The CHO miRNA Transcriptome - MicroRNAs as Tools in Biopharmaceutical Production - page 53

through which the performance of cell factories for recombinant protein production
can be enhanced; second, the use of miRNAs as indicators of CHO cell pheno-
types, eventually resulting in their application as predictors of specific cellular traits
such as productivity during cell line development—as been recently achieved using
mRNA levels (Clarke et al. 2011 )—or as indicators of cellular state during process
optimization.
In the light of published studies in other species, both applications seem realistic:
genetic mutations, deletions or heterologous expression of miRNAs in mammalian
cell lines can severely impact cellular behavior in terms of growth (Yu et al. 2012 ),
metabolism (Krützfeldt and Stoffel 2006 ), and protein translation (Nottrott et al.
2006 ), as well as stress resistance and cell death (Frank et al. 2011 ). This was
shown for several miRNAs, but most remarkably in the case of miR-21. Initially
miR-21 drew attention as it was repeatedly found overexpressed in many different
tumour types. In vitro studies that followed these observations attributed miR-21
oncogenic activity by showing that it negatively regulates the expression of numerous
tumor suppressor protein mRNAs (Zhu et al. 2008 ). Finally, also in vivo miR-
21 was confirmed as an oncogene ( oncomiR ) by Frank Slack's lab, showing that
conditional overexpression of miR-21 in mice results in the formation of malignant B-
Lymphocytes. In contrast, inactivation of miR-21 overexpression caused the tumors
to regress, presumably due to increased levels of apoptosis (Medina et al. 2010 ).
Similarly, other miRNAs have been implicated in the regulation of glucose/glutamine
metabolism (Zhu et al. 2011 ; Gao et al. 2009 ) and cell cycle regulation (Cloonan
et al. 2008 ), p53 mediated stress response (Hermeking 2010 ), or protein secretion
(Lovis et al. 2008 ). For an in-depth discussion of putative engineering candidates
please consult Chap. 5 of this topic and respective reviews (Barron et al. 2011 ; Müller
et al. 2008 ).
The possibility of applying miRNAs as biomarkers of cellular state during cell
line development and process optimization, derives from technologies that are be-
ing developed for prognosis and diagnosis of diseases: by mining the plentitude of
miRNA transcriptome data generated from several disease specimens, the levels of
single or sets of miRNAs (so-called expression signatures) can be correlated to the
presence or absence of a disease or even to determine the disease state for a given
patient (Jeffrey 2008 ). Initally RNA extracts for biomarker discovery were obtained
from tissue samples. Recently, researchers also discovered that miRNAs along with
other types of RNAs and proteins are secreted from cells into the bloodstream (or
find their way into the bloodstream after lysis of diseased cells). The field has now
been exteneded to plasma and serum biomarker discovery (Gilad et al. 2007 ). Poten-
tial analogous bioprocessing applications lie in the identification of intracellular or
extracellular miRNAs that can indicate early on whether a cell harbors a favorable
phenotype or not, thus shortening the timelines during cell line development and
making screening more targeted and efficient.
Although the road towards these future applications is paved by innovations in the
field of diagnosis and treatment of disease, what is missing is a profound character-
ization of miRNA expression in common recombinant protein production cell lines,
including both CHO, human (HEK293, PerC6) or other animal derived cell lines;
and while few CHO miRNA profiling studies have already been undertaken using