Baumeister, S. approach == During the scale up of a cell culture process for a late stage project we noticed that the cell age might influence the non-fucose level and thus the ADCC from the recombinant monoclonal antibody negatively. To ensure a higher product quality even at a high cell age we investigated the correlation between cell age group and non-fucose level in more detail by identifying underlying mechanisms with focus on the glycosylation enzymes GntIII and ManII that are overexpressed in this cell range. For this purpose a method was established in collaboration with Roche Diagnostics to quantify the gene expression degree of the glycosylation enzymes using RT-qPCR based on the RealTime ready technology. At the beginning of the project a cell age group study was conducted using shake flasks in serial culture mode to generate cells with different and especially high cell age. The cells were cultured under different selective conditions: (1) a combination of puromyin and MSX (2) only with puromycin, (3) only with MSX and (4) without selective pressure. Cells were freezing at diverse time points up to a cell age of 97 days. Later on a fed-batch experiment with almost all cell banking institutions of different cell age and selective conditions simultaneously was run. The fed-batch experiment was conducted with our in-house developed robotic cell culture system that enables a fully automated workflow based on shaken multiwell plates [1]. == Results and discussion == The data from the cell age group study verified the finding that the cell age negatively influences the non-fucose levels. We could show that the combination of puromycin and MSX stabilizes the non-fucose level at a high cell age up to 110 days whereas the use of puromycin or MSX only provides only a slight Rabbit Polyclonal to Mst1/2 stabilization. The cultivation without selective pressure resulted in the lowest ADX88178 non-fucose levels. Running the automated fed-batch experiment we could verify the results from the cell age study and we could also show the results from the ADX88178 automated system are predictive for a bioreactor. To understand the role from the glycosylation enzymes in this context we quantified ADX88178 the gene transcription degree of the recombinant glycosylation enzymes ManII and GnTIII. Since a suitable RT-qPCR method was not available we developed a customized method based on the RealTime ready technology in collaboration with colleagues from Roche Diagnostics. The mRNA levels of GntIII and ManII were measured over the course of the seed train study (shake flask) using the developed RT-qPCR method and related to corresponding glycosylation data of the mAb at the end from the fed-batch production run in our cell culture robotics facility (Figure1). At high cell age a direct correlation between non-fucose level and GntIII gene transcription level could be shown, whereby the highest mRNA levels were obtained intended for the cultures that used the combination of puromycin and MSX. The absence of selective pressure resulted in the lowest GntIII mRNA levels and thus the lowest non-fucose levels. The correlation between ManII mRNA and non-fucose level is not as clear because seen intended for the GntIII however the stabilizing effect of selective pressure could be shown. == Figure 1 . == (A) GntIII gene transcription data; (B) ManII gene transcription data; (C) Non-fucose level. For (C): Colored squares represent data from automated cell culture system and black squares from 10. 000 L bioreactor. The stabilizing effect of selective pressure on non-fucose level as well as the direct correlation between GntIII mRNA and non-fucose levels could be verified with a second recombinant cell line. Based on the results of this study cultivation recommendations regarding the in vitro cell age and the selection pressure for the seed- and inoculum-train of a production process could be provided. Also a direct correlation between selective pressure addition, GntIII and ManII transcription level and the non-fucose level intended for cells.