Knockout of a difficult-to-remove CHO host cell protein, lipoprotein lipase, for improved polysorbate stability in monoclonal antibody formulations

Abstract

While the majority of host cell protein (HCP) impurities are effectively removed in typical downstream purification processes, a small population of HCPs are particularly challenging. Previous studies have identified HCPs that are challenging for a variety of reasons. Lipoprotein lipase (LPL)-a Chinese hamster ovary (CHO) HCP that functions to hydrolyze esters in triglycerides-was one of ten HCPs identified in previous studies as being susceptible to retention in downstream processing. LPL may degrade polysorbate 80 (PS-80) and polysorbate 20 (PS-20) in final product formulations due to the structural similarity between polysorbates and triglycerides. In this work, recombinant LPL was found to have enzymatic activity against PS-80 and PS-20 in a range of solution conditions that are typical of mAb formulations. LPL knockout CHO cells were created with CRISPR and TALEN technologies and resulting cell culture harvest fluid demonstrated significantly reduced polysorbate degradation without significant impact on cell viability when compared to wild-type samples. Biotechnol. Bioeng. 2017;114: 1006-1015. © 2016 Wiley Periodicals, Inc.

Keywords: CHO cell; CRISPR; host cell protein; lipase; polysorbate.

Figure 1

Venn diagram of difficult-to-remove HCP impurities and the mechanism by which they challenge clearance in biopharmaceutical manufacturing. Diagram not drawn to scale.

Figure 2

Average degradation rate of PS-80 in solutions containing recombinant CHO LPL (produced in E. coli) in different solution conditions at 37 °C for 24 hours.

Figure 3

Relative LPL protein expression measured with a LPL-specific MRM assay for initial 26 TALEN-treated and 15 CRISPR-treated CHO-K1 cells. LPL expressions are normalized to the wildtype control, n = 1. Wildtype is shown in blue, TALEN in red, and CRISPR in green.

Figure 4

Characterization of LPL protein expression from CHO lpl knockout cell lines by western analysis and MRM. (A) Expression of LPL protein probed by the N-terminus LPL antibody (Anti-LPL(N)) and C-terminus LPL antibody (Anti-LPL (C)). The predicted molecular weight of LPL is 53 kDa. The red arrow indicates the native LPL band. The yellow arrow indicates the altered LPL protein expressed by cell line 42, with predicted molecular weight of 52.2 kDa. The white arrow indicates the altered LPL protein expressed by cell line 45, with predicted molecular weight of 29.9 kDa. (B) Relative expression of LPL ITG peptide by MRM. (C) Relative expression of LPL GLG peptide by MRM. LPL expressions are normalized to the wildtype control. Error bars represent the standard error of the mean from three technical replicates.

Figure 5

Cell culture performance of CHO-K1 lpl knockout cell lines. (A) Integrated viable cell density (IVCD), relative to the wild-type control, (B) viability cell culture profile of CHO-K1 wild-type, and cell lines 42, 43, 44, 45, and 46. Error bars represent the standard error of the mean from two biological replicates. (C) Total extracellular protein expression at day 4 of cell culture. Error bars represent the standard error of the mean from three biological replicates.

Figure 6

Effect of LPL on the degradation of (A) PS-80 and (B) PS-20 by CHO HCCF from CHO-K1 wild-type control (WT), CHO-K1 lpl knockout cell lines (42, 43, 44, 45, and 46), and apoC-III (I). Error bars represent the standard error of the mean from three biological replicates.