Physiological effects of conditioned medium and passage number on Spodoptera frugiperda Sf9 serum free cultures

Abstract: The aim of this study was to better understand the role of conditioned medium (CM) in Spodoptera frugiperda Sf9 insect cell proliferation and recombinant protein production using the baculovirus expression system.CM was found to stimulate cell proliferation. Addition of CM and 10 kDa CM filtrate to an Sf9 culture decreased the lagphase and the maximum cell density was reached earlier than for cultures in fresh medium. The positive effect of 10 kDa CM filtrate showed that CM contains at least one small growth promoting factor. The effect was not eliminated by trypsin treatment. Addition of CM or 10 kDa CM filtrate to Sf9 cultures was found to have a negative effect on the recombinant protein production. The effect was thought to be indirect and most probably via the impact of CM on cell physiology. CM was also found to contain proteinase activity. The proteinase was identified as Sf9 cathepsin L. A proform with a molecular mass about 49 kDa and two active forms at about 39 and 22 kDa were found. The role of cathepsin L in Sf9 cultures is not yet clear. However, the knowledge of the presence of this proteinase in CM can be of great value for improving product quality and yield. Further, CM was found to have other properties as well: a concentrated fraction of CM exhibited strong antibacterial activity towards Bacillus megaterium and a weaker activity towards Escherichia coli. B. megaterium lysed rapidly after incubation in the CM fraction.Repeated subculturing of Sf9 cells provoked a switch in growth kinetics. After 30-45 passages the cells started to proliferate earlier after inoculation and addition of CM had no longer a growth stimulating effect. However, CM still stimulated growth of a culture with low passage (LP) number (up to 45 passages). High passage cells (HP cells, over 100 passages) displayed a shorter lagphase than LP cells and the culture reached the maximum cell density 24-48 h earlier. Cell cycle analysis showed that the Sf9 cells were transiently synchronised in the G2/M phase 10 h after inoculation, before proliferation was initiated. This synchronisation was more pronounced for HP cells than for LP cells, which correlated to a higher recombinant protein production in baculovirus infected HP cells than in LP cells. Synchronisation of cells in G2/M by yeastolate-limitation before infection with baculoviruses suggested that the degree of synchronisation is connected to the cell density dependent decrease in recombinant protein production of Sf9 cultures.