RHIZOBIUM MELILOTI RHIZOPINE CATABOLISM GENES: DISTRIBUTION, ROLE IN COMPETITION AND POTENTIAL AS MARKER GENE TO TRACK MICROBES

RHIZOBIUM MELILOTI RHIZOPINE CATABOLISM GENES: DISTRIBUTION, ROLE IN COMPETITION AND POTENTIAL AS MARKER GENE TO TRACK MICROBES

Silvia Rossbach^a,b, Brian McSpadden^a, Michelle Ganoff^a, and Frans J. de Bruijn^a,b,c

^aMSU-DOE Plant Research Laboratory, ^bNSF Center for Microbial Ecology, and ^cDepartment of Microbiology, Michigan State University, East Lansing, MI 48824, (517)353-2229, fax (517)353-9168, deBruijn@msu.edu

SUMMARY

We are using the Rhizobium meliloti rhizopine catabolism (moc) and synthesis (mos) genes to construct a selectable marker cassette and to create biased rhizospheres. Rhizopine (L-3-O-methyl-scyllo-inosamine, 3-O-MSI) is a symbiosis-specific compound, which is synthesized in nitrogen-fixing nodules of Medicago sativa induced by R. meliloti strain L5-30. Strain L5-30 is also able to specifically catabolize 3-O-MSI. It has been postulated that the ability to utilize rhizopine may confer a competitive advantage upon this strain. We are using the mos and moc genes to create a model system consisting of transgenic plants excreting the rhizopine into the rhizosphere (Mos⁺), and combining these with beneficial soil bacteria able to utilize the rhizopine (Moc⁺). It is expected that Moc⁺ microbes may gain a specific competitive advantage in the rhizosphere of Mos⁺ partner plants. In addition, we have tested the distribution of rhizopine catabolism genes among common soil bacteria. This is a precondition for the use of rhizopine catabolism genes as a selectable marker cassette to monitor genetically modified microorganisms in soil.

Key words: Biased rhizosphere, rhizopine, nutritional mediator, marker gene, competition

INTRODUCTION

A large number of beneficial microorganisms have been identified, including bacterial strains involved in bio-remediation, nitrogen fixation or bio-control. These microbes have been isolated from their natural habitat or have been improved by genetic modification in the laboratory. However, when they are reintroduced into the environment, they are normally out-competed by the already present, well-adapted microflora in this habitat. In order to overcome this problem, we have suggested to develop a model system, the "biased rhizosphere" (Rossbach et al., 1994; 1995a; 1995b). The biased rhizosphere consists of a transgenic plant, excreting an unusual compound (rhizopine) into its rhizosphere, matched with a beneficial soil bacterium able to utilize rhizopine as sole nitrogen (N) and carbon (C) source.

Using this model system we are studying whether bacteria, able to catabolize a specific substrate (nutritional mediator), can gain a competitive advantage in the rhizosphere of plants secreting such mediators; and whether it will be therefore feasible to manipulate the composition of microbial communities in the rhizosphere. Moreover, this project has implications for the risk assessment of the release and containment of genetically modified microorganisms (GMO's). With the model system of the biased rhizosphere it should be possible to predict whether GMO's can be contained in their specific environment and/or conditionally lost from it by removing the nutritional mediator producing partner plants at the end of the growing season, and thereby causing the loss of the selective advantage of the microbe.

We have chosen the R. meliloti rhizopine catabolism (moc) and synthesis (mos) genes to construct a selectable marker cassette and biased rhizospheres. The ability to synthesize and catabolize rhizopine (L-3-O-methyl-scyllo-inosamine; 3-O-MSI) was first identified in R. meliloti strain L5-30 (Tempe et al., 1982). It appeared to be a unique feature, since other R. meliloti strains tested did not show this ability (Tempe et al., 1982). By genetic manipulations and the use of one of the R. meliloti strains, which is normally not able to synthesize or catabolize rhizopine, Murphy et al. (1987) were able to identify the genetic region of R. meliloti strain L5-30 responsible for rhizopine synthesis and catabolism. The genetic loci for rhizopine synthesis and catabolism of rhizopine were found to be linked and located on the symbiotic (sym-)plasmid. Subsequently, these loci were cloned and termed mos (synthesis of 3-O-MSI) and moc (catabolism of 3-O-MSI; Murphy et al., 1987). In addition, it was shown that strain L5-30 is able to utilize rhizopine as sole nitrogen and carbon source, and it has been suggested that this ability may give this strain a selective advantage in the nodules and rhizosphere of plants.

MATERIALS AND METHODS

DNA Manipulations. DNA manipulations were carried out according to standard protocols (Sambrook et al., 1989)

Tn5 Mutagenesis and DNA Sequence Determination. Tn5 mutagenesis and DNA sequence determination were performed as described in Rossbach et al. (1994).

Rhizopine Catabolism Assays and High-Voltage Paper Electrophoresis (HVPE). Rhizopine catabolism assays and high-voltage paper electrophoresis (HVPE) were carried out as described in Rossbach et al. (1994).

Oligonucleotide Primers and PCR. The design of the oligonucleotide primers and the conditions for PCR are described in Rossbach et al. (1995a).

Nodule Occupancy Test. Surface-sterilized seeds of Medicago sativa var. Cardinal were grown as described in Rossbach et al. (1995a). After one week, the germinated seedlings were inoculated with a 1:1 mixture of late logarithmic cultures of the wild type R. meliloti strain L5-30 and a Moc^- mutant derivative (mocC::Tn5). Nodules were harvested after 4, 8, and 12 weeks. Each nodule was harvested separately, surface-sterilized, and bacteria were reisolated from the nodules. Bacteria were plated onto non-selective medium, followed by a transfer to selective medium containing kanamycin. Since the MocC^-mutant carries a Tn5 insertion, it is resistant to the antibiotic kanamycin, thus allowing an easy determination of ratio between wild type and Moc^--mutant.

DNA Extraction from Soil. DNA extraction from soil was carried out according to Zhou et al. (1995).

RESULTS AND DISCUSSION

Characterization of the Rhizopine Catabolism (moc) Genes. The first objective of our project is the construction of a selectable marker cassette which is based on the catabolism of the rare substrate rhizopine. In order to construct this cassette, the genes for the rhizopine catabolism had to be delimited. By employing Tn5-mutagenesis and DNA-sequence determination, four open reading frames (ORF's) were identified as being essential for rhizopine catabolism (mocA,B,C, and R; Rossbach et al., 1994). The deduced protein products of these four ORF's were analyzed and they were compared to proteins stored in databases. Their putative roles in rhizopine degradation are discussed here briefly.

MocA contains a putative FAD-binding domain and shows similarity to dehydrogenases, for example to an inositol-dehydrogenase which is involved in inositol degradation in Bacillus subtilis. MocB contains a putative signal sequence at its N-terminus, characteristic for exported proteins. Moreover, it shows high similarity to periplasmic sugar-binding proteins. No obvious motifs or similarities to proteins stored in databases could be identified for MocC. MocR contains a helix-turn-helix motif at its N-terminus and shows similarity to a class of bacterial regulator proteins, called the GntR family. We concluded that MocB is probably involved in sensing and binding the rhizopine, thus facilitating the uptake and transport into the cell. MocR probably plays the role of a positive regulator and is activating the rhizopine catabolism genes. MocA is a putative dehydrogenase involved in the enzymatic degradation of the rhizopine (which is an inositol related compound.) MocC is essential for rhizopine degradation, however, its sequence does not reveal any hints about its function. These results are further discussed in detail by Rossbach et al. (1994).

Expression of the Rhizopine Catabolism (moc) Genes in Beneficial Soil Bacteria. We previously designed a moc cassette, which conferred the ability to catabolize rhizopine to another R. meliloti strain (strain 1021; Rossbach et al., 1995b). We transferred this moc-cassette, which consists of mocBAC expressed from a constitutive promoter, to a number of different bacterial strains, in order to test its functionality in a wide range of different (soil) bacteria. Among those were Pseudomonas fluorescens, Alcaligenes eutrophus, Azorhizobium caulinodans, Bradyrhizobium japonicum and Escherichia coli. However, the minimal moc-cassette could not confer the ability to degrade rhizopine to any of these strains. We postulate that R. meliloti strain 1021 provides additional functions (for example transport functions), enabling this strain, but not the others, to catabolize rhizopine. In order to search for these additional functions, we are using the following approach:

A fragment carrying the mocABCR genes will be integrated into the genome of P. fluorescensstrain AK15, with the help of a Mini-Tn5 construct (de Lorenzo et al., 1990). In a second step, a clone bank consisting of cosmid clones carrying fragments of the genome of R. meliloti 1021 will be conjugated en masse into P. fluorescens::mocABCR. Then, it will be screened for a strain which is able to catabolize rhizopine via the rhizopine catabolism assay.

Distribution of moc Genes among Soil Bacteria. A marker cassette based on the ability to catabolize a rare substrate is only useful, if common soil microorganisms do not possess this trait. Therefore, we started a screening program to examine the distribution of the rhizopine catabolism genes. First, we examined several laboratory collections, containing microorganisms common to the soil and the rhizosphere, including several Pseudomonas, Agrobacterium and Rhizobium species. Total chromosomal DNA of these organisms was isolated and hybridized with the moc genes. No positive hybridization signals were found in the strains analyzed. We also examined a collection of 100 so far unidentified strains, isolated from an agricultural field plot (LTER100; Kellog Biological Station, Hickory Corners, MI). Using DNA hybridization techniques and oligonucleotides corresponding to mocA as primers in PCR, we were not able to detect any strain in these two collections which harbored moc gene sequences. Based on these results we conclude that moc genes appear to be infrequently distributed in nature.

In order to examine more closely related strains to R. meliloti L5-30, we analyzed a collection of 250 different R. meliloti strains, isolated from Medicago species all over the world, and grouped into 50 different electrophoretic type strains (Eardly et al., 1990). Via the PCR approach, we identified three additional R. meliloti strains, carrying moc genes and being able to catabolize and synthesize rhizopine. These results are described in detail by Rossbach et al. (1995a). Since we could only identify three additional Moc⁺ strains among very closely related R. meliloti strains, we concluded that the ability to utilize rhizopine indeed represents an uncommon trait among soil microorganisms.

Competition Studies Using the Nodule Occupancy Test. In order to test the hypothesis that the ability to catabolize rhizopine plays a role in competition in the rhizosphere, we used the naturally occurring assay system of Medicago sativa plants, infected with rhizopine producing (Mos⁺), but either Moc⁺ or Moc^- R. meliloti strains. We performed a nodule occupancy assay in order to test whether bacteria able to degrade rhizopine would obtain a competitive advantage in terms of infecting and inhabiting the alfalfa nodule.

Alfalfa plants were inoculated with the wild type strain L5-30 and a Moc^- mutant (mocC::Tn5) in a 1:1 ratio. After 4, 8, and 12 weeks, nitrogen-fixing nodules that developed were harvested and bacteria reisolated from them. After 4 weeks the MocC^- mutant coinhabited 10 out of 12 nodules tested. After 8 and 12 weeks, the number of coinhabited nodules dropped considerably (8 weeks: 2 out of 16; 12 weeks: 4 out of 20). Figure 1 shows that during the experimental time-course the Moc^- mutant is out-competed by the wild-type strain. In addition, 100 colonies of reisolated bacteria from each nodule were tested for their antibiotic resistance. After 4 weeks, 24% of all colonies tested represented the MocC^- mutant. The number declined to 8% after 8 weeks and 10% after 12 weeks. Thus, the MocC^- mutant was not as competitive as the wild type strain with respect to the persistence in the nodule over time. These data speak in favor of a role of rhizopine in competition.

Development of Methods to Detect Moc⁺ Bacteria in Soil. Since the rhizopine catabolism (moc) genes confer the ability to catabolize rhizopine as sole nitrogen (N) and carbon (C) source, the moc cassette should be able to be used as a selectable marker. By plating soil samples onto minimal medium containing rhizopine as sole N and C source, it should be possible to track Moc⁺ bacteria. Another method to detect Moc⁺ bacteria in soil, without having to plate of bacteria, is the method of direct DNA isolation from soil, with subsequent use of the DNA in PCR mediated amplification of moc specific DNA sequences. We inoculated sterilized and non-sterilized greenhouse soil with R. meliloti L5-30 (Moc⁺), and employed a soil DNA isolation protocol developed by Zhou et al. (1995), as well as primers corresponding to the mocAgene developed earlier (Rossbach et al., 1995a). Using PCR mediated amplification, we were able to detect moc-gene containing bacteria in sterile and unsterile soil samples, which had previously been inoculated with 10⁶ L5-30 cells/g soil (Figure 2).

The results presented by Rossbach et al. (1994; 1995a; b), and the experiments summarized here, provide the basis for pursuing our final goal, the development of the biased rhizosphere. When the biased rhizosphere consisting of Mos⁺ plants and Moc⁺ bacteria is assembled, we will have created a model system, which will enable us to answer interesting questions concerning the survival, competition, and persistence of soil bacteria in the rhizosphere of plants.

ACKNOWLEDGEMENTS

We thank Dr. J. Zhou and Mary Ann Bruns for making the unpublished soil DNA extraction protocol available to us. This work was supported by the US Department of Agriculture (92-39210-8224) and the Michigan State University Rackham Foundation.

REFERENCES

Eardly BD, Materon LA, Smith NH, Johnson D A, Rumbaugh MD, Selander RK (1990) Genetic structure of natural populations of the nitrogen-fixing bacterium Rhizobium meliloti. Appl. Env. Micro. 56:187-194.

Egelhoff TT, Long SR (1985) Rhizobium meliloti nodulation genes: Identification of nodDABC gene products, purification of NodA protein, and expression of nodA in Rhizobium meliloti. J Bacteriol 164:591-599.

de Lorenzo V, Herrero M, Jakubzik U, Timmis KN (1990) Mini-Tn5 transposon derivatives for insertion mutagenesis, promoter probing, and chromosomal insertion of cloned DNA in Gram-negative eubacteria. J Bacteriol 172:6568-6572.

Murphy PJ, Heycke N, Banfalvi Z, Tate ME, de Bruijn FJ, Kondorosi A, Tempe J, Schell J (1987) Genes for the catabolism and synthesis of an opine-like compound in Rhizobium meliloti are closely linked and on the Sym plasmid. Proc Natl Acad Sci USA 84: 493-497.

Murphy PJ, Heycke N, Trenz SP, Ratet P, de Bruijn FJ, Schell J (1988) Synthesis of an opine-like compound, a rhizopine, in alfalfa nodules is symbiotically regulated. Proc Natl Acad Sci USA 85: 9133-9137.

Murphy PJ, Trenz SP, Grzemski W, de Bruin FJ, Schell J (1993) The Rhizobium meliloti rhizopine moslocus is a mosaic structure facilitating its symbiotic regulation. J Bacteriol 175:5193-5204.

Rossbach S, Kulpa DA, Rossbach U, de Bruijn FJ (1994) Molecular and genetic characterization of the rhizopine catabolism (mocABRC) genes of Rhizobium meliloti L5-30. Mol Gen Genet 245:11-24.

Rossbach S, Rasul G, Schneider M, Eardley B, de Bruijn FJ (1995a) Structural and functional conservation of the rhizopine catabolism (moc) locus is limited to selected Rhizobium meliloti strains and unrelated to their geographical origin. Mol Plant-Microbe Interact; in press.

Rossbach S, McSpadden B, Kulpa D, de Bruijn FJ (1995b). Rhizopine synthesis and catabolism genes for the creation of "Biased Rhizospheres" and as marker system to detect (genetically modified) microorganisms in the soil. In: Levin M., Grim C, Angle S (eds) Biotechnology Risk Assessment: Risk Assessment Methodologies. Univ. of Md., College Park, MD, pp 223-249.

Saint CP, Wexler M, Murphy PJ, Tempe J, Tate ME, Murphy PJ (1993) Characterization of genes for synthesis and catbolism of a new rhizopine induced in nodules by Rhizobium meliloti Rm220-3: extension of the rhizopine concept. J Bacteriol 175:5205-5215.

Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning. A laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

Tempe J, Petit A, Bannerot H (1982) Presence de substances semblables a des opines dans des nodosites de Luzerne. C R Acad Sc Paris Serie III 295: 413-416.

Zhou J, Bruns MA, Tiedje JM (1995) Rapid method for DNA recovery from soils of diverse compositions. Appl Environ Microbiol; submitted.

Figure 1. Nodule occupancy assay. The diagram shows the number of nodules occupied by R. meliloti wild type strain L5-30 versus the MocC^- mutant strain after 4, 8, and 12 weeks, respectively.

Figure 2. Photograph of an agarose gel after electrophoresis and ethidium bromide straining, showing PCR products corresponding to the expected size of mocA. mocA specific primers were used in PCR with DNA directly isolated from soil as template. mocA PCR products could be detected in soil samples, to which R. meliloti strain L5-30 (Moc⁺) had been previously added.