In our previous work, using indirect analysis, we established that the glyoxylate shunt, the TCA cycle and acetate uptake by acetyl-CoA synthetase are more active in E.coli B than in E.coli K. By implementing the system biology approach, we showed that in addition to what we knew, other metabolic pathways are active differently in the two strains. These are glucoenogensis, sfcA shunt, ppc shunt, glycogen biosynthesis and fatty acid degradation. It was found that in E.coli JM109, acetate is produced by pyruvate oxidase (poxB) using pyruvate as a substrate rather than by phosphotransacetylase-acetate kinase (Pta-AckA) system which uses acetyl-CoA. The inactivation of the gluconegensis enzyme phosphoenolpyruvate synthase (ppsA), the activation of the anaplerotic sfcA shunt, and low and stable pyruvate dehydrogenase (aceE, aceF), cause pyruvate accumulation which is converted to acetate by pyruvate oxidase B. The behavior of the ppsA, acs and aceBAK in JM109 was dependent on the glucose supply strategy. When the glucose concentration was high, no transcription of these genes was observed and acetate concentration increased, but at low glucose concentrations, these genes were expressed and the acetate concentration decreased. It is possible that there is a major regulatory molecule that controls not only ppsA and aceBAK but also acs. The gluconeogenesis genes (fbp, pckA, and ppsA) lead to glycogen accumulation and are constitutively active in E. coli BL21 regardless of glucose feeding strategy, our assumption was that the Cra protein (catabolite repressor/activator, formally called FruR) is responsible for this effect. To further understand this phenomenon we decided to check the effect of the Cra on the growth and acetate production by E. coli B and K, and evaluate the gene transcription pattern between the Cra positive and the Cra negative strains by using microarrays. This work is currently on going. The growth and acetate production at high glucose concentration was evaluated;the results indicated that in E. coli B there is no considerable difference between the Cra positive strain and the Cra negative strain, acetate production is a bit lower in the Cra positive strain but the growth kinetics and the glucose consumption are similar. In E. coli K, however, there is a significant difference, the Cra negative strain is stop growing at a concentration about one third of the final cell concentration of the Cra positive strain, as for the acetate production, there is no difference between the two strains. Another approach for possible understanding the differences between the two strains is to look at the role of small RNA, a regulatory RNA. We decided to concentrate on the Sgrs a small RNA which affects the expression of the glucose transporter IICBGLU by inactivating the corresponding mRNA ptsG. The initial results so far indicate that the Sgrs is transcribed in E. coli B in response to high glucose concentration but at the same growth conditions is not transcribed in E. coli K. The higher level of Sgrs in E. coli B is an indication that there is less PTSG in this strain and consequently less glucose transport, the results of which is probably lower load on the TCA cycle and lower acetate concentration. The fact that E. coli B is a better grower than E. coli K triggered us to look for a better way to produce Plasmid DNA (pDNA). The E. coli strains currently used for pDNA production, JM109 and DH5, are both suitable for production of stable pDNA due to the deletion of recA and endA, however, these two E. coli K strains are sensitive to growth conditions such as high glucose concentration. On the other hand E. coli BL21 is less sensitive to growth conditions than E. coli JM109 or DH5, this strain grows to higher densities and is not sensitive to high glucose concentration. This strain is used for recombinant protein production but not for pDNA production because of its inability to produce stable pDNA. To adapt E. coli BL21 for stable pDNA production, the strain was mutated by deleting both recA and endA, and a proper growth and production strategy was developed. Production values, exceeding 2 grams per liter were obtained using glucose as a carbon source. The produced plasmid, which was constructed for HIV clinical study, was found to have identical properties to the plasmid currently produced by E. coli DH5.
