Entête

Chapitre 2 Functional diversity of antibiotics-producing Pseudomonas spp. from spruce rhizosphere in nursery and natural forest.

Table des matières

Les espèces de Pseudomonas qui produisent des antibiotiques comme le 2,4-diacéthylphloroglucinol et les phénazines ont une capacité de biocontrôle envers plusieurs pathogènes racinaires fongiques. Nous avons isolé à partir de racines d’épinettes noires ( Picea mariana ) et d’épinettes blanches ( Picea glauca ) provenant de deux pépinières et d’une forêt naturelle, plusieurs souches de Pseudomonas portant des gènes impliqués dans la synthèse du DAPG et du PCA. L’analyse de la séquence du gène phlD a révélé qu’un même génotype est dominant dans les deux pépinières alors qu’un génotype différent est présent dans la forêt naturelle. De plus, les souches récoltées en forêt naturelle possèdent les gènes responsables de la synthèse de la pyrrolnitrine mais l’opéron responsable de la biosynthèse de la pyoluthéorine est absent de ces souches. L’analyse des gènes de production de phénazines a révélé 3 groupes distincts de producteurs de phénazines. Un d’entre eux est présent à la fois dans une pépinière et en forêt naturelle alors que les deux autres sont présents en forêt naturelle seulement. Des tests de confrontation in vitro envers Cylindrocladium floridanum ont également démontré un pouvoir inhibiteur plus grand chez les souches isolées en milieu naturel que chez celles isolées en pépinières.

Fluorescent Pseudomonas spp. which produce antibiotics such as 2,4-diacetylphloroglucinol and phenazines have biocontrol activity against many important soil-borne fungal pathogens. We isolated several strains of Pseudomonas carrying genes for DAPG and PCA synthesis from black spruce ( Picea mariana ) and white spruce ( Picea glauca ) rhizosphere in two different conifers nurseries and in one natural stand. Sequence analysis of a portion phlD revealed that one dominant genotype was present in both nurseries and that a different genotype was dominant in the natural forest. Strains from the natural forest also have the genes for pyrrolnitrin synthesis but lack pyoluteorin biosynthesis operon. Furthermore in vitro anitfungal assay against Cylindrocladium floridanum showed a stronger inhibition by strains isolated from the natural forest than from nurseries. Analysis of phenazine genes revealed 3 distinct groups of phenazine producers. One of them was present in the natural forest and in one nursery and the other two were only present in the natural forest.

Large-scale conifer seedling production for reforestation can be severely affected by root rot and damping-off fungal pathogens, such as Cylindrocladium floridanum, Cylindrocarpon destructans, and Fusarium spp (MRFNP 2003). In Quebec, where over 150 millions seedlings are produced annually, from 10 to 80% of the production can be destroyed, or has to be culled out during phytosanitary inspections (MRFNP 2003). In addition, the disease can significantly lower the survival rate after outplanting to reforestation site (Juzwik et al. 1988, Sanders et al. 1992). In the perspective of a reduction of the use of fungicides and with the phasing-out of methyl bromide (Martin 2003), it is now imperative to find alternatives. The development of biocontrol agents adapted to conifer nurseries is a promising approach.

Strains of Pseudomonas spp. of worldwide origin have shown their abilities to suppress a wide variety of fungal root pathogens in many different agricultural crops around the world (Haas and Keel 2003, Walsh et al. 2001, Thomashow 1996). In most cases, the biocontrol ability of those strains is directly correlated with production of antibiotics such as 2,4-diacethylphloroglucinol (DAPG), phenazines (Phz), pyrrolnitrin (PRN), pyoluteorin (PLT) and hydrogen cyanide (HCN) (Blumer et al. 2000, Haas ans Keel 2003, Raaijmakers et al. 2002, 1997, Thomashow and Weller 1988). The genetic loci responsible for the biosynthesis of these antibiotics have been well characterized and the sequences are available (Bangera et al. 1996, 1999, Mavrodi et al. 1998, Hammer et al. 1997, Nowak-Tompson et al. 1997, 1999, Laville et al. 1998). It is therefore possible to detect and isolate candidate biocontrol agents on the basis of the presence of antibiotic biosynthesis genes in their genome.

The broad-spectrum antibiotic DAPG is one of the most important antibiotics produced by biocontrol Pseudomonas and is implicated in the suppression of several fungal root diseases and in suppressive soils (Bender et al. 1999, Keel et al. 1992, Mazzola 2002, Raaijmakers et al. 1998, 2002, Ramette et al. 2003, Weller et al. 2003). Several distinct groups of DAPG-producing Pseudomonas have been identified from strains of various origins. Two main phenotypic groups can be differentiated based on their antibiotic production. One producing only DAPG and HCN and the other produces in addition pyoluteorin and, in some cases pyrrolnitrin (Keel et al. 1996). While members of this last group are known to be genetically very similar (De La Fuente et al. 2004, Keel et al. 1996, Sharifi-Tehrani et al. 1998, McSpadden Gardener et al. 2000), extensive genomic diversity was highlighted in DAPG + PLT- Pseudomonas by different fingerprinting methods such as ARDRA, RAPD and BOX-PCR (Mavrodi et al. 2001, McSpadden Gardener et al. 2000, Keel et al. 1996). This led to the subdivision of DAPG-producing Pseudomonas in as many as 17 different genomic groups (Landa et al. 2002, Mazzola et al. 2005, McSpadden Gardener et al. 2000). Yet, at the gene level, the diversity is also very important, since phlD , coding for the key enzyme in DAPG synthesis, displayed significant polymorphism, easily revealed by PCR-RFLP (Mavrodi et al. 2001, Picard et al. 2003, Ramette et al. 2001), and DGGE (Bergsma-Vlami et al. 2005). These phlD groups correlated very closely with those obtained by genomic fingerprinting (Landa et al. 2002, Mavrodi et al. 2001, Ramette et al. 2001). These observations led to the conclusion that the phlD gene evolved in concert with the rest of the bacterial genome. However, this was not confirmed on a collection of 144 Pseudomonas isolated from maize at different stages of plant growth which included 4 ARDRA and 59 RAPD groups but presented only one PCR- Hae III pattern (Picard et al. 2000, Picard and Bosco 2003).

This diversity among DAPG producing Pseudomonas is relevant to biocontrol, because strains belonging to certain genotypes can have better biocontrol capacities, such as superior root colonizing abilities (Landa et al. 2002, Mavrodi et al. 2002, Raaijmakers et al. 2001) or higher antibiotic production (Ramette et al. 2001). To a finer scale, even a single base substitution in a gene involved in antibiotic synthesis can be associated with higher antibiotic production. It was highlighted that such a substitution at position 520 of the phlD gene can be a marker of good DAPG producers for the DAPG+ PLT- group (Picard and Bosco 2003). Determination of the phlD genotype, and analysis of the gene sequence can be very informative, and can help in the screening of large banks of potential biocontrol agent.

Pyrrolnitrin (PRN) is also an antibiotic relevant to biocontrol, being produced by Pseudomonas fluorescens, Bulkholderia cepacia, Bulkholderia pyrrocinia, and Myxococcus fluvus . This active metabolite was first used as a clinical antifungal agent for the treatment of skin mycoses, and a derivative of PRN was also developed as an agricultural fungicide (Ligon et al. 2000). PRN producing strains have shown their activity against several bacteria and fungi, in particular Rhizoctonia solani (Hill et al. 1994), and this molecule is also effective against post-harvest diseases caused by Botrytis cinerea (Hammer et al. 1993, Janisiewicz et al. 1988). The prn operon has been completely sequenced and contains 4 genes: prnA,B,C,D (Hammeret al. 1997, Kirner et al. 1998). The operon is well conserved between the 4 species except for prnA which is transcribed divergently in M. fluvus (Hammer et al. 1999). Polymorphism was revealed in prnD , over a collection of Pseudomonas and Burkholderia of different origins. Cluster analysis resulted in seven distinct groups, and the largest was formed by strains that produce both PRN and PLT. Moreover a sequence from B. pyrrocinia DSM 10685 clustered closer to Pseudomonas than to other Burkholderia strains (de Souza et al. 2003).

Pyoluteorin is an aromatic chlorinated polyketide produced by several P. fluorescens strains. This compound is especially active against oomycetes like Pytium ultimum . PLT requires 10 genes for its biosynthesis, pltL,A,B,C,D,E,F,G and pltR,M transcribed divergently (Nowak-Thompson et al. 1997, 1999). Until recently, PLT production seemed to be limited to a selected group of Pseudomonas spp. also producing DAPG and sometimes PRN, which are genotypically very similar (Keel et al. 1996, McSpadden gardener et al. 2000, Sharifi-Tehrani et al 1998). Indeed, RFLP analysis of a pltC fragment showed no polymorphism among 12 Pseudomonas strains of different origins (de Souza et al 2003). Recently however, a strain producing PLT and phenazines has been isolated (Ge et al. 2004), possibly providing new polymorphism within the ptl genes sequences.

The phenazines are another important class of secondary metabolites produced by many strains of Pseudomonas . They cover a large family of heterocyclic nitrogen-containing, brightly coloured pigments, with broad-spectrum antibiotic activity (Chin-A-Woeng et al. 2003). Over 50 naturally occurring phenazines have been described and some organisms can produce as many as 10 different derivatives at a time. The implication of these antibiotics in biocontrol have been shown when bacterization of wheat and tomato seeds with P. fluorescens 2-79 and P. Chlororapis PCL1391 provided protection against Gaeumannomyces graminis and Fusarium oxisporum f. sp. radicis-lycopersici (Thomashow and Weller 1988, Chin-A-Woeng et al. 1998). In addition, phenazines play an important role in bacterial competition and survival in the rhizosphere (Mazzola et al. 1992).

Seven genes phzA,B,C,D,E,F,G , are involved in the synthesis of PCA, and the Phz biosynthetic loci are well conserved in P. fluorescens 2-79, P. aeruginosa PAO1 and P. chlororapis PCL 1394, with nucleotide homology ranging from 70 to 95 % between the different species (Chin-A-Woeng et al. 2003). Despite this homology, individual species differ in the range of compound they produce. In all cases, PCA is the first compound formed in the biosynthetic pathway, but some species can have other genes coding for modifying enzymes. For example, in P. aureofaciens 30-84 PhzO is responsible for conversion of PCA in 2-OH-PCA (Delaney et al. 2001). And in P. chlororapis, phzH is coding for an enzyme responsible for the conversion of PCA in phenazine carboxamide (PCN) (Chin-A-Woeng et al. 2001).

Finally, hydrogen cyanide is also a secondary metabolite produced by pseudomonads that can be involved in disease suppression (Blumer and Haas 2000). Production of HCN by biocontrol Pseudomonas have an implication in the suppression of several diseases, like black root-rot of tobacco caused by Thielaviopsis basicola (Laville et al. 1998). Genetic diversity was also present in the genes responsible for cyanide synthesis. This diversity led to the classification of the HCN-producing DAPG+ pseudomonas in 4 different groups based on hcnBC sequence, and it was shown that the different HCN groups differ in their HCN production level (Ramette et al. 2003) ; groups 1, 2, and 3 were the most efficient producers, while members of group 4 produced significantly less hydrogen cyanide in vitro . These differences also correlated well with biocontrol efficacy in planta .

Antibiotic-producing Pseudomonas have been extensively studied for their use as biocontrol agents in agriculture. Yet little is currently known about their presence and diversity in conifer nurseries and forest ecosystems. Recently, a large range of diversity was observed in microorganism present in black spruce rhizosphere based on analysis of the 16s genes (Filion et al. 2004). Also, the genetic diversity of P. fluorescens in the Douglas fir- Laccaria bicolor was shown to be higher than in the bulk soil (Frey-Klett et al. 2005). However, their functional diversity in this environment has not yet been widely investigated. The objective of the study reported here was to test the hypothesis that antibiotic-producing bacteria are present in the rhizospheres of conifers, and that unique polymorphisms can be identified by characterization of genes involved in antibiotic biosynthesis. We analyzed the sequences of the phlD , phzC, hcnBC, and prnC genes to investigate the functional diversity of antibiotic-producing pseudomonads present in these environments, and to select candidate biocontrol agents to be used in conifer nurseries. We also tested the inhibitory potential of those strains against Cylindrocladium floridanum in vitro.

All Pseudomonas strains used in this study were isolated from the rhizosphere of black spruce and white spruce ( Picea mariana, Picea glauca ). Five seedlings where collected at the St-Modeste provincial nursery (410 rue Principale Saint-Modeste, Quebec, Canada), and three at the Normandin provincial nursery (134 chemin Alfred Villeneuve, Normandin, Quebec, Canada). Also, three seedlings were collected in each natural forest, in Saint-Hippolyte (Quebec, Canada) and in Valcartier (Quebec, Canada). One and two-year-old seedlings were collected, placed in pots, and taken to the laboratory, 5g of root and closely adhering soil was suspended in a flask containing 50 ml of sterile deionized water. Flasks where placed on a rotary shaker at 200 rpm for 120 minutes. Serial dilutions of the mixture were plated on King’s B agar (KMB) (per litter: 20g proteose peptone, 1.2g KH2PO4, 1.5g MgSO4, 10ml glycerol) supplemented with ampicillin (40µg/ml), chloramphenicol (13µg/ml) and cycloheximide (200µg/ml). After 4 days of incubation at room temperature, individual colonies were picked and suspended in the wells of a 96-well plate containing 100µl of water. Then, 20µl aliquots were transferred to the corresponding wells of another plate containing 80µl of liquid King’s B medium and incubated for 48h at room temperature. The plates with suspended cells were boiled in a thermocycler for 8 minutes at 99°C to lyse the cells. The plates were then spun to pellet cell debris, and 2 µl of this whole cell template was used directly for the PCR reaction. Following PCR, cells were harvested from the second plate in the wells corresponding to the samples giving a positive amplification, and were plated on KMB plates. The bacteria plate was then conserved at –80°C after addition of glycerol to a final concentration of 40% (v/v). All strains were otherwise routinely maintained on KMB agar.

The phlD gene was amplified with primer pairs B2BF, BPR2 and Phl2a, Phl2b described previously (McSpadden Gardener et al. 2001, Raaijmakers et al. 1997). The first pair was used for detection and the second pair, yielding a longer fragment, was used for sequencing. For phzC , primers PCA2A and PCA3B were used (Raaijmakers et al. 1997). Primers pyr1 (forward) and pyr2 (reverse), targeting prnC, involved in pyrrolnitrin synthesis, were designed based on an alignment of prnC from Pseudomonas fluorescens BL915 (Genebank U74493), Burkolderia cepacia LT4-12-W (Genebank AF161183) , Burkolderia pyrrocinia DSM 10685 (Genebank AF161186) and Myxococcus fluvus Mx fl47 (Genebank AF161185). The HcnBC gene was amplified with primers Aca and Acb designed by Ramette et al. (2003). The whole cell template used for the PCR reactions was made by suspending one colony in 100µl of distilled water, followed by cell lysis by one treatment at 99°C for 8 minutes and a 4 fold dilution in water. The reactions were carried out in a thermal cycler (model PTC-100; MJ Research Inc., Watertown, MA), in a 25 µl volume containing 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 200 µM of each dNTP (Amersham Biosciences Corp.), 1 µM of each oligonucleotide (QIAGEN), 1 unit of Taq DNA polymerase (Invitrogen), and 2 µL of whole cell template. For amplification of the 800pb prnC fragment, with the pyr1 and pyr2 primers, the reaction started with 2 minutes of denaturation at 95°C followed by 35 cycles of 1 minute denaturation at 95°C, 30 seconds annealing at 60°C and one minute of elongation at 72°C. The reaction ended with a 5 minute elongation at 72°C. For the others primers, we used the reaction conditions previously described except for PCA2A/PCA3B which were used at a Tm. of 58°C instead of 67°C. Finally, the 16s was amplified using the universal primers 8F and 1398R. All the PCR products where run on a 1.5% agarose gel and stained with ethidium bromide. All primers are listed in table 1.

Bayesian inference of phylogeny was performed using a Metropolis-coupled Markov chain Monte Carlo technique as implemented in the program MrBayes version 3.0B4 (Huelsenbek and Ronquist 2001). Four Markov chains were run for one million generations, sampling a tree every 100 generations. The trees were visualized with Treeview version 1.6.6. Two trees were constructed with an alignment of a 746 base pair phzC fragment (figure 2.), and corresponding deduced amino acid fragment (figure 3.), from all the phenazine producing strains isolated in this study and the 4 phzC sequences available in GeneBank: P. fluorescens 2-79 (L48616), P. chlororapis PCL1391 (AF195615), P. aureofaciens 30-84 (L48339), and P. aeruginosa PAO1 (AE004838). In both case, the sequence of P. aeruginosa was used as an outgroup. Furtermore, a tree was constructed with a 1300 base pairs fragment of the 16s RNA gene from all the phenazine producing strains isolated in this study. In this case however, sequences from other phenazine producing strains were not included because they are not available on GeneBank.

In addition, two trees were constructed with an alignment of a 756 bases pairs prnC fragment (figure 4.), and deduced amino acids fragment (figure 5.). Since all the prnC sequences from our isolates are identical, only one was used (6ASH1). The other sequences used were the 4 prnC sequences available on GeneBank : Pseudomonas fluorescens BL915 (U74493), Burkolderia cepacia LT4-12-W (AF161183) , Burkolderia pyrrocinia DSM 10685 (AF161186) and Myxococcus fluvus Mx fl47 (AF161185) and sequence data from the genome of P. fluorescens Pf-5 (contig 3337) obtained from: The Institute for Genomic Research website (http://www.tigr.org). In both cases, the sequence of M. fluvus was used as an outgroup.

Strains of Pseudomonas carrying the phlD gene were detected using the specific primers B2BF and BPR4 (McSpadden Gardener et al. 2001), and sequenced with Phl2a, Phl2b (Raaijmakers et al. 1997) (table 1.). Sequencing of the fragments revealed 3 distinct genotypes, one of which was represented by only one nursery strain. The 43 other nursery strains represent the second genotype and all 26 natural forest strain are from the third genotype. In silico restriction analysis of phlD permitted to classify the natural forest strains in the Hae III-A genotype (like strain CHA0 and Pf-5) while the nursery strains are all members of the Hae III-D genotype (like strain Q8r1-96) when using the classification set up by McSpadden Gardener et al. (Mazzola et al. 2004, McSpadden Gardener et al. 2001) (Table 3.). However, a closer look at the sequences, showed that within this group, 3 strains (7GSM1, A6SM7, A6SM3) had one nucleotide polymorphism. At position 817 of the gene, they have a G instead of a T. And for the strain A6SM3, another mutation is present, at position 811, this strain have a T rather than a C.

Moreover, one strain from the St-Modeste nursery (B6RN) had a totally different genotype. This strain display 32 bases substitutions with the other nursery isolates and share the highest similarity with strains F113 and 7MA12 with 96% similarity in both cases (Table 3.). However, some base substitutions appear in the Hae III restriction sites and this strain cannot be classified in the same group. The isolate B6RN can therefore be considered as a new phlD genotype.

Strains of antibiotic-producing Pseudomonas have been isolated from the rhizosphere of a large number of agricultural crops worldwide, yet, to our knowledge, this is the first report of antibiotic-producing pseudomonads isolated from conifer rhizospheres in nurseries and natural forests. This study also provides a further example of the usefulness of primers targeting antibiotic biosynthesis genes for the isolation of biocontrol agents, rather than the time consuming isolation of strains and screening for antifungal activity.

We conducted this screening toward DAPG and PCA production genes on the rhizobacterial community of seedlings grown in two nurseries and two natural forest sites. Bacteria harbouring these genes were readily isolated from both nurseries and in one natural forest. Interestingly, the population of DAPG producers was very different between the two origins, since they belonged to different phlD groups. Isolates from the natural forest had all the same phlD genotype (99% identity) as strains of the PLT+ PRN+ group, such as CHA0 and Pf-5, while the Pseudomonas isolated in the nurseries, had a phlD sequence very similar (99% identity) to the PRN-, PLT- strain Q8r1-96. Furthermore, these two genomic groups are dominant in their respective locations, and they differed in their capacity to inhibit C. floridanum in vitro .

For the nursery strains, the dominance of one genotype is not surprising, considering the genotypic group of the isolates and the culture practices of the nurseries. These strains come from the rhizosphere of spruce growing in soils that has been subjected to spruce monoculture since 1927 (Normandin) and 1961 (St-Modeste). The phlD genotype of these nursery isolates is similar to strain Q8r1-96 (99% homology). Considering that phlD groups are known to correlate very well with genomic groupings, we can assume that these nursery strains would cluster in the same genomic group as Q8r1-96. Strains from this genotype were found to be dominant on the roots of wheat and pea subjected to many years of monoculture (Landa et al. 2002, Mc Spadden Gardener et al. 2000, Raaijmakers et al. 2001), and they are known for their superior root colonizing abilities (Landa et al. 2002, Mavrodi et al. 2002). The presence of such a dominant root colonizer over a long period would therefore lead to the dominance of strain of this genotype. Finally, only one strain of a different genotype was found, this strain (B6RN) share 24 substitutions with the closest related phlD sequence, and can be considered as a new genotype (table 3.). A larger sampling for each rhizosphere might provide more strains of different genotypes.

On the other hand, this difference in phlD genotype between the two origins can also be caused by the plant host. Plant specie and even plant genotype are known to influence the composition of phlD + Pseudomonas (Bergsma-Vlami et al. 2004, Mazzola et al. 2004, Picard et al. 2004). This could have an effect, since the natural forest seedling are white spruces ( Picea glauca ) and the strains isolated in nurseries come from black spruces ( Picea mariana ).

One could think that the presence of these bacteria would result in a soil suppressive to C. floridanum in these nurseries, because strains of the same genotype have been shown to be responsible for the suppression of take-all of wheat in soils subjected to extensive wheat monoculture (Raaijmakers et al. 1998, Weller et al. 2002). However, in the two nurseries sampled in this study, the disease was present and strains of this dominant genotype are unable to effectively inhibit C. floridanum in vitro . This lack of inhibition might be explained by a level of DAPG production too low to affect the pathogen. This is supported by the absence of the base substitution at position 520 highlighted by Picard et al. as a marker of good DAPG producers (2003). Another possible explanation, is that the co-occurrence of C. floridanum and DAPG producing bacteria in the same environment for a long period have made the pathogen tolerant to this antibiotic rather than evolving toward a suppressive soil. However, DAPG+ pseudomonads were also recoverd from a conductive soil in another study (Ramette et al. 2003). Furthermore, HCN is not believed to be highly effective against C. floridanum , considering that the nursery strains, as members of the HCN-1 group, and HCN-2 in the case of B6RN, should be good cyanide producers (Ramette et al. 2003).

For the natural forest strains, we have one single genotype dominant in the rhizospheres, but the isolates belong to a totally different phlD genotype and a different Pseudomonas group. All phlD sequences from strains isolated at this site are very similar to the PLT+, PRN + strains CHA0 and Pf-5 (99% homology). Strains from this particular group are known to be genotypically very similar or even identical (de Souza et al. 2003, Keel et al. 1996, Mavrodi et al. 2001). The high similarity of the phlD, prnC , hcnBC and 16s ribosomal RNA gene sequences of the strains isolated in this study with the others PLT+, PRN+ strains support this affirmation. However, despite this high homology for four genes, those newly isolated strains lack the plt operon usually found in this group. Attempts to amplify pltC and pltB , failed, even with 2 sets of primers directed toward each gene. Considering that no polymorphism has been discovered so far in the plt operon (de Souza et al. 2003), it is unlikely that the genes are present but too divergent to be recognized by the primers, and we can reasonably assume that the Plt operon is absent in those PRN+, DAPG+ strains.

Analysis of the prnC fragments showed polymorphism between strains isolated in this study and the other sequences available, but no polymorphism was found within the isolates and with P. fluorescens Pf-5. In this sense, the results are parallel to those obtained on prnD by de Souza et al. (2003) and by Garbeva and coworkers (2004). In all cases, the sequence from B. pyrrocinia DSM 10685, is closer to the Pseudomonas sequences than to B. cepacia , and strains producing DAPG and PRN have similar prn sequences (de Souza et al. 2003) (Table 4). And the phylogenetic analysis clearly place our DAPG+, PRN+ strains on the same group as P. fluorescens Pf-5 and on a separate branch than B. pyrrocinia DSM 10685 and P. fluorescens BL915 (figure 3). The deduced amino acid fragment provided similar results (figure 4.). The strong similarity of prnC , phlD and hcnBC sequences isolated in this study with the other DAPG+, PRN+, PLT+ Pseudomonas confirm once again the high degree of similarity found within this group. However, the absence of the plt operon represent a major difference and raise questions about their phylogenetic relationship.

The lack of pyoluteorin production, and the phlD sequence corresponding to the Hae III - RFLP group A, which is known to have a low level of DAPG production (Ramette et al. 2001), would point to pyrrolnitrin as the determining factor for the strong inhibition of C. floridanum in vitro . However, it is hazardous to positively conclude on this, because the absence of pyoluteorin can have an effect on DAPG production. This was shown in a mutant of Pf-5 deficient in PLT production, which had a higher production of DAPG (Brodhagen et al. 2004). Also addition of pyoluteorin in growth medium repressed the expression of phlA in strain CHA0 (Schnider-Keel et al. 2000).

The DAPG+ pseudomonad are known to be cosmopolitan at a worldwide scale and the different phlD clusters can be identified in Pseudomonas associated with different dicotyledonous crops (Wang et al. 2001). This is supported by our result since we harvested from spruce rhizosphere, some Pseudomonas having the same phlD genotype as other strains isolated from different plants host, like strains CHA0 and Q8r1-96 isolated respectively from tobacco and wheat (Keel et al. 1992, Raaijmakers et al. 1998). However, the authors also suggest that pseudomonad belonging to different phlD clusters may coexist in the rhizosphere of dicots at a given geographic location (Wang et al. 2001). In our study, with the exception of one single strain, only one phlD genotype was present at each location in both nurseries and in the natural forest. In this sense, our results are in accordance with those of Mavrodi et al., McSpadden Gardener et al., and Picard et al. who found a single phlD genotype dominant in each geographic location. Yet, in these cases, the plant host were monocots and our results come from the dicotyledonous conifer. However, a wider sampling in other natural forest sites would be necessary to draw clear conclusions.

The phenazine producers isolated in this study also show diversity within one of their antibiotic biosynthesis gene ( phzC ), but to a different extent than the DAPG producers. First, it is difficult to conclude about their presence in the nurseries, because only two strains were isolated from one single nursery seedling. However, considerable diversity was found at the natural forest site. In fact, strains belonging to every cluster were present in this environment, two of them where even present in the same rhizosphere, and this rhizosphere was also harbouring DAPG +, PRN + strains (table 2.). This is interesting because it raises the possibility of using different antibiotic-producing strains all adapted to the same environment and plant host.

The percentage of identity ranging from 94.4% to 91.6% and the strong differences in phenotypes suggest that members of each group belong to different species. Analysis of the 16s RNA gene strongly support this idea by grouping the strains in the same three different clusters than with phzC . For the two nursery strains (A8B1, C4B1), two base substitutions in phzC differentiate them with the other members of this group isolated in the natural forest. This is enough to place them on two different branches for the tree constructed with DNA alignment. However, these two branches are not supported by high statistical value and they can be considered as forming one group. One the other hand, one of these base substitution is translated in an amino acid substitution and thus they are seen as two different groups on the tree constructed with amino acid fragment. For the two other groups, the clustering obtained with DNA is consistent with the results obtained with amino acids. In both cases, group 1 and 2 are placed close to each other and close to P. fluorescens , while members of group 3 are positioned between P. fluorescens and P. chlororapis.

The different clusters, belonging to different species, might produce different phenazine derivatives. Further analysis will be necessary to define exactly which phenazinederivatives are produced by strains of each group. But for group 1, the production of the blue-green pigment can be an indication of the presence of pyocyanin or a related molecule.

The isolates from the different groups differ slightly in their inhibition level toward Cylindrocladium floridanum . The most efficient seem to be the group 3 strains (A7SH), while the three members of group 2 were less efficient (figure 1.). These differences can either be caused by differences in antibiotic production level, or by the production of different derivatives by each strains.

We also tested 100 strains randomly selected among those that were not carrying antibiotic-biosynthesis genes. Only two of those strains were able to effectively inhibit the growth of C. floridanum , four gave weak inhibition and the rest were completely overgrown by the fungus. The first inhibitory strain was identified by 16s RNA gene sequencing, as Pseudomonas spp. Since this strain is not likely to produce any of the antibiotics that we studied in this work, its inhibitory mechanism must have another explanation. It could either be related to the production of other secondary metabolite like viscosinamide, tensin, amphisin or related antifungal lipopeptide (de Souza et al. 2003, Harder et al. 2002, Nielsen et al. 2002) or, the inhibition could be caused by the release of extracellular lytic enzymes like chitinase (Nielsen et al.1998, Nagarajkumar et al. 2004). For the second inhibitory strain the identification is more surprising, as it seems to be a member of the recently discovered genus Collimonas (de Boer et al. 2004). One member of this genus: Collimonas fungivorans has the peculiar ability of growing at the expense of living fungal hyphae (de Boer et al. 2004). This mycophageous lifestyle coupled with chitinolytic activity could naturally lead to an inhibition of fungal growth. It would therefore be interesting to further investigate the mechanism of action of this strain, and its possible use in biocontrol, since it could have an effect complementary to antibiotic producing strains and could be used in a more holistic approach in fighting root pathogens.

Taken as a whole, these results not only confirm the presence of antibiotic-producing pseudomonads in spruce rhizosphere, but also suggest that their functional diversity is greater in the natural forest than in the nurseries. Also, a novel group of Pseudomonas capable of producing DAPG and pyrrolnitrin but not pyolutheorin was identified, strains of this group are the most potent inhibitors of C. floridanum and can be seen as promising biocontrol agents.

Bangera, M. G., and L. S. Thomashow. 1996. Characterization of a genomic locus required for synthesis of the antibiotic 2,4-diacetylphloroglucinol by the biological control agent Pseudomonas fluorescens Q2-87. Mol Plant Microbe Interact 9 :83-90.

Bangera, M. G., and L. S. Thomashow. 1999. Identification and characterization of a gene cluster for synthesis of the polyketide antibiotic 2,4-diacetylphloroglucinol from Pseudomonas fluorescens Q2-87. J Bacteriol 181:3155-3163.

Bender, C. L., and V. Rangaswamy. 1999. Polyketide production by plant-associated pseudomonas . Ann. Rev. Phytopathol. 37 : 175-196.

Bergsma-Vlami, M., M. E. Prins, M. Staats, and J. M. Raaijmakers. 2005. Assessment of genotypic diversity of antibiotic-producing Pseudomonas species in the rhizosphere by denaturing gradient gel electrophoresis. Appl Environ Microbiol. 71  : 993-1003.

Bergsma-Vlami, M., M. E. Prins, and J. M. Raaijmakers. 2004. Influence of plant spicies on population dynamics, genotypic diversity and antibiotic production in the rhizosphere by indigenous Pseudomonas spp. FEMS Microbio. Ecol. 52 : 59-69.

Blumer, C., and D. Haas. 2000. Mechanism, regulation, and ecological role of bacterial cyanide biosynthesis. Arch Microbiol 173 :170-177.

Brodhagen, M., M. D. Henkels, and J. E. Loper. 2004. Positive autoregulation and signaling propreties of pyoluteorin, an antibiotic produced by biological control organism Pseudomonas fluorescens Pf-5. Appl Environ Microbiol. 70  : 1758-1766.

Chin-A-Woeng, T. F., G. V. Bloemberg. and B. J. Lugtenberg. 2003. Phenazines and their role in biocontrol by Pseudomonas bacteria. New. Phyto. 157 : 503-523.

Chin-A-Woeng, T. F., J. E. Thomas-Oates, B. J. Lugtenberg, and G. V. Bloemberg. 2001. Introduction of the phzH gene of Pseudomonas chlororaphis PCL1391 extends the range of biocontrol ability of phenazine-1-carboxylic acid-producing Pseudomonas spp. strains. Mol Plant Microbe Interact 14 :1006-1015.

Chin-A-Woeng, T. F., G. V. Bloemberg, I. H. Mulders, L. C. Dekkers, and B. J. Lugtenberg . 2000. Root colonization by phenazine-1-carboxamide-producing bacterium Pseudomonas chlororaphis PCL1391 is essential for biocontrol of tomato foot and root rot. Mol Plant Microbe Interact. 13:1340-1345.

Cook, R. J., L. S. Thomashow, D. M. Weller, D. Fujimoto, M. Mazzola, G. Bangera, and D. S. Kim. 1995. Molecular mechanisms of defense by rhizobacteria against root disease. Proc Natl Acad Sci U S A 92 :4197-4201.

de Boer W., J. H. Leveau, G. A. Kowalchuk, P. J. Klein Gunnewiek, E. C. Abeln, M. J. Figge, K. Sjollema, J. D. Janse and J. A. van Veen. 2004. Collimonas fungivorans gen. nov., sp. nov., a chitinolytic soil bacterium with the ability to grow on living fungal hyphae. Int J Syst Evol Microbiol. 54 (Pt 3):857-864.

De La Fuente, L., L. Thomashow, D. Weller, N. Bajsa, L. Quagliotto, L. Chernin, and A. Arias. 2004. Pseudomonas fluorescens UP61 isolated from birdsfoot trefoil rhizosphere produces multiple antibiotics and exerts a broad spectrum biocontrol activity. Eur. J. Plant Pathol. 110 : 671-681.

Delaney, S. M., D. V. Mavrodi, R. F. Bonsall, and L. S. Thomashow. 2001. phzO, a gene for biosynthesis of 2-hydroxylated phenazine compounds in Pseudomonas aureofaciens 30-84. J Bacteriol 183 :318-327.

de Souza, J. T. and J. M. Raaijmakers. 2003. Polymorphism within the prnD and pltC genes from pyrrolnitrin and pyoluteorin-producing Pseudomonas and Burkholderia spp. FEMS Microbio. Ecol. 43 . 21-34.

de Souza, J. T., M. De Boer, P. De Waard, T. A. Van Beek, and J. M. Raaijmakers. 2003. Biochemical, genetic, and zoosporicidal properties of cyclic lipopeptide surfactants produced by Pseudomonas fluorescens . Appl Environ Microbiol 69 :7161-7172.

Filion M, Hamelin RC, Bernier L, St-Arnaud M. 2004. Molecular profiling of rhizosphere microbial communities associated with healthy and diseased black spruce (Picea mariana) seedlings grown in a nursery. Appl Environ Microbiol. 2004 Jun;70(6):3541-3551.

Frey-Klett, P., M. Chavatte, M.L. Clausse, S. Courrier, C. Le Roux, J. Raaijmakers, M. Giovanna Martinotti, J.C. Pierrat, and J. Garbaye. 2005. Ectomycorrhizal symbiont affects functional diversity of rhizosphere fluorescent pseudomonads. New phytol. 165 : 317-328.

Garbeva, P., K. Voesenek, and J.D. van Elsas. 2004. Quantative detection and diversity of the pyrrolnitrin biosynthetic locus in soil under different treatments. Soil Bio. Biochem. 36 . 1453-1463.

Ge, Y., X. Huang, S. Wang, X. Zhang, and Y. Xu. 2004. Phenazine-1-carboxylic acid is negatively regulated and pyoluteorin positively regulated by gacA in Pseudomonas sp. M18. FEMS Microbiol Lett. 1; 237 (1):41-47.

Haas, D., C. Blumer, and C. Keel. 2000. Biocontrol ability of fluorescent pseudomonads genetically dissected: importance of positive feedback regulation. Curr Opin Biotechnol 11 :290-297.

Haas, D., and C. Keel. 2003. Regulation of antibiotic production in root-colonizing Peudomonas spp. and relevance for biological control of plant disease. Annu Rev Phytopathol 41 :117-153.

Hammer, P. E., D. S. Hill, S. T. Lam, K. H. Van Pée, and J. M. Ligon. 1997. Four genes from Pseudomonas fluorescens that encode the biosynthesis of pyrrolnitrin. Appl Environ Microbiol 63 :2147-2154.

Hammer, P. E., W. Burd, Hill, S. D., Ligon, J. M. and K. H. van Pée 1999. Conservation of the pyrrolnitrin gene cluster among six pyrrolnitrin-producing strains. FEMS Microbio. Lett. 180 . 39-44.

Hill, D. S., J. I. Stein, N. R. Torkewitz, A. M. Morse, C. R. Howell, J. P. Pachlatko, J. O. Becker and J. M. Ligon. 1994. Cloning of Genes Involved in the Synthesis of Pyrrolnitrin from Pseudomonas fluorescens and Role of Pyrrolnitrin Synthesis in Biological Control of Plant Disease. Appl. Environ. Microbiol., 78-85, Vol 60 , No. 1.

Huelsenbeck, J. P. and F. Ronquist. 2001. MRBAYES : Bayesian inference of phylogentic trees. Bioinfo. Vol. 17 no. 8.

Juzwik, J., C. Honhart, and N. Chong. 1988. Cylindrocladium root rot in Ontario bare-root nurseries : estimate of spruce seedlings losses. Can. J. For. Res. Vol. 18 1493-1496.

Keel, C., U. Schinder, M. Maurhofer, C. Voissard, J. Laville, P. Burger, P. Wirthner, D. Haas, and G. Défago. 1992. Supression of root diseases by Pseudomonas fluorescens CHA0 : importance of the secondary metabolite 2,4-diacethylphloroglucinol. Mol. Plant-Microbe Interact. 5 , 4-13.

Keel, C., D. M. Weller, A. Natsch, G. Defago, R. J. Cook, and L. S. Thomashow. 1996. Conservation of the 2,4-diacetylphloroglucinol biosynthesis locus among fluorescent Pseudomonas strains from diverse geographic locations. Appl Environ Microbiol 62 :552-563.

Kirner, S., P. E. Hammer, D. S. Hill, A., Altmann, I., Fischer, L. J. Weislo, M. Lanahan, K. H. van Pée, and J. M. Ligon. 1998. Functions encoded by pyrrolnitrin biosynthetic genes from Pseudomonas fluorescens. J. of Bact. Vol. 180 , No. 7. 1939-1943.

Landa, B. B., O. V. Mavrodi, J. M. Raaijmakers, B. B. McSpadden Gardener, L. S. Thomashow, and D. M. Weller. 2002. Differential ability of genotypes of 2,4-diacetylphloroglucinol-producing Pseudomonas fluorescens strains to colonize the roots of pea plants. Appl Environ Microbiol 68 :3226-3237.

Laville, J., C. Blumer, C. Von Schroetter, V. Gaia, G. Défago, C. Keel, and D. Haas. 1998. Characterization of the hcnABC gene cluster encoding hydrogen cyanide synthase and anaerobic regulation by ANR in the strictly aerobic biocontrol agent Pseudomonas fluorescens CHA0. J Bacteriol 180 :3187-3196.

Ligon, J. M., D. S. Hill, P. E. Hammer, N. R. Torkewitz, and D. Hofmann. 2000. Natural products with antifungal activity from Pseudomonas biocontrol bacteria. Pest Manage. Sci. 56 :688-695.

Martin, F. N. 2003. Development of alternative strategies for management of soilborne pathogens currently controlled with methyl bromide. Ann. Rev. Phytopath. Vol. 41 . 325-350.

Mavrodi, D. V., R. F. Bonsall, S. M. Delaney, M. J. Soule, G. Phillips, and L. S. Thomashow. 2001. Functional analysis of genes for biosynthesis of pyocyanin and phenazine-1-carboxamide from Pseudomonas aeruginosa PAO1. J Bacteriol 183 :6454-6465.

Mavrodi, D. V., V. N. Ksenzenko, R. F. Bonsall, R. J. Cook, A. M. Boronin, and L. S. Thomashow. 1998. A seven-gene locus for synthesis of phenazine-1-carboxylic acid by Pseudomonas fluorescens 2-79. J Bacteriol 180 :2541-2548.

Mavrodi, D. V., O. V. Mavrodi, B. B. McSpadden-Gardener, B. B. Landa, D. M. Weller, and L. S. Thomashow. 2002. Identification of differences in genome content among phlD-positive Pseudomonas fluorescens strains by using PCR-based subtractive hybridization. Appl Environ Microbiol 68 :5170-5176.

Mazzola, M. 2002. Mechanism of natural soil suppressiveness to soil-borne diseases. Antonie Van Leeuwenhoek. 81 : 557-564.

Mazzola, M., D. L. Funnell and J. M. Raaijmakers. 2005. Weath cultivar-specific selection of 2,4-diacethylphloroglucinol-producing fluorescent Pseudomonas species from resident soil populations. Microb. Ecol. 48 : 338-348.

Mazzola, M., R. J. Cook, L. S. Thomashow, D. M. Weller, and L. S. Pierson, 3rd. 1992. Contribution of phenazine antibiotic biosynthesis to the ecological competence of fluorescent pseudomonads in soil habitats. Appl Environ Microbiol 58 :2616-2624.

McSpadden Gardener, B. B., Mavrodi D. M., Thomashow, L.S. and D. M. Weller. 2001. A rapid polymerase chain reaction-based assay characterizing rhizosphere populations of 2,4-diacetylphloroglucinol-producing bacteria. Phytopath. 91 :44-54. 

McSpadden Gardener, B. B., K. L. Schroeder, S. E. Kalloger, J. M. Raaijmakers, L. S. Thomashow, and D. M. Weller. 2000. Genotypic and phenotypic diversity of phlD-containing Pseudomonas strains isolated from the rhizosphere of wheat. Appl Environ Microbiol 66 :1939-1946.

Ministère des Ressources Naturelles de la Faune et des Parcs . 2003. Insectes, Maladies et Feux dans les forêts québécoises en 2003. http://www.mrnfp.gouv.qc.ca/publications/forets/fimaq/BROCH_72.PDF

Nielsen, T. H., D. Sorensen, C. Tobiasen, J. B. Andersen, C. Christophersen, M. Givskov, and J. Sorensen. 2002. Antibiotic and biosurfactant properties of cyclic lipopeptides produced by fluorescent Pseudomonas spp. from the sugar beet rhizosphere. Appl Environ Microbiol 68 :3416-3423.

Nowak-Thompson, B., N. Chaney, J. S. Wing, S. J. Gould, and J. E., Loper. 1999. Caracterization of the pyoluteorin biosynthetic gene cluster of Pseudomonas fluorescens Pf-5. J. of Bact. Vol. 181 , No. 7. 2166-2174.

Nowak-Thompson, B., S. J. Gould, and J. E., Loper. 1997. Identification and sequence analysis of the genes encoding a polyketide synthase required for pyoluteorin biosynthesis in Pseudomonas fluorescens Pf-5. Gene. 204 . 17-24.

Osborn, A. M., E.R.B. Moore and K.N. Timmis. 2000. An evaluation of terminal-restriction fragment length polymorphism (T-RFLP) analysis for the study of microbial community structure and dynamics. Environ. Microbiol. 2(1): 39-50.

Picard, C., and M. Bosco. 2003. Genetic diversity of phlD gene from 2,4-diacetylphloroglucinol-producing Pseudomonas spp. strains from the maize rhizosphere. FEMS Microbiol Lett 219 :167-172.

Picard, C., E. Frascaroli, and M. Bosco. 2004. Frenquency and biodiversity of 2,4-diacethylphoroglucinol-producing rhizobacteria are differentially affected by the genotype of two maize inbred lines and their hybrid. FEMS Microbiol. Ecol. 49 , 207-215.

Picard, C., F. Di Cello, M. Ventura, R. Fani, and A. Guckert. 2000. Frequency and biodiversity of 2,4-diacetylphloroglucinol-producing bacteria isolated from the maize rhizosphere at different stages of plant growth. Appl Environ Microbiol 66 :948-955.

Raaijmakers, J. M., D. M. Weller., and L. S. Thomashow. 1997. Frequency of antibiotic-producing Pseudomonas spp. in natural environments. Appl Environ Microbiol 63 , No. 3, 881-887.

Raaijmakers, J. M., M. Vlami, and J. T. de Souza. 2002. Antibiotic production by bacterial biocontrol agents. Antonie Van Leeuwenhoek 81 :537-547.

Raaijmakers, J. M., and D. M. Weller. 2001. Exploiting genotypic diversity of 2,4-diacetylphloroglucinol-producing Pseudomonas spp.: characterization of superior root-colonizing P. fluorescens strain Q8r1-96. Appl Environ Microbiol 67 :2545-2554.

Ramette, A., M. Frapolli, G. Défago, and Y. Moënne-Loccoz. 2003. Phylogeny of HCN synthase-encoding hcnBC genes in biocontrol fluorescent pseudomonads and its relationship with host plant species and HCN synthesis ability. Mol Plant Microbe Interact 16 :525-535.

Ramette, A., Y. Moënne-Loccoz, and G. Défago. 2003. Prevalence of fluorescents pseudomonads producing antifungal phloroglucinols and/or hydrogen cyanide in soils naturally suppressive or conductive to tobacco black root rot. FEMS. Microbio. Ecol. 44 . 35-43.

Ramette, A., Y. Moënne-Loccoz, and G. Défago. 2001. Polymorphism of the polyketide synthase gene phID in biocontrol fluorescent pseudomonads producing 2,4-diacetylphloroglucinol and comparison of PhID with plant polyketide synthases. Mol Plant Microbe Interact 14 :639-652.

Saunders, J.E., J. Juzwick, and R. Hutchinson. 1992. Outplanting survival of cylindrocladium root rot affected black spruce seedlings. Can. J. For. Res. 22. 1204-1207.

Schnider-Keel, U., A. Seematter, M. Maurhofer, C. Blumer, B. Duffy, C. Gigot-Bonnefoy, C. Reimmann, R. Notz, G. Défago, D. Haas, and C. Keel. 2000. Autoinduction of 2,4-diacetylphloroglucinol biosynthesis in the biocontrol agent Pseudomonas fluorescens CHA0 and repression by the bacterial metabolites salicylate and pyoluteorin. J Bacteriol. 182 :1215-1225.

Thomashow, L. S. 1996. Biological control of plant root pathogens. Curr Opin Biotechnol 7 :343-347.

Thomashow, L. S., and D. M. Weller. 1988. Role of a phenazine antibiotic from Pseudomonas fluorescens in biological control of Gaeumannomyces graminis var. tritici. J Bacteriol 170 :3499-3508.

Wang, C., A. Ramette, P. Punjasamarnwong, M. Zala, A. Natch, Y. Moënne-Locoz, and G. Défago. 2001. Cosmopolitan distribution of phlD -contaning dicotyledonus crop-associated biocontrol pseudomonads of worldwide origin. FEMS Microbio. Eco. 37 . 105-116.

Walsh, U. F., J. P. Morrissey, and F. O'Gara. 2001. Pseudomonas for biocontrol of phytopathogens: from functional genomics to commercial exploitation. Curr Opin Biotechnol 12 :289-295.

Weller, D. M., J. M. Raaijmakers, B. B. Gardener, and L. S. Thomashow. 2002. Microbial populations responsible for specific soil suppressiveness to plant pathogens. Annu Rev Phytopathol 40 :309-348.

HaeIII groups are based on the classification made by McSpadden Gardener et al. (2001). HCN groups are based on the classification made by Ramette et al. (2003).

© Mathieu Allaire, 2005