SEARCH

SEARCH BY CITATION

Keywords:

  • rhizosphere;
  • mRNA-SIP;
  • microbial community structure;
  • gene expression

Abstract

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

The rhizosphere is an active compartment where plant and microorganisms establish a molecular dialogue. In this study, we analysed the impact of Arabidopsis thaliana on bacterial community structure and the expression of certain beneficial genes using DNA- and mRNA-SIP in the rhizosphere of plantlets grown under 13CO2 for 13, 21 and 27 days. DNA- and rRNA-SIP revealed changes in bacterial communities inhabiting the rhizosphere soil that were probably related to modification of root exudates, while root-colonizing populations were maintained over time suggesting their metabolic versatility and adaptation. The impact of the plant via root exudates on the expression of the noncoding RNAs rsmZ,acdS gene encoding 1-aminocyclopropane-1-carboxylate deaminase and nosZ gene encoding nitrous oxide reductase, in the root-adhering soil and on the roots of A. thaliana was determined using mRNA-SIP. Results showed that these genes were present and expressed by bacteria inhabiting roots and by those that derive nutrients from the breakdown of organic matter in soils or from root exudates. The expression of rsmZ under natural conditions indicates the importance of noncoding RNAs in bacterial adaptation to their ecological niches.


Introduction

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

Interactions between plants and microorganisms in the rhizosphere are complex and varied. They include the general transfer of nutrients and specific interactions mediated by the release of signalling molecules from plant roots (Prosser et al., 2006). Many studies have analysed the structure of bacterial communities associated with plant roots, and different bacterial species have been identified in root tissues and rhizosphere soil (Nunan et al., 2005; Costa et al., 2006). However, prerequisites for understanding the ecology of the rhizosphere are to (1) identify microbial communities that inhabit this ecosystem, (2) link their structure and function and (3) monitor their metabolic properties. Isotope-labelling techniques combined with molecular detection tools are appropriate to achieve such goals. In fact, the use of stable isotope probing (SIP) has provided the potential for characterizing microorganisms actively assimilating carbon derived from plant root exudates (Vandenkoornhuyse et al., 2007; Haichar et al., 2008; Bressan et al., 2009). Indeed, Lu et al. (2006) used this approach to identify bacterial species actively incorporating root exudates from rice plants, and more recently Rasche et al. (2009) used SIP approach to identify active bacterial endophytes that use 13C-enriched photosynthates in potatoes. Although laboratory studies provide information on bacterial functions, we lack knowledge of whether these functions and associated functional genes are expressed in situ and on plant roots.

Detection of mRNAs, which typically have short half-lives, provides a strong indication of specific gene expression at the time of sampling (Robinson et al., 1998). Recently, reverse transcription PCR (RT-PCR) has been used to investigate gene expression in different environments, including the expression of nirK and nirS in the rhizosphere of legumes (Sharma et al., 2005), pmoA in rice rhizosphere (Shrestha et al., 2010) and nirK, nosZ and nirS in adjacent riparian and agricultural zones (Dandie et al., 2011). Interestingly, Huang et al. (2009) coupled mRNA stable isotope probing with single-cell Raman-fluorescence in situ hybridization to link microbial identity and function concerning naphthalene degradation in groundwater.

The aim of this study is to analyse the dynamics, structure and composition of bacterial community in the rhizosphere of Arabidopsis thaliana by combining DNA- and RNA-SIP approaches and to determine the impact of plant via root exudates in controlling the expression of certain bacterial genes considered as relevant for plant–bacteria interaction and rhizosphere functioning by performing mRNA-SIP approach. A. thaliana plants were grown under 13CO2 for 13, 21 and 27 days. The expression of the noncoding regulatory RNA rsmZ, which is part of the GacS-GacA regulatory cascade that controls the production of secondary metabolites, acdS, which is encoding 1-aminocyclopropane-1-carboxylate (ACC) deaminase involved in ethylene metabolism (Glick et al., 1998), and nosZ, coding for dissimilatory bacterial nitrous oxide reductase (Rösch et al., 2002), was analysed from root mRNA and 12C- and 13C-mRNA issued from the rhizosphere.

Materials and Methods

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

Plant growth

The experiment was carried out in the laboratory with A. thaliana (ecotype Colombia) on Eutric cambisol soil (Derrien et al., 2004). The soil was sieved (1 mm mesh size), air-dried and adjusted to 0.17 g water g−1 dry weight. Forty grams dry weight soil was then placed into polypropylene cylinders. Seeds of A. thaliana were sterilized according to Achouak et al. (2004) and germinated on half-strength Hoagland medium and 0.8% phytagel (Sigma, St Louis) plates at 25 °C for 48 h. After germination, one seedling was planted per pot. Plants were grown in triplicate in a growth chamber (developed and managed by Groupe de Recherche Appliquées en Phytotechnologies, CEA Cadarache, France) as previously described (Haichar et al., 2008). Triplicate cylinders containing soil without plants (bulk soil treatment) were also incubated under the same conditions.

13C labelling

Continuous labelling started 1 week after seeds were sown in pots. CO2 partial pressure was assured by injection of pure (> 99% atom 13C) 13CO2 (Cortec Net, Paris, France). The isotope excess of CO2 and the partial pressure in the chamber were both continuously monitored by near infrared spectroscopy. To avoid dilution of 13CO2 by 12CO2 from soil respiration, the CO2 concentration of the chamber was lowered to 400 μL L−1 at the end of the night by gas trapping. 13CO2 was then quickly injected so that CO2 concentration reached 350 μL L−1. The isotope excess in the chamber was maintained at > 80% atom 13C during the first 10 days and > 90% atom 13C thereafter. Arabidopsis thaliana plants and soil microcosms (bulk soil) were harvested in triplicate after 13, 21 and 27 days of 13C labelling. Plants were also grown under 12CO2, but were not analysed.

DNA extraction and gradient fractionation

At the end of labelling, as Arabidopsis roots are fine, the root systems were separated manually from the root-adhering soil (RS) fraction and carefully washed with sterilized water. Root material was frozen in liquid N2 and stored at −80 °C. The RS fraction without any root contamination was also frozen in liquid N2 and stored at −80 °C. DNA was extracted from 5 g RS and from 500 mg of root system from each plant as described by Ranjard et al. (2003). DNA samples were visualized and quantified by electrophoresis in a 1% agarose gel. DNA extracted form roots was derived from endophytic microorganisms and from microorganisms firmly attached to the root surface. DNA extracted from the RS was fractionated by CsCl equilibrium density gradient centrifugation (Lueders et al., 2004). Buoyant density was determined by weighing, and DNA was quantified in CsCl density fractions with the Picogreen assays (Molecular Probes). Nucleic acids were purified from CsCl salts using Geneclean kit (Qbiogene, Montreal, QC). For each gradient, one fraction representative of 13C-labelled (‘heavy’, fraction N° 10) DNA and one fraction representative of unlabelled (‘light’, fraction N° 5) DNA were chosen according to previous studies in the laboratory (Haichar et al., 2007, 2008).

RNA extraction and cesium trifluoroacetate (CsTFA) centrifugation

RNA was extracted from the RS and root using an RNA Power Soil isolation kit (MO BIO) that yielded nondegraded total RNA. DNA was removed from RNA extracted from RS and root tissues by treatment with DNase for 1 h at 37 °C according to the manufacturer's recommendations (Promega). 16S rRNA gene amplifications were carried out using treated RNA as template to ensure that DNA was totally removed. The amount of RNA was checked by agarose gel electrophoresis, and 2 μg of RNA from each RS and at each date was fractionated by CsTFA equilibrium density gradient centrifugation (Rangel-Castro et al., 2005). Buoyant density was determined by weighing, and RNA was quantified in CsTFA density fractions with the RiboGreen assays (Molecular Probes). For each gradient, one fraction representative of 13C-labelled RNA and one fraction representative of unlabelled 12C-RNA were chosen according to Rangel-Castro et al. (2005). Nucleic acids were purified from CsTFA salts by isopropanol precipitation, and RNA pellets were air-dried and resuspended in 20 μL of RNase-free sterile water.

13C analysis

13C content of heavy- and light-DNA/RNA fractions from triplicates at each date was measured by isotope ratio mass spectrometry (IRMS; Delta+, Thermofinnigan) coupled with an elemental analyser (Delta+ and Conflo, Thermofinnigan, Thermo-electron corp, Bremen) according to Haichar et al. (2007).

δ 13C (‰) was determined using the equation:

  • display math

where R = 13C/12C. The Rstandard was Pee Dee Belemnite (Wang & Hsieh, 2002).

PCR amplification of universal 16S rRNA gene fragments for denaturing gradient gel electrophoresis (DGGE) analysis

16S rRNA gene fragments were amplified from root DNA and heavy- and light-DNA fractions from RS and bulk soil using a nested PCR approach as described by Haichar et al. (2008). Briefly, the first PCR amplification step was performed using universal bacterial primers fD1 (5′-AGAGTTTGATCCTGGCTCAG-3′, position 8–27 of the Escherichia coli rrs gene) and S17 (5′-GTTACCTTGTTACGACTT-3′, position 1492–1509 of the E. coli rrs gene). For the second PCR step, PCR products were amplified using primers 375f-GC (Muyzer et al., 1993) and S10 (Supporting Information, Table S1) to generate 584 bp products for DGGE analysis. Products were checked by electrophoresis in 2% agarose gels.

PCR amplification of group-specific 16S rRNA gene fragments

PCR amplification of group-specific 16S rRNA gene fragments was applied for root DNA and heavy- and light-DNA fractions from RS of A. thaliana and bulk soil after 27 days of 13CO2 labelling. A nested PCR approach was used to amplify 16S rRNA gene fragments of Alpha-, Beta-, Gammaproteobacteria, Firmicutes and different bacterial subgroups by DGGE. This approach consisted of a first, group-specific PCR amplification of 16S rRNA gene fragments using specific primers (Table S1) followed by a 375f-GC/S10 PCR amplification for Alpha-, Gammaproteobacteria and F984GC/R1378 PCR amplification for Betaproteobacteria and Firmicutes as used for DGGE analysis. PCR amplification conditions for each group are mentioned in Table S1, and products were checked by electrophoresis in 2% agarose gels.

Reverse transcription polymerase chain reaction (RT-PCR)

Reverse transcription of RNA to complementary DNA (cDNA) was performed according to Rangel-Castro et al. (2005) using the Superscript II RNase H- reverse transcriptase kit (Invitrogen, Paisley, UK). PCR amplification of cDNA template was performed as for DNA using specific primers listed in Table S2 (see below) followed by DGGE analysis.

PCR amplification of specific genes

PCR amplification of rsmZ, nosZ and acds was performed using primers listed in Table S2. Five μl of RT-PCR product from RS 12C-, 13C- and root RNA was used as template to amplify these genes. For rsmZ, RT-PCR was carried out from total RNA extracted from RS without gradient density separation. PCR amplification conditions were the same as those for 16S rRNA amplification.

DGGE fingerprinting and principal component analysis (PCA)

DGGE analysis of PCR products from root DNA and RNA, light- and heavy-DNA and RNA fractions at each date was carried out using the Dcode Universal Mutation Detection System (BIO-Rad Laboratories, France) according to Haichar et al. (2007). Following electrophoresis, the gels were silver-stained and scanned (McCaig et al., 2001). ImageQuant TL one-dimensional gel analysis software (V2003; Amersham Biosciences, France) was used to determine band presence and intensity as described by McCaig et al. (2001). The DGGE data matrices obtained from each triplicate were analysed using PCA ADE-4 software (Thioulouse et al., 1997). Statistical ellipses representing 90% confidence intervals on PCA plots were used to compare the DGGE profiles statistically. If two ellipses representing two different treatments do not overlap, the treatments have significant different DGGE profiles with an alpha risk of 10%.

Recovery of DNA template from DGGE bands, sequencing and phylogenetic analysis

Certain bands of interest were excised, and DNA was eluted and amplified according to Haichar et al. (2007). Resulting products were analysed on an agarose gel to estimate product concentration before DGGE analysis. In most cases, as mentioned by Mahmood et al. (2005), the PCR product yielded a number of bands in addition to the band of interest, necessitating further rounds of band excision, PCR amplification and DGGE analysis to ensure purity. The correct migration positions of PCR products of purified bands were confirmed by DGGE analysis of these products and environmental PCR products on the same gel.

PCR products were sequenced at Genome express (Meylan, France), and sequences were analysed by the blastn search tool (Altschul et al., 1990) to determine sequence homology and to search for similar sequences in the GenBank database. Sequences from this study were deposited to GenBank under the accession numbers (FJ439137FJ439172) for rRNA 16S bacterial group sequences, (GQ323722GQ323729) for rsmZ, (GQ323730GQ323736) for acdS and (GQ323749GQ323766) for nosZ.

Results

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

13C enrichment of microbial communities

To confirm assimilation of 13C root exudates by soil microorganisms, DNA and RNA extracted from the rhizosphere of A. thaliana at each date were fractionated by isopycnic centrifugation, and fractions along the gradient were analysed for ∂13C and subsequently by DGGE. In all cases, two fractions were selected for SIP analysis. For DNA, a fractions from ultracentrifuge tubes corresponding to buoyant density of 1.63 g mL−1 CsCl and giving ∂13C values of −15‰, −7‰ and 2 ‰ after 13, 21 and 27 days of 13C-labelling, respectively, were selected as representative of the light-DNA fraction corresponding to microbial populations not involved in the consumption of 13C-root exudates. Fractions corresponding to buoyant density of 1.67 g mL−1 CsCl and giving ∂13C values of 46‰, 121‰ and 156‰ after 13, 21 and 27 days of 13C-labelling, respectively, were selected as representative of the heavy-DNA fraction corresponding to microbial populations involved in the assimilation of 13C-root exudates.

The same procedure was performed for extracted RNA from the RS after 13, 21 and 27 days of 13C-enrichment. The RNA density profile showed a unique peak ranging at 1.70 g mL−1 density (12C-RNA), while 13C-labelling measurement by IRMS showed the presence of labelled RNA (13C-RNA) at 1.74 g mL−1 density (Fig. 1). Similar profiles were obtained for RNA extracted from the RS at each date. 13C levels were significantly greater in heavy fractions of DNA/RNA than in light fractions of DNA/RNA values, indicating 13C-labelled incorporation into RS microbial community.

image

Figure 1. Incorporation of 13C-root exudates into RNA of microbial communities from the rhizosphere soil of Arabidopsis thaliana after 27 days of 13CO2 labelling. 12C- and 13C-RNA were separated by CsTFA density gradient centrifugation, total RNA was quantified fluorimetrically (■), and 13C values were measured by IRMS (●) within gradient fractions (1–14).

Download figure to PowerPoint

Structure and dynamics of bacterial communities associated with A. thaliana root exudates

To assess the impact of root exudates on the structure and dynamics of total bacterial community in the RS of A. thaliana, we combined RNA and DNA-SIP analysis to determine the bacterial community structure involved in the assimilation of 13C-root exudates.

DNA/RNA extracted from root tissues, fractions taken from a gradient of light DNA–heavy DNA and from a gradient of 12C-RNA-13C-RNA obtained after 13, 21 and 27 days of A. thaliana 13C-labelling were analysed by DGGE (Fig. 2). Experiments were performed in triplicate for each sampling date and similar DGGE profiles were observed with triplicates from different microcosms showing a good reproducibility of the three experiments (data not shown).

image

Figure 2. DGGE banding profiles of bacterial 16S rRNA genes amplified from fractions taken from a gradient of light-DNA (L) to heavy-DNA (H) fractions (fractions 4–11) obtained by density gradient centrifugation of extracted DNA from RS of 13CO2 labelled Arabidopsis thaliana at 13, 21 and 27 days. Black left-facing arrowheads indicate examples of bands that appeared or increased in relative intensity in heavy- or light-DNA fractions.

Download figure to PowerPoint

DGGE profiles generated from heavy- and light-DNA fractions from bulk soil showed different patterns from those generated from heavy- and light-DNA fractions from rhizosphere soil and from the roots indicating a rhizosphere impact on microbial community structure (Fig. S1).

PCA of DGGE profiles was performed for A. thaliana for root DNA and heavy- and light-DNA RS fractions and for heavy- and light-DNA bulk soil, over time and statistical ellipses were generated to compare treatments (Fig. 3).

image

Figure 3. Principal component (PC1 × PC2) plots generated from DGGE profiles of bacterial communities obtained from heavy (H), light (L) DNA fractions from bulk soil (BS) and from rhizosphere soil (RS) and root (R) DNA fractions of 13CO2 labelled Arabidopsis thaliana for (a) 13 days, (b) 21 days and (c) 27 days. Ellipses represent 90% confidence limits.

Download figure to PowerPoint

DGGE profiles from heavy-DNA fractions (fractions 8–11) of Athaliana RS after 13 days of 13C-labelling differed significantly from those of the corresponding light-DNA RS fractions (fractions 4–7) (Fig. 2) with the increase in relative intensities of some bands and the appearance of others. These differences persisted until 21 days of 13C-labelling (Fig. 2), suggesting that some specific bacterial populations were actively assimilating 13C-root exudates. However, at 27 days of 13C-labelling, significant changes in bacterial community structure were observed (Fig. 2). Indeed, some bands disappeared in favour of new ones while others increased in relative intensity suggesting succession of bacterial populations (Fig. 2).

DGGE profiles issued from root DNA differed significantly from those obtained from heavy- and light-DNA RS fractions during 13C-labelling (Fig. 3). The root tissue-colonizing bacterial community already established at 13 days was maintained over the time (Fig. 3). However, no significant changes were observed for the bulk soil. The major changes observed between compartments over time were confirmed by RNA-SIP (data not shown). Moreover, additional bacterial species were shown to be active at 27 days of 13C-labelling (Fig. S2).

Genetic structure of bacterial community after 27 days of plant 13C-labelling

Community fingerprints of four bacterial groups (Alpha-, Beta-, Gammaproteobacteria and Firmicutes) were generated from heavy- and light-DNA fractions from the rhizosphere soil and from root DNA (Fig. 4).

image

Figure 4. DGGE fingerprints of five bacterial groups (a) Alpha-, (b) Beta-, (c) Gammaproteobacteria and (d) Firmicutes amplified from light-DNA (L), heavy-DNA (H) RS fractions and root DNA after 27 days of Arabidopsis thaliana 13CO2 labelled. Black left-facing arrowheads indicate examples of bands that appeared or increased in relative intensity in heavy- or light- RS or root DNA fractions.

Download figure to PowerPoint

Alphaproteobacteria that benefited from root exudates of A. thaliana at 27 days included Devosia and Mycoplana species (Table 1). Other bacterial species, such as the diazotrophs Rhodoplanes and Afipia species, specifically colonized root tissues (Fig. S4). Whereas Betaproteobacteria (closely related to Ralstonia, Massilia and Pelomonas species) appeared to assimilate root exudates as they were retrieved only in heavy-DNA RS fraction, other bacteria were probably using soil organic matter as they were retrieved only from light-DNA RS fraction (Table 1). In contrast, bacteria closely related to Variovorax and Azospira species preferentially colonised root tissue (Fig. S5).

Table 1. Phylogenetic affiliation of partial bacterial 16S rRNA gene sequences corresponding to prominent bands from DGGE gels of Alpha-,Gamma-,Beta-proteobacteria and Firmicutes retrieved from root DNA (R), heavy (H)- and light (L)- DNA fractions of Arabidopsis thaliana plant
Phylogenetic groupBLAST closest match% of similarityCompartment/fraction
Gammaproteobacteria
 Enterobacteriales
  EnterobacteriaceaeSerratia sp.100H
Enterobacter sp.100L
 Xanthomonadales
  XanthomonadaceaeLysobacter brunescens99H
Stenotrophomonas maltophilia99H
Xanthomonadaceae bacterium99H
Lysobacter ginsengisoli99H
 Pseudomonadales
  PseudomonadaceaePseudomonas sp.99R
Pseudomonas nitroreducens99R
Alphaproteobacteria
 Rhizobiales
  HyphomicrobiaceaeUncultured Hyphomicrobiaceae bacterium98H, L, R
Rhodoplanes roseus99R
Uncultured Devosia sp.98H, L
  RhizobiaceaeRhizobium sp. Lv6.1Se99L
Sinorhizobium sp.99H
Agrobacterium sp.99R, H, L
  BradyrhizobiaceaeAfipia sp.100R
Uncultured bacterium98R, H
  MethylocystaceaePleomorphomonas koreensis100R, H, L
  MethylobacteriaceaeMirovirga sp.98L
  BrucellaceaeMycoplana dimorpha99H
 Rhodospirillales
  RhodospirillaceaeMagnetospirillum sp.100H, L
B-proteobacteria
 Burkholderiales
  OxalobacteraceaeUncultured Janthinobacterium sp.99H
Duganella sp.100R, H, L
Massilia sp.100H, L
Duganella violaceinigra99R
  ComamonadaceaeComamonadaceae bacterium98H
Comamonas sp.99L
Variovorax sp.99R
Acidovorax facilis100H
Pelomonas saccharophila99L, H
  BurkholderiaceaeRalstonia solanacearum100R, H, L
 Rhodocyclales
  RhodocyclaceaeAzospira oryzae100R
Firmicutes
 Bacillales
  BacillaceaeBacillus cereus H, R
Bacillus megaterium99L
Uncultured bacterium100L
Bacillus luciferensis99L, R
Geobacillus sp.98L
  SporolactobacillaceaeTuberibacillus calidus97H
  PaenibacillaceaePaenibacillus sp.99H

For Gammaproteobacteria, DGGE profiles generated from RS and from root tissue showed a complex pattern, suggesting a high diversity of 16S rDNA sequences (Fig. S6). Moreover, each compartment was specifically inhabited by different bacterial species. Bacteria closely related to Stenotrophomonas and Lysobacter (Table 1) appeared to assimilate root exudates in the rhizosphere, while root tissues were preferentially colonised by bacterial species closely related to Pseudomonas genus (Table 1).

Unlike Gammaproteobacteria, DGGE profiles of Firmicutes generated from RS and from root tissue were much less complex than those generated from light-DNA factions (Table 1). In the rhizosphere, bacteria closely related to Bacillus cereus, Paenibacillus and Tuberibacillus seemed to assimilate root exudates, while bacteria closely related to Bacillus and Geobacillus species did not use root-derived carbon (Table 1, Fig. S7).

Gene expression analysis of certain genes relevant for plant–bacteria interaction

We investigated the impact of root exudates assimilation on the expression of certain bacterial genes in the rhizosphere of A. thaliana. Three genes were selected: rsmZ, acdS and nosZ. The presence of these genes was investigated in light and heavy RS DNA fractions and root DNA fraction, and their expression was analysed in RS 12C-, 13C- and root RNA fractions of A. thaliana after 27 days of 13C-enrichment. PCR and RT-PCR products were analysed by DGGE.

For the noncoding regulatory RNA rsmZ, the gene fragment was effectively amplified light and heavy RS DNA fractions and root DNA fraction (Fig. 5b). However, the low yield of these small RNAs isolated from RS RNA did not allow their fractionation into 12C- and 13C-RNA fractions. For this reason, the expression of rsmZ was investigated from total RNA extracted from RS and from roots of Athaliana. rsmZ expression was detected in the RS and in root tissues (Fig. 5, Fig. S3). Certain rsmZ bands from DNA and cDNA retrieved from roots and RS were sequenced, and they showed homology to the corresponding gene of Pseudomonas species (Table 2).

image

Figure 5. (a) Presence and expression analysis of genes, rsmZ,acdS,nosZ and rrs. (b) PCR and RT-PCR were performed respectively from heavy (H), light (L) and root DNA fractions and from 12C-, 13C- and root RNA fractions. For rsmZ, RT-PCR was performed from roots RNA and from nonfractionated RS RNA.

Download figure to PowerPoint

Table 2. Phylogenetic affiliation of acdS,nosZ and rsmZ sequences corresponding to DGGE prominent bands retrieved from root DNA (R-DNA) and RNA (R-RNA), heavy (H) and light (L) DNA fractions, 12C- and 13C-RNA fractions and from total RNA extracted from RS (R-RS) for rsmZ amplification
GenesBLAST closest match% of similarityCompartment/fraction
acdSPseudomonas fluorescens strain CM1′A297H
Pseudomonas brassicacearum99H
Burkholderia caledonica100R-RNA
Pseudomonas fluorescens strain TM1A3100L
Pseudomonas strain 6G598H, L, R-RNA, R-DNA
nosZPseudomonas migulae PD 19812C-RNA
Pseudomonas fluorescens9812C-RNA
Pseudomonas brassicacearum PD 59912C-RNA
Bradyrhizobium japonicum99R-DNA
Pseudomonas brassicacearum75R-DNA, 12C-RNA
Uncultured bacterium clone P3C1595L
Uncultured bacterium clone PS369812C-RNA
Unidentified bacterium clone 7z3593L
Uncultured bacterium clone ZD1999H
Uncultured soil bacterium HJALMZE1294H
Uncultured soil bacterium clone JR125p2c2 R-DNA
Unidentified bacterium clone 19z290R-DNA
Uncultured Solibacter usitatus Ellin60769813C-RNA
rsmZPseudomonas brassicacearum98L, R-DNA, R-RNA
Pseudomonas fluorescens strain 2P2498H, L, R-DNA, R-RNA, R-RS
Pseudomonas sp. P97.3097R-RS, R-RNA

The presence of acdS was detected light and heavy RS DNA fractions and root DNA fraction (Fig. 5a), and transcripts were also detected from RS 12C-, 13C- and root RNA (Fig. S3). Sequencing of certain bands from DNA and cDNA showed homology with acdS homologues in Pseudomonas and Burkholderia species (Table 2).

The nosZ gene was amplified from RS and roots of A. thaliana, and its expression was successfully detected from RS 12C-, 13C- and root RNA (Fig. 5a, Fig. S3). Sequencing of certain bands from cDNA showed homology with nosZ homologues in uncultured bacteria, Solibacter usitatus, Bradyrhizobium japonicum and Pseudomonas species (Table 2).

Discussion

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

SIP-based approaches have found widespread application in microbial ecology and enable investigating into which microbes ‘eat what, where and when’. However, there remains a need to determine how they do this. In this study, we aimed to determine the impact of root exudates on bacterial community structure and dynamics in the A. thaliana rhizosphere as well as their influence on expression of certain bacterial genes considered as relevant in plant-bacterial interaction and rhizosphere functioning.

The ∂13C value of heavy fractions of DNA/RNA demonstrated 13C-enrichment of bacterial DNA, confirming in situ assimilation and growth at the expense of 13C-root exudates in the RS.

Bacterial community structure and dynamics in A. thaliana rhizosphere

Root exudates may activate a chemotactic response in soil bacteria and guide their motility toward rhizosphere and root tissue, hence shaping microbial community structures in this ecological niche.

Data from the present study show significant differences between DGGE profiles from heavy-DNA fractions and from light-DNA fractions of A. thaliana RS after 13 days of 13C-labelling (Figs 2 and 3), suggesting an impact of root exudates on bacterial communities. These differences persisted until 21 days of 13C-labelling (Figs 2 and 3) and changed significantly after 27 days of 13C-labelling (Figs 2 and 3). The bacterial shift observed at 27 days post-labelling may indicate that new colonizers appeared and out-competed with established populations for derived nutrients from the breakdown of root exudates. Indeed, at 27 days post-labelling, plantlets were 6 weeks old, as they had been growing for 2 weeks under 12CO2 before labelling started. The growth stage of A. thaliana at this time point corresponds to the reproductive phase. This bacterial shift is probably related to plant growth stage, as transition from vegetative to reproductive phase in plants is accompanied by a modification of carbon allocation and partitioning. Indeed, Dunfield & Germida (2003) demonstrated that the composition of the microbial community in the rhizosphere of Brassica napus changes with time in response to the changing root exudates patterns that vary during the life cycle of plants. More recent data by Houlden et al. (2008) suggest that the plant developmental stage influence microbial community structure and activity in the rhizosphere of three field crops (wheat, pea and sugar beat). Moreover, the proportion of photosynthates released in the rhizosphere and the composition of corresponding rhizodeposits have been shown to vary during the plant's life cycle according to changes in plant physiology during its development (Hamlen et al., 1972).

Significant differences were observed between DGGE profiles from root DNA and those obtained from heavy- and light-DNA fractions from RS during 13C-labelling (Fig. 3), suggesting selection of bacterial communities on root tissues prior to diffusion of lower amounts of nutrients into the rhizosphere. The bacterial community associated with root tissues at 13 days of 13C-labelling persisted over time (Fig. 3). Although the composition of root exudates may have changed over time, root-colonizing populations were maintained, which illustrates their metabolic versatility and adaptation ability.

In this study, we targeted, at 27 days of 13C-labelling, members of Alpha-, Beta-, Gammaproteobacteria and Firmicutes to detect less abundant ribotypes that were not easily detectable in universal profiles (Fig. 4). In addition, evaluating universal and group-specific fingerprints resulted in a more comprehensive approach for studying the structure of active bacterial community in the rhizosphere. Some Alphaproteobacteria assimilating root exudates were closely related to Devosia and Mycoplana species (Table 1), while some strains belonging to these species were already described to promote plant growth by improving availability of mineral nutrients to plants (Valverde et al., 2007). Certain Betaproteobacteria found in association with A. thaliana roots, such as Variovorax and Azospira species, were also shown to be specifically associated with rape roots (Haichar et al., 2008). Several Gammaproteobacteria closely related to Stenotrophomonas, Pseudomonas and Lysobacter species (Table 1), which appeared to assimilate root exudates in the rhizosphere of A. thaliana, have also been found to assimilate root exudates in the rhizosphere of maize, wheat and rape (Haichar et al., 2008).

Transcription analysis of functional bacterial genes involved in plant-bacteria interactions

In the rhizosphere, plant growth-promoting rhizobacteria (PGPR) are expected to increase plant growth by producing phytohormones and antimicrobial metabolites against phytophatogens, solubilizing minerals such as phosphate, regulating ethylene production by roots and decreasing heavy metal toxicity (Whipps, 2001). However, these functions occur only if the corresponding genes are expressed or have been expressed, under natural conditions. To investigate the impact of the assimilation of root exudates on bacterial gene expression, we developed the technique of mRNA-SIP. Three genes were selected: the noncoding regulatory RNA rsmZ involved in the activation of secondary metabolism production, acdS encoding ACC deaminase involved in ethylene metabolism and nosZ encoding for dissimilatory bacterial nitrous oxide reductase.

Antimicrobial properties of rhizosphere bacterial community

In Pseudomonas spp., the GacS/GacA two-component system activates the expression of three small regulatory RNAs (rsmX, rsmY and rsmZ) and thereby counters translational repression exerted by the RsmA and RsmE proteins on target mRNAs required for the synthesis of secondary metabolites (Lapouge et al., 2008) such as 2,4-diacetylphloroglucinol and hydrogen cyanide (HCN) components that may reduce harm caused by phytopathogenic fungi and therefore act as biopesticides (Haas & Keel, 2003).

It is more advantageous for bacteria that are exposed to sudden environmental changes to use the noncoding regulatory RNAs (ncRNAs) as they are less costly to the cell and faster to produce, do not require the extra step of translation and may regulate many different genes. The expression of rsmZ was detected on roots and rhizosphere compartments (Fig. 5b), suggesting the importance of this regulatory ncRNA in the bacterial adaptation to its ecological niche.

Plant growth promotion properties of rhizosphere bacterial community

Ethylene is an important hormone that profoundly alters root growth and development in plants. Reduction of ethylene concentration by microbial deamination of the ethylene precursor ACC (Glick et al., 1998) has been identified as a phytobeneficial mode of action exhibited by certain PGPR and nitrogen-fixing bacteria (Ma et al., 2003). Bacterial ACC deaminase lowers the amount of ethylene synthesized by the plant, and thus reduces the inhibitory effect of this phytohormone on root elongation. In addition, bacteria may use products of ACC hydrolysis as nutrients for growth (Glick et al., 1998). In fact, they take up the ethylene precursor ACC and convert it into 2-oxobutanoate and NH3. The acdS gene was present in bacterial populations inhabiting RS and roots. acdS transcripts were detected from 12C-, 13C- and root RNA. However, only a weak transcription was observed with organic matter-degrading bacteria compared to root exudates assimilating bacteria in the rhizosphere and root compartment. It has been postulated that much of the ACC produced by ACC synthase activity in plant roots may be exuded in the rhizosphere (Contesto et al., 2008). Bacteria that benefit from root exudates may incorporate ACC that probably induces ACC deaminase expression and consequently promote plant root elongation.

Denitrification in the rhizosphere of A. thaliana

Denitrification is mediated by physiologically diverse groups of prokaryotes. Analysis of their activity is essential for understanding the ecosystem-level controls on the biogeochemical process of denitrification. The nosZ gene that catalyses conversion of nitrous oxide (N2O) into dinitrogen (N2) under suboxic conditions was carried by bacterial community inhabiting the rhizosphere soil and the roots of A. thaliana, and it was expressed by a bacterial communities present in RS and root tissue (Fig. 4). This indicates that denitrifying bacteria are actively expressing nosZ in the rhizosphere. These findings support and improve pervious knowledge of the impact of plants on the activity and diversity of denitrifiers (Philippot et al., 2006; Henry et al., 2008). Interestingly, nosZ expression was observed in organic matter-degrading bacteria. Until recently, denitrification activity in the rhizosphere was positively correlated with the presence of easily mineralizable carbon such as sugar, organic acids and amino acids (Henry et al., 2008). In this study, we showed that in addition to root-derived carbon, soil organic matters do influence denitrification.

Conclusion

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

DNA-SIP in the rhizosphere allows detailed descriptions of microbial community structures and to identify populations involved in root-derived carbon assimilation from those degrading soil organic matter. This allows us to formulate hypotheses on the relative contributions of different groups of microbes to rhizosphere functioning. Additionally, using mRNA-SIP technique to investigate gene expression shows great promise for understanding the impact of the plant via roots exudation in controlling bacterial gene expression. In the future, combining mRNA-SIP with qRT-PCR method will provide quantitative data, which will allow to better describing the transcriptional activity of the studied genes.

Nonetheless, our recognition of the extraordinary microbial species diversity in the rhizosphere highlights the need for innovative and new experimental techniques to understand the signalling interactions between microorganisms and between plants and microorganisms. Moreover, understanding the physiological relevance and ecological significance of noncoding regulatory RNA is an exciting challenge for environmental microbiology.

Acknowledgements

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

We are grateful to Groupe de Recherche Appliquées en Phytotechnologies, CEA Cadarache, for labelling facilities, and C. Marol, for 13C analysis. This work was supported by MICROGER program and CNRS PhD grant (PDI-PED).

References

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

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. Acknowledgements
  9. References
  10. Supporting Information
FilenameFormatSizeDescription
fem1345-sup-0001-FigureS1.pdfapplication/PDF1977KFig. S1.DGGE banding profiles of bacterial 16S rRNA genes amplified from fractions taken from a gradient of light DNA (L) to heavy DNA (H) fractions of extracted DNA from bulk soil and from rhizosphere soil of 13CO2 labelled A. thaliana at 13, 21 and 27 days.
fem1345-sup-0002-FigureS2.pdfapplication/PDF784KFig. S2.DGGE banding profiles of bacterial 16S rRNA genes RT-PCR amplified from 12C- and 13C-RNA issued from rhizosphere soil of A. thaliana after 13CO2 labelling for 27 days.
fem1345-sup-0003-FigureS3.pdfapplication/PDF257KFig. S3.DGGE patterns of nosZ,acdS and rsmZ gene fragments amplified by RT-PCR of RNA extracted form root tissues (R), from 12C- (L) and 13C-RNA (H) rhizosphere soil fractions and from rhizosphere soil (RS) RNA from the rhizosphere of A. thaliana after 27 days of 13CO2 labelling.
fem1345-sup-0004-FigureS4.pdfapplication/PDF155KFig. S4. Phylogenetic tree constructed from rrs genes of Alphaproteobacteria sequences retrieved from DGGE profiles of heavy (H), light (L) and root (R) DNA fractions.
fem1345-sup-0005-FigureS5.pdfapplication/PDF142KFig. S5. Phylogenetic tree obtained from rrs genes of Betaproteobacteria sequences retrieved from DGGE profiles of heavy (H), light (L) and root (R) DNA fractions.
fem1345-sup-0006-FigureS6.pdfapplication/PDF123KFig. S6. Phylogenetic tree obtained from rrs genes of Gammaproteobacteria sequences retrieved from DGGE profiles of heavy (H), light (L) and root (R) DNA fractions.
fem1345-sup-0007-FigureS7.pdfapplication/PDF130KFig. S7. Phylogenetic tree obtained from rrs genes of Firmicutes sequences retrieved from DGGE profiles of heavy (H), light (L) and root (R) DNA fractions.
fem1345-sup-0008-TableS1.docWord document46KTable S1. Primers and PCR conditions used in this study to target 16S rRNA genes.
fem1345-sup-0009-TableS2.docWord document28KTable S2. Primers used in this study for targeting functional genes.

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