Maintenance of a Pseudomonas fluorescens plasmid in heterologous hosts: metabolic burden as a more reliable variable to predict plasmid instability.

The stability of a large, multiresistance plasmid, pSCL of P. fluorescens CAS102 was studied in Pseudomonas putida and E. coli under various non-stress conditions. Both the strains lost the plasmid within 25 days when repeatedly subcultured in LB broth without any antibiotic. The transformants survived in sterile soil and water without any marked reduction in the viability. In sterile soil, P. putida lost 93% and E. coli, 98% of their plasmid containing population in 30 days, while in sterile water the plasmid loss was 92.5% and 97% respectively. The two variables, viz. the efficiency of plasmid-partitioning during cell division and measurement of relative specific growth rates of plasmid-plus and plasmid-minus cells which are used to predict plasmid instability cannot be used to predict plasmid loss during starvation. The utility of a third variable, viz. the metabolic burden due to plasmid maintenance in predicting plasmid instability in different hosts is discussed. The rate of plasmid loss was found to be comparatively faster in E. coli than in P. putida. The biosynthetic burden due to plasmid maintenance was also more in E. coli than in P. putida when compared to the plasmid-plus and plasmid-minus cells of the two strains which was evident from the increased nutrient uptake rates (glucose, O2, and amino acid) and increased protein content of the plasmid-plus cells of E. coli. From the results, a correlation could be found between the degree of metabolic burden and the rate of plasmid loss. The reliability of metabolic burden, to predict plasmid instability versus the relative specific growth rates is discussed.