SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Here we report on the presence of sulfated lipopolysaccharide molecules in Azospirillum brasilense, a plant growth-promoting rhizosphere bacterium. Chemical analysis provided structural data on the O-antigen composition and demonstrated the possible involvement of the nodPQ genes in O-antigen sulfation.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Azospirillum sp. are the model organisms for associative plant growth promotion (Steenhoudt and Vanderleyden, 2000; Somers et al., 2004). Azospirillum enhances plant growth via an efficient interaction with the plant roots: the bacterium colonizes the plant root surface and stimulates plant root proliferation via the production of auxins, resulting in enhanced plant root exudation.

The plant root colonization process involves a number of features, present on the cell surface of Azospirillum, that are essential for bacterial adhesion. These features include cell surface proteins, such as the polar flagellin and major outer membrane protein, and cell surface polysaccharides, such as exopolysaccharides (EPS) and lipopolysaccharides (LPS). As major components of the cell surface, these polysaccharides were shown to play a role in host recognition and in plant root attachment (Michiels et al., 1991; Skvortsov and Ignatov, 1998; Matora et al., 2001).

Knowledge on the composition of LPS, and in particular the O-repeating unit, in Azospirillum species is increasing (Fedonenko et al., 2002; 2004; 2005), but genetic information regarding LPS biosynthesis is still limited. Sequencing and annotation of the pRhico plasmid of Azospirillum brasilense Sp7 revealed a high number of genes, possibly involved in LPS biosynthesis (Vanbleu et al., 2004). This information will help to determine, via mutational and chemical structural analysis, the LPS biosynthetic pathway in Azospirillum and to identify specific epitopes involved in plant root colonization.

Two of the genes, encoded on the pRhico plasmid, are similar to nodPQ genes in other plant-associated bacteria. NodPQ homologues have been identified as proteins forming a sulfate-activating complex, which converts sulfate together with ATP into an activated sulfur donor, which is then used for incorporation in different pathways (Schwedock et al., 1994). In many rhizobial strains, different copies of nodPQ homologues are present with a distinct function (Snoeck et al., 2003): cysD and cysN convert sulfate and ATP into adenosine-5′-phosphosulfate (APS), which is used in the biosynthesis of the amino acids cysteine and methionine, whereas the actual nodPQ genes function in sulfation of Nod factors (bacterial signalling molecules) via production of 3′-phosphoadenosine-5′-phosphosulfate (PAPS). In addition, nodPQ homologues in Sinorhizobium meliloti (Cedergren et al., 1995) and Xanthomonas oryzae pv. oryzae (Shen et al., 2002) were shown, respectively, to be required for sulfation of LPS molecules and to be involved in avirulence activity.

In Azospirillum, however, the function of the nodPQ genes is unclear. Previous study on an A. brasilense nodPQ mutant showed no significant growth difference or change in N2 fixation or IAA production as compared with the wild type. Also, Azospirillum does not produce Nod factors and is not auxotroph after deletion of the nodPQ genes (Vieille and Elmerich, 1990). But the presence of the nodPQ genes in a region on the pRhico plasmid, carrying polysaccharide biosynthesis genes, made us postulate that they are involved in (lipo)polysaccharide biosynthesis (Vanbleu et al., 2004). Moreover, analysis of the phylogenetic relationship among nodP homologues showed the A. brasilense nodP gene to cluster in a single group together with nodP of Bradyrhizobium elkanii. The other nodP homologues, functioning in Nod factor sulfation and amino acid biosynthesis, clearly belonged to distinct clusters (Snoeck et al., 2003). This supports the idea that the nodP homologue in A. brasilense belongs to another functional group.

To confirm this hypothesis, the LPS structure of A. brasilense Sp7 was subjected to chemical analysis. The analysis provided data on the glycosyl composition of the O-repeating unit and demonstrated the presence of sulfate groups on specific sugar moieties. In the A. brasilense nodPQ mutant 7803 (Vieille and Elmerich, 1990), LPS sulfation is drastically reduced.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Purification of A. brasilense LPS

Lipopolysaccharides of the A. brasilense strains were extracted via the hot phenol/water procedure. Deoxycholate-polyacrylamide gel electrophoresis (DOC-PAGE) analysis and gas chromatography-mass spectrometry (GC-MS) analysis of the trimethylsilyl (TMS) methyl glycosides from the water and phenol phase revealed the presence of LPS only in the water layer (data not shown). Further analysis was hence performed on this phase.

In order to remove RNA, DNA and/or protein contaminants, the water layer was treated with RNase, DNase I and proteinase K. Deoxycholate-PAGE analysis of purified LPS of A. brasilense wild-type Sp7 and mutant nodPQ revealed a different band pattern (Fig. 1), possibly indicating a difference in LPS composition between wild type and mutant strain.

image

Figure 1. Deoxycholate-polyacrylamide gel electrophoresis (DOC-PAGE) analysis of purified LPS of Azospirillum brasilense Sp7 (lane 1) and nodPQ mutant (lane 2).

Download figure to PowerPoint

An attempt was made to separate these distinctive bands via size-exclusion chromatography on a Sephacryl S400 column, DOC-equilibrated. However, no separation of LPS bands could be obtained (data not shown).

Separation and glycosyl analysis of the oligosaccharide fraction of A. brasilense LPS

To further analyse the oligosaccharide moiety of the A. brasilense LPS molecule, LPS molecules were subjected to mild acetic acid cleavage. The precipitating lipid A was removed by centrifugation and the remaining oligosaccharide was lyophilized before further analysis. The oligosaccharide fractions of A. brasilense Sp7 wild type and nodPQ mutant strain were separated via gel filtration chromatography on a BioGel P30 column. After gel filtration chromatography, two major oligosaccharide portions were detected for both the wild type and the mutant strain. These two fractions were subjected to glycosyl composition analysis as well as glycosyl linkage analysis.

The glycosyl composition analysis (GC-MS-TMS) of the different oligosaccharide fractions of the two different strains is presented in Table 1. According to the glycosyl composition, the overall sugars (rhamnose, fucose, galactose) are present in both the wild type and the mutant strain. A difference in sugar ratios is apparently present between wild type and mutant strain but this difference is most likely due to the mode of construction of the nodPQ mutant 7803 (Vieille and Elmerich, 1990): the nodPQ genes were knocked out via deletion of an area comprising not only the nodPQ genes but also a region, upstream of nodPQ, containing in particular a glycosyl transferase gene. Alditol acetate derivatives (data not shown) demonstrated the presence of two types of rhamnose residues, non-methylated rhamnose and rhamnose methylated at the two positions. Moreover, compared with the parent the oligosaccharide from the mutant strain contains increased levels of heptose (Table 1). This might be due to a truncation of the LPS molecule in the nodPQ strain, leading to increased levels of the sugar residues present towards the reducing end of the LPS molecule, e.g. due to a relatively increased amount of core glycosyl residues. However, as the current methods used for composition analysis would not detect phosphorylated glycosyl residues, there could be a difference between the mutant and parent strain in the level of heptose phosphorylation. Therefore, if the parent oligosaccharide contained a significant level of phosphorylated heptose while the mutant heptose lacked phosphate substitution, the level of heptose in the mutant would appear larger than in the parent.

Table 1.  Glycosyl composition (mole ratios) of Azospirillum brasilense wild-type Sp7 and nodPQ mutant LPS.
Glycosyl residueA. brasilense Sp7 wild typeA. brasilense nodPQ mutant
Fraction 1Fraction 2Fraction 1Fraction 2
  1. Ratios are normalized to GlcNAc and are determined from the GC-MS total ion current peak areas with response factor correction to authentic standards. Mole ratios of KDO or 2-Me-rhamnose are not shown. The rhamnose ratio in the nodPQ mutant is slightly overestimated as residual ribose was present in the nodPQ LPS extract.

Rhamnose5.62.01.01.4
Fucose4.32.51.10.8
Xylose0.60.50.40.1
Mannose0.30.10.1 
Galactose1.60.70.50.7
Glucose0.1   
Heptose0.1 0.50.2
GalNAc 0.2  
GlcNAc1111

The glycosyl linkages of the oligosaccharide fractions from the wild-type Sp7 and mutant nodPQ strain are presented in Fig. 2. The major sugar linkages present in both wild type and mutant strain are 3-linked rhamnose and 3-linked galactose, with deoxysugars (fucose or rhamnose) as terminal sugar moieties and a 1:1 ratio of 4-linked xylose in the pyranose form (or 5-linked xylose in the furanose form) and a doubly linked 3,4-fucose residue. The major differences between Sp7 wild type and nodPQ mutant are the presence of 2,3-linked rhamnose and 3,6-linked galactose in the wild-type strain. Doubly linked galactose and rhamnose residues are both lacking in the mutant strain. Moreover, the amount of 3-linked galactose present in the wild type is less than in the nodPQ mutant strain. These differences possibly reflect the presence of a substituent on the rhamnose and galactose moieties in the wild-type strain, which is missing in the nodPQ mutant strain.

image

Figure 2. Glycosyl linkage analysis of the permethylated alditol acetate (PMAA) derivatives from Azospirillum brasilense Sp7 wild type and nodPQ mutant oligosaccharide fraction 1. The peaks are as indicated in the GC profiles. Those peaks labelled * are non-carbohydrate contaminants. Sugars in bold print represent major differences between both strains. t, terminal sugar residue; f, furanose form; p, pyranose form.

Download figure to PowerPoint

Sulfate analysis of A. brasilense Sp7 wild type and nodPQ mutant oligosaccharides

In order to determine the presence of sulfate residues on A. brasilense LPS molecules, the oligosaccharide fractions 1 of both wild type and mutant strain were subjected to sulfate determination via pyrolysis and anion exchange chromatography (Glycotech, University of California, San Diego, CA, USA). The wild-type strain contained 0.605 nM S per mg sample while the nodPQ strain showed 0.145 nM S per mg sample. Thus, LPS of A. brasilense Sp7 wild-type strain contain sulfate as a substituent in its oligosaccharide portion, while the oligosaccharide fraction of the nodPQ mutant has more than four times less sulfate incorporated.

Both oligosaccharide fractions 1 of wild type and mutant strain were then subjected to a desulfation reaction, performed in dimethyl sulphoxide (DMSO) containing methanol and subsequently analysed by conversion into their permethylated alditol acetate (PMAA) derivatives.

Interestingly, the desulfated oligosaccharide fraction 1 from the wild-type strain no longer contains the doubly linked sugars 3,6-galactose and 2,3-rhamnose (Fig. 3). Both these sugar linkages were also absent in the non-desulfated oligosaccharide fraction 1 of the nodPQ mutant strain (Fig. 2). This strongly suggests the presence of sulfate groups on the 6-position of galactose residues and the 2-position of rhamnose residues in the wild-type strain, and the absence of sulfation at these sites in the mutant oligosaccharide.

image

Figure 3. Glycosyl linkage analysis of the permethylated alditol acetate (PMAA) derivatives from desulfated Azospirillum brasilense Sp7 oligosaccharide (OS) fraction 1. The peaks are as indicated in the GC profiles. Those peaks labelled * are non-carbohydrate contaminants.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Chemical analysis on the LPS structure of A. brasilense Sp7 revealed a complex glycosyl composition, consisting of several different hexose and deoxyhexose residues. From the results of the glycosyl compositional and linkage analysis, it can be hypothesized that the O-repeating unit consists of rhamnose and galactose residues with a 3,4-linked fucose at the branching point and a xylose unit present in a ratio 1:1 with fucose. The oligosaccharide portion of the LPS of A. brasilense Sp7 is very likely a heterogeneous mixture of sugar molecules, differing not only in the length of the O-repeating unit but also in its composition.

Recent analysis of the pRhico plasmid genome of A. brasilense Sp7 revealed the presence of two clusters of genes, involved in rhamnose and fucose biosynthesis (Vanbleu et al., 2004). The rhamnose gene cluster was partially analysed (Jofréet al., 2004) and an rmbA mutant showed a different LPS profile, indicating a change in O-antigen and core composition. However, detailed information on the actual composition or the glycosyl linkage of the monosaccharides in the core moiety and the O-polysaccharide of the mutant was not provided.

The LPS structure of a nodPQ mutant of A. brasilense Sp7 was compared with that of the wild-type strain. Mass spectrometric analysis of the lipid A from Sp7 and its nodPQ mutant gave identical results indicating that there was no difference in the lipid A structures of these strains (data not shown). However, the oligosaccharide fraction from the nodPQ mutant differed from that of the wild-type strain in that it lacks sulfate groups on the O-2 and O-6 positions of the rhamnosyl and galactosyl residues respectively.

The presence of sulfate substituents on bacterial cell surface polysaccharides is not common. So far, sulfate-containing polysaccharides have only been described in a strain of Pseudoalteromonas isolated from a deep-sea hydrothermal vent (Rougeaux et al., 1999), in the plant symbionts S. meliloti and Rhizobium sp. NGR234 (Cedergren et al., 1995) and in the human pathogen Mycobacterium tuberculosis (Mougous et al., 2004). In M. tuberculosis, the major sulfated molecule is a glycolipid, consisting of sulfated trehalose-esters with acyl chains at various positions on the trehalose moiety. It makes part of the cell wall and is implicated in virulence (Rivera-Marrero et al., 2002).

In S. meliloti and Rhizobium sp. NGR234, sulfated LPS were found (Cedergren et al., 1995; Keating et al., 2002). Previous studies suggested that sulfation of LPS molecules competes with the sulfation of Nod factors for a limited pool of sulfate molecules. These activated sulfate groups are produced by nodPQ homologues involved in different biosynthetic pathways. The ultimate transfer of sulfate groups to an acceptor molecule is provided by a pathway-specific sulfotransferase. For instance, in S. meliloti, the sulfotransferase LpsS, which specifically transfers sulfate groups to LPS molecules, was recently identified (Cronan and Keating, 2004).

In the case of A. brasilense Sp7, a reduced sulfation of LPS molecules was found after deletion of nodPQ and the LPS sulfation was not completely abolished. This indeed indicates the presence of a residual pool of activated sulfate groups, available for transfer to LPS molecules but Azospirillum sulfotransferase genes have not yet been identified.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains

Six litres of cultures of the A. brasilense Sp7 wild-type strain (Tarrand et al., 1978) and the A. brasilense Sp7 nodPQ mutant, 7803 (Vieille and Elmerich, 1990), were grown in LB medium, supplemented with 2.5 mM CaCl2 and 2.5 mM MgSO4. Kanamycin (50 µg ml−1) was added when growing the nodPQ mutant. Bacteria were harvested by centrifugation (25 min, 8000 r.p.m.) at late log/early stationary phase, washed twice with physiological water and lyophilized.

Lipopolysaccharide purification

The parent strain, Sp7, and mutant strain, nodPQ, were extracted using the hot phenol/water procedure (Westphal and Jann, 1965). Briefly, the cell pellet was suspended in 45% aqueous phenol and stirred at 68°C for 30 min. After cooling down (< 10°C), the water and phenol layers were separated by centrifugation (45 min, 5000 r.p.m., 4°C) and the upper aqueous phase was transferred. The extraction procedure was repeated twice and the water as well as the phenol phase was dialysed for 2 days against water (1000 MW cut-off, regenerated cellulose dialysis membrane, Spectra/Por), concentrated and lyophilized. The extracted LPS were subsequently treated with ribonuclease, deoxyribonuclease and proteinase K and then dialysed and lyophilized.

Electrophoresis

Polyacrylamide gel electrophoresis (PAGE) in the presence of DOC and Alcian-blue silver stain were performed according to Reuhs and colleagues (1998). The separating and stacking gel contain, respectively, 18% and 4% acrylamide.

Lipopolysaccharide fractionation

Fractionation of intact LPS molecules was performed by gel filtration chromatography on a Sephacryl S400 column (Amersham Biosciences) with DOC buffer (0.2 M NaCl, 0.25% DOC, 1.0 mM EDTA, 10 mM Tris base, pH 9.2) as eluent and monitored by a refractometer. Each fraction was subjected to DOC-PAGE analysis and different fractions were pooled, dialysed and lyophilized.

Oligosaccharide isolation

Twenty micrograms of LPS molecules were subjected to mild acetic acid cleavage (2% HOAc, 100°C) for 5.5 h (Ryan and Conrad, 1974). The lipid A fraction was removed by centrifugation (10 min, 3000 r.p.m., 15°C) and the supernatant, containing the oligosaccharides, was lyophilized.

The oligosaccharide fraction was fractionated using size-exclusion chromatography on a Biogel P30 column (Bio-Rad), equilibrated with degassed, filtered, deionized water. Fractions (1 ml) were collected and the elution pattern was monitored with a refractometer. The different fractions were pooled and lyophilized.

Glycosyl composition analysis

Glycosyl compositions of LPS, oligosaccharides or lipid A were determined by preparation of the trimethylsilyl methyl glycosides (TMS) or alditol acetate derivatives and GLC-MS (electron-impact) analysis. For TMS analysis, the samples were subjected to methanolysis in methanolic 1 M HCl at 80°C for 18 h, N-acetylated with pyridine and acetic anhydride in methanol, trimethylsilylated (Tri-Sil, Pierce) and then analysed, using a 30 m DB5 fused silica capillary column (J&W Scientific). For alditol acetates, the samples were hydrolysed in 4 M trifluoroacetic acid for 4 h at 100°C. The resulting glycoses were reduced with NaBH4, acetylated and analysed by using a 30 m SP2330 (Supelco) capillary column.

Glycosyl linkage analysis

Linkage analysis of neutral sugars was performed by permethylation with methyl iodide in dimethyl sulfoxide containing dimethyl sulfoxide anion (Hakomori method), conversion to the PMAAs as described previously (York et al., 1985) and GLC-MS analysis using a 30 m SP2330 (Supelco) capillary column. NAc sugars were identified on a 30 m DB5 fused silica capillary column (J&W Scientific).

Sulfate determination

Sulfate determination was performed following a procedure, adapted from Nagasawa and colleagues (1977). One millilitre of oligosaccharide was suspended in 400 µl of distilled water and a few mg of cationic exchange resin was added and mixed well. After filtration and washing of the resins, the sample was lyophilized. Desulfation was performed in 14% methanol in DMSO for 1.5 h at 100°C. After cooling down, water was added and the pH was adjusted to 9 with 0.1 M NaOH. The desulfated oligosaccharides were dialysed for 2 days against water and lyophilized.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank C. Elmerich for providing the A. brasilense Sp7 nodPQ mutant strain, 7803. E.V. is the recipient of a predoctoral fellowship of the Fonds voor Wetenschappelijk Onderzoek-Vlaanderen. This work was also supported in part by a US Department of Energy grant (DE-FG02-93ER20097) to the Complex Carbohydrate Research Center and a National Institutes of Health grant (GM39583) to R.W.C.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Cedergren, R.A., Lee, J., Ross, K.L., Hollingsworth, R.I. (1995) Common links in the structure and cellular localization of Rhizobium chitolipopolysaccharides and general Rhizobium membrane phospholipid and glycolipid components. Biochemistry 34: 44674477.
  • Cronan, G.E., Keating, D.H. (2004) Sinorhizobium meliloti sulfotransferase that modifies lipopolysaccharide. J Bacteriol 186: 41684176.
  • Fedonenko, Y.P., Zatonsky, G.V., Konnova, S.A., Zdorovenko,  E.L.,  Ignatov,  V.V. (2002)  Structure of the O-specific polysaccharide of the lipopolysaccharide of Azospirillum brasilense Sp245. Carbohydr Res 337: 869872.
  • Fedonenko, Y.P., Konnova, O.N., Zatonsky, G.V., Shashkov, A.S., Konnova, S.A., Zdorovenko, E.L., et al. (2004) Structure of the O-polysaccharide of the lipopolysaccharide of Azospirillum irakense KBC1. Carbohydr Res 339: 18131816.
  • Fedonenko, Y.P., Konnova, O.N., Zatonsky, G.V., Konnova, S.A., Kocharova, N.A., Zdorovenko, E.L., Ignatov, V.V. (2005) Structure of the O-polysaccharide from the Azospirillum lipoferum Sp59b lipopolysaccharide. Carbohydr Res 340: 12591263.
  • Jofré, E., Lagares, A., Mori, G. (2004) Disruption of dTDP-rhamnose biosynthesis modifies lipopolysaccharide core, exopolysaccharide production, and root colonization in Azospirillum brasilense. FEMS Microbiol Lett 231: 267275.
  • Keating, D.H., Willits, M.G., Long, S.R. (2002) A Sinorhizobium meliloti lipopolysaccharide mutant altered in cell surface sulfation. J Bacteriol 184: 66816689.
  • Matora, L.Y., Serebrennikova, O.B., Shchyogolev, S.Y. (2001) Structural effects of the Azospirillum lipopolysaccharides in cell suspension. Biomacromolecules 2: 402406.
  • Michiels, K., Croes, C., Vanderleyden, J. (1991) Two different modes of attachment of Azospirillum brasilense Sp7 to wheat roots. J Gen Microbiol 137: 22412246.
  • Mougous, J.D., Petzold, C.J., Senaratne, R.H., Lee, D.H., Akey, D.L., Lin, F.L., et al. (2004) Identification, function and structure of the mycobacterial sulfotransferase that initiates sulfolipid-1 biosynthesis. Nat Struct Mol Biol 11: 721729.
  • Nagasawa, K., Inoue, Y., Kamata, T. (1977) Solvolytic desulfation of glycosaminoglycuronan sulfates with demethyl sulfoxide containing water or methanol. Carbohydr Res 58: 4755.
  • Reuhs, B.L., Geller, D.P., Kim, J.S., Fox, J.E., Kolli, V.S., Pueppke, S.G. (1998) Sinorhizobium fredii and Sinorhizobium meliloti produce structurally conserved lipopolysaccharides and strain-specific K antigens. Appl Environ Microbiol 64: 49304938.
  • Rivera-Marrero, C.A., Ritzenthaler, J.D., Newburn, S.A., Roman, J., Cummings, R.D. (2002) Molecular cloning and expression of a novel glycolipid sulfotransferase in Mycobacterium tuberculosis. Microbiology 148: 783792.
  • Rougeaux, H., Guezennec, J., Carlson, R.W., Kervarec, N., Pichon, R., Talaga, P. (1999) Structural determination of the exopolysaccharide of Pseudoalteromonas strain HYD 721 isolated from a deep-sea hydrothermal vent. Carbohydr Res 315: 273285.
  • Ryan, J.M., Conrad, H.E. (1974) Structural heterogeneity in the lipopolysaccharide of Salmonella newington. Arch Biochem Biophys 162: 530535.
  • Schwedock, J.S., Liu, C., Leyh, T.S., Long, S.R. (1994) Rhizobium meliloti NodP and NodQ form a multifunctional sulfate-activating complex requiring GTP for activity. J Bacteriol 176: 70557064.
  • Shen, Y., Sharma, P., Da Silva, F.G., Ronald, P. (2002) The Xanthomonas oryzae pv. oryzae raxP and raxQ genes encode an ATP sulphurylase and adenosine-5′-phosphosulphate kinase that are required for AvrXa21 avirulence activity. Mol Microbiol 44: 3748.
  • Skvortsov, I.M., Ignatov, V.V. (1998) Extracellular polysaccharides and polysaccharide-containing biopolymers from Azospirillum species: properties and the possible role in interaction with plant roots. FEMS Microbiol Lett 165: 223229.
  • Snoeck, C., Verreth, C., Hernandez-Lucas, I., Martinez-Romero, E., Vanderleyden, J. (2003) Identification of a third sulfate activation system in Sinorhizobium sp. strain BR816: the CysDN sulfate activation complex. Appl Environ Microbiol 69: 20062014.
  • Somers, E., Vanderleyden, J., Srinivasan, M. (2004) Rhizosphere bacterial signalling: a love parade beneath our feet. Crit Rev Microbiol 30: 205240.
  • Steenhoudt, O., Vanderleyden, J. (2000) Azospirillum, a free-living nitrogen-fixing bacterium closely associated with grasses: genetic, biochemical and ecological aspects. FEMS Microbiol Rev 24: 487506.
  • Tarrand, J.J., Krieg, N.R., Döbereiner, J. (1978) A taxonomic study of the Spirillum lipoferum group with description of a new genus Azospirillum gen. nov. and two species Azospirillum lipoferum (Beijerinck) comb. nov. and Azospirillum brasilense sp. nov. Can J Microbiol 24: 967980.
  • Vanbleu, E., Marchal, K., Lambrecht, M., Mathys, J., Vanderleyden, J. (2004) Annotation of the pRhico plasmid of Azospirillum brasilense reveals its role in determining the outer surface composition. FEMS Microbiol Lett 232: 165172.
  • Vieille, C., Elmerich, C. (1990) Characterization of two Azospirillum brasilense Sp7 plasmid genes homologous to Rhizobium meliloti nodPQ. Mol Plant–Microbe Interact 3: 389400.
  • Westphal, O., Jann, K. (1965) Bacterial lipopolysaccharides: extraction with phenol-water and further applications of the procedure. Methods Carbohydr Chem 5: 8391.
  • York, W.S., Darvill, A.G., McNeil, M., Stevenson, T.T., Albersheim, P. (1985) Isolation and characterization of plant cell walls and cell wall components. Methods Enzymol 118: 330.