Volumetric power input as a reliable parameter for scale-up from shake flask to stirred-tank bioreactor: production of a recombinant glycoprotein by Streptomyces lividans

In Streptomyces lividans, filamentous morphology varies according to hydrodynamics/aeration culture conditions, playing important roles in the production, secretion, and post-translational modification of recombinant glycoproteins. Volumetric power input (P/V) is normally used to scale-up bioprocesses. Here, we analyzed the effect of P/V on S. lividans growth and morphology as well as the production and O-mannosylation of the recombinant glycoprotein APA from Mycobacterium tuberculosis. Previously, we measured the P/V in conventional normal (NF, 0.20 kW/m3) and coiled (CF, 0.44 kW/m3) shake flasks. Here, these P/V were matched in 1.0 L bioreactors by adjusting the stirring speed. Similar specific growth rates were observed between shake flask and bioreactor cultures. In addition, similar rAPA yields (~25%) were observed in both the bioreactor and NF cultures, but these were lower than the yield in the CF culture (~36%). Furthermore, up to seven mannose residues were attached to the C-terminal of rAPA in both bioreactor cultures, which was one more than in the NF and CF cultures. However, at similar P/V, the morphology and biomass kinetics of growth were not similar. Our data indicate that scaling-up the production of recombinant glycoproteins using S. lividans by matching the P/V can be successful in terms of some, but not all stequeometric parameters, suggesting that the cellular responses could be differently affected by hydrodynamics/aeration differences between shake flasks and bioreactors.

Keywords: S. lividans, power input, morphology, rheology, scale-up, shake flasks.