Information Technology Reference
In-Depth Information
computing problems, it will be necessary to introduce another technology to
rewrite the plasmids. However, little is known about rewriting DNA information
with high efficiency in free-living cells.
Previously, various computing models have been proposed based on chemi-
cal reactions [7] and DNA [2]. Compared to these models, the reaction speed of
our information gate is rather slow. Our information gate may be more suitable
for use in applications such as biological sensors than for mainstream comput-
ing devices. Nevertheless, this bacterial information gate seems to have some
advantages. First, it is plasmid DNA that serves as a medium for exchanging
information between individuals. The specific signal, which is composed of
pheromones and inhibitors, regulates the system execution. Finally, the signal
itself can be written in the plasmid DNA. The plasmid leaves some room to
integrate various genes other than the gate signal. Taking these factors into
consideration, the information gate seems to be scalable.
Of course, many unsolved problems remain. For example, it is difficult to
rewrite the plasmid
in vivo
. The membrane structures of bacteria are generally
simple. In the case of
E. faecalis
, all of the cell contents, including the genomic
DNA, are located inside a single cell-surface membrane. Rewriting plasmids
within the cell involves the unexpected alteration of the genome DNA, usually
resulting in cell death. Pheromones released by recipients affect every donor in
culture. The recipient may choose the type of donor to conjugate, but cannot
choose any individual specifically. Homogenous conjugation between donors
may sometimes arise. This type of conjugation merely results in the exchange to
plasmids between two donors. It is difficult to transfect the conjugative plasmids
into the cells because of their large size. Unlike ordinary plasmid vectors, the
conjugative plasmids have few unique restriction sites. Alternative methods
such as homologous recombination are required to transform these large-scale
plasmids.
The technology of protein engineering is still under development. We have
virtually no capability to design artificial proteins with a desired function
de
novo
, even less so for artificial cells. All we can do at the moment is to combine
artifacts found in nature and perhaps improve them.
Acknowledgments
All of the
E. faecalis
strains and plasmids used in this project were
obtained from the H. Nagasawa laboratory. We thank J. Nakayama for his contributions
and helpful advice.
References
[1] H. Abelson, D. Allen, D. Coore, C. Hanson, G. Homsy, T. F. Knight, R. Nagpal,
E. Rauch, G. J. Sussman, and R. Weiss. Amorphous computing.
Comm. ACM
,
43(5):74-82, 2000.
