Endogenous BiP reporter system for simultaneous identification of ER stress and antibody production in Chinese hamster ovary cells

Highlights

• Endogenous tagging of mGFP at BiP generated an ER stress monitoring system.

• BiP-mGFP cell line was functional in detecting UPR activation upon ER stress.

• Basal expression levels of BiP affect recombinant protein productivity.

• A correlation was found between productivity and fluorescence intensity.

Abstract

As the biopharmaceutical industry expands, improving the production of therapeutic proteins using Chinese hamster ovary (CHO) cells is important. However, excessive and complicated protein production causes protein misfolding and triggers endoplasmic reticulum (ER) stress. When ER stress occurs, cells mediate the unfolded protein response (UPR) pathway to restore protein homeostasis and folding capacity of the ER. However, when the cells fail to control prolonged ER stress, UPR induces apoptosis. Therefore, monitoring the degree of UPR is required to achieve high productivity and the desired quality. In this study, we developed a fluorescence-based UPR monitoring system for CHO cells. We integrated mGFP into endogenous HSPA5 encoding BiP, a major ER chaperone and the primary ER stress activation sensor, using CRISPR/Cas9-mediated targeted integration. The mGFP expression level changed according to the ER stress induced by chemical treatment and batch culture in the engineered cell line. Using this monitoring system, we demonstrated that host cells and recombinant CHO cell lines with different mean fluorescence intensities (MFI; basal expression levels of BiP) possess a distinct capacity for stress culture conditions induced by recombinant protein production. Antibody-producing recombinant CHO cell lines were generated using site-specific integration based on host cells equipped with the BiP reporter system. Targeted integrants showed a strong correlation between productivity and MFI, reflecting the potential of this monitoring system as a screening readout for high producers. Taken together, these data demonstrate the utility of the endogenous BiP reporter system for the detection of real-time dynamic changes in endogenous UPR and its potential for applications in recombinant protein production during CHO cell line development.

Introduction

Biotherapeutics, such as therapeutic proteins, show remarkable market growth in the biopharmaceutical industry with three to four of the top 10 best-selling drugs being produced using Chinese hamster ovary (CHO) cell lines (Urquhart, 2021). Accordingly, increasing the production of therapeutic proteins in CHO cell lines is an important research subject. While monoclonal antibodies (mAbs), vaccines, and hormones have been mainly used as biological therapeutics, complex protein therapeutics, such as viral antigens and bi- or tri-specific mAbs, which are difficult to express (DTE) proteins, have been recently developed and produced (Tihanyi and Nyitray, 2021). However, excessive and complicated protein production causes protein unfolding and misfolding, triggering endoplasmic reticulum (ER) stress. When ER stress occurs, cells mediate the unfolded protein response (UPR) pathway to restore protein homeostasis and folding capacity of the ER. UPR is a signaling pathway involving several types of ER-localized chaperones, folding enzymes, and transcription factors to maintain ER homeostasis and enhance protein-folding capacity. When UPR is activated, protein folding, translation, and translocation are regulated, and simultaneously, unfolding and misfolding proteins are degraded through ER-associated protein degradation (ERAD) or autophagy, such that the ER capacity can be maintained (Hetz et al., 2020).

There are three main pathways identified in the UPR, which are mediated by ER stress sensors, namely IRE1α, ATF6, and PERK. IRE1α activates the spliced form of X-box binding protein 1 (XBP1s) by cutting the 26-nucleotide intron of the mRNA encoding for the transcription factor XBP1 and inducing shifting of the translational open reading frame. Activated XBP1s are involved in ER protein translocation, folding, secretion, and degradation (Kaufman, 2002; Shen et al., 2001). ATF6 migrates to the Golgi apparatus under ER stress where it is cleaved by site-1 and site-2 proteases and converted into the transcription factor ATF6p50 (Ye et al., 2000). ATF6p50 migrates from the Golgi back to the nucleus and is involved in protein folding, secretion, and ERAD (Shoulders et al., 2013). PERK phosphorylates eIF2α to temporarily decrease protein synthesis preventing unfolded and misfolded protein accumulation in the ER (Kaufman, 2002). Simultaneously, it initiates the expression of ATF4, a transcription factor involved in protein synthesis and apoptosis. This transcription factor can restore protein synthesis or induce apoptosis by regulating the expression of GADD34 or one of the proapoptotic factors, CCAAT/enhancer-binding protein homologous protein (CHOP) (Han et al., 2013; Marciniak et al., 2004). In resting cells, the three ER stress sensors are maintained in an inactive state through physical interactions with the ER chaperone GRP78/BiP. Under ER stress conditions, such as unfolded protein accumulation, BiP preferentially binds to unfolded proteins, thereby releasing the ER stress sensors to allow them to activate downstream transcription factors (Bertolotti et al., 2000). When the UPR pathway is activated and ER stress is resolved, ER homeostasis is restored. However, cell death is induced if the UPR pathway fails to relieve ER stress and continues excessively. Thus, the UPR pathway is an important pathway involved in protein synthesis and cell death.

Previous studies have revealed a correlation between the expression level of BiP and protein productivity (Bakunts et al., 2017; Du et al., 2013; Kober et al., 2012). Therefore, monitoring the UPR degree by the amount of BiP could be valuable for achieving high productivity of therapeutic proteins with the desired quality. However, to date, there is no monitoring system for checking ER stress in real time using UPR molecules. Thus, we engineered CHO cell lines using the endogenous tagging method and clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) to construct a UPR monitoring system. We integrated mGFP into endogenous HSPA5 encoding BiP, a major ER chaperone, and the primary ER stress activation sensor. Thus, we confirmed that the mGFP expression level changed according to the ER stress induced by chemical treatment and batch culture in the engineered cell lines. These results showed that HSPA5 is an efficient reporter gene. Unlike previous studies (Poulain et al., 2019; Roy et al., 2017; Samali et al., 2010), this monitoring system has the advantage of being able to check the entire UPR level in real time, and not just a specific UPR branch, which can help construct highly productive cell lines efficiently. Finally, we validated the feasibility and efficacy of the present fluorescence-based UPR monitoring system by providing a method for generating and selecting highly stable mAb-expressing CHO cell lines while continuously monitoring endogenous UPR levels.