Investigations of Plasmid-Host Cell Interactions in Recombinant Escherichia coli Populations

Citation

Seo, Jin-Ho (1986) Investigations of Plasmid-Host Cell Interactions in Recombinant Escherichia coli Populations. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/RGTN-X895. https://resolver.caltech.edu/CaltechETD:etd-03102008-104251

Abstract

Plasmid-host cell interactions have been characterized experimentally for recombinant Escherichia coli populations. The plasmids used contain pMB1 replication origins and propagate in E. coli at different copy number levels ranging from 12 to 408. Host E. coli HB101 strains transformed with those plasmids were used throughout this research.

The specific growth rate and amount of cloned-gene product (β-lactamase) were determined in batch cultivations as a function of plasmid copy number. Maximum specific growth rates in LB and M9 media were reduced monotonically for increasing plasmid content per cell. The maximum specific growth rate for a recombinant strain with copy number 408 was reduced by 25% relative to the plasmid-free strain. The ratio of β-lactamase specific activity to plasmid content, as a measure of the overall efficiency of plasmid-gene expression, declines by a factor of 7 as the copy number increases from 12 to 408. The relationship between copy number and cloned-gene product activity can be reasonably approximated by a parabolic equation, with approximately linear proportionality for copy numbers up to 60 but subsequently with reduction in the product/copy number ratio.

Investigations of environmental effects on plasmid content and β-lactamase activity have also been done in batch and continuous bioreactors. Plasmid copy number decreased when specific growth rate was increased by varying medium composition in a batch fermentor, or when dilution rate (inverse of residence time) was increased by increasing medium flow rate in a continuous fermentor. The response of plasmid-gene product formation to growth conditions in these two different reactor types is interesting. Batch experiments showed that the specific activity of β-lactamase is a gradual decreasing function of specific growth rate, whereas product activity has a maximum with respect to dilution rate in chemostat growth. Comparative analysis of genetic and environmental effects implies that both plasmid content and growth conditions should be taken into account in determining the productivity of recombinant cell populations.

Further insight into the host-vector interactions was obtained by exploring plasmid amplification behavior and cell-cycle characteristics in the recombinant strains. Chloramphenicol-derived amplification kinetics were measured using flow cytometry, which allows direct determination of the total cellular DNA content after staining with mithramycin, a DNA-specific fluorescent dye. Both rates of plasmid amplification and final contents of amplified plasmids were found to be proportional to the average initial plasmid copy number obtained at steady-state growth.

The measurements of the cell-cycle parameters in E. coli cells were performed using experimentally determined single-cell DNA content frequency functions in connection with mathematical analysis of cell growth. Plasmid-containing cells exhibit C and D times that are approximately 14 and 24 percent, respectively, smaller than corresponding values in the plasmidfree host. Mean cell sizes and cell mass at the point of replication initiation decrease 38 and 18 percent, respectively, relative to plasmid-free cells. These results agree qualitatively with the previous findings with E. coli B/r carrying F plasmids of various sizes. Interestingly, the effects of plasmid presence on C, D, average cell size, and initiation mass were found to be approximately independent of plasmid copy number in these host-vector systems.

Finally, mathematical description of plasmid propagation in cells which reproduce by binary fission was presented using a segregated, population balance model. With the population balance equations, the probability of plasmid loss, distribution of cellular plasmid content, and specific rate of product synthesis were determined based on different single-cell models of plasmid replication, partition, and gene expression. Control modes of plasmid replication during cell growth markedly affect the distribution of plasmid content and accordingly the specific rate of gene product synthesis. The degree of selection required for stable maintenance of plasmidharboring populations was derived by analyses that focused on the transient growth of plasmid-containing and plasmid-free cells in partially selective medium.

Thesis Files