202 10 Programming Gene Expression by Engineering Transcript Stability Control and Processing in Bacteria
Considering transcript stability, and how it may impact system function, is
especially important when introducing secondary structure into a transcript, as
this may cause unwanted RNase binding or result in premature transcription
termination. See Section 10.3.4 for an example detailing issues arising from add-
ing cis-repressor riboregulator RNA into a 5′ UTR in E. coli. Another example
comes from efforts to obtain detectable signals in vivo from RNA aptamer-based
fluorescent biosensors (i.e., “Spinach” aptamer conjugates). To do this, Paige
et al. had to employ an RNase E-deficient E. coli strain [100] to circumvent limi-
tations likely resulting from an otherwise short aptamer half-life.
There are other potentially confounding effects that are more difficult to
account for, but that should still be considered in the course of genetic device
engineering. Large amounts of synthetic mRNA and/or regulatory RNA from
complex circuit designs could lead to overloading the degradosome or associated
enzymes such as Hfq, causing cell-wide RNA stability changes. A phenomenon
of this sort is difficult to study, but the work of Hussein and Lim on competition
for Hfq [49] suggests it is worth attempting to understand. They found that sRNA
expressed without a target binding partner reduced sRNA effectiveness cell-wide
by binding Hfq and limiting its accessibility. Expression of the target mRNA
removed this problem, suggesting that balance in expressing synthetic sRNA can
be critical.10.3.3 Uniformity of 5′ and 3′ Ends
Variations in UTR sequence context may elicit differences in local secondary
structure, which in turn may alter transcript stability and gene expression
[78, 88]. One way to guard against such context-dependent transcript stability
problems is to attenuate UTR variability effects by removing 5′ and/or 3′ RNA
that may form undesired secondary structure (Figure 10.3b). Several studies
have utilized removal of 5′ UTR secondary structures as a mechanism for
minimizing context-dependent differences in gene expression. One involved a
screen of “insulator” sequences and structures placed within the 5′ UTR. A
ribozyme–hairpin combination, termed RiboJ, produced nearly identical
transfer functions for two different genes under the control of three different
promoters (several other ribozymes had similar effects) [78]. A second study
used the Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)
processing system from Pseudomonas aeruginosa strain UCBPP-PA14 to
remove both 5′ and 3′ UTR sequences. At both ends, a 28-nucleotide repetitive
sequence, recognized by the Csy4 endonuclease, was added, which resulted in
efficient transcript cleavage and UTR sequence removal. Using the CRISPR
system, they were able to show similar levels of protein production in the
context of different promoter and RBS combinations in mono- and bicistronic
systems, with green fluorescent protein (GFP) and red fluorescent protein
(RFP) outputs [88].
Mutalik et al. recently published a scheme for minimizing 5′ UTR-induced
variations in gene expression that involves introducing a standby RBS in a bicis-
tronic design (BCD) [101]. The standby RBS is designed to cause ribosome bind-
ing upstream of the real RBS (i.e., the RBS from which translation of the desired