From left: Dr Wong Sze Yue, Dr Lin Pao Chun and Dr Song Zhiwei
Pao-Chun Lina, Kah Fai Chana, Irene A. Kiessa, Joselyn Tana, Wahyu Shahreela, Sze-Yue Wonga and Zhiwei Songa,b
a Bioprocessing Technology Institute, Agency for Science, Technology and Research (A*STAR), Singapore
b Department of Biochemistry, Yong Loo Lin School of Medicine, National University of Singapore, Singapore
Published in mAbs 2019 11(5): 965-976 (Online Version)
Monoclonal antibody (mAb), an important class of cancer biotherapeutic, is produced in mammalian cells such as Chinese hamster ovary (CHO) cells. The glutamine synthetase (GS) selection system is commonly employed in CHO cells to select for cells stably producing biologics such as mAb. The selection system for stable biologic-producing cells is crucial as it affects the efficiency of identifying high producing cells. To increase selection stringency, the use of the GS inhibitor, methionine sulfoximine(MSX) is frequently employed such that only high producing cells can survive the inhibition.
However, the use of MSX complicates downstream purification processes and causes regulatory concerns. Furthermore, due to the loss of selection pressure,the productivity of a selected high producing clone is often lost upon MSX removal during scaling up to large scale bioreactor. One method to achieve high and stable productivity without the use of MSX is to drive the expression of GS under an attenuated promoter to reduce GS expression. However, the use of a separate promoter for the selection marker may generate cells which only carry the GS selection marker system, while the biologic could be lost or integrated into less active sites.
We thus sought an alternative method to improve selection stringency by attenuating GS activity. We expressed both attenuated GS and the biologic in the same expression cassette, which reduces the chances of selecting GS-positive, but biologic-negative cells. We evaluated the use of a congenital mutation, R324C, as the attenuated GS selection marker for the generation of stable cell line expressing mAb. Our system enhanced the bulk productivity of stable cells generated in the absence of MSX to levels higher than or similar to MSX selection while maintaining stability of mAb production.
We further identified novel GS mutations through structure-based sequence alignment that resulted in varying degrees of attenuated GS activity. These were similarly able to enhance productivity in correlation to their GS activity levels, confirming the principle of our selection system. More importantly, our system resulted in improved efficiency of identifying stable clones, where at least 80% of the single clones isolated were stable mAb producers over at least 60 generations, which is a clear time advantage. Taken together, we have an attenuated GS selection system that can be used to efficiently produce biologics at high titer and stability.
Figure 1. Comparison of attenuated GS R324C with GS wild type (WT) and with MSX for the stable production of mAb. (A) Antibody titer measurement of independently transfected stable cell pools. Stability assessment of mAb production over 8 weeks(>60 generations) of single clones generated by (B) GS WT+MSX and (C) GS R324C selection.