Macromolecular and elemental composition analysis and extracellular metabolite balances of Pichia pastoris growing at different oxygen levels

Background: Analysis of the cell operation at the metabolic level requires collecting data of different types and to determine their confidence level. In addition, the acquired information has to be combined in order to obtain a consistent operational view. In the case of Pichia pastoris, information of its biomass composition at macromolecular and elemental level is scarce particularly when different environmental conditions, such as oxygen availability or, genetic backgrounds (e.g. recombinant protein production vs. non production conditions) are compared. Results: P. pastoris cells growing in carbon-limited chemostat cultures under different oxygenation conditions (% O-2 in the bioreactor inlet gas: 21%, 11% and 8%, corresponding to normoxic, oxygen-limiting and hypoxic conditions, respectively), as well as under recombinant protein (antibody fragment, Fab) producing and non-producing conditions, were analyzed from different points of view. On the one hand, the macromolecular and elemental composition of the biomass was measured using different techniques at the different experimental conditions and proper reconciliation techniques were applied for gross error detection of the measured substrates and products conversion rates. On the other hand, fermentation data was analyzed applying elemental mass balances. This allowed detecting a previously missed by-product secreted under hypoxic conditions, identified as arabinitol (aka. arabitol). After identification of this C5 sugar alcohol as a fermentation by-product, the mass balances of the fermentation experiments were validated. Conclusions: After application of a range of analytical and statistical techniques, a consistent view of growth parameters and compositional data of P. pastoris cells growing under different oxygenation conditions was obtained. The obtained data provides a first view of the effects of oxygen limitation on the physiology of this microorganism, while recombinant Fab production seems to have little or no impact at this level of analysis. Furthermore, the results will be highly useful in other complementary quantitative studies of P. pastoris physiology, such as metabolic flux analysis.