Biology Reference
In-Depth Information
The result is a set of candidate reactions for knockout. OptKnock was subsequently
expanded through OptStrain, which considered not only knockouts but also addition of
nonnative pathways to
E. coli
;
56
OptReg, which further allowed for up- and down-regulation
of reactions;
57
and OptForce, which incorporates flux data to identify all of the necessary
reaction modifications that force the system towards the target production level.
58
Other
groups have since devised their own complementary optimization tools, such as CASOP,
which evaluates every reaction in the host organism via weighted elementary modes,
59
and OptORF, which explicitly focuses on gene deletion and overexpression targets rather
than on reactions.
60
Finally, the OptFlux framework was recently developed to make such
tools easily usable by researchers without significant expertise in bioinformatic analysis.
61
While such tools have predominantly been applied to the synthesis of platform and
specialty chemicals, the underlying principles are readily applicable to drugs and drug
candidates as well.
In addition to the systems-level approaches described above, smaller-scale computational
tools have also been devised for pathway optimization. For example, Salis and coworkers
have developed a method for the in silico design of ribosome binding sites to control
protein expression levels.
62
To demonstrate the utility of their approach, they rationally
tuned an AND gate genetic circuit to express green fluorescent protein only in the presence
of arabinose and salicylate. Subsequently, they have expanded their toolkit to include the
design of diverse RBS libraries, bacterial operons, and small RNAs for control of translation
initiation. Another example of a computational tool that operates at the protein level is
Gene Designer, which can be used to optimize protein expression via codon optimization,
as well as optimization of promoter strength and mRNA stability.
63
One step beyond pathway optimization is the goal of complete de novo pathway design.
Several tools, including BNICE, ReBiT, and UM-BBD, have been developed that propose
biosynthetic connections between two given compounds. For a detailed description of these
and other tools refer to Chapter 3. At present, the reported applications of de novo design
software to complex natural product pathways are limited. A notable example, however,
was presented by Gonzalez-Lergier and coworkers, who employed the BNICE computational
framework to explore routes to novel polyketides.
64
Otherwise, general application of these
tools to natural product biosynthesis has yet to be realized, in part due to the difficulty
in evaluating the feasibility of a proposed pathway with a given set of enzymes. One
computational tool that begins to address this problem is DESHARKY, which not only can
predict routes from a host
188
s metabolism to a desired product, but also evaluates the impact
of the proposed routes via mathematical modeling.
65
As such tools continue to develop,
they will no doubt prove invaluable to the engineering of designer pathways and organisms
for drug biosynthesis.
'
Experimental Tools
ASSEMBLY OF LARGE DNA CONSTRUCTS
The biosynthesis of drugs and drug candidate molecules typically requires the coordinated
actions of many gene products. As a result, the DNA constructs of interest to the synthetic
biologist encompass many genes and regulatory elements, and can be quite large, ranging
from tens of kilobases to hundreds of kilobases or more. Perhaps the most straightforward
route to such constructs is through direct chemical synthesis. However, chemical synthesis
of large constructs can introduce sequence errors and can also be cost-prohibitive, although
significant strides have been made in this direction.
66
Traditionally, designer DNA constructs
have been prepared through sequential digestion and ligation-based cloning techniques,
but this can be a laborious, time-consuming, and inefficient process. As a result, some
powerful new techniques have emerged in recent years for the assembly of large
DNA constructs, up to and including entire genomes.
