Biology Reference
In-Depth Information
Motivated by these advantages, in this work, we have explored the use of
bacterial inversion recombination systems to construct genetic switches with the
above properties and show how more complicated finite-state machine-like de-
vices could be constructed out of such systems. For our purposes invertases are a
nice starting point since they act very much as simple switches that may invert a
sequence of DNA in place. To summarize briefly, inversion recombination hap-
pens between two short inverted repeated DNA sequences, typically less than 30
basepairs (bp) long. The recombinases bind to these inverted repeated sequences,
which are recombinase specific. A DNA loop formation, assisted by DNA bend-
ing proteins, brings the two repeat sites together, at which point DNA cleavage
and ligation occur. This reaction is ATP independent, but requires super-coiled
DNA. The end result of the recombination is that the stretch of DNA between
the repeated sites inverts. That is, the stretch of DNA switches orientation—what
was the coding strand is now the non-coding strand, and vice versa. In this reac-
tion, the DNA is conserved, and no gain or loss of DNA occurs. Additionally,
there is great flexibility in the distance between the inverted repeats, ranging from
only a few hundred up to 5 kilobases (kb) of additional DNA between the repeats
possible (refer to the work of Johnson [18] and Blomfield [19] for a detailed
treatment of inversion systems). The advantages of site-specific inversion are the
binary dynamics, the sensitivity of output, the efficiency of DNA usage, and its
persistent DNA encoding, even after cell death. The disadvantages are the possible
interference between multiple recombinases, DNA loss by excision, and revers-
ibility of the reaction.
Previously, we have described a tightly regulated expression switch using the
FimE protein of the fim system of E. coli [20] which demonstrated the leak-less
properties of this system and persistence of state after removal of recombinase
input. Here, we have constructed an artificial overlapped inversion switch by in-
tegrating two recombination systems (using the FimB protein of the fim system
from E. coli [19] and the hin system from Salmonella [21]) to form an intercalat-
ed double inversion system that implements a heritable memory with finite state
machine-like behavior with four states dependent on the sequence of invertase
activity inputs. We expand on how this works below.
There exist only a few known examples of natural systems that utilize multiple
overlapping DNA inversions for diverse gene expression. Some examples are the
R64 plasmid shufflon [22], which uses inversion to select among different ver-
sions of PilV gene, and the Min system from the p15B plasmid [23], which can
make 240 different isomeric forms of a phage protein. The natural example most
parallel to our own is the nested inversion system of Campylobacter fetus, where
a promoter is moved around via inversion near various S-layer protein genes for
expression [24]. All of these systems are thought to be involved in extending the
Search WWH ::




Custom Search