Monitoring of transcriptome and proteome profiles to investigate the cellular response of E. coli towards recombinant protein expression under defined chemostat conditions.

The use of strong promoter systems for recombinant protein production generates high product yields, but also overburdens the host cell metabolism and compromises production. Escherichia coli has highly developed regulatory pathways that are immediately responsive to adverse conditions. To gain insight into stress response mechanisms and to detect marker genes and proteins for stress specific monitoring time course analysis of controlled chemostat cultivations was performed using E. coli total microarray and difference gel electrophoresis (Ettan DIGE). In order to detect differences and consistencies of stress response as well as the impact of the inducer isopropyl-beta-d-thiogalactopyranosid on cells, expression of two recombinant proteins (hSOD and GFPmut3.1) was investigated. Genes involved in aerobic metabolism under control of the ArcB/ArcA two component system were found to be down-regulated, and the interplay of the psp operon, ArcA system and guanosine tetraphosphate is suggested to be involved in stress regulatory mechanisms. A distinct impact of the two recombinant proteins was observed, particularly on levels of known stress regulatory genes and proteins, as well as on the response associated with ArcA and psp. Altogether, 62 genes as well as seven proteins showed consistent expression levels due to recombinant gene expression, and are therefore suggested to be appropriate monitoring targets.

Effect of CyDye minimum labeling in differential gel electrophoresis on the reliability of protein identification.

Differential 2-DE (DIGE) is a widely applied tool for the quantitative analysis of differentially represented proteins. The method involves covalent minimal labeling of proteins prior to their electrophoretic separation using CyDye DIGE Fluor minimal dyes. This methodology creates two different species per protein, the labeled (approx. 1-2%) and unlabeled (approx. 98-99%) ones, which differ in their molecular masses by 434-464 Da, depending on the attached dye. DIGE followed by automated spot picking according to the CyDye coordinates misses in many instances the exact positions where the maximum amounts of the considered proteins are located. This fact leads to a loss in sensitivity of the subsequent MALDI-MS analyses and results in a reduced reliability of protein identification and sequence coverage. In this paper, the migration differences of labeled and unlabeled species are quantified together with the impact of this effect on the certainty of protein identification and sequence coverage investigating proteins up to 90 kDa.

Monitoring of protein profiles for the optimization of recombinant fermentation processes using public domain databases.

The expression of human superoxide dismutase in fed-batch fermentation of E. coli HMS174(DE3)(pET3ahSOD) was studied as model system. Due to the frequently used strong T7 promoter system a high metabolic load is exerted, which triggers stress response mechanisms and finally leads to the differentiation of the host cell. As a consequence, host cell metabolism is partly shifted from growth to survival accompanied by significant alterations of the protein pattern. In terms of process optimization two-dimensional electrophoresis deserves as a powerful tool to monitor these changes on protein level. For the analysis of samples derived from different states of recombinant protein production wide-range Immobiline Dry Strips pH 3-10 were used. In order to establish an efficient procedure for accelerated process optimization and to avoid costly and time-consuming analysis like mass spectrometry (MS), a database approach for the identification of significant changes of the protein pattern was evaluated. On average, 935 spots per gel were detected, whereby 50 are presumably stress-relevant. Out of these, 24 proteins could be identified by using the SWISS-2DPAGE database (www.expasy.ch/ch2d/). The identified proteins are involved in regulatory networks, energy metabolism, purine and pyrimidine nucleotide synthesis and translation. By this database approach, significant fluctuations of individual proteins in relation to recombinant protein production could be identified. Seven proteins show strong alterations (>100%) directly after induction and can therefore be stated as reliable marker proteins for the assessment of stress response. For distinctive interpretation of this highly specific information, a bioinformatic and statistic tool would be essential in order to perceive the role and contribution of individual proteins in stress response.