10.3 Managing Transcript Stability 203open reading frame predominantly initiates), disrupting secondary structure
that may cause variability in ribosome binding to the real RBS. The use of a BCD
resulted in a large improvement of transfer function correlations between multi-
ple gene contexts.
With the three individual systems in this section as starting points, it may be
possible to combine them into a general-use template for attenuating 5′ UTR-
and 3′ UTR-induced variations in gene expression. Further reductions in coding
sequence-induced variability could come from work that identifies sequences
and structures that, through targeted codon changes, minimize RNase E direct
entry and RNase III binding.
10.3.4 RBS Sequestration by Riboregulators and Riboswitches
As mentioned in the Introduction, RBS sequestration can hasten mRNA degra-
dation by leaving the transcript open to binding from RNases. Thus, riboregula-
tors, riboswitches, and similar structures that sequester the RBS can be used to
dynamically control mRNA stability, at least in part, via induction (or repression)
by a trans-activating RNA or small molecule (Figure 10.3c). Riboregulators are
functional RNA structures with two components: a cis-repressor and a trans-
activator. The cis-repressor is generally 5′ UTR RNA that folds into a conforma-
tion that base-pairs with the RBS, making it unavailable to ribosomes. This
repression can be relieved by the trans-activator RNA, an RNA that interacts
with the cis-repressor RNA so that the RBS is revealed for translation [54]. The
mechanistic design can also be reversed, whereby trans-RNA binding changes
the cis-RNA conformation to sequester the RBS [54]. Like riboregulators, ribos-
witches can function to bind the RBS in cis and prevent or enhance translation
by hindering, or allowing, ribosome docking. In the place of trans-RNA, ribos-
witches control RBS sequestration with conformational changes mediated by
small molecule ligand binding. Functional synthetic riboswitches have been
developed to bind a variety of ligands [102–105].
As both riboregulators and riboswitches involve adding secondary structure
into the 5′ UTR, their presence is likely to cause altered transcript stability due to
changes in RNase binding site accessibility. It may prove useful to think of these
systems in terms of their dual effects on translation rate and transcript degrada-
tion rate, which will require gathering data related to half-life, and not just final
gene expression and protein output. Both sets of data would be necessary to
decouple the contributions of translation rate changes from transcript stability
changes.
Efforts by the Collins lab to engineer synthetic riboregulators show how tran-
script stability changes can impact device outputs. The addition of cis-repressor
RNA in the 5′ UTR of a particular GFP expression cassette significantly reduced
protein expression, and repression was largely alleviated by the trans-activator
RNA [9]. The synthetic riboregulator system was subsequently expanded to
develop a microbial kill switch [106] and a genetic switchboard to regulate four
carbon-utilization genes [107]. Though this system has been utilized success-
fully, it is worth noting that inserting cis-repressor RNA into their constructs led
to a 40% reduction in mRNA levels versus no cis-repressor RNA [9], which the