSM13661
Escherichia coli JM109pB10IncP-1βH. Heuer, Biologische Bundesanstalt (BBA), Braunschweig, Germany
Table 2.   PCR primers used, target and PCR product size
PrimerSequenceAnnealing
temperature (°C)
Target site/geneProduct
size (bp)
References
  • *

    Unpublished data.

trfA2Fw: 5′-cgaaattcrtrtgggagaagta-3′57Replication protein241Götz et al. (1996)
Rv: 5′-cgyttgcaatgcaccaggtc-3′    
IncPoriTFw: 5′-cagcctcgcagagcaggat-3′57Origin of transfer110Götz et al. (1996)
Rv: 5′-cagccgggcaggataggtgaagt-3′    
IncWoriTFw: 5′-tctgcatcattgtagcacc-3′51Origin of transfer317Götz et al. (1996)
Rv: 5′-ccgtagtgttactgtagtgg-3′    
IncNrepFw: 5′-agttcaccacctactcgctccg-3′55Replication protein164Götz et al. (1996)
Rv: 5′-caagttcttctgttgggattccg-3′    
IncQoriVFw: 5′-ctcccgtactaactgtcacg-3′57Origin of replication436Götz et al. (1996)
Rv: 5′-atcgaccgagacaggccctgc-3′    
IncQoriTFw: 5′-ttcgcgctcgttgttcttcgagc-3′57Origin of transfer191Götz et al. (1996)
Rv: 5′-gccgttaggccagtttctcg-3′    
pMVoriTFw: 5′-ctacctgtcccttgctgat-3′55Origin of transfer142Alexandrino (2006)
Rv: 5′-gcagtgccgaccaaaacc-3′   Grohmann et al. (1999)
ampCFw: 5′-gtgaccagatactggccaca-3′60Ampicillin resistance822H. Dorries & E. Grohmann*
Rv: 5′-ttactgtagcgcctcgagga-3′    
tet (O)Fw: 5′-ggattgcatacaggcacaga-3′60Tetracycline resistance738H. Dorries & E. Grohmann*
Rv: 5′-gtttggatcatagggagaggat-3′    
ermBFw: 5′-gcatttaacgacgaaactggct-3′60Erythromycin resistance573H. Dorries & E. Grohmann*
Rv: 3′-gacaatacttgctcataagtaatggt-3′    
mecAFw: 5′-taatagttgtagttgtcgggtttg-3′60Methicillin resistance733H. Dorries & E. Grohmann*
Rv: 5′-taacctaatagatgtgaagtcgct-3′    
SHV-5Fw: 5′-tgttagccaccctgccgct-3′60Extended spectrum beta lactamases825H. Dorries & E. Grohmann*
Rv: 5′-gttgccagtgctcgatcag-3′    
vanAFw: 5′ gaaatcaaccatgttgatgtagca-3′60Vancomycin resistance572H. Dorries & E. Grohmann*
Rv: 5′-ttgccggtttcctgtatccgt-3′    

DNA extractions and amplification of 16S rRNA genes

Total community DNA was directly extracted from soil with the FastDNA® Spin Kit for Soil applying mechanical disruption of the cells as described by the manufacturer (Qbiogene, Carlsbad, CA). Cell lysis by this kit proved to be superior to other lysis procedures tested in the authors' laboratory on environmental samples, e.g. the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) (Alexandrino et al., 2004).

Plasmid DNA from the transconjugants was extracted with the Qiagen Plasmid Midi kit (Qiagen, Hilden, Germany) and analysed on 0.7% agarose gels with a λ/HindIII/EcoRI marker as molecular weight standard.

PCR was performed with a Primus 96 plus Thermocycler (MWG Biotech, Ebersberg, Germany). Amplification of DNA from soil samples was performed in a volume of 50 μL containing 5 μL 10 × Gentherm reaction buffer supplied by the manufacturer (Rapidozym, Berlin, Germany), 160 nM of each primer, 1 μL of each dNTP (10 mM) and 100–200 ng of purified DNA extracted from soil and 1.5 U Taq DNA polymerase (Rapidozym, Berlin, Germany). Dimethyl sulfoxide was added to the reaction mixture to a final concentration of 2%.

To check PCR amplifiability of the extracted DNA, PCR of 16S rRNA genes was performed. Primer pair 616 f and 1425 r (Lane, 1991) was chosen for amplification of the 16S rRNA genes. PCR conditions were as follows: initial denaturation at 95 °C for 2 min, followed by 10 cycles each consisting of 95 °C for 30 s, 55 °C for 30 s, 72 °C for 3 min, followed by 35 cycles 95 °C for 30 s, 55 °C for 30 s, 72 °C for 3 min plus 10 s per cycle, terminated by a final extension at 72 °C for 7 min.

PCR of plasmid specific sequences and antibiotic resistance genes

Oligonucleotide primers to amplify plasmid backbone regions related to replication (IncP trfA2, IncN rep, IncW oriV) and transfer (IncP oriT, IncW oriT) are published in Götz et al. (1996), pMV158 oriT in Alexandrino (2006). The oligonucleotide primers for the antibiotic resistance genes ampC, tet (O), ermB, SHV-5, mecA, and vanA were designed by Dorries & E. Grohmann (unpublished data, Table 2). The following sequences (GenBank accession No.) were used to design the oligonucleotides: AJ005633 (ampC), U00453 (ermB), X52593 (mecA), X55640 (SHV-5), M18896 [tet(O)], AF516335 (vanA). Primer specificity was determined with sequence alignments using blast and NCBI entries. They were tested for cross-reactions with the respective sensitive strains, and seven to ten strains for each resistance gene were applied (U. Böckelmann & E. Grohmann, unpublished data). All PCR primers were synthesized by MWG Biotech (Ebersberg, Germany). For plasmid specific sequences, PCRs were performed in a 50 μL reaction mixture containing 5 μL 10 × Gentherm reaction buffer (Rapidozym, Berlin, Germany), 1 μL of each dNTP (10 mM), 3.75 mM MgCl2 (except for IncQ oriT, 4.75 mM), 2 μL of each primer (20 pmol μL−1) and 1.5 U Gentherm DNA polymerase (Rapidozym, Berlin, Germany). The amplification program consisted of the following steps: 94 °C for 5 min, followed by 35 cycles of amplification divided as 1 min of denaturation at 94 °C, 1 min of primer annealing at the temperature according to Table 2 and 1 min of primer extension at 72 °C, followed by a 10 min final extension step at 72 °C. PCRs were carried out in a Primus 96 plus PCR-Cycler (MWG Biotech, Ebersberg, Germany). For antibiotic resistance genes, amplification was also carried out in a 50 μL reaction mixture containing 2 μL of each primer (20 pmol μL−1), 1 μL of each dNTP (10 mM), 1.5 U Gentherm DNA polymerase (Rapidozym, Berlin, Germany), 2 mM MgCl2, and 5 μL 10 × reaction buffer (Rapidozym, Berlin, Germany). The amplification program consisted of the following steps (1) initial denaturation at 95° C for 2 min, (2) 35 cycles of 95 °C for 30 s, 60 °C for 45 s and 72 °C for 1 min and (3) a final elongation at 72  °C for 7 min. PCRs were performed with 1 μL of 1 : 10 diluted and 1 μL undiluted DNA obtained from soil extractions with the FastDNA® Spin Kit for Soil.

PCR products were purified for digoxigenin-labelling with the Wizard SV Gel and PCR Clean-UP System (Promega, Mannheim, Germany). Digoxigenin labelling of the purified PCR product was carried out using the PCR digoxigenin probe synthesis kit (Roche, Mannheim, Germany). For product size confirmation and yield estimation 10 μL of the PCR products were loaded onto 1.5% agarose gels submitted to electrophoresis for 3 h at 60 V and stained with ethidium bromide.

For the German test sites three samples each were investigated by the plasmid specific and antibiotic resistance gene specific PCR. Every experiment was repeated at least three times. For the Indian test sites two composite samples each were investigated by the plasmid specific and the antibiotic resistance gene specific PCR, each with three repetitions. There were no differences detected according to the presence of plasmid specific genes and antibiotic resistance genes in different samples from the same site.

Southern blot and dot blot

Dot blot or Southern blots were performed for all plasmid and antibiotic resistance gene sequences amplified regardless of detection of the PCR products in agarose gels. Southern blotting and the subsequent chemiluminescence detection of the digoxigenin-labelled DNA hybrids were carried out according to the digoxigenin system users guide for filter hybridization (Roche, Mannheim, Germany). For Southern blot, 10 μL of the PCR product was run on a 1.5% agarose gel in 0.5 × TBE buffer. The DNA was denatured and transferred to a Hybond-N+ membrane as described in the users guide. The chemiluminescence detection of the digoxigenin-labelled DNA hybrids was performed with the digoxigenin luminescent detection kit (Roche, Mannheim, Germany) and CSPD [3-(4-methoxyspiro{1,2-dioxetane- 3,2′-(5′-chloro)tricyclo[3.3.3.1.3,7]decan}-4-yl)phenyl phosphate] as substrate. The membrane was exposed to a high performance chemiluminescence film (Amersham Biosciences, Freiburg, Germany) and the X-ray film was developed.

Dot blots of the amplified PCR products were also carried out according to the digoxigenin system users guide for filter hybridization (Roche, Mannheim, Germany). 20 μL of PCR product was boiled for 10 min (100 °C) and then 1 μL of each PCR product was dotted on a Hybond-N+ membrane (Roche, Mannheim, Germany). For both dot blot and Southern blot, DNA was fixed to the membrane by baking at 120 °C for 30 min. Digoxigenin-labelled DNA probes specific for trfA2 (241 bp) and IncP oriT (110 bp) were used in both blots. For both sequences RP4-derived (IncPα) and pB10-derived (IncPβ) probes were applied. The dot blot detection procedure was the same as described for Southern blotting.

Plasmid DNA extracted from transconjugants was analysed by dot blot for the presence of IncP plasmids using digoxigenin-labelled probes derived from pB10 and RP4 as described above. The detection limit of the Southern and dot blots was determined to be 1–10 pg DNA by applying different amounts of the positive control.

Exogenous plasmid isolation in biparental matings

Exogenous plasmid isolations were essentially done as described previously (e.g. Bale et al., 1988; Smalla et al., 2000). From the ten test sites bacteria were detached from the soil particles by dispersion in a solution containing 7.5 mM sodium pyrophosphate and 0.5% Tween 80 under vigorous shaking as described in Böckelmann et al. (2003). The supernatant was applied as plasmid donor, while rifampicin resistant Escherichia coli HMS served as the recipient in the matings. One milliliter of each of donor cells and recipient cells was mixed and filtered through sterile nitrocellulose filters (0.22-μm pore size). The filters were incubated overnight at 28 °C on TSA plates. Controls were made by incubating the bacterial fraction from the soil samples and the recipients separately. After overnight incubation, the cell lawn was resuspended in 1 mL of sterile saline, and transconjugants were obtained after plating serial 10-fold dilutions on selective media. After 48 h incubation at 28 °C CFU were counted and putative transconjugants picked for further characterization.

Results and discussion

Total community DNA was isolated from ten soil samples (six samples from German soil and four from Indian soil). Three each of the German and Indian soil samples were contaminated with wastewater.

In this study, the DNA derived from Indian and German soils was PCR amplified and plasmid-specific sequences (IncP, IncN, IncW, IncQ and pMV158-type) and antibiotic resistance gene sequences [ampC, tet (O), ermB, SHV-5, mecA, and vanA] were analysed by Southern and dot blot hybridization, respectively. PCR amplifications with two sets of primers for IncP specific genes (trfA2 and oriT), IncQ (oriT and oriV) and one set of primers for IncN (rep), IncW (oriT) and pMV158-type rolling circle plasmids (oriT) were performed with all the samples. 70% of the samples gave PCR products with trfA2 and oriT primers of the IncP group (Table 3). These PCR products also hybridized with the RP4-derived (IncPα) probes (Figs 1a–b and 2a). The samples were also hybridized with trfA2 and oriT probes derived from the IncPβ plasmid pB10. With the exception of the B2 sample (Buch wettable sample WDPT 2 min 40 s) which was positive for pB10-derived oriT but negative for pB10-derived trfA2 by Southern blot, all samples positive for the IncPα sequences were also positive with both IncPβ probes (Figs 1c and 2b.) It has to be noted that dot and Southern blots of PCR-negative samples did not hybridize with digoxigenin-labelled trfA2 and oriT probes derived from RP4 or pB10.

Table 3.   PCR detection of conjugative plasmids in DNA extracts from German and Indian soil samples with subsequent Southern hybridization for IncPα, IncQ, IncN, IncW plasmids and pMV158-type rolling circle plasmids
Soil originIncPIncQIncNIncWpMV158-type
trfA2oriToriToriVreporiToriT
PCRBlotPCRBlotPCRBlotPCRBlotPCRBlotPCRBlotPCRBlot
  1. B1, Berlin Buch WDPT 40 min; B2, Berlin Buch WDPT 2 min 40 s; B3, Berlin Buch WDPT <1 s; NW, nonwettable Berlin Tiergarten; W, wettable Berlin Tiergarten; Rz, rhizospheric Berlin Tiergarten

  2. Hybridization with digoxigenin-labelled probe. trfA2 and oriT were detected by Southern and dot blot, and others (IncW, N, Q, pMV158) were analysed by dot blot hybridization. +, weak hybridization signal; +, strong hybridization signal.

German soil samples
 B1 Berlin Buch++++
 B2 Berlin Buch++++
 B3 Berlin Buch++++
 NW Berlin Tiergarten
 W Berlin Tiergarten
 Rz Berlin Tiergarden
Indian soil samples
 M Mathura Road, Aligarh++++
 K Jajmau area, Kanpur++++
 G Industrial area, Ghaziabad++++
 C.Aligarh Bypass, Aligarh++++
Figure 1.

 (a) Southern blot of IncP trfA2 PCR products of Indian soil samples with a digoxigenin-labelled RP4-derived probe. Lanes 1–4, contain the PCR products of the Indian soil samples, M, K, G, and C; lane 5, Escherichia coli SCS1 (RP4) as positive control; lane 6, negative control (no DNA was applied to the PCR). (b) Dot blot of IncP trfA2 PCR products of German soil samples from Berlin-Buch with a digoxigenin-labelled RP4-derived probe. Spot 1, PCR product of Escherichia coli SCS1 (RP4) as positive control; Spot 2, negative control (no DNA was applied to the PCR); Spots 3–5 contain the PCR products of the Buch soil samples (B1, B2 and B3). (c) Southern blot of IncP trfA2 PCR products of Indian and German soil samples with a digoxigenin-labelled pB10-derived probe. Lanes 1, Tiergarten rhizosphere; 2, Tiergarten non wettable; 3, Tiergarten wettable; 4–6, Buch samples, B1, B2, B3; 7–10, Indian samples (M, K, G, C); 11-12: PCR product of Escherichia coli JM109 (pB10) as positive control; 11, 0.1 ng; 12, 0.01 ng; 13, negative control (no DNA was applied to the PCR).

Figure 2.

 (a) Dot blot of IncP oriT PCR products with a digoxigenin-labelled RP4-derived probe. Spot 1, PCR product of Escherichia coli SCS1 (RP4) as positive control; spot 2, negative control (no DNA was applied to the PCR); spots 3–6, Indian soil samples, M, K, G, C; and spots 7–9, German soil samples from Berlin-Buch (B1, B2, B3). (b) Dot blot of IncP oriT PCR products with a digoxigenin-labelled pB10-derived probe. Spot 1, Tiergarten rhizosphere; 2, Tiergarten non wettable; 3, Tiergarten, wettable; 4, Buch 1 (B1); 5, B2; 6, B3; 7–10, Indian soil samples (M, K, G, C); 11–15, positive controls (PCR product of Escherichia coli JM109 (pB10): 11, 0.5 ng; 12, 1 ng; 13, 0.1 ng; 14, 0.01 ng; 15, 0.001 ng; 16, negative control (no DNA was applied to the PCR). (c) Dot blot of plasmids isolated by exogenous plasmid isolation from German soils. Before hybridization plasmid DNA was cut with EcoRV. A digoxigenin-labelled oriT probe derived from pB10 was used. Spots 1–5, positive controls [PCR product of Escherichia coli JM109 (pB10)]; 1, 1 ng; 2, 0.1 ng; 3, 0.01 ng; 4, 1 pg; 5, 0.1 pg; 6, negative control (no DNA was applied to the PCR); lanes 7–14, four plasmids isolated from transconjugants from selective ampicillin (10 μg mL−1) plates, each in two different concentrations: 7, plasmid 1 (1 μL); 8, plasmid 1 (3 μL); 9, plasmid 2 (1 μL); 10, plasmid 2 (3 μL); 11, plasmid 3 (1 μL); 12, plasmid 3 (3 μL); 13, plasmid 4 (1 μL); 14, plasmid 4 (3 μL).

However, all the samples were negative for all the other investigated plasmids as proved by PCR, Southern and dot blots (data not shown). Only plasmids belonging to the IncP group were detected in all the soil samples from India and Berlin Buch (Figs 1a–c, 2a and b). These soils are contaminated with various inorganic and organic pollutants (Aleem et al., 2003; Aleem & Malik, 2003; Täumer et al., 2005). Several heavy metal and antibiotic resistant gram-positive and gram-negative bacteria have been isolated from the investigated Indian soils (Malik et al., 2002; Aleem et al., 2003; Ansari & Malik, 2007). The detection of IncP trfA2 and oriT sequences by PCR amplification and hybridization in total DNA from soil samples clearly showed that IncP plasmids are prevalent in these habitats. trfA2 and IncP oriT detection as well as IncP plasmid isolation from polluted environments has also been described by others (Götz et al., 1996; Gstadler et al., 2003). Götz et al. (1996) reported that 60% of the IncP plasmids gave a positive PCR result with oriT primers and 53% with trfA2 primers. The presence of conjugative/mobilizable IncP plasmids in soils indicates that these habitats are likely to have a gene-mobilizing capacity with implications for potential dissemination of introduced recombinant DNA. Plasmids from enteric bacteria and Pseudomonas fall into more than 30 incompatibility groups. Plasmids from four of these groups (IncP, W, N, and Q) can transfer to and maintain themselves in both enteric bacteria and Pseudomonas. IncP plasmids are especially widely distributed in gram-negative bacteria e.g. in Escherichia coli, Pseudomonas, Klebsiella aerogenes and Sphingomonas (Thomas, 2000; Harada et al., 2006). In contrast to IncP specific sequences, IncQ, IncN, IncW and pMV158-type rolling circle plasmid specific sequences could not be detected in the soil samples.

Exogenous plasmid isolation by biparental matings with bacteria detached from German and Indian soil samples as donors and rifampicin resistant Escherichia coli HMS as recipients was performed by selecting for ampicillin (10 μg mL−1) or kanamycin (30 μg mL−1) resistance conjugative plasmids. Tetracycline (20 μg mL−1) and chloramphenicol (50 μg mL−1) were also tested in the selective plates but no transconjugants were obtained. In the Berlin soils in all six test sites conjugative plasmids were detected. The frequencies of kanamycin resistance gene transfer were in the range of 2.4 × 10−10–6.6 × 10−8 transconjugants per recipient, with the lowest rate for the Tiergarten rhizospheric soil and the highest rate for the hydrophobic Berlin-Buch soil. For the ampicillin resistance, transfer frequencies of 5.4 × 10−8 (hydrophobic sample Berlin-Buch) to 1.6 × 10−7 transconjugants (hydrophilic sample Berlin–Buch) were obtained. Twelve putative transconjugants were picked each from the ampicillin and the kanamycin selective plates (two for each test site) for further characterization. Nine out of the 12 transconjugants from the ampicillin plates were PCR-positive for IncP oriT, seven were positive for IncP trfA2, and six were positive for both oriT and trfA2. Four of the transconjugants positive for oriT and trfA2 were selected for plasmid DNA isolation which revealed four plasmids of approximately the same size, considerably bigger than 21 kb as compared with the λ/HindIII/EcoRI marker. The plasmids were hybridized with IncPα- and IncPβ-derived oriT and trfA2 probes. All four plasmids gave a signal with oriTpB10 and also a very weak signal with trfA2pB10 (Fig. 2c and data not shown). None of the plasmids gave a signal with the RP4-derived probes. From the kanamycin plates also 12 putative transconjugants were selected, seven of them were PCR-positive for IncP oriT, six of them for trfA2, four for both sequences.

In the Indian soils, in three of the four test sites conjugative plasmids were detected. From Aligarh Bypass, the agricultural field irrigated with ground water instead of wastewater no conjugative plasmids were isolated from plates supplemented with kanamycin or ampicillin. The frequencies of kanamycin resistance gene transfer for the Indian soil samples were in the range of 5.9 × 10−10 (Kanpur soil) to 1.2 × 10−9 (Ghaziabad soil) transconjugants per recipient. For the ampicillin resistance transfer for the Ghaziabad soil (only the exogenous isolation from Ghaziabad soil resulted in ampicillin resistant transconjugants) the transfer rate was two to three orders of magnitude higher, namely 2.7 × 10−7. It is not surprising that highest gene transfer frequencies were observed in Ghaziabad soil, as this field has been constantly irrigated with industrial wastewater until now. Five putative transconjugants were picked from the ampicillin plates and 12 from the kanamycin selective plates (from Aligarh Mathura Road, Kanpur and Ghaziabad soils) for further characterization. Four out of the five transconjugants from the ampicillin plates were PCR-positive for oriT, two for trfA2 and two for both sequences. Eleven out of the 12 transconjugants from the kanamycin plates were PCR-positive for tfrA2, nine for oriT and nine for both of them.

Transfer frequencies of ampicillin resistance were generally higher than for kanamycin resistance, typically encoded by IncPα plasmids (e.g. Haines et al., 2007). This data could indicate a prevalence of IncPβ plasmids in the investigated strongly polluted German soils (abandoned sewage field Berlin-Buch) and Indian soils (industrial area Ghaziabad). This would be in agreement with the hybridization data of the exogenously isolated plasmids from the German soils.

Tests were also carried out on the soil samples for the presence of the antibiotic resistance genes, ampC, tet (O), SHV-5, ermB, mecA, and vanA. Only three Indian soil samples irrigated with wastewater (samples M, K, and G) gave a positive PCR result for ampC as well as with dot blots (Table 4). None of the samples from groundwater irrigated soils contained the ampC gene. This observation could indicate that antibiotic resistance genes might be transferred from the sewage system to soil. Clinically relevant antibiotic resistance genes like ampC have been detected in municipal wastewater, groundwater (Kümmerer, 2004) as well as in soil (Malik et al., 2002; Agreso et al., 2004). All the other samples were negative for all tested antibiotic resistance genes.

Table 4.   Antibiotic resistance genes analysed in total DNA extracts from Indian soil samples by PCR and dot blot hybridization
Soil originAntibiotic resistance genes
ampCtet(O)ermBSHV-5mecAvanA
PCRdotPCRdotPCRdotPCRdotPCRdotPCRdot
Indian soil samples
M Mathura Road, Aligarh++
K Jajmau area, Kanpur++
G Industrial area, Ghaziabad++
C Aligarh Bypass, Aligarh

To the authors' knowledge, this is the first molecular analysis of various differently contaminated Indian soils with regard to the prevalence of plasmid-specific and antibiotic resistance genes. The presence of heavy metal resistance plasmids in bacterial isolates from the investigated Indian soil samples has been analysed with regard to mercury, chromium, nickel and copper resistance (Malik et al., 2002; Aleem et al., 2003). A variety of gram-negative bacteria including Pseudomonas, Enterobacter and Azotobacter spp. have been isolated from the Indian soil samples and tested for mercury resistance. Mercury resistance was found to be plasmid-encoded in the isolates (Malik & Jaiswal, 2000; Aleem et al., 2003). Recently, Smalla et al. (2006) investigated two highly mercury polluted and two nonpolluted sites for the presence of mercury resistance genes and broad host range plasmids. They found an increased abundance of mercury resistance genes and IncP replicon specific sequences in the total community DNA of mercury polluted sites (sediments) based on PCR and hybridization.

The combined PCR/hybridization method applied in this work proved to be a rapid and specific approach for the detection of plasmid specific sequences and antibiotic resistance genes in soil. The continuous development of plasmid replicon probes will greatly facilitate future studies of plasmid diversity and distribution as well as gene flow in natural microbial communities.

Acknowledgements

A.M. is thankful to the Department of Science & Technology, Ministry of Science & Technology, Govt. of India, New Delhi (India) for providing the BOYSCAST fellowship. The authors highly acknowledge the DFG research group INTERURBAN for helpful discussions and financial support, the Department of Soil Science, TU Berlin, for providing the Berlin Buch samples and M.Y.A. for help with Southern hybridizations.

Ancillary