72 4 Rational Efforts to Streamline the Escherichia coli Genome
cleavage tools (TALENS, CRISPS/Cas), tailored to target specific genomic sites,
might be used to devise novel schemes to rapidly perform various manipula-
tions [8, 122–124].
On the other hand, rational, targeted streamlining, and optimization could be
complemented by random engineering coupled with directed evolution. Devising
efficient, forced random deletion-creating schemes, applying cyclic multiplex
genomic alteration techniques, or shuffling different genomes would vastly
increase the number of genomic variants, from which the fittest versions could
be identified by proper selection.
The plummeting cost of DNA synthesis is continuously increasing the rele-
vance and reality of synthesizing streamlined genomes. Originally, bottom-up
synthesis and top-down reduction of genomes were viewed as two competing
and opposite approaches to simplify bacterial cells. In current practice, these
two strategies seem to harmonically complement each other: reduced genomes
are used as starting points of complete genetic rewiring using synthetic DNA
cassettes [16, 111], and deletion construction has also been demonstrated with
plasmids carrying synthetic DNA fragments [74]. Furthermore, the boundary of
the two strategies is blurred ab ovo, for the gene sets of minimal genomes syn-
thesized to date are all subsets of the genetic repertoire of extant bacterial strains
[15]. It is possible, however, that in the future, minimal genomes will be synthe-
sized by combining genes originating from multiple species.
How far should genome size reduction extend? In general, the relatively small
effects of the extensive genomic perturbations represented by streamlined
genomes attest to a remarkable robustness of the cellular physiology and genome
architecture. However, reduced complexity inevitably comes at the expense of
robustness and adaptability to external factors [23]. Observations from practical
genome streamlining works ([97] and our observations) also suggest that large-
scale elimination of genes, while initially resulting in improvements, may reduce
robustness and cause deterioration of basic cellular physiology (growth proper-
ties, adaptability, nucleoid structure, cell morphology) beyond a certain point
that roughly corresponds to the core-genome size (Figure 4.5).
Beyond the complexity issue, physical constraints on genome size might
also limit reduction efforts. Despite decades of research, little is known on the
homeostatic mechanisms coordinating DNA replication, transcription, and
translation to maintain a constant DNA to cell mass ratio [125]. Significantly
altering the genome size may perturb these mechanisms. Moreover, while our
knowledge regarding gene and network functions is getting even more com-
plete, constraints of genome architecture per se are less understood [126].
The specific and relative localization of some genes (e.g., ribosomal RNA
operons) and specific chromosomal sites (e.g., binding sites for proteins par-
ticipating in cell division), superhelicity of the genome, and macro- and
microdomain structure are all influenced in a largely unknown way by genome
reduction.
However, synthetic biology provides us just the appropriate tools to address
these issues [9]: streamlined genomes can be specifically designed and con-
structed to elucidate the constraints of genome size and architecture.