Introduction
Mammalian cell culture, most notably Chinese Hamster Ovary (CHO) cells,
are commonly used to produce recombinant proteins such as monoclonal
antibodies (mAbs) by the biopharmaceutical industry for use in both
therapy and diagnosis (Goulet & Atkins, 2020). Such proteins offer
enormous advances in disease therapy (or diagnosis) and are products of
high commercial value. Over the last several decades there have been
significant advances in the efficiency of CHO cells to produce these
recombinant proteins (Budge et al., 2020; Kunert & Reinhart, 2016;
Sharker & Rahman, 2020; Srirangan, Loignon, & Durocher, 2020).
However, the rate of protein production remains a significant
bottleneck, particularly for many of the novel format based
biotherapeutics now in development, and therefore there remains
substantial interest in further advancing the productivity of CHO cells
(Budge et al., 2020; Sharker & Rahman, 2020).
The rate of protein synthesis (mRNA Translation) is one important
limiting determinant of recombinant protein production (Godfrey et al.,
2017; Khoo & Al-Rubeai, 2009; O’Callaghan et al., 2010; Roobol et al.,
2020; Smales et al., 2004). One key regulator of protein synthesis is
the mechanistic target of rapamycin complex 1 (mTORC1) (X. Wang &
Proud, 2006). mTORC1 is a protein kinase that is activated by a variety
of upstream signals, most notably amino acids (Kim, Goraksha-Hicks, Li,
Neufeld, & Guan, 2008; Sancak et al., 2008) and growth factors (Inoki,
Li, Zhu, Wu, & Guan, 2002; B. D. Manning, Tee, Logsdon, Blenis, &
Cantley, 2002), and acts as a master regulator of anabolic processes
including protein synthesis (Proud, 2019) and ribosomal biogenesis
(Iadevaia, Liu, & Proud, 2014; Saxton & Sabatini, 2017), both of which
are critical for efficient protein production. This is achieved through
phosphorylation of a number of downstream effectors such as ribosomal
protein S6 kinase 1 (S6K1) (Chung, Kuo, Crabtree, & Blenis, 1992),
eukaryotic elongation factor 2 kinase (eEF2K) (X. Wang et al., 2014),
eIF4E binding protein 1 (4E-BP1) (Beretta, Gingras, Svitkin, Hall, &
Sonenberg, 1996), and Maf1 (Michels et al., 2010), a regulator of
ribosomal RNA transcription. mTORC1 signalling is thus potentially a
major positive regulator of efficient, high-level production of
recombinant proteins in mammalian cells. Increased phosphorylation of
4E-BP1, which permits increased translation initiation, has indeed been
shown to increase production of interferon-γ (Kaur et al., 2007). We
have also previously shown that the ratio of eIF4E to 4E-BP1 correlates
to higher cell productivity (Jossé, Xie, Proud, & Smales, 2016). In
addition to this, mTORC1 drives other anabolic pathways that contribute
to cell growth and faster protein production, including lipid synthesis
(Caron, Richard, & Laplante, 2015)and ribosome biogenesis (Iadevaia et
al., 2014).
mTORC1 is activated by the small GTPase Rheb (Ras homologue enriched in
brain) when it is in its GTP-bound form; its conversion to the inactive
GDP-bound state is promoted by the tuberous sclerosis complex which
includes the proteins TSC1 and TSC2, the latter acting as a
GTPase-activator protein (GAP) for Rheb (Garami et al., 2003; Inoki,
Zhu, & Guan, 2003; Tee, Manning, Roux, Cantley, & Blenis, 2003). In
turn, the ability of TSC1/2 to impair Rheb function is inhibited by
signalling events activated by hormones, mitogenic stimuli and growth
factors (Huang & Manning, 2008; Inoki et al., 2002; B.D. Manning &
Cantley, 2003; Zhang et al., 2003). We have recently discovered that
several mutants of Rheb (which occur in certain human cancers) are
resistant to the GAP activity of TSC2 and are thus ‘constitutively
active’, promoting high levels of mTORC1 activity in human cells
(Jianling Xie et al., 2020).
The folding, assembly and maturation of most secreted proteins occurs,
with the assistance of chaperones, within the endoplasmic reticulum
(ER). Homeostatic control of the ER is mediated by the unfolded protein
response (UPR) (Preissler & Ron, 2019). In response to protein
disequilibrium in the ER, the protein kinase RNA-like ER kinase (PERK)
undergoes homodimerization and becomes active (Cui, Li, Ron, & Sha,
2011). PERK then phosphorylates eukaryotic initiation factor 2α (eIF2α)
at serine-51 (Harding, Zhang, Bertolotti, Zeng, & Ron, 2000; Harding,
Zhang, & Ron, 1999). eIF2α is a component of the heterotrimeric
initiation factor eIF2 which is required to deliver the initiator
methionyl-tRNA to the 43S preinitiation complex in order to initiate
mRNA translation (Merrick & Pavitt, 2018), a process that requires the
hydrolysis of eIF2-bound GTP to GDP (Kapp & Lorsch, 2004). In order to
facilitate subsequent initiation events, the guanine exchange factor
(GEF) eIF2B binds eIF2 and catalyses the exchange of the GDP for GTP to
regenerate active eIF2.GTP. When eIF2α is phosphorylated in response to
upstream stress, eIF2 is unable to be released from eIF2B and thus is no
longer able to initiate translation (Bogorad, Lin, & Marintchev, 2018).
This has two contrasting consequences which in combination mediate
homeostasis of the ER. Firstly, it leads to global inhibition of protein
synthesis, decreasing the ‘load’ of new proteins to be folded within the
ER (Wek, 2018). However, secondly, certain mRNAs utilise their upstream
open reading frames (uORFs) to undergo preferentially translation in
response to P-eIF2α mediated inhibition of general protein synthesis
(Harding, Novoa, et al., 2000). One such protein is activating
transcription factor 4 (ATF4). Translation of ATF4 is thus selectively
upregulated in response to phosphorylation of eIF2α. ATF4 is a
transcription factor that drives the expression of genes responsible for
protein homeostasis which are collectively called ER quality control
genes (ERQC) (Preissler & Ron, 2019). ERQC genes consist mainly of
chaperones and other protein folding genes; thus upregulation of ATF4
counters ER stress by increasing the protein folding capacity of the
cell (Shaffer et al., 2004; Sriburi, Jackowski, Mori, & Brewer, 2004;
M. Wang & Kaufman, 2016).
Here we show that two such Rheb mutants, T23M and E40K, drive
constitutive mTORC1 signalling in CHO cells and enhance the production
of recombinant protein including, importantly, its secretion from the
cells. Manipulation of mTORC1 signalling by these Rheb mutants therefore
has the clear potential to enhance the production of proteins of high
commercial value.