Biotechnological processes are distinguished from classical chemistry by employing bio-molecules or whole cells as the catalytic element, providing unique reaction mechanisms with unsurpassed specificity. Whole cells are the most versatile \'factories\' for natural or non-natural products, however, the conversion of e.g. hydrophobic substrates can quickly become cytotoxic. One host organism with the potential to handle such conditions is the gram-negative bacterium Pseudomonas putida, which distinguishes itself by solvent tolerance, metabolic flexibility, and genetic amenability. However, whole cell bioconversions are highly complex processes. A typical bottleneck compared to classical chemistry is lower yield and reproducibility owing to cell-to-cell variability. The intention of this work was therefore to characterize a model producer strain of P. putida KT2440 on the single cell level to identify non-productive or impaired subpopulations. Flow cytometry was used in this work to discriminate subpopulations regarding DNA content or productivity, and further mass spectrometry or digital PCR was employed to reveal differences in protein composition or plasmid copy number. Remarkably, productivity of the population was generally bimodally distributed comprising low and highly producing cells. When these two subpopulations were analyzed by mass spectrometry, only few metabolic changes but fundamental differences in stress related proteins were found. As the source for heterogeneity remained elusive, it was hypothesized that cell cycle state may be related to production capacity of the cells. However, subpopulations of one, two, or higher fold DNA content were virtually identical providing no clear hints for regulatory differences. On the quest for heterogeneity the loss of genetic information came into focus. A new work flow using digital PCR was created to determine the absolute number of DNA copies per cell and, finally, lack of expression could be attributed to loss of plasmid in non-producing cells. The average plasmid copy number was shown to be much lower than expected (1 instead of 10-20). In conclusion, this work established techniques for the quantification of proteins and DNA in sorted subpopulations, and by these means provided a highly detailed picture of heterogeneity in a microbial population.

Biotechnological processes are distinguished from classical chemistry by employing bio-molecules or whole cells as the catalytic element, providing unique reaction mechanisms with unsurpassed specificity. Whole cells are the most versatile \''factories\'' for natural or non-natural products, however, the conversion of e.g. hydrophobic substrates can quickly become cytotoxic. One host organism with the potential to handle such conditions is the gram-negative bacterium Pseudomonas putida, which distinguishes itself by solvent tolerance, metabolic flexibility, and genetic amenability. However, whole cell bioconversions are highly complex processes. A typical bottleneck compared to classical chemistry is lower yield and reproducibility owing to cell-to-cell variability. The intention of this work was therefore to characterize a model producer strain of P. putida KT2440 on the single cell level to identify non-productive or impaired subpopulations. Flow cytometry was used in this work to discriminate subpopulations regarding DNA content or productivity, and further mass spectrometry or digital PCR was employed to reveal differences in protein composition or plasmid copy number. Remarkably, productivity of the population was generally bimodally distributed comprising low and highly producing cells. When these two subpopulations were analyzed by mass spectrometry, only few metabolic changes but fundamental differences in stress related proteins were found. As the source for heterogeneity remained elusive, it was hypothesized that cell cycle state may be related to production capacity of the cells. However, subpopulations of one, two, or higher fold DNA content were virtually identical providing no clear hints for regulatory differences. On the quest for heterogeneity the loss of genetic information came into focus. A new work flow using digital PCR was created to determine the absolute number of DNA copies per cell and, finally, lack of expression could be attributed to loss of plasmid in non-producing cells. The average plasmid copy number was shown to be much lower than expected (1 instead of 10-20). In conclusion, this work established techniques for the quantification of proteins and DNA in sorted subpopulations, and by these means provided a highly detailed picture of heterogeneity in a microbial population.