Development and optimisation of a procedure for the production of Parapoxvirus ovis by large-scale microcarrier cell culture in a non-animal, non-human and non-plant-derived medium.

For the production of a chemically inactivated Parapoxvirus ovis (PPVO), an adherent bovine kidney cell line was cultivated on Cytodex-3 microcarriers in suspension culture. The inactivated and purified virus particles have shown immune modulatory activity in several animal models. PPVO was produced by a biphasic batch process at the 3.5 and 10 L scale. Aeration was realised by bubble-free membrane oxygenation via a tube stator with a central two-blade anchor impeller. In order to increase efficiency, process robustness and safety, the established process was optimised. The cell line was adapted to a protein-free medium (except recombinant insulin) in order to increase biosafety. A scale up to a 50 L pilot plant with direct cell expansion was performed successfully. In parallel, the biphasic batch process was optimised with special emphasis on different operating conditions (cell number, Multiplicity of Infection (MOI), etc.) and process management (fed-batch, dialysis, etc.). The quality and concentration of the purified virus particles was assessed by quantitative electron microscopy, residual host cell protein and DNA-content and, finally, biologic activity in a transgenic mouse model. This integrated approach led to a new, safe, robust and highly productive large-scale production process, called "Volume-Expanded-Fed" Batch with cell densities up to 6-7e06 cells/mL. By subsequent dilution of infected cells into the next process scale, an increase in total productivity by a factor of 40 (related to an established biphasic batch process) was achieved.

Using buried water molecules to explore the energy landscape of proteins.

Buried water molecules constitute a highly conserved, integral part of nearly all known protein structures. Such water molecules exchange with external solvent as a result of protein conformational fluctuations. We report here the results of water (17)O and (2)H magnetic relaxation dispersion measurements on wild-type and mutant bovine pancreatic trypsin inhibitor in aqueous solution at 4-80 degrees C. These data lead to the first determination of the exchange rate of a water molecule buried in a protein. The strong temperature dependence of this rate is ascribed to large-scale conformational fluctuations in an energy landscape with a statistical ruggedness of approximately 10 kJ mol(-1).

Residence times of the buried water molecules in bovine pancreatic trypsin inhibitor and its G36S mutant.

The three-dimensional structure of the bovine pancreatic trypsin inhibitor (BPTI) contains 4 internal water molecules, denoted W111, W112, W113, and W122, the latter being replaced by a seryl side chain in the BPTI(G36S) analogue. To investigate the effect of the exchange between these explicit water sites and the bulk solvent, we have measured water 17O and 2H nuclear magnetic relaxation in solutions of BPTI and the G36S mutant over the Larmor frequency range 2.6-49 MHz. A comparison of the data from the two nuclei shows unequivocally that the isolated buried water molecule, W122, of BPTI contributes only to 2H, but not to 17O relaxation, while the other 3 waters contribute fully to the relaxation of both nuclei. This demonstrates that the residence time of W122 is in the range 10-200 microseconds, while the residence times of W111-W113 are in the range 15 ns-1 microseconds. The slower exchange of W122 indicates that the functionally active region of BPTI, near the Cys14-Cys38 disulfide bond, is less flexible than the central region of BPTI, where the other 3 buried waters are located.