The CHO miRNA Transcriptome - MicroRNAs as Tools in Biopharmaceutical Production - page 61

cases) rapidly degraded (the “star” strand, e.g. miR-221 ∗ ) while the other strand is
incorporated into a protein machinery and hence more stable and easier to detect (the
mature strand, e.g. miR-221). The other possible outcome of miRNA processing is
that a hairpin actually gives rise to two equally abundant mature miRNAs; in this
case the affixes “-5p” and “-3p” are added to the names to distinguish both forms—
a much clearer nomenclature, which will soon be installed for all miRNAs, thus
replacing the often confusing mature/star nomenclature (personal communication
with Dr. Griffiths-Jones).
In case of miRNAs originating from the mir-221 hairpin, miR-221-3p was ob-
served in several species to be higher abundant than miR-221-5p, and therefore
became known as miR-221, while miR-221-5p was further regarded as miR-221 ∗ .
In CHO cells, however, not only equal expression levels for both mir-221 derived
miRNAs were reported (Hackl et al. 2011 ), but also that the log-ratio in levels of
miR-221-3p and miR-221-5p changes from positive (i.e. more miR-221-3p miR-

221) for both serum-free DHFR and K1 host cells to negative for the respective
DHFR and K1 recombinant IgG producing subclones (i.e. less miR-221-3p). With
this, two independent studies using different CHO cell types have identified a re-
lationship between the regulation of miR-221 and recombinant cell phenotype. As
functional characterization of mir-221 is most likely underway, it will be exciting
to see whether targeted reduction of miR-221-3p levels lead to an improvement in
recombinant protein production.

=

4.5 Perspectives on the Use of miRNA Tools in Biopharmaceutics

The data summarized here on miRNA expression and relevance in CHO cells clearly
point to (i) a high number of expressed miRNAs, (ii) a dynamic in miRNA levels in
response to culture conditions and cellular state and (iii) clear evidence that individual
miRNAs like miR-7 and miR-466h can modify growth and recombinant protein
production capacities or alter the stress responses of CHO cells, respectively. These
first evidences for the opportunities of miRNA research in CHO cells also sets the
need for developing further tools that will facilitate miRNA research in CHO cells:
first, this encompasses the complete characterization of the miRNA transcriptome
including the prediction and verification of novel (CHO-specific) small RNAs as well
as large-scale analysis of miRNA transcription in various CHO cell lines. Secondly,
the recent publication of the CHO mRNA transcriptome in the form 15,000 complete
cDNA sequences with > 90 % coverage of homologous mouse transcripts (Becker
et al. 2011 ) will allow a detailed analysis of miRNA target sites in CHO cells and
whether they have evolved compared to its close rodent relatives or not. Based on such
analyses as well as experimental approaches employing transcriptomics, proteomics
and a combination of deep-sequencing and miRNA-target crosslinking (Chi et al.
2009 ; Nonne et al. 2010 ) it will be possible to gain insights into CHO-specific
miRNA

∼

mRNA target interactions.
Together, these tools will provide the basis to better understand individual or sets
of miRNAs that can be used to modulate (engimiRs) or predict (biomarkers) CHO