assessment of the diversity and antagonism towards rhizoctonia solani ag3 of pseudomonas species in soil from different agricultural regimes

FEMS Microbiology Ecology 47 (2004) 51^64 www.fems-microbiology.org Assessment of the diversity, and antagonism towards Rhizoctonia solani AG3, of Pseudomonas species in soil from di¡erent agricultural regimes Paolina Garbeva b a;b , Johannes Antonie van Veen b , Jan Dirk van Elsas a;Ã a Plant Research International (PRI), Wageningen, The Netherlands Netherlands Institute of Ecology (NIOO-CTE), Heteren, The Netherlands Received 24 January 2003; received in revised form 26 August 2003; accepted 27 August 2003 First published online 27 October 2003 Abstract The genus Pseudomonas is one of the best-studied bacterial groups in soil, and includes numerous species of environmental interest Pseudomonas species play key roles in soil, for instance in biological control of soil-borne plant pathogens and in bioremediation of pollutants A polymerase chain reaction-denaturing gradient gel electrophoresis system that specifically describes the diversity of Pseudomonas spp in soil was developed On the basis of this molecular method as well as cultivation-based approaches, the diversity of Pseudomonas species in soil under different agricultural regimes (permanent grassland, arable land either under rotation or under monoculture of maize) was studied Both types of approaches revealed differences in the composition of Pseudomonas populations between the treatments Differences between the treatments were also found based on the frequency of isolation of Pseudomonas strains with antagonistic properties against the soil-borne pathogen Rhizoctonia solani AG3 Higher relative numbers of isolates either with antagonistic activity toward this pathogen or with chitinolytic activity were obtained from permanent grassland or from the short-term arable land than from the arable land The results obtained in this study strongly indicate that agricultural regimes influence the structure of Pseudomonas populations in soil, with specific antagonistic subpopulations being stimulated in grassland as compared to arable land ß 2003 Federation of European Microbiological Societies Published by Elsevier B.V All rights reserved Keywords : Pseudomonas ; Bacterial diversity ; Soil; Polymerase chain reaction-denaturing gradient gel electrophoresis; Antifungal activity Introduction As a result of an increasing interest in environmentally friendly agricultural practices, it has become necessary to assess how di¡erent cropping regimes a¡ect microbial diversity in soil In fact, our knowledge on how plant communities and their management in£uence microbial communities in soil is still limited [1^3] even though exudation from roots is known to be a key factor in these interactions [4^8] One of the most important and best-studied bacterial taxa in soil is the genus Pseudomonas This genus includes several functional groups of environmental interest, such * Corresponding author Present address: Department of Microbial Ecology, Biological Center, Groningen University, Kerklaan 30, P.O Box 14, 9750 AA Haren, The Netherlands Tel.: +31 (50) 3632151; Fax : +31 (50) 3632154 E-mail address : j.d.van.elsas@biol.rug.nl (J.D van Elsas) as plant growth promoters [9], plant pathogens [10] and xenobiotic degraders [11] Moreover, Pseudomonas species can also play important roles as biological control agents against soil-borne plant pathogens Di¡erent mechanisms may be involved, such as the production of secondary metabolites (antibiotics, Fe-chelating siderophores), cellulolytic and chitinolytic activity, and the induction of systemic resistance against phytopathogens in the host plant [12^14] For example, antibiotic-producing Pseudomonas species have been isolated from soil that was naturally suppressive to di¡erent plant diseases, including take-all disease of wheat, black rot of tobacco and fusarium wilt [15^17] Root-associated £uorescent Pseudomonas spp producing the antibiotic 2,4-diacetyl phloroglucinol, the key component of the speci¢c suppression of the take-all disease agent, were shown to be enriched in numbers in take-all-suppressive soils [17,18] This suppression was lost when these Pseudomonas spp were eliminated Conversely, conducive soil regained its suppressiveness to take-all dis- 0168-6496 / 03 / $22.00 ß 2003 Federation of European Microbiological Societies Published by Elsevier B.V All rights reserved doi:10.1016/S0168-6496(03)00234-4 FEMSEC 1587 22-12-03 52 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 ease when antibiotic-producing Pseudomonas strains were introduced The realization that the soil micro£ora may be responsible for soil suppressiveness led to the idea that it should be possible to make a conducive soil suppressive by manipulating the soil microbial balance and diversity [19] The aim of this study was to gain a better understanding of the in£uence of di¡erent agricultural management regimes applied at an experimental ¢eld site in The Netherlands on the diversity of Pseudomonas spp in soil, as determined by both culture-dependent and culture-independent methods We based our study on previous work that assessed pseudomonads in environmental settings either by cultivation [20] or by cultivation-independent methods [21^24] In particular, a newly adapted polymerase chain reaction-denaturing gradient gel electrophoresis (PCR-DGGE) system was tested and used for the assessment of Pseudomonas diversity directly from soil Moreover, we studied the in£uence of the di¡erent agricultural regimes on the prevalence of Pseudomonas antagonistic (antifungal and chitinolytic) activity against the potato pathogen Rhizoctonia solani AG3 Materials and methods 2.1 Bacterial strains The strains and isolates used in this study and their origins are listed in Tables and All strains were stored at 380‡C in 20% glycerol 2.2 Fungal pathogen R solani AG3 (basidiomycete with a chitin/glucan-containing cell wall) was originally isolated from potato plants exhibiting symptoms of potato rot The culture was kept on potato dextrose agar (PDA; Oxoid, Hampshire, UK) at 4‡C and subcultured every second month 2.3 Field treatments, soil and soil sampling Soil samples were collected from a long-term ecological site at the Wildekamp ¢eld, located in Bennekom, The Netherlands The soil in this ¢eld is a loamy sand rich in organic matter (2.5%) with slightly acidic pH (5.5^ 6.5) The site has been under permanent grassland (treatment G) for approximately 50 years, but part of it was turned into agricultural land about 20 years ago (longterm arable land ^ A) The long-term arable land was divided into arable land under common agricultural rotation (A-R; including oats, maize, barley and potato) and arable land under continuous maize (A-M) The site also contained short-term arable land (GA), which consisted of plots turned from permanent grassland into arable land in 2000 This short-term arable land was either kept under maize monoculture (GA-M), or placed under the same rotation as in the A-R plots (GA-R) In the year 2000, oats were grown in the arable land under rotation, followed by maize in 2001 In the ¢rst year, samples were taken from triplicate (10U10-m) plots per treatment at four points in time (February, May, September and November) Two treatments, A-R and G, were studied In the second year, samples were taken at the end of the growing season, i.e in September, from ¢ve di¡erent treatments (A-R, A-M, GA-R, GA-M and G), in triplicate Sampling from all plots proceeded as follows About 100 (3^5-g) samples from the soil top layer (0^10 cm) per plot were taken randomly throughout that plot, using a sterilized auger (2 cm diameter) and mixed thoroughly in a plastic bag, to yield one composite sample per plot The soil samples were used for analysis within 24 h after sampling Rhizospheres of grass (G plots), oats and maize (A-R, GA-R, A-M and GA-M plots) were also sampled and processed as described previously [19] 2.4 Bacterial isolations, media and growth conditions For isolation of bacterial cells, 10 g of soil was suspended in 95 ml of 0.1% tetrasodium pyrophosphate (Na4 P2 O7 W12H2 O, Merck) containing 10 g common aquarium gravel (2^4 mm diameter), and shaken for 25 at 250 rpm One ml of this soil suspension was used to prepare serial 10-fold dilutions in 0.8% NaCl For the enumeration of Pseudomonas spp., 100 Wl from the 1032 and 1033 dilutions were used for plating on Gould’s S1 agar [20] Plates were incubated at 27‡C and enumerations of total as well as £uorescent (under UV light) Pseudomonas colonies were done after 48 h of incubation Results are presented as the log numbers of CFU per g (dry weight) soil 2.5 Screening of Pseudomonas isolates for antagonistic activity toward R solani AG3 Antagonistic activity of 500 Pseudomonas isolates against R solani AG3 was tested by dual culturing on 0.1UPDA Using sterile toothpicks, four Pseudomonas sp isolates were placed approximately mm from the edges of the 0.1UPDA plates and a 6-mm agar disk containing grown fungal mycelium was placed in the center of each plate The plates were analyzed after days of incubation at 25‡C, by measuring the extension of mycelia and zones of inhibition (haloes without mycelial growth) around the isolates A strain with known antagonistic activity (P syringae AM20) served as the positive control (halo s cm), whereas a strain without that activity (Agrobacterium radiobacter IPO-At2, no halo) was the negative control Isolates that formed halo zones over mm were accepted as antagonists to R solani AG3 The scoring was performed as follows : halo v cm (++); halo 1^3 cm (+), and halo cm or no halo (3) FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 2.6 Screening of Pseudomonas isolates for chitinolytic activity Chitinase activity was measured by modi¢cation of the Schales procedure with colloidal chitin as an assay substrate [25] Colonies were screened for chitinolytic activity by plating on two media: CY (chitin yeast extract) medium containing 2.5 g NaCl, 0.02 g MgSO4 W7H2 O, 0.02 g CaCl2 W2H2 O, 0.05 g yeast extract, 50 mg Delvocid (contains 50% natamycin; DSM, Delft, The Netherlands), 40 ml chitin stock (2.5%) and 10 g agar, and CTSB (chitin^ tryptone soy broth) medium containing 2.5 g NaCl, 1.5 g TSB (tryptone soy broth), 50 mg Delvocid and 40 ml chitin stock (2.5%) plus 10 g agar Clearance haloes indicating the enzymatic degradation of chitin were measured after days of incubation at 27‡C 2.7 Soil DNA extraction For soil DNA extraction, a cell homogenizer (Bead beater; B Braun, Melsungen, Germany) was used prior to extraction with the MoBio Ultraclean soil DNA extraction kit (Biozym TC, Landgraaf, The Netherlands) Glass beads (1.5 g, 0.11 mm diameter) were added to 0.25-g soil samples in 0.5 ml bu¡er and the mixture was bead-beaten four times for 90 s each time After bead beating, DNA extraction proceeded in accordance with the protocol furnished by the manufacturer (MoBio extraction protocol) A ¢nal DNA puri¢cation step was performed using the Wizard DNA cleanup kit (Promega, Leiden, The Netherlands) DNA purity and quality were assessed after electrophoresis of subsamples in 0.8% agarose gels in 0.5UTBE bu¡er [26] DNA quality was checked by electrophoresis in 0.8% agarose gels and DNA was quanti¢ed by comparison to a standard (1-kb ladder, Invitrogen, Cat no 15615-024) DNA yields were, on average, about 25 Wg g31 soil 2.8 Pure culture DNA extraction DNA extraction from pure cultures (Tables and 2) grown in 0.1UTSB was commenced by harvesting cells from 1.5-ml overnight cultures into ml of 0.8% NaCl Lysis was performed by bead beating (1 g of 0.11 mm diameter beads in ml, four times 90 s) The lysate was extracted with phenol^Tris^HCl (pH 8.0) and chloroform/ isoamyl alcohol (24:1) followed by precipitation with 96% ethanol in the presence of M NaCl [26] The DNA pellets were washed with 70% ethanol, vacuum-dried and dissolved in 50 Wl sterile Milli-Q water 2.9 PCR ampli¢cation of 16S rRNA gene fragments A semi-nested system was used for ampli¢cation of the V6/V7 region of the 16S ribosomal RNA gene The ¢rst 53 PCR reaction was performed by applying the Pseudomonas-speci¢c primers PsF and PsR (Escherichia coli positions 298 and 1258, respectively ; Table 3) described by Widmer et al [21] PCR was performed in an MJ Research PT-200 thermal cycler in 50-Wl reaction volumes containing 0.2 WM of each primer, 3.75 mM MgCl2 (Perkin-Elmer, Nieuwersluis, The Netherlands), 200 WM of each dNTP (Boehringer, Almere, The Netherlands) and 0.25 Wg T4 gene 32 protein (Boehringer, Mannheim, Germany) using U AmpliTaq Sto¡el fragment in 1USto¡el bu¡er The thermal cycling was as follows: denaturation at 94‡C for min, followed by 35 cycles of 94‡C for min, 66 or 68‡C (for soil DNA and pure culture, respectively) for min, and 72‡C for min, and a ¢nal extension step at 72‡C for 10 The PCR products (expected sizes 760 bp) were analyzed by running 5-Wl aliquots of the reaction mixtures in 1.2% agarose gels The remaining 45 Wl of the PCR volumes were precipitated with 1/10 volume of M NaCl and ice-cold 96% ethanol for 15 at 320‡C After centrifugation, washing with 70% ethanol and airdrying, the pellets were resuspended in 100 Wl sterile MilliQ water The precipitated PCR products served as templates for a second PCR with conserved bacterial forward primer F968 with attached GC clamp (F968-GC [27]), and the Pseudomonas-speci¢c primer PsR The program used for the second PCR was as follows : initial denaturation at 94‡C for min, one cycle of at 94‡C, at 60‡C, and at 72‡C, followed by 10 times the same cycle with every subsequent one using a 0.5‡C lower annealing temperature (until 55‡C), 20 cycles of 94‡C (1 min), 55‡C (1 min) and 72‡C (2 min), and ¢nal extension at 72‡C (10 min) The PCR products (expected sizes 290 bp) were ¢rst analyzed by running 5^10-Wl aliquots of the reaction mixtures in 1.2% agarose gels and secondly by running 15 Wl on 45%^65% DGGE gels For comparison, selected soil-derived DNA samples and Pseudomonas pure cultures were also analyzed by applying a PCR-DGGE system recently described by Gyam¢ et al [23] 2.10 DGGE analysis DGGE was performed using 6% polyacrylamide gels (ratio of acrylamide to bis-acrylamide 37:1) with a gradient of 45^65% denaturant (100% denaturant was de¢ned as M urea plus 40% formamide) The gels were electrophoresed at 60‡C at 100 V for 15 h in a PhorU2 apparatus (Ingeny, Goes, The Netherlands) and stained with SYBR gold (Molecular Probes, Leiden, The Netherlands) For analysis of the molecular community pro¢les, the Molecular Analyst Fingerprinting software (version 1.61, BioRad, Veenendaal, The Netherlands) was used Clustering was determined by the unweighted pair group with mathematical averages (UPGMA) method FEMSEC 1587 22-12-03 54 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 Table Bacterial strains tested with Pseudomonas-speci¢c primers PsF and PsR Strains and soil isolates PCR product with PsF and PsR primers Pseudomonas strains from collection at PRI P aeruginosa PA1 P aeruginosa PAO5 P aeruginosa PU21 P corrugata PD704 P corrugata MS1 P syringae Ps6 P cichorii Pc4 P chlororaphis Pc8 P glycinea Pg1 P putida Pp1 P putida Pp2 P stutzeri Ps1 P £uorescens Pf2 Pseudomonas sp PRI-PCA1 Pseudomonas sp PRI-DAPG1 Pseudomonas soil isolates from this work Presented in Table Non-Pseudomonas strains Burkholderia cepacia LMG1222 and 18941 Ralstonia solanacearum IPO267 and IPO1609 Acinetobacter calcoaceticus DSM586 Agrobacterium radiobacter IPO-At2 Agrobacterium tumefaciens Gmi9023 Alcaligenes eutrophus IPO2 Alcaligenes faecalis a1501R Bacillus sp Sal1 Bacillus subtilis Bs2 Enterobacter agglomerans PS23S Erwinia amylovora P4-17-8 Flavobacterium sp ATCC39723 Klebsiella aerogenes 418 Listeria innocua ALM105 Mycobacterium chubuensis ATCC33609 Xanthomonas maltophilia PD1484 Streptomyces griseus ISP5236 Staphylococcus aureus 6538P Serratia plymuthica PRI-2c Serratia liquefaciens PRI-K1.5 a + + + + + + + + + + + + + + + + 3 3 3 3 3 3 3 3 3 3a Positive at low annealing temperature 2.11 Cloning and sequencing of amplicons For direct cloning of the PCR products obtained with primers F968 and PsR, the pGEM-T Easy vector system (Promega, Leiden, The Netherlands) was used Prior to cloning, PCR products were puri¢ed with High-Pure PCR product puri¢cation kit (Boehringer Mannheim, Almere, The Netherlands) and cloned into the pGEM-T vector according to the manufacturer’s instructions Plasmid extraction was performed using the Wizard Plus SV miniprep DNA puri¢cation kit (Promega Benelux) Clones with the correct insert (as judged by size) were subjected to (one strand) sequencing with universal M13 primers using the services of BaseClear, The Netherlands (http:// www.baseclear.nl) For sequencing of DGGE bands, bands were ¢rst excised from the gel, re-ampli¢ed and analyzed on DGGE for purity and correct migration behavior, after which PCR products were sequenced Two clones per band were sequenced, which gave identical sequences throughout 2.12 PCR with primers for genes encoding the synthesis of antibiotics Three sets of primers received from Dr J Raaijmakers (Phytopathology Department, Wageningen University, The Netherlands) were used for the detection of genes encoding the production of pyrrolnitrin (PRN) [28], 2,4diacetylphloroglucinol (DAPG) [18] and phenazine-carboxylic acid (PCA) [18] in isolates Primer sequences, annealing temperatures and references are listed in Table Approximately 5^10 ng of genomic DNA was used per strain The PCR products were analyzed by running 25- FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 55 Table Bacterial isolates obtained from Gould’s S1 medium identi¢ed by partial 16S rRNA gene sequence Origin (and numbers) of isolates a Group I G (2), A-R (4), A-M (1) AM (1) A-R (2) G (1), GA-M (31), A-M (1) G (2), GA-M (2), GA-R (2), A-R (6) G (1), GA-R (1) Group II GA-M (4), GA-R G (1), GA-M (2), Group III G (5), GA-M (3), G (5), GA-M (3), R (1) G (5), GA-M (3), (4) G (1), A-M (3) GA-R (3), A-M (1) Closest hits to sequences in NCBI database % similarity Ant.b Chit.c Antbd PCRe P syringae AF511511 (P orientalis AF064457) P syringae pv mori AB001446 P syringae pv savastori AB021402 (P chlororaphis AJ492826) P veronii strain CA-4 AY081814 (P brennerii AF268968) P rhodesiae AY043360 (P lini AY035996) P tolaasii NCPPB 741 AF320992 (Pseudomonas sp SMCC B0361 AF500621) 100 99 99 98^99 99^100 100 + 3 + + 3 3 + + + + + + + P libaniensis AF057645 (P marginalis AF364098) 100 P marginalis strain NZCX27 AF364098 (P syringae AF511511) 98^100 + + + + + GA-R (2), A-R (4) Pseudomonas sp NZ031 AY014807 (P tolaasii AF094750) GA-R (2), A-M (2), A- Pseudomonas sp Sau7 AF511510 (P putida AB016428) 99^100 100 + ++ + GA-R (6), A-M (2) 98^100 ++ + DAPG* + PRN/ + PCA PRN + + G (2), A-R (1) G (1), GA-R (2), A-R (2) G (4), GA-M (3), GA-R (3) Pseudomonas sp E102 AF451270 (P £uorescens AJ308303 ; P gessardii AF074384) Pseudomonas sp NZ065 AY014815 (Pseudomonas sp NZ124 AY014829) Pseudomonas sp NZ081 AF388206 (P migulae AY047218) Pseudomonas sp FSL-D1-045 AF20513 (P corrugata AF348508) 100 + 99 99^100 + + + PRN/ PCA + + G (2), GA-R (1) GA-R (3), A-R (6) Group IV G (6), GA-M (4), GA-R (2), A-R (1) Pseudomonas sp A-07-10 AY136523 (P £uorescens AY014829) Pseudomonas sp 9-1 AF52165 (P veronii AY144583) 99 100 + + + P £uorescens LCSA0TU2 AF506042 (P jessenii AF068259) 99^100 ++ + G (1), A-M (2), A-R (3) P £uorescens ATCC13525 AJ308308 (P £uorescens AJ308303) 99^100 + + G (1), GA-M (2), A-M (2) Group V G(3), A-R (2) Group VI G(1), GA-R (3), GA-M (2) A-R (3), A-M (2) P £uorescens ATCC49642 AF094732 (P £uorescens AF094732) 99^100 + + P migulae AY047218 (P putida AF094743) + + + 3 3 + + 98^99 P orientalis AF064457 (P syringae AF511511) 98^99 P chlororaphis LMG 5004T Z76657 (P chlororaphis AJ308301) 98^100 + + PRN/ PCA PRN/ PCA + + a Codes indicate from which treatment the isolates originate (see Section 2.3) Grouping is in accordance with groups de¢ned in Fig Ant : presence of antagonistic activity toward R solani AG3 ++, halo s cm (diameter) ; +, halo 1^3 cm ; 3, halo cm c Chit.: presence or absence of chitinolytic activity d Antb: PCR-based detection of antibiotic synthesis genes *DAPG was detected in only two isolates (G2, G3) e PCR: Positive PCR signal obtained with Pseudomonas-speci¢c primers PsF and PsR b Wl aliquots of the reaction mixtures in 1.2% agarose gels and by hybridization under high-stringency conditions with speci¢c probes prepared by PCR on the basis of the reference strains Burkholderia cepacia LMG 1222 PRN (for PRN), Pseudomonas spp PRI-DAPG1 (for DAPG) and Pseudomonas sp PRI-PCA1 (for PCA) Probe preparation, hybridization and signal detection were performed as described by the PCR DIG Probe Synthesis Kit (Roche, Cat no 1636090) 2.13 DNA sequence analysis The partial 16S rDNA sequences (length about 290 bp) obtained from 175 isolates and 25 DGGE bands were compared against those available in the database using BLAST-N provided by the Plant Research International server (http://lx10003.plant.dlo.nl), NCBI (http://www ncbi.nlm.nih.gov) or the Ribosomal Database Project (http://rdp.cme.msu.edu/cgis/seq_match.cgi) The sequences were further aligned and clustered using ClustalW provided by the Institut Pasteur (http://bioweb pasteur.fr) A phylogenetic tree was constructed from these aligned sequences by neighbor joining, using Treecon (version 1.3b, Yves van de Peer, Ghent, Belgium) Bootstrapping was performed using the bootstrap modus of the program and values above 50% are reported Sequences obtained were deposited in GenBank under numbers AY365075 to AY365106 2.14 Statistics For each treatment, samples were obtained from three replicate plots per treatment in the completely randomized block design used in the ¢eld Bacterial counts (cfu g31 FEMSEC 1587 22-12-03 56 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 Table Primers used in this study Primer Sequence Annealing temperature (‡C) PsR PsF F968GC GGTCTGAGAGGATGATCAGT TTAGCTCCACCTCGCGGC CGCCCGGGGCGCGCCCCGGGCGGGGCGGGGGCACG GGGGGAACGCGAAGAACCTTAC 66 66 66^55 PRND1 GGGGCGGGCCGTGGTGATGGA PRND2 YCCCGCSGCCTGYCTGGTCTG Phl2a GAGGACGTCGAAGACCACCA Phl2b ACCGCAGCATCGTGTATGAG PCA1a TGCCAAGCCTCGCTCCAAC PCA2b CCGCGTTGTTCCTCGTTCAT Detection of Reference 960 960 290 Pseudomonas spp Pseudomonas spp Universal bacterial primer [21] [21] [36] 67 786 Pyrrolnitrin production locus (PRN) [28] 67 745 2,4-Diacetylphloroglucinol production locus (DAPG) [18] 67 1150 Phenazine 2-carboxylic acid production locus (PCA) [18] soil) were analyzed after logarithmic transformation The data were statistically analyzed using Genstat 5, Release 4.2 (Rothamsted Exp Stat., UK) Data were considered to be signi¢cantly di¡erent at P 0.05 Results 3.1 Number of total and £uorescent Pseudomonas spp on Gould’s S1 agar The numbers of total and £uorescent Pseudomonas spp PCR product size (bp) recovered on Gould’s S1 agar in the ¢rst sampling year are presented in Table 4, A The total counts from bulk soil did not show large variations over the season or between the treatments (arable land versus grassland), ranging from log 5.5 to log 6.1 cfu g31 of dry soil On the other hand, the numbers of £uorescent pseudomonads increased signi¢cantly over the growing season in all treatments (P 0.05), ranging from log 4.1 (A-R, bulk) to log 5.5 (G, rhizosphere) cfu g31 of dry soil Both counts were generally higher, albeit not signi¢cantly, in the bulk soil of the grassland than in that of the arable land In the November samples, higher numbers of total and £uores- Table Pseudomonas populations (log CFU g31 of dry soil) in bulk and rhizosphere soils under di¡erent management regimes Soil treatmenta A: First sampling year A-R, Bulk A-R, Rhizb G, Bulk G, Rhiz B: Second sampling year G, Bulk G, Rhiz GA-R, Bulk GA-R, Rhizc GA-M, Bulk GA-M, Rhizc A-R, Bulk A-R, Rhizc A-M, Bulk A-M, Rhizc February May September a a b a b a 5.6 (0.2) 5.5 (0.2) 4.1 (0.1) 5.5 (0.3) 4.3 (0.1) 6.1 (0.1) 6.0 (0.1) 5.8 (0.3) 5.6 (0.1) 4.6 (0.2) 4.8 (0.2) 5.9 (0.1) 5.9 (0.3) 4.7 (0.2) 4.8 (0.2) 5.4 6.1 6.1 6.3 6.0 6.2 5.8 5.9 5.9 5.9 5.1 5.3 5.2 5.4 5.0 5.7 4.6 5.2 4.8 5.3 4.1 4.7 4.2 4.4 (0.3) (0.1) (0.2) (0.2) (0.2) (0.3) (0.1) (0.3) (0.2) (0.2) November b (0.2) (0.3) (0.2) (0.1) 5.0 5.5 5.4 5.8 (0.4) (0.2) (0.1) (0.2) (0.1) (0.5) (0.4) (0.2) (0.1) (0.2) (0.3) (0.2) (0.3) (0.3) a = number of total cfu; b = number of £uorescent Pseudomonas cfu The numbers in parentheses represent the standard deviation Total cfu: G, GA-R, GA-M s A-R, A-M (P 0.05) Fluorescent cfu: Bulk: G, GA s A (P 0.05) Rhiz: G, GA s A (not signi¢cant) a For soil treatments see see Section 2.3 b Rhizosphere of oat c Rhizosphere of maize FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 cent Pseudomonas spp were detected in the rhizospheres than in the corresponding bulk soils However, this was only signi¢cant (P 0.05) in the A samples The results obtained during the second sampling year (one sampling) are presented in Table 4, B Again, di¡erences in Pseudomonas numbers between the treatments were observed Signi¢cantly higher numbers of total and £uorescent pseudomonads (P 0.05) were measured in the permanent grassland (log 6.0 and log 5.0 cfu g31 of dry soil) and short-term arable land (log 5.8 and log 4.6 cfu g31 of dry soil) than in the long-term arable land (log 5.1 and log 4.1 cfu g31 of dry soil) In all treatments, rhizosphere e¡ects were observed In particular, the numbers of £uorescent pseudomonads were signi¢cantly higher in the rhizosphere than in the bulk soil samples (P 0.05) 3.2 Screening for Pseudomonas isolates with antagonistic activity toward R solani AG3, chitinolytic and/or antibiotic production capacity In total, 500 Pseudomonas isolates originating from ¢ve di¡erent treatments (G, A-R, A-M, GA-M, GA-R; 100 isolates per treatment) were screened in an in vitro assay for their ability to suppress R solani AG3 Approximately 17.4% (87 of 500) of all isolates showed this antagonistic capacity The highest isolation frequencies were found in the short-term arable land, i.e GA-M (25 ỵ 3%) and GA-R (22 ỵ 2%), which was followed by permanent grassland (19 þ 4%) and long-term arable land, i.e A-M (15 þ 2%) and A-R (6 ỵ 1%) These isolation frequencies were statistically similar between the samples derived from grassland (G, GA), whereas those from (long-term) arable land were lower, albeit only signi¢cantly for A-R (P 0.05) All 87 isolates that showed antagonistic activity were identi¢ed by partial 16S rRNA gene sequencing (Table 2) The most frequently found antagonists had 16S rRNA gene sequences that were highly (98^100% similarity) related to sequences of Pseudomonas sp SaU7, P £uorescens LCSA0TU2, P migulae, Pseudomonas sp E102, P rhodesiae, P syringae and P libaniensis (Table 2) To assess their potential action on the R solani AG3 cell wall by chitinolysis, all 500 Pseudomonas isolates were also screened for chitinolytic activity Chitinolysis was detected in 79 strains, or 15.8% of the total Several, but not all, of the chitinolytic isolates also had shown anti-AG3 activity Chitinolytic isolates were most prevalent in shortterm arable land, i.e GA-M (22 ỵ 1%) and GA-R (20 ỵ 3%), and permanent grassland (17 ỵ 2%) In the long-term arable land, lower percentages (A-M: 14 ỵ 4%; A-R : ỵ 2%) of the tested isolates were chitinolytic Thus, the isolation frequencies of chitinolytic strains were signi¢cantly higher in the grassland-derived samples (G, GA) than in those from arable land (P 0.05), except for those from A-M All 79 chitinolytic Pseudomonas spp were identi¢ed by partial sequencing of their 16S rRNA genes (Table 2) The data showed that the isolates were distrib- 57 uted, at high (98^100%) similarity values, among a range of closest database hits, notably with (numbers of isolates in parentheses): Pseudomonas sp SaU7 (13), Pseudomonas sp E102 (15), Pseudomonas sp FSL-D1-045 (10), Pseudomonas sp A-07-10 (3), P £uorescens LCSA0TU2 (9), P £uorescens ATCC13525 (3), P rhodesiae (7), P libaniensis (10), P marginalis (7) and P migulae (2) A selection (210) of the 500 Pseudomonas isolates was screened for evidence of the presence of the PRN, DAPG and PCA synthetic operons by PCR followed by hybridization with the appropriate probes About 18% (38/210) of the tested isolates showed positive signals with the PRN primers, 1% (2/210) with the DAPG primers, and 12% (25/ 210) with the PCA primers (Table 2) Some isolates of the latter group also showed a positive signal with the PRN detection system 3.3 Pseudomonas-speci¢c PCR-DGGE analysis 3.3.1 Validation of the PCR system Speci¢c PCR ampli¢cation of Pseudomonas 16S rDNA genes directly from soil samples was performed by applying the highly selective PCR system described by Widmer et al [21] First, the speci¢city of the PCR system was rechecked with a range of Pseudomonas and non-Pseudomonas strains (Table 1) All Pseudomonas strains used produced a PCR product with this system Among the 22 non-Pseudomonas strains tested, including representatives of Burkholderia sp., Ralstonia solanacearum, Agrobacterium sp., Alcaligenes sp., Bacillus sp and Xanthomonas sp., only one non-Pseudomonas strain (identi¢ed as Serratia plymuthica) showed a positive PCR signal This aspeci¢c PCR ampli¢cation was avoided when the annealing temperature was increased, from 66 to 68‡C The Pseudomonas-speci¢c PCR products were used as targets for a second, semi-nested, PCR performed with the forward bacterial primer F968 (with a GC-clamp) and the Pseudomonas-speci¢c primer PsR The ampli¢ed 290-bp products were successfully separated on 45^65% DGGE gels, yielding separated bands from di¡erent selected Pseudomonas strains (Fig 1A, see below) The detection limit of the Pseudomonas-speci¢c PCR, assessed by performing PCR with 10-fold dilutions of Pseudomonas sp DNA (containing from ng to fg DNA) was 100 fg DNA, estimated to represent 15^20 genome equivalents per PCR Inhibition of the PCR reaction by soil DNA was not observed, as evidenced by performing PCR with a mixed template containing 10 ng of soil DNA and 100 fg of Pseudomonas sp DNA (not shown) 3.3.2 Evaluation of two Pseudomonas-speci¢c PCR-DGGE systems We compared the PCR-DGGE system used in this study, which was based on ampli¢cation of the highly variable V6 and V7 regions of 16S rDNA, with the one recently published by Gyam¢ et al based on the V1^V3 16S FEMSEC 1587 22-12-03 58 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 Fig Evaluation of two Pseudomonas-speci¢c PCR-DGGE systems using as targets pure culture and soil DNA A: Pure cultures Lanes 1^5: ampli¢ed with the primers located in the V6/V7 region of the 16S rRNA gene; 6^10 : ampli¢ed with the primers located in the V1^V3 region of the 16S rRNA gene Lanes 1, 6: Pseudomonas sp E102; 2, 7: P libaniensis; 3, 8: Pseudomonas sp SaU7; 4, 9: P £uorescens LCSA0UT2 ; 5, 10: Pseudomonas sp Fa8 B: Five soil DNA samples Lanes 1^5: soil DNA ampli¢ed with primers located in the V6/V7 region of the 16S rRNA gene; 6^10: same samples ampli¢ed with primers located in the V1^V3 region of the 16S rRNA gene M: marker (from top to bottom): amplicons of Enterobacter cloacae BE1, Listeria innocua ALM105, Rhizobium leguminosarum bv trifolii R62, Arthrobacter sp., Burkholderia cepacia P2 rDNA regions [23] The results of the DGGE analysis based on ampli¢cation of both regions obtained with pure cultures as well as with soil DNA by both systems are presented in Fig 1A,B First, when pure culture DNA of di¡erent pseudomonads was used, a better separation by DGGE was observed with the system amplifying the V6/V7 regions of the 16S rDNA than with that based on the V1^V3 region (Fig 1A) In fact, some Pseudomonas isolates, which were di¡erentiated by DGGE of the V6/V7 region products, were not distinguished by DGGE of those of the V1^V3 region Speci¢cally, isolates identi¢ed as P libaniensis and Pseudomonas sp E102 and those related to Pseudomonas sp SaU7 and P £uorescens LCSA0TU2 gave bands at the FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 59 same position when analyzed by V1^V3 region-based DGGE, whereas they were separated by the DGGE system based on the V6/V7 regions (Fig 1A) In addition, the DGGE patterns obtained from soil DNA with the primers amplifying the V6/V7 regions showed higher numbers of bands than those obtained with the primers amplifying the V1^V3 regions (Fig 1B) The maximum number of bands detected by DGGE of the V1^V3 regions was ¢ve, while, on the basis of the V6/V7 regions, maximally eight bands were detected in the pro¢les obtained from the same soil samples (Fig 1B) 3.3.3 PCR-DGGE of soil DNA The numbers of dominant DGGE bands obtained from Fig Dendrogram constructed with UPGMA representing the similarity between Pseudomonas-speci¢c PCR-DGGE patterns obtained from soil samples under di¡erent management regimes over one growing season (2000) Codes: F: February; M: May; S: September; N : November ; A: arable land under rotation; G: permanent grassland; 1, 2, 3: replicates 1, and Fig DGGE analysis of PCR products from DNA extracted from soil samples under di¡erent treatments (¢rst year sampling) : A: long-term arable land under rotation; B: permanent grassland ; F: February; M: May; S: September ; N: November For every sampling three replicate plots are presented (1, and 3) m : marker (from top to bottom, amplicons of Enterobacter cloacae BE1, Listeria innocua ALM105, Rhizobium leguminosarum bv trifolii R62, Arthrobacter sp., Burkholderia cepacia P2) DGGE bands (closest hits in database): a: P rhodesiae (99%, AY043360); b: P £uorescens (100%, AY092072) ; c: P migulae (100%, AY047218); d: P syringae (99%, AF511511); e: P orientalis (99%, AF064457); f: Pseudomonas sp E102 (100%, AF451270); g: Pseudomonas sp SaU7 (100%, AF511510); h: P rhodesiae (98%, AY043360); i: P putida (97%, AF094743); j : P tolaasii (98%, AF320986) soil samples of the ¢rst sampling year varied from three to nine, and depended mainly on soil treatment (Fig 2A,B) Based on the DGGE banding patterns, clear di¡erences were observed between the treatments, notably permanent grassland vs arable land Per treatment, very little variation was found between the DGGE pro¢les obtained from the three replicate plots However, one plot of the permanent grassland showed slightly di¡erent DGGE patterns in comparison with the other two plots of that treatment (Fig 2B) Clustering of the DGGE patterns using UPGMA con¢rmed the clear separation of all patterns in two main clusters (Fig 3) All pro¢les from the permanent grassland formed one cluster, which grouped together with the pro¢les obtained from arable land of November, at 40% similarity A second main cluster was formed by all other arable land samples, at 21% similarity to the former cluster Whereas there was no clear e¡ect of sampling time in the grassland samples, those from arable land showed a subclustering in accordance with sampling date In order to identify the most dominant Pseudomonas species that made up the DGGE patterns, selected dominant DGGE bands were excised, re-ampli¢ed and sequenced The single strong band apparent in the samples from the arable land of February and May 2000 (Fig 2A, band a) was a⁄liated with a 16S rRNA gene sequence of P rhodesiae (99%, accession number AY043360) The bands detected in the arable land later in the same sampling year, i.e in September and November 2000 (Fig 2A, bands b^d), were a⁄liated with sequences of P £uorescens (100%, AY092072), P migulae (100%, AY047218) and P syringae (99%, AF511511) The numbers of DGGE bands detected in the permanent grassland were higher than FEMSEC 1587 22-12-03 60 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 those in the arable land in every sample during the season (Fig 2B), and the banding patterns were relatively stable The most dominant DGGE bands (Fig 2B, bands e^j) were a⁄liated with sequences from P orientalis (99%, AF064457), Pseudomonas sp E102 (100%, AF451270), Pseudomonas sp SaU7 (100%, AF5111510), P rhodesiae (98%, AY043360), P putida (97%, AF094743) and P tolaasii (91%, AF320986) In the second sampling year, the e¡ects of ¢ve treatments, i.e G, A-R, A-M, GA-R and GA-M, were analyzed The DGGE patterns were consistent between the replicates per treatment, based on both the intensities and numbers of bands The numbers of bands ranged from ¢ve to 12, depending on the treatment (Fig 4) In the long-term arable land, only ¢ve dominant bands were detected, while in the short-term arable land and the permanent grassland, the numbers of dominant DGGE bands were nine and 12, respectively Analysis of the DGGE patterns by UPGMA showed a clear clustering along treatment, with 40% similarity between all pro¢les (Fig 5) The DGGE patterns obtained from the short-term arable land were most similar to those obtained from the permanent grassland (56% similarity), whereas those from arable land under maize monoculture clustered with these at 50% similarity Three DGGE bands present in all treatments (Fig 4, bands b, d and e) were homologous to sequences of Pseudomonas sp E102 (100%, AF451270), Pseudomonas £uorescens LCSA0TU2 (100%, AF506042) and Pseudomonas sp SaU7 (100%, AF5111510) One other band, not de- Fig PCR-DGGE banding patterns representing the dominant Pseudomonas populations in soil samples under di¡erent treatments during the second sampling year For soil treatments see Section 2.3 Two or three replicates per treatment, numbered 1, and 3, are shown Where only two are shown, the third replicate was equal M: Marker (from top to bottom, amplicons of Enterobacter cloacae BE1, Listeria innocua ALM105, Rhizobium leguminosarum bv trifolii R62, Arthrobacter sp., Burkholderia cepacia P2) DGGE bands (closest hits in database shown): a: P orientalis (99%, AF064457); b: Pseudomonas sp E102 (100%, AF451270); c: P libaniensis (100%, AF057645); d: Pseudomonas sp SaU7 (100%, AF511510); e: P £uorescens LCSA0TU2 (100%, AF506042); f: P rhodesiae (98%, AY043360); g: P tolaasii (98%, AF320986) Fig Dendrogram constructed with UPGMA representing the similarity between Pseudomonas-speci¢c PCR-DGGE patterns obtained from soil samples under di¡erent management regimes (as shown in Section 2.3) Numbers 1, and indicate replicates tected in the ¢rst sampling year, was detected in the second year in the samples from A-M This band (Fig 4, band c) was found to be related to P libaniensis (100%, AF057645) In the second sampling year, no signi¢cant change in the DGGE patterns was observed from the ¢rst year for the permanent grassland The most dominant DGGE bands making up these patterns (Fig 4, bands a, b, e^g) had sequences a⁄liated with P orientalis (99%, AF064457), Pseudomonas sp E102 (100%, AF451270), P £uorescens LCSA0TU2 (100%, AF506042), P rhodesiae (98%, AY043360) and P tolaasii (99%, AF320992) 3.4 Phylogenetic and DGGE-based analysis of selected Pseudomonas isolates About 165 Pseudomonas isolates selected among the 500 (33 per treatment ; ¢ve treatments) obtained from Gould’s S1 agar were analyzed by DGGE and partial sequencing of their 16S rRNA genes All sequences showed between 96 and 100% similarity with 16S rRNA gene sequences of Pseudomonas spp from the database The sequences were pre-grouped on the basis of their a⁄liations, as well as on internal alignments Then, a selection (minimally one sequence per pre-group) was used to generate a phylogenetic tree This analysis showed that the sequences clustered into six di¡erent groups, with very high levels of relatedness between them (Fig 6) The ¢rst group (I) included sequences that showed a⁄liation to the highly related species P veronii, P syringae, P rhodesiae and P tolaasii The second group (II) included all sequences that were a⁄liated with P marginalis and P libaniensis Group III was the largest one, including most of the sequences a⁄liated with organisms denoted Pseudomonas sp (NZ031, Sau7, E102, NZ065, NZ081 and Fa2) All isolates identi¢ed as P £uorescens formed group IV Groups V and VI were formed by sequences a⁄liated with organisms denoted P migulae and P orientalis, respectively All isolates of the six sequence groups produced bands on DGGE that migrated to a limited number of di¡erent positions The isolates generally produced single bands, with the exception of two isolates identi¢ed as P tolaasii (100%, AF057645) and P syringae (100%, AF511511), FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 61 detected, i.e P migulae (100%, AY047218), P £uorescens (99%, AF094732) and P syringae (100%, AF511511) In the second sampling year in the same plots (maize as the crop), the most dominant types were related to Pseudomonas sp E102 (99%, AF541270), P £uorescens (99%, AJ308308), and P libaniensis (100%, AF057645) Several types were detected in every treatment, but with large differences in the isolation frequencies (Table 2) For example, 50% of the isolates related to P £uorescens LCSA0TU2 (99^100%, AF506042) originated from the permanent grassland, 41.6% from the short-term arable land and 8.4% from the arable land One particular isolate, related to Pseudomonas sp E102 (98^100%, AF451270), which possessed strong antifungal and chitinolytic activity and produced PRN, was found mainly in the permanent grassland (33%) or short-term arable land (GA-M 20%; GA-R 33%) and less in the arable land (13%) Conversely, 54% of isolates identi¢ed as P rhodesiae (99^100%, AY043360) originated from the arable land under rotation and only 15% from the permanent grassland Fig Neighbor-joining phylogenetic tree based on partial 16S rDNA sequences from soil-derived Pseudomonas isolates In parentheses, the closest hit from the database is shown ( s 97^100% similarity) Representatives of each group of sequences shown here: group I (4 representatives), group II (3), group III (8), group IV (3), group V (1) and group VI (2) The sequence of Serratia plymuthica was used as an outgroup A bootstrap analysis was performed with 100 repetitions and values greater than 50% are indicated Codes of the isolates correspond with the codes of Table which each produced two DGGE bands Speci¢cally, about seven migration positions were found There was no absolute match between sequence group and migration position in DGGE, as fragments representing some groups, e.g V and VI, migrated to the same position, whereas within other groups, fragments were at di¡erent banding positions (e.g groups I, three positions, and III, two positions) Among the most dominant Pseudomonas strains detected in this study was a group of isolates related to Pseudomonas sp (NZ031, E102, FSL-D1, SaU7, NZ065, NZ081) Except for one (Pseudomonas sp Sau7), all of these isolates produced DGGE bands at the same position (corresponding to band f in Fig 2B) Other isolates, related to P £uorescens strains ATCC13525, ATCC17556, ATCC49642 and LCSA0TU2, also produced single DGGE bands at a di¡erent migration position All other isolates, related to P rhodesiae, P tolaasii, P orientalis and P libaniensis, produced di¡erent single DGGE bands Some isolates apparently became dominant as a result of the di¡erent soil treatments For example, during the time when no crop was present in the arable land under rotation, only one dominant Pseudomonas type (related to P rhodesiae, 99%, AY043360) was detected Later, when oats were grown in the same plots, three more types were Discussion Due to the wide distribution of di¡erent Pseudomonas species in the environment, and as a result of the relatively easy cultivation of the strains, the genus Pseudomonas is among the best-studied bacterial groups worldwide In our study, we selected Gould’s S1 as the selective medium for studying culturable Pseudomonas populations, as several previous studies [20,29] had proved this medium to be most selective, least susceptible to overgrowth by nonpseudomonads and most reproducible, allowing the growth of a broad diversity of Pseudomonas spp Using this medium, di¡erences in the numbers of total and £uorescent Pseudomonas cfu were found between the di¡erent soil treatments These di¡erences had most likely been induced by the di¡erences in the soil management regimes, although a range of other factors may also have a¡ected them Most strikingly, grassland and arable land recently derived from grassland appeared to maintain a larger population density of pseudomonads than long-term arable land However, the fact that the numbers of pseudomonads in the short-term arable land were similar to those in the permanent grassland may be due to remnants of grass roots that were still present in the short-term arable land A similar observation was recently made for Bacillus populations studied in the same treatments [30] More importantly, in the current study, the treatments were found to have di¡erent e¡ects with respect to the frequencies of isolation of Pseudomonas strains with R solani AG3 antagonistic properties Hence, the abundances of the antiAG3 types, within Pseudomonas populations of varying sizes per agricultural regime, were a¡ected by treatment, clearly indicating that agricultural management in£uenced both the total and the speci¢c (antagonistic) pseudomo- FEMSEC 1587 22-12-03 62 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 nads Most of the isolates with either direct (unspeci¢ed) antagonistic activity toward the soil-borne plant pathogen R solani AG3 or with chitinolytic or antibiotic production capacity originated from the permanent grassland or from the short-term arable land, whereas isolation of such strains from long-term arable land was more rare Hence, the plant (grass), by the action of its roots, most likely in£uenced the Pseudomonas community structure in soil, in the sense that Pseudomonas types with antimicrobial activity were enriched in numbers Similar e¡ects of plants on the abundance of antagonists of Verticillium dahliae were found in a recent study by Berg et al [31] These authors observed an e¡ect of di¡erent plants on P putida populations, which were enriched, in particular by strawberry In our study, the di¡erences in the sizes of the Pseudomonas spp populations between di¡erent treatments were observed on the basis of not only culture-dependent, but also culture-independent analyses In this study, the discriminative value of a PCR-DGGE system based on the Pseudomonas-speci¢c primers described by Widmer et al [21] was evaluated On the basis of the clear patterns obtained, the system proved to be very useful for studying the Pseudomonas diversity directly on the basis of soil-derived DNA Thus, discrimination between Pseudomonas populations in ¢elds under di¡erent agricultural management regimes became possible The PCR-DGGE system, which includes the most variable region of 16S rDNA, i.e V6, was compared to the one recently proposed by Gyam¢ et al on the basis of the V1^V3 regions of 16S rDNA [23] The banding patterns obtained by the two systems showed clear di¡erences Since the Pseudomonas-speci¢c PCR-DGGE system proposed in this work showed higher diversity and better separation of bands than the PCR-DGGE system of Gyam¢ et al [23], we recommend it for further work on the diversity of Pseudomonas spp in soil On the other hand, we found that there was no clear match between sequence type and band (migration) position, and, hence, we cannot make robust inferences about the total numbers of Pseudomonas types present solely on the basis of the patterns obtained A combination of PCR-DGGE analysis with 16S rRNA gene sequencing on the basis of clone libraries and/or excised DGGE bands, such as applied in this study, is actually recommended, as sequencing should clearly complement the limited discrimination in di¡erent migration-based Pseudomonas ‘types’ achieved by PCRDGGE Such a combined analysis will, thus, provide a more complete picture of the diversity of pseudomonads in soil than what can be achieved by ¢ngerprinting alone One striking observation was the high level of relatedness between all Pseudomonas sequences found from DGGE bands, from the clone library analyzed or from the isolates on Gould’s S1 agar Apparently, the pseudomonads, targeted by these methods, are all taxonomically tightly knit, although numerous di¡erent types with clearly di¡erent ecological roles can be shown to be part of this tight group This close relatedness, at the 16S rRNA gene level, is also supported by the literature [24] The results obtained with the direct molecular analyses in this study supported the general view that agricultural management and crop type strongly in£uence the structure of Pseudomonas populations in soil For example, permanent grassland consistently showed the presence of more diverse Pseudomonas populations than arable land under rotation (with initially no rhizosphere present, and a developing rhizosphere in the course of the season) Whereas this enhanced diversity in grassland may have been a longterm e¡ect, also in the arable land itself the presence of a crop had clear e¡ects on the Pseudomonas populations During the time when no crop was present in the arable land under rotation, only one dominant Pseudomonas type, a⁄liated with P rhodesiae, was detected by sequence analysis, but the number of di¡erent sequences clearly increased as soon as a developing crop was present Hence, plants are indeed major factors that can quickly, i.e within sub-periods of crop growth over one growing season, in£uence the structure of Pseudomonas communities in soil Such plant e¡ects on Pseudomonas populations corroborate results obtained before in several studies in other soil or agricultural systems Latour et al [32] showed that plant, as well as soil type, had a selective in£uence on the populations of £uorescent Pseudomonas spp Clays-Josserand et al [33] found that P £uorescens was the numerically most abundant species in the rhizosphere of £ax and tomato plants, followed by P putida and P chlororaphis Lottmann and Berg [34] found that the most dominant pseudomonads associated with the potato rhizosphere were P putida, P £uorescens, P chlororaphis and P syringae According to Misko and Germida [35], P putida and P chlororaphis were the predominant bacteria associated with the roots of ¢eld-grown canola plants in Canada As virtually all Pseudomonas spp are known to be r-strategists, copiotrophic organisms which are able to quickly respond to enhanced nutrient availability in their environment, it is expected that plants with roots exuding di¡erent compounds will indeed quickly select di¡erent Pseudomonas communities in their rhizospheres Taken together, the results obtained in the present study support the hypothesis that plants are major factors determining the Pseudomonas community structure and composition in soil, at both the cultivation-based and cultivationindependent levels In particular, the selection in grassland of the group of pseudomonads de¢ned by Pseudomonas sp E102 and SaU7 was striking These isolates were strongly antagonistic to the target phytopathogen R solani AG3, produced chitinases and possessed the genetic machinery for the production of PRN and PCA We suggest that the abundance and activity of these organisms, stimulated by grass roots, is one of the underlying causes of the suppression of R solani AG3-induced potato rot Further detailed analyses into the ecological role of these organisms are certainly warranted FEMSEC 1587 22-12-03 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 Acknowledgements We thank Krysta Voesenek for excellent technical assistance J Raaijmakers is acknowledged for providing primer sequences for detection of the PRN, PCA and DAPG production loci, and relevant reference strains, and W de Boer for his pertinent advice We further thank the Dutch Organization for Pure Scienti¢c Research (NWO), the Dutch Ministry of Agriculture, Directorate DWK, Agrobiodiversity Program (352) and the EU POTATOCONTROL project for ¢nancial support [15] [16] [17] [18] [19] References [1] O’Donnell, A.G., Seasman, M., Macrae, A., Waite, I and Davies, J.T (2001) Plants and fertilizers as drivers of change in microbial community structure and function in soil Plant Soil 232, 135^145 [2] Lupwayi, N.Z., Rice, W.A and Clayton, G.W (1998) Soil microbial diversity and community structure under wheat as in£uenced by tillage and crop rotation Soil Biol Biochem 30, 1733^1741 [3] Lemanceau, P., Corberand, T., Gardan, L., Latour, X., Laguerre, G., Boeufgras, J.-M and Alabouvette, C (1995) E¡ect of two plant species, £ax (Linum usitatissimum L.) and tomato (Lycopersicon esculentum Mill.), on the diversity of soil populations of £uorescent pseudomonads Appl Environ Microbiol 61, 1004^1012 [4] Duineveld, B.M., Kowalchuk, G.A., Keijzer, A., Van Elsas, J.D and Van Veen, J.A (2001) Analysis of bacterial communities in the rhizosphere of chrysanthemum via denaturing gradient gel electrophoresis of PCR-ampli¢ed 16S rDNA as well as DNA fragments coding for 16S rRNA Appl Environ Microbiol 67, 172^178 [5] Grayston, S.J., Wang, S., Campbell, C.D and Edwards, A.C (1998) Selective in£uence of plant species on microbial diversity in the rhizosphere Soil Biol Biochem 31, 145^153 [6] Yang, C.-H and Crowley, D.E (2000) Rhizosphere microbial community structure in relation to root location and plant iron nutritional status Appl Environ Microbiol 66, 345^351 [7] McCaig, A.E., Glover, L.A and Prosser, J.I (2001) Numerical analysis of grassland bacterial community structure under di¡erent land management regimes by using 16S ribosomal DNA sequence data and denaturing gradient gel electrophoresis banding patterns Appl Environ Microbiol 67, 4554^4559 [8] Hutchinson, S.W (1998) Current concepts of active defense in plants Annu Rev Phytopathol 36, 59^90 [9] Patten, C.L and Glick, B.R (2002) Role of Pseudomonas putida indole acetic acid in development of the host plant root system Appl Environ Microbiol 68, 3795^3801 [10] Samson, R., Sha¢k, H., Benjamo, A and Gardan, L (1998) Description of the bacterium causing blight of leek as Pseudomonas syringae pv porri (pv nov.) Phytopathology 88, 844^850 [11] Clausen, G.B., Larsen, L., Johnsen, K., de Lipthay, J.R and Aamand, J (2002) Quanti¢cation of the atrazine-degrading Pseudomonas sp strain ADP in aquifer sediment by quantitative competitive polymerase chain reaction FEMS Microbiol Ecol 41, 221^229 [12] Kloepper, J.W., Leong, J., Teintze, M and Schroth, M.N (1980) Pseudomonas siderophores: a mechanism explaining disease-suppressive soil Curr Microbiol 4, 317^320 [13] Naseby, D.C and Lynch, J.M (2001) E¡ect of 2,4-diacetylphloroglucinol producing, overproducing and nonproducing Pseudomonas £uorescens F113 in the rhizosphere of pea Microb Ecol 42, 193^ 200 [14] Lim, H.-S., Kim, Y.-S and Kim, S.-D (1991) Pseudomonas stutzeri YPL-1 genetic transformation and antifungal mechanism against Fu- [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] 63 sarium solani, an agent of plant root rot Appl Environ Microbiol 57, 510^516 Thomashow, L.S., Weller, D.M., Bonsall, R.F and Pierson, L.S (1990) Production of the antibiotic phenazine-1-carboxylic acid by £uorescent Pseudomonas species in the rhizosphere of wheat Appl Environ Microbiol 56, 908^912 De Boer, M., Van der Sluis, I., Van Loon, L.C and Bakker, P.A.H.M (1999) Combining £uorescent Pseudomonas spp strains to enhance suppression of fusarium wilt of radish Eur J Plant Pathol 105, 201^210 Raaijmakers, J.M and Weller, D.M (1998) Natural plant protection by 2,4-diacetylphloroglucinol-producing Pseudomonas spp in take-all decline soils Mol Plant-Microbe Interact 11, 144^152 Raaijmakers, J.M., Weller, D.M and Thomashow, L.S (1997) Frequency of antibiotic producing Pseudomonas spp in natural environments Appl Environ Microbiol 63, 881^887 Van Elsas, J.D., Garbeva, P and Salles, J (2002) E¡ects of agronomical measures on the microbial diversity of soil as related to the suppression of soil-borne plant pathogens Biodegradation 13, 29^40 Gould, W.D., Hagedorn, C., Bardinelli, T.R and Zablotowicz, R.M (1985) New selective media for enumeration and recovery of £uorecent pseudomonads from various habitats Appl Environ Microbiol 49, 28^32 Widmer, F., Seidler, R., Cillevet, P.M., Watrud, L.S and Di Giovanni, G.D (1998) A highly selective PCR protocol for detecting 16S rRNA genes of the genus Pseudomonas (sensu stricto) in environmental samples Appl Environ Microbiol 64, 2545^2553 Johnsen, K., Enger, O., Jacobsen, C.S., Thirup, L and Torsvik, V (1999) Quantitative selective PCR of 16S ribosomal DNA correlates well with selective agar plating in describing population dynamics of indigenous Pseudomonas spp in soil hot spots Appl Environ Microbiol 65, 1786^1789 Gyam¢, S., Pfeifer, U., Stierschneider, M and Sessitsch, A (2002) E¡ects of transgenic glufosinate-tolerant oilseed rape (Brassica napus) and the associated herbicide application on eubacterial and Pseudomonas communities in the rhizosphere FEMS Microbiol Ecol 41, 181^190 Anzai, Y., Kim, H., Park, J.-Y., Wakabayashi, H and Oyaizu, H (2000) Phylogenetic a⁄liation of the pseudomonads based on 16S rRNA sequence Int J Syst Evol Microbiol 50, 1563^1589 Imoto, T and Yagishita, K (1971) A simple active measurement of lysozyme Agric Biol Chem 35, 1154^1156 Sambrook, J., Fritsch, E.F and Maniatis, T (1998) Molecular Cloning: A Laboratory Manual, 2nd edn Cold Spring Harbor Laboratory, Cold Spring Harbor, NY Garbeva, P., Van Overbeek, L.S., Van Vuurde, J.W.L and Van Elsas, J.D (2001) Analysis of endophytic bacterial communities of potato plants by plating and denaturing gradient gel electrophoresis (DGGE) of 16S rDNA based PCR fragments Microb Ecol 41, 369^ 383 De Souza, J.T (2002) Distribution, Diversity, and Activity of Antibiotic-producing Pseudomonas spp Ph.D Thesis, Wageningen University, Wageningen Johnsen, K and Nielsen, P (1999) Diversity of Pseudomonas strains isolated with King’s B and Gould’s S1 agar determined by repetitive extragenic palindromic polymerase chain reaction, 16S rDNA sequencing and Fourier transformed infrared spectroscopy characterization FEMS Microbiol Lett 173, 155^162 Garbeva, P., Van Veen, J.A and Van Elsas, J.D (2003) Predominant Bacillus spp in agricultural soil under di¡erent management regimes detected via PCR-DGGE Microb Ecol 45, 302^316 Berg, G., Roskot, N., Steidle, A., Eberl, L., Zock, A and Smalla, K (2002) Plant-dependent genotypic and phenotypic diversity of antagonistic rhizobacteria isolated from di¡erent Verticillium host plants Appl Environ Microbiol 68, 3328^3338 Latour, X., Corberand, T., Laguerre, G., Allard, F and Lemanceau, P (1996) The composition of £uorescent Pseudomonas populations FEMSEC 1587 22-12-03 64 P Garbeva et al / FEMS Microbiology Ecology 47 (2004) 51^64 associated with roots is in£uenced by plant and soil type Appl Environ Microbiol 62, 2449^2456 [33] Clay-Josserand, A., Ghiglione, J.F., Philippot, L., Lemanceau, P and Lensi, R (1999) E¡ect of soil type and plant species on the £uorescent pseudomonads nitrate dissimilating community Plant Soil 209, 275^282 [34] Lottmann, J and Berg, G (2001) Phenotypic and genotypic characterization of antagonistic bacteria associated with roots of transgenic and non-transgenic potato plants Microbiol Res 156, 75^82 [35] Misko, A.L and Germida, J.J (2002) Taxonomic and functional diversity of pseudomonads isolated from the roots of ¢eld-grown canola FEMS Microbiol Ecol 42, 399^407 [36] Heuer, H., Krsek, M., Baker, P., Smalla, K and Wellington, E.M.H (1997) Analysis of actinomycete communities by speci¢c ampli¢cation of genes encoding 16S rRNA and gel-electrophoretic separation in denaturing gradients Appl Environ Microbiol 63, 3233^3241 FEMSEC 1587 22-12-03 ... system was tested and used for the assessment of Pseudomonas diversity directly from soil Moreover, we studied the in? ?uence of the di¡erent agricultural regimes on the prevalence of Pseudomonas antagonistic... determining the Pseudomonas community structure and composition in soil, at both the cultivation-based and cultivationindependent levels In particular, the selection in grassland of the group of. .. (antifungal and chitinolytic) activity against the potato pathogen Rhizoctonia solani AG3 Materials and methods 2.1 Bacterial strains The strains and isolates used in this study and their origins are