Abstract
Microbial structural and expression profiles of the rhizospheres of three legumes, faba beans, peas and white lupin, were compared by RNA-arbitrarily primed PCR technique. Two different primers, M13 reverse and 10-mer primers, were used in the amplification and products resolved on non-denaturing polyacrylamide gel. With both DNA and RNA profiles Lupinus and Pisum rhizospheres were more similar to each other than to Vicia rhizosphere. The RAP-PCR products were also dot blotted and probed for bacterial peptidase transcripts. Plant-dependent rhizosphere effect was evident by the marked absence of transcripts for bacterial neutral metallopeptidase in Lupinus rhizosphere. The results of dot blot were further confirmed by RT-PCR for the expression of bacterial neutral metallopeptidase in the three rhizospheres.
1 Introduction
The cultivation of legumes in crop rotation practice is an important way to improve soil quality in agricultural soil. Legumes possess all the characteristic properties of green manures like improving soil by adding large amounts of organic material and valuable nutrients to it, preventing soil erosion, preventing the leaching of nutrients and improving soil's ability to hold water. Most important, however, is their contribution to the overall N economy of soil by sequestering atmospheric N through symbiotic N2-fixation and through subsoil N retrieval [1].
Plant roots influence soil borne microbial communities via several mechanisms, including excretion of organic compounds, competition for nutrients, and providing a solid surface for attachment. The nature of this influence is highly variable and depends upon both the amount and the composition of organic materials released by the plants [2,3]. Numerous studies reveal differences in rhizodeposition by different grain legumes [4,5,6,7,8] thereby possibly influencing their respective rhizosphere structure and function. Recent study by Sharma et al. [9], addressing the effect of legume rhizodeposition on bacterial communities, showed a distinct plant-dependent rhizosphere effect on the distribution of different bacterial groups present in legume rhizospheres. However, reports on plant-dependent rhizosphere effect on microbial community functions are limited.
Changes in the active microbial communities may lead to changes in the functions performed by the community. Direct extraction of mRNA from soil [10] and quantification of mRNA by an RNase protection assay [11] have been used for naphthalene dioxygenase in soil and for soluble methane monooxygenase in aquifer sediments [12]. Reverse transcriptase-PCR (RT-PCR) amplification of mRNA for soluble methane monooxygenase in aquifer sediments [13] and for lignin peroxidase in soils [14] has also been performed. One of the major prerequisites for such methods is knowledge of the sequence flanking the target region for the design of amplification primers for PCR. This limits the choice of primers and thus methods that bypass the requirement of prior sequence information are needed to investigate the functional changes in the rhizosphere microbial communities in response to changes in the root exudates.
DNA profiles can be generated using the Arbitrarily Primed-PCR (AP-PCR) where an arbitrarily chosen primer is used in the PCR [15]. Aneja et al. [16] described the non-radioactive application of RNA Arbitrarily Primed (RAP)-PCR protocol to investigate the metabolic profiles of microbial community involved in leaf litter degradation without using selective primer systems for PCR. Both mRNA and rRNA transcripts could be targeted in the described protocol, which has the advantage of simultaneous analysis of both active taxa (rRNA) and expression of functional genes (mRNA). This technique is useful if no prior knowledge is available about functional genes of the microbial community to be analysed.
The main objective of the present study was to identify plant dependent rhizosphere effects on the DNA and RNA pool in the rhizospheres of three grain legumes using the described AP-PCR and RAP-PCR technique.
2 Materials and methods
2.1 Soil samples
Rhizosphere samples from the previous experiments by Sharma et al. [9] were used in the present study.
2.2 Nucleic acid extraction
Nucleic acid extraction from the rhizosphere soil material was performed using the method of co-extraction of DNA and RNA described by Griffiths et al. [17] with a few modifications as described elsewhere [9]. The method principally involved bead beating and solvent extraction of the nucleic acids. To prevent the degradation of RNA, all incubations were performed on ice. To obtain pure DNA, incubation at 37 °C with RNaseA (Sigma, Munich, Germany) at a final concentration of 100 μg ml−1 for 10 min, was performed. To obtain pure RNA, DNA was removed from RNA by treatment with DNase (1 U μl−1; RNase free; Promega) according to manufacturer's instructions.
2.3 cDNA synthesis
RNA was reverse transcribed using the Omniscript RT Kit (Qiagen GmbH, Hilden, Germany). Two microlitres of total-RNA sample was added to an 18 μl RT mixture containing 20 pmole of random hexamers (RT mixture prepared following the manufacturer's instructions). The reaction was incubated at 37 °C for 90 min, then heated at 93 °C for 5 min for enzyme inactivation followed by rapid cooling on ice. The synthesized cDNA was stored at −20 °C.
2.4 AP-PCR and RAP-PCR
DNA and RNA fingerprints for the three rhizosphere soil samples were generated using AP-PCR and RAP-PCR methods. Two primers were used to generate the fingerprints; M13 reverse primer, 5′-CAGGAAACAGCTATGAC-3′ and a 10-mer primer, 5′-TCACGATGCA-3′ described by Williams et al. [15]. Amplification reactions were performed in volumes of 25 μl containing 2.5 μl of 10X reaction buffer, 0.5 μl of 100 pmole primer (M13 reverse or 10 mer), 1.25 μl of 2 mM deoxyribonucleoside triphosphate mixture, 1 μl of 50 mM MgCl2, 2.5 μl of 3% bovine serum albumin, 1.25 μl of dimethyl sulfoxide, 1 μl of DNA/cDNA, 14.75 μl of water and 1.25 U of Taq DNA Polymerase (5 U μl−1, Invitrogen, Karlsruhe, Germany). Thermal cycler was programmed as: 94 °C for 5 min to denature, 37 °C for 5 min for low stringency annealing of the primer and 72 °C for 5 min for extension for 2 cycles. This was followed by 45 cycles of 1 min at 94 °C, 1 min at 37 °C, 2 min at 72 °C. Final extension was given at 72 °C for 10 min followed by cooling at 4 °C.
Controls to rule out the possibility of DNA contamination in the RNA preparation were performed by using DNase treated but not reverse transcribed (DNase+RT−) nucleic acids in the assays.
2.5 Polyacrylamide gel electrophoresis
Non-denaturing polyacrylamide gels (8% with a 29:1 ratio of acrylamide to bisacrylamide) were prepared as described by Sambrook et al. [18]. Appropriate volumes containing about 2 μg of AP-PCR and RAP-PCR products, measured by absorbance at 260 nm, were loaded. The gels were electrophoresed at 50 V for 17 h using D-Gene system (Bio-Rad Laboratories, Munich, Germany). The gels were silver stained using the protocol described by Heukeshoven and Dernick [19]. Dried gels were scanned using HP Scanjet 7400c. Profiles obtained were analysed by clustering via the unweighted pair group method with mathematical averages (UPGMA; Dice coefficient of similarity) using GelCompar II Software (Applied Maths, Kortrijk, Belgium). The position tolerance was set at 1% and background subtraction was applied. Both strong and weak bands were included in the analysis, thus taking into account only the presence and absence of bands at specific positions. Cophenetic correlation values, which are a parameter to express the consistency of acluster, were calculated using the same software.
2.6 Detection of bacterial peptidase genes and transcripts
Dot blot hybridization for bacterial serine peptidases (sub) and neutral metallopeptidases (npr) was performed on positively charged nylon membranes (Roche Diagnostics, Mannheim, Germany) using 10 μl of RAP-PCR products according to the protocol described by Bach et al. [20]. PCR products for sub and npr genes of Bacillus cereus were generated using the protocol described by Bach et al. [20] and used as controls for specificity of the probes.
PCR and RT-PCR targeting npr, using primers and protocol described by Bach et al. [20], were used to check for the presence of the gene and its expression. Amplification products were resolved on 1.5% agarose gels in 1X TAE buffer and detected by ethidium bromide staining.
3 Results
3.1 AP-PCR and RAP-PCR with M13 reverse and 10-mer primers
Rhizosphere soil DNA and cDNA were used to generate fingerprints using M13 reverse primer and 10-mer primers in separate reactions. Multiple nucleic acid extractions from the same plant rhizosphere sample generated AP-PCR and RAP-PCR profiles that were more than 98% similar to each other (data not shown). AP-PCR and RAP-PCR profiles of 30 plants for each rhizosphere soil type (ten pots per legume and three plants per pot) were compared to screen for sampling variations. No significant differences between the profiles of the three plants in each pot and also for the ten different pots for each legume, could be observed (>95% similar; data not shown). Furthermore, clear plant-dependent effects on the microbial structural and transcription profiles were visible.
Using the M13 reverse primer, a clear separation between DNA and cDNA profiles was visible, forming the two major clusters. DNA profiles of Pisum and Lupinus rhizospheres were more than 90% similar to each other. The similarity of both Pisum and Lupinus DNA fingerprints to Vicia DNA profile was 70%. The number of bands for all DNA fingerprints was similar (10–13 bands) independent of the plant species. In contrast, the obtained RNA patterns were highly variable between the three legumes, with a maximum homology of 50%. When compared to the respective DNA profiles, differences could be observed both in the number and position of bands. Interestingly, also for the RNA profiles, Lupinus and Pisum were closer than Lupinus and Vicia or Pisum and Vicia. The data are summarized in Fig. 1.
AP-PCR and RAP-PCR products with the M13 reverse primer resolved on non-denaturing polyacrylamide gel. Lanes 1–3 –Vicia, Lupinus and Pisum DNA fingerprints with the M13 reverse primer, lanes 4–6 –Vicia, Lupinus and Pisum cDNA fingerprints with M13 reverse primer. UPGMA tree representing the similarity of the microbial community structure and expression profiles obtained by AP-PCR and RAP-PCR using M13 reverse primer from the three rhizospheres. Scale represents percent similarity. Number at branch point shows the cophenetic correlation value, which is a parameter to express the consistency of a cluster.
AP-PCR and RAP-PCR products with the M13 reverse primer resolved on non-denaturing polyacrylamide gel. Lanes 1–3 –Vicia, Lupinus and Pisum DNA fingerprints with the M13 reverse primer, lanes 4–6 –Vicia, Lupinus and Pisum cDNA fingerprints with M13 reverse primer. UPGMA tree representing the similarity of the microbial community structure and expression profiles obtained by AP-PCR and RAP-PCR using M13 reverse primer from the three rhizospheres. Scale represents percent similarity. Number at branch point shows the cophenetic correlation value, which is a parameter to express the consistency of a cluster.
Similar to results with the M13 reverse primer, DNA and RNA fingerprints obtained using 10-mer primer for the same rhizosphere soil differed significantly and formed the main clusters. Higher number of discreet bands was obtained in the high molecular weight region. Interestingly, the variability between DNA profiles, when the three legume rhizospheres were compared, was higher compared to the profiles obtained with M13 reverse primer. However, the pattern of similarity was the same with high homologies between Lupinus and Pisum patterns and increased differences between the two and Vicia. The data are summarized in Fig. 2.
AP-PCR and RAP-PCR products with 10-mer primer resolved on non-denaturing polyacrylamide gel. Lanes 1–3 –Vicia, Lupinus and Pisum DNA fingerprints with 10-mer primer, lanes 4–6 –Vicia, Lupinus and Pisum cDNA fingerprints with 10-mer primer. UPGMA tree representing the similarity of the microbial community structure and expression profiles obtained by AP-PCR and RAP-PCR using 10-mer primer from the three rhizospheres. Scale represents percent similarity. Number at branch point shows the cophenetic correlation value, which is a parameter to express the consistency of a cluster.
AP-PCR and RAP-PCR products with 10-mer primer resolved on non-denaturing polyacrylamide gel. Lanes 1–3 –Vicia, Lupinus and Pisum DNA fingerprints with 10-mer primer, lanes 4–6 –Vicia, Lupinus and Pisum cDNA fingerprints with 10-mer primer. UPGMA tree representing the similarity of the microbial community structure and expression profiles obtained by AP-PCR and RAP-PCR using 10-mer primer from the three rhizospheres. Scale represents percent similarity. Number at branch point shows the cophenetic correlation value, which is a parameter to express the consistency of a cluster.
3.2 Detection of functional genes in the RAP-PCR products using specific probes
RAP-PCR products with M13 reverse primer and 10-mer primer were transferred to positively charged nylon membrane and probed for bacterial peptidase (serine peptidase and neutral metallopeptidase) transcripts using DIG labelled probes. It was observed that while serine peptidase was expressed in all the three rhizosphere soils (Fig. 3(a)), neutral metallopeptidase could be detected in Vicia and Pisum rhizospheres only (Fig. 3(b)).
(a) Dot blot for serine peptidase. Lanes 1–3 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with M13 reverse primer, lanes 4–6 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with 10-mer primer, lane 7 – positive control: PCR product for sub gene amplified from Bacillus cereus, lane 8 – negative control: PCR product for npr gene amplified from Bacillus cereus. (b) Dot blot for neutral metallopeptidase. Lanes 1–3 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with M13 reverse primer, lanes 4–6 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with 10-mer primer, lane 7 – positive control: PCR product for npr gene amplified from Bacillus cereus, lane 8 – negative control: PCR product for sub gene amplified from Bacillus cereus.
(a) Dot blot for serine peptidase. Lanes 1–3 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with M13 reverse primer, lanes 4–6 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with 10-mer primer, lane 7 – positive control: PCR product for sub gene amplified from Bacillus cereus, lane 8 – negative control: PCR product for npr gene amplified from Bacillus cereus. (b) Dot blot for neutral metallopeptidase. Lanes 1–3 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with M13 reverse primer, lanes 4–6 – RAP-PCR products of Vicia, Lupinus and Pisum rhizosphere RNA with 10-mer primer, lane 7 – positive control: PCR product for npr gene amplified from Bacillus cereus, lane 8 – negative control: PCR product for sub gene amplified from Bacillus cereus.
To confirm the absence of neutral metallopeptidase transcripts in the Lupinus rhizosphere, RT-PCR using specific primers for the corresponding gene was performed. The results clearly indicate the absence of npr specific mRNA in the rhizosphere of Lupinus. A distinct band of the size of 230 bp was visible for Vicia and Pisum; no band of the expected size could be detected in Lupinus rhizosphere. To show that the effect was on the expression of the gene and not on its presence in the gene pool of the microbial community, DNA from the three rhizospheres was also PCR amplified using the same primers. To rule out the possibility of DNA contamination in the RNA preparations, DNase+RT− samples for all the three rhizosphere soils, were included in the assay (Fig. 4).
PCR and RT–PCR image showing the lack of expression of bacterial neutral metallopeptidase in Lupinus rhizosphere soil. lane M – 100 base pair ladder, lanes 1–3 –Vicia, Lupinus and Pisum DNA, lanes 4–6 –Vicia, Lupinus and Pisum DNase+RT− nucleic acids, lanes 7–9 –Vicia, Lupinus and Pisum cDNA, lane 10 – negative control.
PCR and RT–PCR image showing the lack of expression of bacterial neutral metallopeptidase in Lupinus rhizosphere soil. lane M – 100 base pair ladder, lanes 1–3 –Vicia, Lupinus and Pisum DNA, lanes 4–6 –Vicia, Lupinus and Pisum DNase+RT− nucleic acids, lanes 7–9 –Vicia, Lupinus and Pisum cDNA, lane 10 – negative control.
4 Discussion
RAP-PCR opens the possibility to analyse samples where no prior knowledge is available about functional genes of the microbial community [16]. Using random primers, profiles can be generated, which enable screening of complex communities for similarities and differences. The present study is the first report of generation of metabolic profiles of soil using RAP protocol.
On comparing the DNA and RNA profiles, major differences between the total DNA and total RNA profiles of the same rhizosphere were revealed. When the three rhizospheres were compared, more differences were observed with respect to their transcription profiles than their DNA-based structural profiles. This observation highlights the fact that though the three different rhizospheres are more similar in their genetic potential (Lupinus and Pisum microbial communities are more than 90% similar to each other and these two are 70% similar to the community of Vicia), as revealed by their DNA based profiles with M13 reverse primer, they differ considerably in their metabolic profiles (Lupinus and Pisum microbial communities are only about 50% similar to each other and these two are about 45% similar to the community of Vicia). A similar trend is observed with AP-PCR and RAP-PCR profiles generated using the 10-mer primer though the proportion of differences and similarities vary. These observations can be attributed to qualitative and quantitative differences between the rhizodeposits of the three legumes as reported by Mayer et al. [8]. Our study provides evidence that differences in rhizodeposition may not necessarily translate to differences in microbial community structure but may affect the transcription profiles.
With each primer, two different clusters were obtained, one comprising of profiles generated by DNA and the other by RNA. In each cluster, Lupinus and Pisum rhizospheres were more similar to each other than to Vicia rhizosphere. In contrast, using the same rhizosphere soil described in this study, Sharma et al. [9] have reported Vicia and Lupinus 16S profiles to be more similar to each other than to Pisum rhizosphere. 16S analysis takes into account only bacterial population excluding other members of the rhizosphere (e.g. fungi, archeabacteria) which may be equally or more important. However, in present study, by taking the total RNA, we do not exclude any members of the rhizosphere microbial community.
Plant dependent rhizospheres effects on the production and activity of enzymes have also been previously reported, e.g., higher levels of both alkaline and acid phosphatase in the rhizosphere as compared to the bulk soil have been observed for various crop species by Tarafdar and Jungk [21]. Knauff et al. [22] reported higher activity of arylsulfatase in rhizospheres of Sinapis album, Lolium perenne, Triticum aestivum and Brassica napus. Their expression was shown to be influenced by crop species. Rhizosphere effect was evident in the present study too, but at the transcriptional level when bacterial neutral metallopeptidase was investigated in the three rhizospheres. It was observed that though transcripts for serine peptidase could be detected for all the three samples, no neutral metallopeptidase transcripts could be detected in Lupinus rhizosphere. However, a very low level of expression, which is below the kit's detection limit (0.1 pg), cannot be ruled out. Absence of neutral metallopeptidase from Lupinus rhizosphere was further confirmed by RT-PCR using specific primers. Detection of serine peptidase transcripts in the three rhizospheres and neutral metallopeptidase transcripts in Vicia and Pisum rhizospheres minimize the possibility of degradation of mRNA during extraction and subsequent processing. In a study aimed at estimating the nitrogen rhizodeposition of grain legumes, Mayer et al. [8] reported that faba beans, peas and lupin did not differ significantly when amount of nitrogen derived from rhizodeposition (NdfR) was measured. In a related study by Mayer et al. [23], where the turnover of Ndfr was followed, it was observed that in lupin the value was 21% whereas for faba beans and peas this value was 26% and 27%. Lack of neutral metallopeptidase expression could be one of the factors for the observed reduced mineralization in lupin. For better understanding of the ecological significance of this observation, additional studies combining techniques like FISH (fluorescence in situ hybridization) and in situ-PCR are needed to correlate the structural and functional aspect of the same.
Acknowledgements
The study was supported by a research Grant mu 831/10-1 from Deutsche Forschungsgemeinschaft (DFG), Bonn, Germany.
