Health Articles

Bia.unibz.it

This article was downloaded by:[ABM Utvikling STM / SSH packages]On: 10 April 2008Access Details: [subscription number 787564630]Publisher: Informa HealthcareInforma Ltd Registered in England and Wales Registered Number: 1072954Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK Microbial Ecology in Health andDisease Publication details, including instructions for authors and subscription information:
Low prevalence of blaTEM genes in Arctic environments
and agricultural soil and rhizosphere
Lorenzo Brusetti a; Trine Glad b; Sara Borin a; Petter Myren bc; Aurora Rizzi a;
Pål J. Johnsen b; Phil Carter c; Daniele Daffonchio a; Kaare M. Nielsen bd
a Department of Food Science, Technology and Microbiology (DISTAM), Universityof Milan, Milan, Italy b Department of Pharmacy, Faculty of Medicine, University of Tromsø, Tromsø,Norway c ESR Kenepuru Science Centre, Porirua, New Zealand d Norwegian Institute of Gene Ecology, Tromsø, Norway First Published on: 07 February 2008
To cite this Article: Brusetti, Lorenzo, Glad, Trine, Borin, Sara, Myren, Petter, Rizzi, Aurora, Johnsen, Pål J., Carter,
Phil, Daffonchio, Daniele and Nielsen, Kaare M. (2008) 'Low prevalence of blaTEM genes in Arctic environments and
agricultural soil and rhizosphere', Microbial Ecology in Health and Disease, 20:1, 27 - 36
To link to this article: DOI: 10.1080/08910600701838244
URL:
PLEASE SCROLL DOWN FOR ARTICLE Full terms and conditions of use: This article maybe used for research, teaching and private study purposes. Any substantial or systematic reproduction,re-distribution, re-selling, loan or sub-licensing, systematic supply or distribution in any form to anyone is expresslyforbidden.
The publisher does not give any warranty express or implied or make any representation that the contents will becomplete or accurate or up to date. The accuracy of any instructions, formulae and drug doses should beindependently verified with primary sources. The publisher shall not be liable for any loss, actions, claims, proceedings,demand or costs or damages whatsoever or howsoever caused arising directly or indirectly in connection with orarising out of the use of this material.
Microbial Ecology in Health and Disease. 2008; 20: 27
36 Low prevalence of blaTEM genes in Arctic environments andagricultural soil and rhizosphere LORENZO BRUSETTI1, TRINE GLAD2, SARA BORIN1, PETTER MYREN2,3,AURORA RIZZI1, PA ˚ L J. JOHNSEN2, PHIL CARTER3, DANIELE DAFFONCHIO1 & KAARE M. NIELSEN2,4 1Department of Food Science, Technology and Microbiology (DISTAM), University of Milan, Milan, Italy, 2Department ofPharmacy, Faculty of Medicine, University of Tromsø, Tromsø, Norway, 3ESR Kenepuru Science Centre, Porirua, NewZealand and 4Norwegian Institute of Gene Ecology, Science Park, Tromsø, Norway AbstractThe prevalence of blaTEM genes conferring ampicillin resistance (Ampr) in different soils was determined to clarify theenvironmental distribution of resistance determinants of major clinical importance. Samples were collected from 14 sites inNew Zealand, mainland Norway, Svalbard, and 2 soil microcosms made of compost purchased in Italy. The Ampr bacteriarepresented 1.7
100% of the cultivable microflora with an average of 28%. Approximately 1200 Ampr isolates were further Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 analyzed. Although
50% of the resistant isolates were capable of b-lactam-ring (nitrocefin) degradation, none carried aPCR-detectable blaTEM gene. The proportion of blaTEM genes in the culturable Ampr isolates was B0.07%. The overallblaTEM gene prevalence was determined by blaTEM-specific PCR of DNA extracted directly from the environmental sample.
DNA hybridization was performed on selected samples with a detection limit of 11 blaTEM genes per PCR sample. Ouranalysis indicates that the prevalence of blaTEM carrying bacteria is B1 per 1000 to 100 000 bacteria in the samplesanalyzed. The study suggests that blaTEM genes are rare in soil environments, in contrast to their increasing prevalence insome clinical and commensal bacterial populations. The frequent observation of nitrocefin-degrading capacity among thesampled isolates suggests that other mechanisms conferring enzyme-mediated resistance to b-lactam antibiotics arewidespread in Arctic and agricultural soil environments.
Key words: ampicillin resistance, blaTEM alleles, non-selective environments, natural reservoirs, soil, geographic distribution fish farms, sewage, drinking water, polluted rivers,and food. The focus of most of these investigations Bacterial resistance to antimicrobials in previously has been on the phenotypic characteristics of identi- susceptible pathogenic species can develop after fied antibiotic-resistant bacteria (4). With the excep- spontaneous mutation or horizontal acquisition of tion of the environmental distribution of some antimicrobial resistance determinants present inother bacterial populations, species, and environ- aminoglycoside resistance genes (5
7), the broader ments (1). Several studies suggest that some resis- natural prevalence of antibiotic resistance genes is poorly described in non-clinical environments.
populations of no clinical significance, i.e. the Among the most clinically and economically determinants are present in environments not be- important antibiotic resistance genes are those en- lieved to encounter significant exposure to, and coding the b-lactamases (bla genes) (8) producing selection imposed by, pharmaceutically produced high level resistance to b-lactam antibiotics, the most antibiotics (1
3). The description of antibiotic- widely used antibiotics in clinical and veterinary resistant bacteria in non-clinical environments has practice (9,10). A variety of transferable genes traditionally been limited to those environments encoding b-lactamase activity have been described important for human health, such as farm animals, in clinical environments including blaCTX-M, blaGES, Correspondence: K.M. Nielsen, Department of Pharmacy, Faculty of Medicine, University of Tromsø, 9037 Tromsø, Norway. Tel:  47 77646165.
Fax:  47 77646151. E-mail: [email protected] (Received 12 February 2007; accepted 2 November 2007) ISSN 0891-060X print/ISSN 1651-2235 online # 2008 Taylor & FrancisDOI: 10.1080/08910600701838244 L. Brusetti et al.
blaHER, blaOXA, blaOXY, blaSED, blaSHV, blaSPM, Arctic Svalbard islands (Norway). In addition, two blaVEB, blaVIM, and ampC alleles. Among the most soil microcosms kept in a greenhouse with bulk soil common bla genes is the blaTEM-1 gene, the first or with maize plants were sampled over time in Italy described bla gene and a representative of the blaTEM (Table I). The soil microcosms were made from group that now consists of almost 150 different organic compost (Technic no. 7, Potgrondbedrijf alleles, all encoding different amino acid polymorph- Vrienzenveen BV, The Netherlands); the composi- isms that extend their substrate range (http:// tion was described by Brusetti et al. (15). The plant soil microcosm was prepared by seeding a commer- variants of the bla cial maize cultivar (Tundra) in circular-section pots TEM alleles have only been found in clinical isolates and are likely emerging as a result containing organic soil compost and cultivated in a of point mutations and directional selection. Specific greenhouse as described previously (15). Bulk soil ampicillin resistance-encoding bla microcosms, without plants, were kept under the TEM alleles are also present in various bacterial cloning vectors such as same conditions.
the pUC series, and have been inserted in sometransgenic plant cultivars including commercially Determination of CFUs of bacteria approved maize lines (e.g. event Bt176).
Colony-forming units (CFUs) were determined for TEM alleles, and the TEM-1 allele in particular, are commonly encountered in clinical the cultivable aerobic copiotrophic bacteria and for environments, among pathogenic and commensal the ampicillin-resistant (Ampr) aerobic copiotrophic Enterobacteriaceae present in the intestinal tract bacteria for all environments. Ampr bacteria were (11,12), in bacterial cloning vectors, and in some grown on 50 mg/l of ampicillin (Sigma) on agar transgenic plant varieties (13), few studies have plates solidified with 15 g/l of BactoAgar (Difco).
determined the non-clinical distribution and poten- Aerobic copiotrophic bacterial CFUs were deter- tial natural reservoirs of this important gene, e.g. in mined on Plate Count Agar medium (Difco) sup- complex environments such as soil and plant rhizo- plemented with 100 mg/l of cycloheximide (Fluka), sphere (14). An expanded study of the ecology of and enumerated after incubation for 2 days at 308C.
Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 The soil samples from Tromsø, Skibotn, and Sval- TEM alleles in non-clinical settings will improve our bard were grown on Plate Count Agar medium understanding of the origins, environmental dy- without cycloheximide and enumerated after incu- namics, and evolution of antibiotic resistance deter- bation for 3 days at 208C. For the CFUs, means and minants. In this study, the prevalence of blaTEM standard deviations (SD) were calculated on the variants in cultivable and uncultivable ampicillin- basis of three replicates (five to nine replicates for resistant bacteria populations was determined from a Tromsø samples), and analysis of variance of CFU/g range of geographically diverse soils, composts, and fresh or dry weight was carried out (Table II).
rhizosphere samples. Material was collected from Individual means in the ANOVA analysis were three different agricultural soils and pasture land in compared using Tukey's test.
New Zealand, six pristine environments in Tromsøand Svalbard, Norway, five farming soils and com-posts made from organic waste from Northern Identification of ß-lactamase activity with the nitrocefin Norway, and two soil microcosms made of compost purchased in Italy. The proportions of ampicillin- A nitrocefin disk test was used to determine if an resistant bacteria, the nitrocefin-positive, and the Ampr isolate had extracellular b-lactamase activity. A blaTEM-specific PCR-positive isolates were deter- stock solution (0.5 g/l) of nitrocefin (chromogenic mined from the cultivable fraction. The potential cephalosporin compound 87/312, Glaxo) was pre- presence of unidentified blaTEM-carrying bacteria pared by adding 9.5 ml of 0.1 M phosphate buffer, was examined with blaTEM-specific PCR of total pH 7.0, to 0.5 ml of nitrocefin stock solution (5 mg DNA extracted from the various samples. We dissolved in 0.5 ml dimethylsulfoxide). Ten ml of the provide an estimate of the maximum prevalence of solution was added to single colonies of each isolate the blaTEM alleles in the environmental samples and a color change from yellow to pink within 30 min after application indicated b-lactamase activity.
PCR amplification of potential bla Materials and methods TEM genes in the Ampr Sampling of bulk and rhizosphere soil Single colonies were used for DNA extraction using Soils and rhizosphere were sampled from 14 differ- a boiling lysis method (16). Genomic DNA of the ent sites in New Zealand, Northern Norway, and the isolates from the Italian microcosms was extracted Prevalence of blaTEM genes in soil Table I. Origin and characteristics of the environmental samples.
Location and description* Bulk commercial organic compost kept in pots Maize grown in pots containing commercial organic compost Ryegrass and white clover grown in untreated soil of volcanic origin Grass grown in untreated sandy loam soil used for pastoral farming Browntop, sweet vernal and flatweed grown in untreated agricultural soil Tromsø; barley, ryegrass, and peas in organic farming agricultural soil Tromsø; barley, ryegrass, and peas in conventional farming, agricultural soil Tromsø; barley in organic farming, agricultural soil Tromsø; barley in conventional farming, agricultural soil Skibotn; compost made of household waste from Northern Norway Skibotn; organic soil in low alpine zone ˚ lesund; free soil without vegetation nearby Italian Arctic Research Base Midtre Love´nbreen; free soil without vegetation in front of glacier moraine Stuphallet plateau; Saxifraga foliolosa grown in pristine soil Midtre Love´nbreen; Silene acaulis grown in pristine soil in front of glacier moraine Glacier sediments Midtre Love´nbreen; sediments in cryoconite holes *Italy: samples from compost soils (Technic no. 7, The Netherlands) kept in a greenhouse in Milan were collected from six pots withcultivated maize (rhizosphere samples) and from three pots without cultivated maize (bulk soil samples). For each pot, two samples weretaken after 30 or 100 days of plant growth or incubation. The soil microcosms were sampled as described by Brusetti et al. (15). NewZealand: for the rhizosphere samples, the grass layer was removed and soil from right under the grass and to 10 cm depth was kept. Soilsamples were homogenized by sieving through a 2 mm sieve and stored at 48C (up to 1 week) in double plastic bags before further analysis.
Samples were obtained in October and November 2003 (late spring and early summer in the Southern hemisphere). Norway: the sampleswere collected as follows. For bulk soil, the surface was removed and three replicate samples were collected from 1
5 cm depth into a sterilecontainer. For the rhizosphere, three plants were collected in sterile containers and transported to the laboratory. The soil fraction attachedto the roots was collected in sterile containers by the use of a sterile spatula. Cryoconite holes (water-filled cylindrical melt-holes on glacialice surface) were sampled by removing water and stones and collecting the bottom sediments with a sterile spatula. The Tromsø and Skibotn Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 samples were analyzed immediately after being returned to the laboratory (within 5 h). The Svalbard samples were stored at 48C untilanalysis (within a few days). The Svalbard sampling was done in August 2004 and the Tromsø and Skibotn samplings in September 2005.
with a CTAB method (17). To confirm the absence 728C for 10 min. For all blaTEM PCR analyses, the of PCR inhibitory substances in the genomic DNA primers BlaF and BlaR (Table III) were used to of the various single isolates, PCR amplification of amplify a product of 828 bp (TEM-1 allele). The 16S rRNA gene was performed on a subset of the specificity of the primers was confirmed by ‘in silico' DNA extracts with the primers 16S-27F and 16S- 1494R (Table III). The genomic DNA extracted infx) and by aligning the primer binding region of through the CTAB method was confirmed as approximately 100 sequence polymorphic TEM- suitable for PCR analysis by performing PCR of alleles (Table IV).
the bacterial 16S-23S rRNA gene intergenic tran- The following controls were used: five strains of scribed spacers (ITS-PCR) with primers ITSF and Escherichia coli carrying the bla alleles TEM-1, ITSReub (18), as described by Daffonchio et al.
TEM-3, TEM-6, TEM-9, and TEM-10 as positive controls, and one strain carrying the SHV-2 allele as The amplification of blaTEM alleles in individual negative control. The blaSHV-2 gene shares about bacterial isolates was performed in a reaction mix- 70% of nucleotide similarity with the blaTEM gene ture containing 1 HotStartTaq DNA master mix, cluster. All strains were kindly provided by A.
0.2 mM of each primer, and 2 ml of the crude DNA Sundsfjord, University Hospital of North-Norway, solution in a final volume of 30 ml. Bacterial isolates Tromsø, Norway. Two strains, Pseudomonas putida from the microcosm samples were in addition ET-B12 and Flavobacterium sp. ET-N11, found to be analyzed by PCR using a different reaction mixture resistant to ampicillin but without an identifiable containing 1  PCR buffer (Pharmacia), 1.5 mM blaTEM gene, and previously isolated from a soil MgCl2, 0.10 mM dNTPs, 0.2 mM of each primer, 1 microcosm (Italy), were used as positive controls U of native Taq polymerase (Pharmacia), and 2 ml of after they had been transformed by electroporation genomic DNA in a final volume of 50 ml. Reactions with plasmid pZR80 carrying the blaTEM-1 gene were denatured at 958C for 15 min and then conferring resistance to ampicillin (17,20). To con- subjected to 30 cycles of 948C for 45 s, 618C for firm the presence of the pZR80 plasmid in the two 45 s, and 728C for 1 min, with a final extension at electroporated strains, direct sequencing of the L. Brusetti et al.
Table II. Colony forming units (log10) of total and ampicillin-resistant bacteria (50 mg/l) obtained from the various environments.
Ampr copiotrophic Italian soil microcosm Bulk soil, day 30 Bulk soil, day 100 Maize rhizosphere, day 30 Maize rhizosphere, day 100 New Zealand agricultural soils Horotiu, rhizosphere Otaki, rhizosphere Templeton, rhizosphere Tromsø, bulk soil, organic farming Tromsø, bulk soil, conventional farming Tromsø, rhizosphere, organic Tromsø, rhizosphere, conventional Skibotn, low alpine soil Svalbard soils and sediments ˚ lesund, bulk soil Lovenbreen, bulk soil Stuphallet, rhizosphere Lovenbreen, rhizosphere Lovenbreen, glacier sediments *Total copiotrophic bacteria and Ampr bacteria measured as log10 CFU/g (dry weight) of soil. For the Svalbard samples, the CFU were Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 calculated as log10 CFU/g (fresh weight) of soil. Mean values are given with standard deviation (SD) based on three replicates; five to ninereplicates for Tromsø samples.
$Percentage of Ampr CFU of the total culturable bacterial population.
flanking regions of the blaTEM gene were performed were deposited in the EMBL nucleotide sequence using the sequencing primers TemI3, TemI5a, or database (GenBank/EMBL/DDBJ) under the acces- TemI5b (Table III).
sion numbers AM261986 to AM262113.
Sequencing of the 16S rRNA gene of Ampr isolates Extraction and analysis of DNA from soil To describe some of the diversity of Ampr bacterial Total DNA from 11 different sample sites was phenotypes present in the Italian soil microcosm, extracted from 0.5 g of material using the Bio101 288 isolates were analyzed with ITS-PCR (see FastDNA Spin Kit for soil (Q-BioGene) according above) (21). One isolate of each of the ITS-PCR to the manufacturer's instructions. Further purifica- haplotypes was further identified by partial 16S tion and concentration of eluted DNA was per- rRNA gene sequence analysis described by Brusetti formed with the QIAamp DNA Stool Mini Kit et al. (22) using primer 16S-926F (Table III). The (Qiagen). Soil DNA was finally eluted in 30 ml of nucleotide sequences of the partial 16S rRNA genes TE buffer, pH 8. The general suitability of DNA for Table III. PCR and sequencing primers used in this study.
Primer sequence (5?-3?) Prevalence of blaTEM genes in soil Table IV. Identified nucleotide polymorphisms between the BlaF and BlaR primers used and some published blaTEM-alleles*.
BlaF primer (5?-3?) blaTEM-1b, 1d, 1e, 1g, 1h, 5, 10, 15, 18-19, 27-28, 33, 35, 47, 50, 55, 61, 68, 70, 76-77, 79, 88, 92, 95, 102, 104-107 BlaR primer (5?-3?) *The primer sequences are identical to the blaTEM-1a allele. In addition to the alleles blaTEM-1a to -1h and blaTEM-116, the alignment includedall alleles up to blaTEM-107, except blaTEM-14, -23, -37, -58, -62, -64, -67, -69, -75, -100, -103. Only alleles with nucleotide polymorphisms in theprimer binding regions are shown.
PCR was confirmed with amplification of the 16S pUC18 molecules. The variation in the pUC18 copy rRNA gene. The PCR mixture contained 1 Hot- number, as affected by variation in individual DNA StartTaq master mix, 0.3 mM of each primer, 1 Q- sample measurements, ranged from 1.4 1011 to solution (Qiagen), 0.1 mg/ml bovine serum albumin, 3.4 1011 for the estimated 1.7 1011 copies, and 5% dimethyl sulfoxide, 2 mM of MgCl2, and 2 ml of from 2 to 5 copies for the estimated 3 copies.
extracted DNA (100 ng/ml) in a final volume of Stochasticity in the pUC18 distribution during 30 ml. The number of cycles was increased to 33.
dilutions makes the lower estimate less precise.
The PCR targeting of blaTEM genes from total DNA The detection limit of the blaTEM-specific PCR was was performed with the primers described in Table also determined without soil DNA present using a III and 2 ng of DNA per reaction.
range of pUC18 dilutions. The detection limit wasdetermined in total DNA extractions of the soilsamples from Tromsø, Skibotn, and Svalbard.
Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 Southern blotting of blaTEM PCR products Agarose gels of blaTEM-specific PCRs performed onsome of the DNA samples were blotted overnight onto positively charged nylon membranes (Pharma- Bacterial counts of the culturable bacterial fraction cia) by capillary transfer (17). The membrane wastreated with UV light for 4 min and hybridized The bacterial counts (CFUs) of the total and the overnight at 428C in 50% (v/v) formamide with the Ampr fraction of aerobic copiotrophic bacteria were corresponding DIG-labelled blaTEM-1 variant. Two determined in all samples (Table II). The cultivable stringent washes were performed at 658C for 15 min.
copiotrophic bacteria ranged between 4.7 and 8.6 Prehybridization, hybridization, washes, and chemi- log10 CFU/g dry weight for Italian compost and luminescent detection were performed using the mainland Norway samples New Zealand soil samples, DIG DNA labelling and detection kit following the and between 4.3 and 6.1 log10 CFU/g fresh weight for supplier's instructions (Boehringer Mannheim). The Svalbard. The proportion of ampicillin-resistant bac- probe was DIG-labeled by PCR using the conditions teria (measured as growth on 50 mg/l ampicillin) described for the blaTEM PCR.
ranged from 1.7% to 100% of the culturable bacterialfraction (Table II). The New Zealand soil sampleswere characterized by a relatively high percentage of Limit of detection for blaTEM genes in soil DNA Ampr bacteria (between 31.9% and 54.9%), whereas To determine the detection limit of the blaTEM the soils from Northern Norway (Tromsø and Ski- amplification of total DNA, 2 ng of total soil DNA botn) had lower levels of resistant bacteria (8.8
was added to each reaction spiked with decreasing 16.1%). The observable proportion of Ampr cultur- amounts of commercially available pUC18 plasmids able bacteria in the samples from Svalbard ranged (Invitrogen) from 1.7 1011 to 3 molecules. To from 20.2% in bulk soil to 100% in glacier sediments.
describe the variation in the calculated number of Compost samples such as the organic plant compost pUC18 molecules, the DNA concentration of un- from Skibotn, Norway, and the Italian bulk soil diluted and 10-fold dilution of pUC18 were mea- microcosm had the lowest percentage of Ampr bacteria (1.7% and 3.5%, respectively). In the bulk (NanoDrop Technologies). The highest and lowest soil microcosm (Italy), the relative percentage of values of 18 samples taken from the DNA stock were Ampr bacteria increased significantly from 3.5% to used to calculate the outer concentration range of 9.4% (p 0.00001), over the 100 days incubation L. Brusetti et al.
period. In the plant soil microcosms, the percentage PCR targeting of blaTEM genes in the Ampr isolates of Ampr bacteria decreased significantly from 17.2% The suitability for PCR analysis of the DNA to 11.6% over the 100 days incubation period (p  prepared from the 1329 Ampr bacterial isolates was 0.00004). Despite the relative percentage changes, confirmed with PCR amplification of the 16S rRNA the overall levels of Ampr bacteria changed little, from gene or by ITS-PCR. Of the 1329 DNA prepara- 5.2
5.6 log10 CFU/g dry weight to 5.0 log10 CFU tions tested, 1198 were positive, indicating that over the time course. Significant differences (p B bacterial DNA was amplifiable in 90.1% of the 0.05) between plate counts with or without ampicillin samples. Subsequently, 1198 isolates were screened were found in all the samples except for two soil and for the presence of bla sediment samples from Svalbard, both with an Ampr genes with primers designed for the TEM-1 allele and derivatives.
proportion
60%.
PCR of the different TEM-harboring E. coli strainsand of the two pZR80-transformed strains was Phenotypic analysis of the Ampr bacterial isolates positive, while amplification of the SHV-2 genegave a faint band (data not shown). Of the 1198 A total of 1329 Ampr bacterial isolates was obtained environmental isolates analyzed by PCR, none from the various environments. The nitrocefin test produced visible bla was performed on 1185 isolates (Table V). The TEM amplicons (Table V). In silico PCR, absence of primer mismatches to the cephalosporin nitrocefin was cleaved by 97.7% of most relevant TEM alleles (Table IV), and amplifi- the bacterial isolates, excluding the New Zealand cation of the positive controls demonstrate the broad soils and the Svalbard glacier sediment samples. For ability of the primers used to detect relevant TEM the New Zealand soil samples, the nitrocefin-de- alleles across phylogenetically diverse groups, if grading Ampr bacteria represented only a minor present in the isolates.
fraction, between 18.6% and 30.7% of the totalAmpr isolates. No nitrocefin-degrading bacteria wereidentified among the isolates in Svalbard cryoconite Identification of Ampr isolates through partial 16S holes containing glacier sediments, despite the ob- Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 served 100% proportion of Ampr cultivable bacteria.
To provide some insight into the species diversity of Table V. Nitrocefin substrate test and blaTEM targeted PCR the Ampr bacteria isolated in a specific soil environ- analysis of single Ampr bacterial isolates from each environment.
ment, a total of 288 Ampr isolates from the soilmicrocosms made of purchased compost (Italy) were Isolates positive/tested grouped into 139 haplotypes by ITS-PCR finger-printing analysis (19). The 16S rRNA gene se- quences subsequently obtained covered between Italian soil microcosm 240 and 570 bp, with an average of 506 bp.
Bulk soil, day 30 At day 30 of the greenhouse incubation, the Ampr Bulk soil, day 100 bulk soil isolates were characterized by the presence Maize rhizosphere, day 30 of 13 different genera. Of the isolates examined, Maize rhizosphere, day 100 44% were b-Proteobacteria, mostly belonging to the New Zealand agricultural soils genera Acidovorax and Pandoraea; 21% of the isolates Horotiu, rhizosphere Otaki, rhizosphere were affiliated to a-Proteobacteria of the genera Templeton, rhizosphere Ochrobactrum, Caulobacter, and Zooglea. g-Proteo- Norwegian mainland soils bacteria was represented by 15% of the isolates. One Tromsø, bulk soil, organic farming isolate of the Actinobacteria (Streptomyces sp.) was Tromsø, bulk soil, conventional found. At day 100, the majority of the isolates belonged to 4 different species of Pseudomonas (P.
Tromsø, rhizosphere, organic alcaligenes, P. jessenii, P. putida, and P. resinovorans), Tromsø, rhizosphere, conventional although 25 different Pseudomonas species were Skibotn, low alpine soil identified among the isolates. The other sequenced Svalbard soils and sediments isolates were grouped to Acidovorax spp., Chryseo- ˚ lesund, bulk soil bacterium sp., Flavobacterium sp., Mesorhizobium sp., Lovenbreen, bulk soil Roseateles sp., and Stenotrophomonas sp.
Stuphallet, rhizosphere At day 30 in the maize rhizosphere microcosm Lovenbreen, rhizosphere samples, the bacterial isolates sequenced mainly Lovenbreen, glacier sediments belonged to b- and g-Proteobacteria and to the Bacteroidetes/Chlorobi group. The Bacteroidetes/ ND, not determined.
Chlorobi group bacteria were mostly represented

Prevalence of blaTEM genes in soil by Dyadobacter fermentens and Flexibacter sanctispecies (10% of the strains). Other genera repre-sented were Acidovorax, Flavobacterium, Pandoraea,Pseudomonas, and Stenotrophomonas. In the rhizo-sphere of maize sampled at day 100, 56% of the 16Ssequenced bacteria were affiliated to the g-Proteo-bacteria class, exclusively to the genera Pseudomonasand Stenotrophomonas. The species P. fluorescens, P.
jessenii, P. monteilii, P. pseudoalcaligenes, P. putida, andP. resinovorans were identified. Other bacterial generawere represented by Azospirillum sp., Chryseobacter-ium sp., Flavobacterium sp., Mesorhizobium sp.,Ochrobactrum sp., and Variovorax sp. Some of theabove species or genera are known to be b-lactamasecarriers (23,24).
Detection limit of the PCR targeting of blaTEM genes Eleven soil, rhizosphere, and sediment samples fromNorway were screened for the presence of blaTEMgenes in extracted total DNA. No visible amplifica-tion products of the blaTEM genes were obtainedusing 2 ng of extracted DNA in the reactions. DNA Figure 1. PCR with blaTEM-specific primers of dilution series of sequence alignment, and the PCR-positive controls pUC18 DNA (estimated copy number of 170 to 5), in the (TEM-1, -3, -6, -9, -10), as well as spiked soil presence of soil DNA extracts. Lanes 1
5, with soil DNA from samples (pUC18) demonstrated the broad ability of organic farming in Tromsø; lanes 6
10, DNA from compost from Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 Northern Norway; 12, DNA fragment size ladder.
the TEM-specific primers used to detect relevantTEM alleles in the DNA samples analyzed. In 10 of genes for experimental field releases of genetically 11 samples, the detection limit of the blaTEM modified organisms (14). However, few experimen- amplification products visualized in agarose gels tal data exist on the environmental distribution of was 21 (range 17
43, see Materials and methods) pUC18 molecules (Figure 1). The extracted DNA TEM genes in natural environments, such as in food, feed, water, bulk soil, rhizosphere, compost, from low alpine soil in Skibotn, Norway, did not and pristine environments (14).
yield a satisfactory detection limit when spiked with In this study, the prevalence of bla the pUC18 plasmid. To further increase the sensi- was surveyed in microbial populations of a variety of tivity, Southern hybridizations were done on thePCR solutions of some of the soil DNA samples environments. The concentrations of cultivable (Figure 2). Very faint bands were observed in the Ampr bacteria in all samples was high, typically blotted PCR solution from Ny A ˚ lesund and Tromsø 103
108 CFU per dry weight gram of sample; in bulk soil, rhizosphere from Lovenbreen, and from some samples they represented
50% of the total conventional farming land from Tromsø. Southern culturable microbial population (Table II). The high blotting of PCR products from solutions with proportion of Ampr bacteria already present in the decreasing amounts of pUC18 target DNA yielded samples suggests that in situ or in vitro enrichment a clear signal with 85 molecules and was positive (with ampicillin selection) would enhance our cap- with as low as 11 pUC18 copies (Figure 2).
ability to detect blaTEM-positive bacteria only ap-proximately twofold, assuming uniform communityresponse to the enrichment conditions.
Although the majority of
1100 isolates tested were nitrocefin-degrading (hence capable of cleav- TEM types are among the earliest described b- lactamase encoding genes (11; http://www.lahey.org/ ing the b-lactam ring), bacterial isolates carrying Studies/), they are frequently used in cloning vectors blaTEM-alleles could not be identified in the analyzed in molecular biology laboratories (13), and are environmental isolates (Table V). Lower proportions present in some of the first generation genetically of nitrocefin-degrading bacteria were seen in New modified plants (14,25). The broad environmental Zealand soil samples and in the sediment samples distribution of specific blaTEM alleles has been used from cryoconite holes in glaciers on Svalbard. In the as an argument for their continued use as marker latter the entire cultivable microbial population was

L. Brusetti et al.
Figure 2. Southern blotting of agarose gel separated PCR amplified DNA with blaTEM-specific primers. (A) Lane 1, Ny A ˚ lesund, bulk soil (Svalbard); 2, Stuphallet, rhizosphere (Svalbard); 3, Lovenbreen, rhizosphere (Svalbard); 4, Tromsø, bulk soil, organic farming; 5, Tromsø,bulk soil, conventional farming; 6, Tromsø, rhizosphere, conventional farming. (B) PCR of dilution series of pUC18 DNA from anestimated copy number of 85 to 0.05. Lanes TEM-1, TEM-3, TEM-6, and TEM-9 are positive controls with 3 ng of PCR products ofdifferent variants of blaTEM genes, and NC lanes are negative controls (no DNA added to the PCR reaction).
Ampr, yet nitrocefin-degrading bacteria were not despite not providing evidence for direct transfer of found (Table V). The high proportion of bacteria resistance elements
they proposed a previously with extracellular b-lactamase activity in environ- under-appreciated density and concentration of en- ments without known or expected prior exposure to vironmental antibiotic resistance determinants.
pharmaceutically produced b-lactam compounds Our data from soil microcosms (made from suggests that antibiotic-degrading capacities are commercially available compost) incubated under also important in microbial communities unaffected controlled greenhouse conditions for 3 months by humans. The molecular and ecological basis for indicate that the relative levels of Ampr bacteria this observation needs to be determined and the change over time. After 1 month, the relative natural substrates for these enzymes need to be percentage of Ampr bacteria in the bulk soil compost identified to advance our understanding and pre- was about four times lower than in the same soil 3 dictive ability of the capacities of broadly distributed, months later. An opposite and also significant Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 naturally occurring enzymes to act on pharmaceuti- difference was observed for the proportion of Ampr bacteria in maize plant-containing soil microcosms sequence-divergent representatives of the chromoso- (see Table II). Temporal fluctuations in the resis- mally localized ampC gene to be responsible for tance proportions were also described for tetracy- some of the resistant phenotypes observed.
The fraction of Ampr bacteria reported here is bacteria (7,30). The observed fluctuations in resis- lower than the percentage of Ampr bacteria found in tance levels over time probably reflect population other soil and rhizosphere samples (26,27). A recent dynamic processes among bacterial species as a study by Badosa et al. (28) indicated that the Ampr result of competition without necessarily reflecting bacterial fraction in some maize fields in Spain directional selection of resistance traits.
represented between 77% and 88% of the total The BLAST analysis of 16S rRNA genes amplified cultivable bacterial population in the rhizosphere, from 139 Ampr haplotypes revealed a wide variety of and between 68% and 88% in bulk agricultural soil.
ampicillin-resistant species (MIC
50 mg ampicil- These latter high proportions are close to those lin/l), belonging to a-, b-, and g-Proteobacteria, observed for the rhizosphere samples from Svalbard.
Sphingobacteria, the Bacteroidetes/Chlorobi group, Several other studies of the distribution of pheno- and to Actinobacteria. Although the TEM-1 allele typic antibiotic resistance patterns have been re- has mainly been linked to commensal and clinical pathogenic isolates identified as Capnocytophaga bacteria in Danish farmlands soils was reported to ochracea, Enterobacter cloacae, E. coli, Klebsiella oxy- be between 0.1% and 7.1%, with an average of 2.1% toca, Klebsiella pneumoniae, Morganella morganii, (30). Similar percentage levels were found for Proteus mirabilis, Providencia stuartii, Pseudomonas streptomycin-resistant bacteria in bulk and rhizo- aeruginosa, Zymomonas mobilis and others, non- sphere soils (7). However, higher proportions of TEM-like b-lactamase-producing genes have been gentamicin-resistant (exact numbers not given) and described in some soil bacteria. The class B b- kanamycin-resistant (between 0.01% and 38.6%) lactamases have been found in B. cereus (31), bacteria among the cultivable copiotrophic or oligo- Chryseobacterium meningosepticum (23), and S. mal- trophic bacteria were found in different European tophilia (24).
bulk and rhizosphere soils (5,6). D'Costa et al. (29) The absence of the blaTEM gene in
1100 found high levels of multi-drug-resistant actinomy- cultured Ampr environmental isolates indicates that cetes in urban, agricultural, and forest soil and
the blaTEM genes are rare in soil environments Prevalence of blaTEM genes in soil despite the reports of high prevalence of blaTEM-1 tion and not convergent evolution. The few single genes in some pathogenic bacterial species (11), and nucleotide polymorphisms (SNPs) present in the the occurrence of the gene in the digestive system of TEM alleles strongly suggest a shared recent origin healthy humans (12). Since no blaTEM-carrying and subsequent vertical and horizontal dissemina- bacteria were identified among the examined Ampr tion among bacterial species. Sequence analysis of a isolates, it is important to be aware of the limit of broader set of bacterial species (from various envir- detection provided by the experimental methods. It onments) carrying TEM-like alleles is necessary to can be inferred that the concentration of culturable resolve the evolution and identity of the original Ampr bacteria is B0.3% in the Italian soil micro- bacterial blaTEM-1 gene donor.
B0.2% in the combined New Zealand agricultural soils, B0.4% in the combined Norwe- gian soil samples, and B0.3% in the Svalbard We thank M. Pajoro, A. Pagliuca, and P. Francia for samples, and not detectable among cultured isolates.
technical assistance in plant cultivation, bacterial The proportion of blaTEM genes in the combined counts, and strain isolation performed at the Uni- samples of all Ampr isolates is B0.07%. The versity of Milan. This work was supported by the phenotypic analysis represents 0.001
32.5% of the Italian Ministry for University and Scientific Re- amount of culturable Ampr bacteria present in 1 g of search project ‘Risposta della comunita microbica sample material. This study is thus comparable to del suolo a differenti pressioni antropiche: effetti su the sample sizes of cultured bacteria analyzed in struttura, dinamica e diversita della microflora' and previous studies (28,32).
the Fondazione Diritti Genetici, Italy, project ‘Or- A detection limit of B21 pUC18 copies per PCR ganismi geneticamente modificati ed alimentazione: sample (established through spiking of the soil DNA valutazione degli effetti diretti sull'ospite e sulla with decreasing concentrations of pUC18) was microflora intestinale' funded by the Cariplo Foun- obtained for the total DNA extracted from soil.
dation, Italy. Partial support also came from the EU Our PCR protocol included 2 ng of DNA from the project TRANSBAC QLK3-CT-2001-02242. T.G soil samples. Assuming that all DNA extracted is of Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 and K.M.N acknowledge financial support from the bacterial origin, it represents the DNA in maximum Research Council of Norway. Support for the New 100 000 bacterial cells (32) or in approximately Zealand work was provided by the Foundation for 0.02
0.2 mg soil (assuming 10
100 mg DNA per g Research, Science and Technology.
soil). Moreover, since the blaTEM gene is likely to belocalized on plasmids (and assuming 11 or morecopies per bacterial cell) it can be inferred that the blaTEM gene is present in the samples analyzed atproportions lower than 1 bla 1. Nwosu VC. Antibiotic resistance with particular reference to TEM-carrying bacterium soil microorganisms. Res Microbiol. 2001;152:421
30.
per 100 000 bacteria analyzed. However, the detec- 2. Mazel D, Davies J. Antibiotic resistance in microbes. Cell Mol tion limit in our experimental approach is probably Life Sci. 1999;56:742
54.
100-fold higher, due to the presence of non-bacterial 3. Seveno N, Smalla K, Van Elsas JD, Collard J-C, Karagouni A, DNA (e.g. of fungal origin) in the DNA extraction Kallifidas D, et al. Occurrence and reservoirs of antibiotic from soil. Thus, our investigation indicates that the resistance genes in the environment. Rev Med Microbiol.
2002;13:1
13.
prevalence of bla TEM-carrying bacteria in the sam- 4. Henschke RB, Schmidt FRJ. Screening of soil bacteria for ples analyzed is B1 per 100 000 to 1 per 1000 plasmids carrying antibiotic resistance. Biol Fertil Soils. 1990;/ This investigation suggests that bla 5. Smalla K, Van Overbeek LS, Pukall R, Van Elsas JD.
Prevalence of nptII and Tn5 in kanamycin-resistant bacteria rare in the non-clinical environments sampled in our from different environments. FEMS Microbiol Ecol. 1993;13: investigations. b-Lactamases such as Tem-1 are hypothesized to have evolved from the penicillin- 6. Heuer H, Kro¨gerrecklenfort E, Wellington EMH, Egan S, binding proteins (PBPs) due to directional selection Van Elsas JD, Van Overbeek L, et al. Gentamicin resistance on the PBP's active site (changing activity from genes in environmental bacteria: prevalence and transfer.
FEMS Microbiol Ecol. 2002;42:289
302.
binding and inhibition by penicillin to penicillinase 7. Van Overbeek LS, Wellington EMH, Egan S, Smalla K, activity and detachment of the enzyme from the Heuer H, Collard J-M, et al. Prevalence of streptomycin cytoplasmic membrane) (33,34). The blaTem-1 allele resistance genes in bacterial populations in European habitats.
was first described in 1965 and is now widespread in FEMS Microbiol Ecol. 2002;42:277
88.
clinical environments and among Enterobacteria- 8. Bush K. Characterization of b-lactamases. Antimicrob Agents ceae (35,36). The increased prevalence of the 9. Singh GS. b-Lactams in the new millennium. Part-I: mono- TEM-1 allele in human-associated bacterial popula- bactams and carbapenems. Mini Rev Med Chem. 2004;4:69
tions is due to directional selection and dissemina- L. Brusetti et al.
10. Kumar K, Gupta SC, Chander Y, Singh AK. Antibiotic use in 23. Rossolini GM, Franceschini N, Riccio ML, Mercuri PS, agriculture and its impact on the terrestrial environment. Adv Perilli M, Galleni M, et al. Characterization and sequence of the Chryseobacterium (Flavobacterium) meningosepticum carba- 11. Livermore DL. b-lactamases in laboratory and clinical penemase: a new molecular class b-lactamase showing a resistance. Clin Microbiol Rev. 1995;8:557
84.
broad substrate profile. Biochem J. 1998;332:145 ˜ as L, Zarazaga M, Sa´enz Y, Ruiz-Larrea F, Torres C. b- 24. Ullah JH, Walsh TR, Taylor IA, Emery DC, Verma CS, Lactamases in ampicillin-resistant Escherichia coli isolates Gamblin SJ, et al. The crystal structure of the L1 metallo- from foods, humans, and healthy animals. Antimicrob Agents beta-lactamase from Stenotrophomonas maltophilia at 17 A resolution. J Mol Biol. 1998;284:125
36.
13. Bensasson D, Boore JL, Nielsen KM. Genes without fron- 25. EFSA. Opinion of the scientific panel on genetically modified tiers? Heredity. 2004;92:483
9.
organisms on the use of antibiotic resistance genes as marker 14. Nielsen KM, Berdal KG, Kruse H, Sundsfjord A, Mikalsen genes in genetically modified plants. EFSA J 2004;48:1
18.
A, Yazdankhah S, et al. An assessment of potential long-term 26. Van Dijk P, Van de Voorde H. Sensitivity of environmental health effects caused by antibiotic resistance marker genes in microorganisms to antimicrobial agents. Appl Environ Micro- genetically modified organisms based on antibiotic usage and resistance patterns in Norway. Oslo, Norway: Norwegian 27. Brønstad K, Drønen K, Øvrea˚s L, Torsvik V. Phenotypic Scientific Committee for Food Safety Report; 2005. p. 1
62.
diversity and antibiotic resistance in soil microbial commu- 15. Brusetti L, Francia P, Bertolini C, Pagliuca A, Borin S, Sorlini C, et al. Bacterial communities associated with the rhizo- nities. J Indust Microbiol. 1996;17:253
9.
sphere of transgenic Bt 176 maize (Zea mays) and its non 28. Badosa E, Moreno C, Montesinos E. Lack of detection of transgenic counterpart. Plant Soil. 2004;266:11
21.
ampicillin resistance gene transfer from Bt176 transgenic corn 16. Glad T, Klingenberg C, Flægstad T, Ericson JU, Olsvik Ø.
to culturable bacteria under field conditions. FEMS Micro- Rapid detection of the methicillin-resistance gene, mecA, in biol Ecol. 2004;48:169
78.
coagulase-negative Staphylococci. Scand J Infect Dis. 2001; 29. D'Costa VM, McGrann KM, Hughes DW, Wright GD.
Sampling the antibiotic resistome. Science. 2004;311:374
7.
17. Ausubel FM, Brent R, Kingston RE, Moore DD, Seidman 30. Sengeløv G, Agersø Y, Halling-Sørensen B, Baloda SB, JG, Smith JA, et al. Current protocols of molecular biology.
Andersen JS, Jensen LB. Bacterial antibiotic resistance levels New York: John Wiley and Sons; 1994.
in Danish farmland as a result of treatment with pig manure 18. Cardinale M, Brusetti L, Quatrini P, Borin S, Puglia AM, slurry. Environ Int. 2003;28:587
95.
Rizzi A, et al. Comparison of different primer sets for the 31. Sabath LD, Abraham EP. Zinc as a cofactor for cephalospor- Automated Ribosomal Intergenic Spacer Analysis (ARISA) of inase from Bacillus cereus 569. Biochem J. 1966;98:11
3.
Downloaded By: [ABM Utvikling STM / SSH packages] At: 10:03 10 April 2008 complex bacterial communities. Appl Environ Microbiol.
32. Nielsen KM, Townsend JP. Monitoring and modeling hor- izontal gene transfer. Nature Biotechnol. 2004;22:1110
4.
19. Daffonchio D, Borin S, Frova G, Manachini PL, Sorlini C.
33. Kirby R. Evolutionary origin of the class A and class C beta- PCR fingerprinting of whole genomes, the spacers between lactamases. J Mol Biol. 1992;34:345
50.
the 16S and 23S rRNA genes and of intergenic tRNA gene 34. Hall B, Barlow M. Evolution of the serine beta-lactamases: regions reveals a different intraspecific genomic variability of past, present and future. Drug Res Updates. 2004;7:111
23.
Bacillus cereus and Bacillus licheniformis. Int J Syst Bacteriol.
35. Amyes SGB. Genes and spectrum: the theoretical limits.
Clinic Infect Dis. 1998;27(Suppl 1):S21
8.
20. Kok RG, Young DM, Ornston LN. Phenotypic expression of 36. Kotra LP, Samama J-P, Mobashery S. b-Lactamases and PCR-generated random mutations in a Pseudomonas putida resistance to b-lactam antibiotics. In: Lewis K, Salyers AA, gene after its introduction into an Acinetobacter chromosome Taber HW, Wax RG. editors. Bacterial resistance to anti- by natural transformation. Appl Environ Microbiol. 1999;65: microbials. New York: Marcel Dekker; 2002. p. 126
9.
37. Ehlers B, Strauch E, Goltz M, Kubsch D, Wagner H, ¨ rtler V, Stanisich VA. New approaches to typing and Maidhof H, et al. Nachweis gentechnischer vera¨nderungen identification of bacteria using the 16S
23S rDNA spacer in mais mittels PCR. Bundesgesundhbl. 1997;4:118
21.
region. Microbiology. 1996;142:3
16.
38. Lane DJ. 16S/23S rRNA sequencing. Nucleic acid techniques 22. Brusetti L, Borin S, Mora D, Rizzi A, Raddadi N, Sorlini C, in bacterial systematics. In: Stackebrandt E, Goodfellow M.
et al. Usefulness of length heterogeneity-PCR for monitoring editors. Modern microbiological methods. Chichester, UK: J lactic acid bacteria succession during maize ensiling. FEMSMicrobiol Ecol. 2006;56:154
64.
Wiley & Sons; 1991. p. 133.

Source: https://bia.unibz.it/bitstream/handle/10863/215/2008-BrusettiMEHDBla-TEM.pdf?sequence=2