Examination of plasmid stability in a recombinant organism used in continuous fermentation

Abstract

The gene for the enzyme D-GPA dehalogenase has been cloned into a multi-copy plasmid to produce the vector pTB316. It was transformed into two strains of Pseu- domonas putida , AJl-23 and NCIB11767 in the hope of increasing production of this enzyme. The clones will be cultured in a continuous fermenter in a media not containing the antibiotic. In the absence of antibiotic selection the plasmid is stable in the host but is unstable in continuous fermentation. The instability was identified as segregational using restriction mapping and Southern blots. It is caused by the high levels of constitutive gene expression in the clone; as a result a population of faster growing P(^-) segregants rise to predominance in the fermenter. Two fermentations were operated but were abandoned due to technical problems. Calculating the degree of plasmid instability was therefore done in repeated shake flask culture. From these results it was found that the cells lose the plasmid within 90 h. Strain 316NCIB11 767 is very unstable and also develops natural antibiotic reistance to strep, very readily. The addition of the substrate analogue, MCA, to a shake flask culture of 316 AJl-23 resulted in the stabilisation of the plasmid over a period of 190 hours. MCA is converted to glycollate, a toxic product in high concentrations. Conclusions regarding this observation are tentative since little is known of the organisms physiology but it may be possible that the MCA represses gene activity thereby stabilising the plasmid. Alternatively, either a glycollate-tolerant mutant has risen in the population, or a more stable variant has arisen, irrespective of the presence of MCA in the medium It is suggested that the plasmid is probably best stabilised by the insertion of a regulatable promoter such as the temperature promoters P(_R)andP(_L), of A and the fermentation operated as a two-stage system separating the active growth phase from the active gene expression phase as suggested by Seigel and Ryu (1986) and Sayadi ct al (1987).