Minimising the time from 'scientific breakthrough'to clinical trial of a 'drug candidate' protein is a critical component leading to a successful product release. Crystallographic characterisation has become a standard requirement prior to clinical trial requiring milligram quantities of protein. The optimisation of protein expression systems is therefore of great commercial and social importance and represents a significant technical challenge. Without it, making enough protein for crystallography can quite literally take years. Baculoviral expression of recombinant protein by infection of an insect cell host is a well established technique in modern biotechnology. Although a limit to recombinant protein production in batch culture exists the mechanism has not been demonstrated. In particular, there has been no discussion of how biomass accumulation kinetics relate to the system limits in terms of final recombinant protein yield. The central aim of this thesis was therefore to quantitatively account for the dynamic behaviour in macromolecular compartments after baculovirus infection of insect cells, the rationale being that a rudimentary level of mechanistic structure can greatly enhance our ability to capture transient behaviour. The catalytic mass dictates the rate of total biomass accumulation in the baculovirus expression vector system (BEVS) and is directly proportional to the total RNA content of both baculovirus infected and uninfected Sf9 cells. During infection the total RNA concentration reaches a catalytic limit causing a switch from exponential to zero order mass accumulation kinetics. Importantly, this extends to individual cells as confirmed using a population balance model for the cell volume distribution after the switch to linear growth. By flow cytometry, a positive correlation between RNA content and cell size post infection validated this modelling assumption. The rate of mass accumulation slows down during the first 12 hours post infection (hpi). This is consistent with the decrease in both specific consumption rates of glucose and oxygen when using cell mass rather than cell number as a basis. A decrease in the geometric standard deviation (óg) of the cell volume distribution during the first 12 hpi indicates that cells enter the lower growth rate at times inversely proportional to their volume. Using several approaches no obvious biological mechanism to account for the empirical model was identified. The use of óg kinetics provides a novel tool for characterising the relative behaviour of infected cells in the BEVS. The effect of multiplicity of infection (MOI) on virus timing events between cultures was also tested. Little or no effect of MOI was observed on the timing of virus induced events during synchronous infections. The óg kinetics did indicate virus events occur 5 hours earlier at a MOI of 100 compared to a MOI of 20 plaque forming units per cell. There was however, no significant evidence of earlier death kinetics when measured using Trypan blue dye exclusion to measure cell membrane integrity. Virtually no effect of MOI on virus timing was observed using â-galactosidase production profiles. The viral DNA mass (vDNA) was measured using real time quantitative PCR (RTQ PCR) and has a doubling time of 2 hours. A vDNA template limited replication model fit the data well. Viral replication proceeds from 6 until 24 hpi with the average infected cell accumulating between 12 000 and 84 000 vDNA copies when replication stops. In theory, a dynamic equilibrium could have been present after the commencement of virus budding but this was not the case. At least 62% of the total DNA increase post infection is viral. No more than 16% of the total vDNA produced actually bud from the infected cell. This overproduction of vDNA is probably due to the wild type history of the virus, which normally occludes virions in a crystalline polyhedrin matrix within the cell nucleus as part of its life cycle. The approach taken here provides a framework for characterisation of both viral and total mass accumulation with the use of a few simple intracellular macromolecular pools. This thesis demonstrates that the BEVS limit in batch culture is not simply a result of the exhaustion of an amino acid using a case study of amino acid consumption by uninfected Sf9 cells for a 300 hour culture period. Future attempts to identify the system limits and will require the linkage of a mechanistic model with a more extensive and accurate analysis of important metabolites and specific intracellular species.
| Identifer | oai:union.ndltd.org:ADTP/253736 |
| Creators | Rosinski, Matthew |
| Source Sets | Australiasian Digital Theses Program |
| Detected Language | English |
Page generated in 0.0025 seconds