Early in this project it was decided that the enteric beta-galactosidase genes (lacZY), in conjunction with the sensitive X-gal indicator are suitable 'marker genes' for use in Rhizobium and Bradyrhizobium spp. In order to distinguish between a lacZY 'marker gene' and any fortuitously acquired lac gene, 2 strategies were developed. In the first, LacZ and mer genes (the latter from Tn1831, coding for mercury resistance) were cloned into the broad host range vector pMP220, thus obtaining pMPT15 that allows expression in different bacterial species. The Tc resistance gene of thevector was inactivated by removing a SmaI fragment within the gene, obtaining pMGM1, which is functional in R. leguminosarum and other gram negative soil bacteria. This system can therefore be used to assess gene transfer as well as to monitor survival of soil inoculant.
In the second strategy the structural gene (dctA) for transport of C4-dicarboxylic acids (dCA) in R. meliloti is induced by the presence of dCA in the growth medium. The dctA promoter was fused to the lac genes. The R. meliloti dctA promoter was cloned and analysed and the dctA gene was fused in frame to the enteric beta-galactosidase genes, located on a broad host range tetracycline resistant plasmid (pCU700). pCU700 was introduced into R. meliloti, B. japonicum and E. coli strains. In R. meliloti the lac genes were found to be highly expressed in the presence of dCA in the medium. In E. coli and B. japonicum however no expression of the lac genes was observed.
To achieve an absolute correlation between the genetic marker and the actual cell number, it was deemed necessary to develop vector systems that are stably maintained throughout the duration of any conceivable experiment.
A key requirement in the development of any vector system (either chromosomal integration systems or free replicating plasmids), was a positively selectable marker. Such a selectable trait must maintain a selective pressure throughout an experiment in absence of additional external factors. An elegant solution was found in the cloned thymidylate synthase (thyA) gene from Lactococcus lactis (a food grade microorganism). A broad host range plasmid containing a copy of this gene was combined with a R. meliloti host, mutated in this gene. Since thymidine is rarely encountered in the natural environment of these microorganisms, the thyA gene is essential for survival and will be stably maintained. The development of a thyA vector system for R. meliloti required a strain deficient in thymidylate synthase activity. Such a strain was obtained by selecting a spontaneous thyA mutant in the presence of antifolate drugs.
Such mutants (CM21 and CM22), were found to be thymidine requiring and incapable of persisting in rooting solution or nodulating alfalfa plants. Such defects were overcome by the introduction of the cloned L. lactis thyA gene on a self replicating plasmid. Due to this internal selective pressure in the thyA mutants, it was found that the plasmids, bearing a copy of the thyA gene, were stably maintained throughout nodulation experiments, unlike similar plasmids in a wild type background. The thy system was exploited to introduce and stably maintain a lac 'marker gene' on a self replicating plasmid, devoid of antibiotic resistance genes, into R. meliloti. The L. lactis thy gene was used to constitutively express the lac 'marker genes'. This genetically engineered microorganism will be employed to monitor competition, survival and persisten ce of R. meliloti strains in the field.