For the production of glycosylated proteins, such as monoclonal antibodies, hormones, and blood clothing factors, generally mammalian cells are used. Mammalian cells are preferred over other expression systems, such as bacteria or yeast, because they are able to glycosylate proteins in a human-like pattern. The latter is necessary for the functionality of these proteins in the human body. However, a major drawback of the use of mammalian cell culture is the low growth rate and low specific productivity of mammalian cells, which results in relatively long processes and low final product titers. The increasing demand for active pharmaceutical ingredients produced by mammalian cells has driven research and process development towards the development of production processes with an increased volumetric productivity. The volumetric productivity is the amount of product produced per liter bioreactor volume per day. An increase in volumetric productivity can be attained by an increase in viable-cell density and/or an increase in specific productivity. An approach to increase the viable-cell density is the use of perfusion processes. In a perfusion system continuously nutrients are refreshed, while the cells are retained in the bioreactor by using a cell-retention device (Figure 1). As a result high viable-cell densities can be reached (20-40x109 viable cells/dm3) and during the production phase product can be harvested for about 15-75 days. Thus, a high volumetric productivity can be reached. Figure 1. Schematic overview of perfusion bioreactor However, these systems are considered more complex than the conventional (fed-)batch process due to the continuous operation, the requirement of a cell-retention device, and the long run time. To develop more robust and optimal perfusion processes more insight in the influence of the key process parameters on growth characteristics, nutrient consumption, metabolism and productivity is needed. The work described in this thesis focuses on the development of robust, high cell-density, acoustic perfusion processes from bench to pilot scale. The acoustic-cell separator, the BioSep, is used as a cell-retention device. In this device the cells are separated from the medium broth using acoustic standing waves. Furthermore, a hybridoma cell line producing an IgG1 antibody is used as a model system. In Chapter 2 the reliability and long-term stability of a pilot-scale, acoustic, cell-retention device (200 dm3/d) in a perfusion process are described. As the BioSep is an external device, it is possible that depending on the recirculation rate nutrient gradients occur in the external loop, which could affect cell metabolism. Therefore, the effect of recirculation rate on nutrient gradients, cell metabolism, viability and productivity has been studied. In this study, it was shown that a perfusion process using a pilot-scale acoustic cell-retention device (200 dm3/d) is reliable and simple to operate, which in this case resulted in a stable, 75-day, cultivation of a hybridoma cell line producing a monoclonal antibody. The recirculation rate had a significant effect on the oxygen concentration in the external loop, with oxygen being depleted within the cell-retention device at recirculation rates below 6 m3/mreactor3/d (=600 dm3/d). The oxygen depletion occurring at low circulation rates correlated with a slightly increased lactate production rate. For all other parameters no effect of the recirculation rate was observed, including the specific productivity of monoclonal antibodies and cell death as measured through the release of lactate dehydrogenase. In Chapter 3 and 4 the effect of the feed and bleed rate, which are key parameters for perfusion systems, on the performance of perfusion processes are described. The viable-cell density, viability, growth rate, death rate, lysis rate and cell-cycle distribution of a hybridoma cell line producing an IgG1 have been studied over a range of specific feed and bleed rates. The results of this study are described in Chapter 3. The feed and bleed rates applied in the different cultures could be divided into two regions based on the viable-cell density and cell-cycle distribution. The cultures in the first region, low feed rates (0.5 and 1.0 1/d) combined with low bleed rates (0.05 and 0.10 1/d), were nutrient limited, as an increase in the feed rate resulted in an increase in the viable-cell density. The cultures in the second region, high feed and bleed rates, were non-nutrient limited. In this region the viable-cell density was independent of the feed rate and decreased more or less linear with an increase in the bleed rate. The last finding suggests that for feed and bleed rates falling in the second region cells were limited by a cell-related factor. Comparison of trypan-blue dye-exclusion measurements and lactate-dehydrogenase activity measurements revealed that cell lysis was not negligible in this bioreactor set-up. Therefore, lactate-dehydrogenase activity measurements were essential to measure the death rate accurately. The specific growth rate was nearly constant for all tested conditions. The viability increased with an increase of the bleed rate and was independent of the feed rate. Furthermore, the specific productivity of monoclonal antibody was constant under all tested conditions. The effect of the feed and bleed rate on cell metabolism was further investigated using metabolic flux analysis, which is described in Chapter 4. The two regions as observed in Chapter 3 were also observed based on the nutrient consumption rates. Under all tested feed and bleed rates the biomass concentration as calculated from the nitrogen balance (biomass-nitrogen) increased linearly with an increase in feed rate as would be expected. However, depending on the size of the feed and bleed rate this increase was attained in two different ways. At low feed and bleed rates (Region I) the increase was obtained through an increase in viable-cell concentration, while the cellular-nitrogen content remained constant. At high feed and bleed rates (Region II) the increase was attained through and increase in cellular-nitrogen content, while the cell concentration remained constant. Per gram biomass-nitrogen the specific consumption and production rates of the majority of the nutrients and products were identical in both regions as were most of the fluxes. The major difference between the two regions was an increased flux from pyruvate to lactate and a decreased flux of pyruvate towards citrate in region II. The decreased in-flux at the level of citrate can either be balanced by a decreased out-flux towards lipid biosynthesis leading to a lower fraction of lipids in the cell or by a decreased out-flux towards the tri-citric-acid cycle resulting in a decreased energy generation, or a combination of these. In conclusion, for the optimal design of a perfusion process it should first be established whether viability is an important parameter. If not, the bleed rate should be chosen as low as possible. If low viabilities should be avoided the bleed rate should be chosen higher, with the value depending on the desired viability. Next, as the specific productivity on a cell basis is constant an increase in biomass through an increase in viable-cell density is preferred over an increase in nitrogen content. Therefore, the feed rate should be set at such a rate that the cells are just in the region II. In Chapter 5 the applicability of the acceleration stat (A-stat) technique to shorten investigations on the relation between dilution rate, and growth behaviour and cell metabolism of hybridoma cells is described. When steady state was reached in a chemostat bioreactor operated at 0.70 1/d the dilution rate was smoothly decreased to 0.10 1/d, and finally returned to its initial rate. Furthermore, steady-state data of three individual chemostat runs (D=0.10, 0.25 and 0.40 1/d) are compared with the A-stat data. Generally, the values obtained in the A-stat were similar to the chemostat values at dilution rates between 0.70 and 0.25 1/d. During the A-stat run (around D=0.25 1/d) a shift in cell physiology occurred, which resulted in a higher viable-cell concentration, lower specific consumption rates, lower residual nutrient concentrations and a lower specific productivity. Nevertheless, the viability, growth rate, yield of lactate on glucose, and the ratio of specific consumption rate of essential amino acids over the specific glucose consumption rate (qaa/qglc) were not affected by the shift in physiology. In conclusion, the A-stat can be a useful technique when studying cell growth and metabolism in hybridoma cells, especially when development time is limited and a rough estimation of metabolic data is acceptable. A whole range of dilution rates can be studied in a time frame in which about 3-4 chemostat runs can be done, and consequently changes in metabolism can be determined more accurate. Another production process that is applied for obtaining high volumetric productivities is the fed-batch process. In Chapter 6 the different aspects determining the choice between a fed-batch and a perfusion system for the production of secreted proteins by mammalian cells are described. First of all, the current state-of-the-art for both the fed-batch and perfusion processes is given. Next, the factors that determine the choice for a production system are discussed one by one. The factors that are taken into consideration are product quality, cell-line stability, performance, consistency and robustness, process development and scale up, economical aspects, and logistics and downstream processing. Last, future research directions are given to further improve both processes. The main characteristics of a fed-batch process, simplicity and short run time, make this process robust and facilitate fast and thorough process development. This, together with the fact that most companies have experience with a form of fed-batch, makes the fed-batch process currently the process of choice. Perfusion processes are important in case the product is instable or product quality is very sensitive to process conditions. It is rather difficult to make a general evaluation of the predicted production costs of both culture principles, because it depends heavily on the assumptions that are made.
|Qualification||Doctor of Philosophy|
|Award date||10 Oct 2007|
|Place of Publication||[S.l.]|
|Publication status||Published - 2007|
- cell culture