ILLUSTRATIVE PROBLEM DOMAINS AT THE INTERFACE OF COMPUTING AND BIOLOGY 305
cause an engineered cell would be designed by human beings, the functions of its various elements
would be much better known. This fact implies that it would be easier to identify critical control points
in the system and to understand the rules by which the system operates.
A second approach is to modify an existing living cell to give it new behaviors or to remove
unwanted behaviors; classical metabolic engineering and natural product synthesis would be relevant
to this approach. One starting point would be to use the membrane of an existing cell, but modification
of these lipid bilayers to incorporate chemically inducible channels, integrated inorganic structures for
sensing and catalysis, and other biopolymer structures for the identification and modification of bio-
logical substrates will provide a greater degree of freedom in the manipulation of the chemical state of
the synthetic cell.
A third approach is to abandon DNA-based cells. Szostak et al.^6 argue that the “stripping-down” of
a present-day bacterium to its minimum essential components still leaves hundreds of genes and
thousands of different proteins and other molecules. They suggest that synthetic cells could use RNA as
the repository of “genetic” information and as enzymes that catalyze metabolism. In their view, the
most important requirements of a synthetic cell from a scientific standpoint are that it replicates autono-
mously and that it is subject to evolutionary forces. In this context, autonomous replication means
continued growth and division that depends only on the input of small molecules and energy, not on
the products of preexisting living systems such as protein enzymes. Evolution in this context means that
the structure is capable of producing different phenotypes that are subject to forces of natural selection,
although being subject to evolutionary forces has definite disadvantages from an engineering perspec-
tive seeking practical application of synthetic cells.
The elements of a synthetic cell are likely to mirror those of simulations (see Box 9.2), except of
course that they will take physical representation. Inputs to the synthetic cell would take the form of
environmental sensitivities of various kinds that direct cellular function. (Another perspective on “arti-
ficial cells” similar to this report’s notion of synthetic cells is offered by Pohorille.^7 In general, synthetic
cells share much with artificial cells, and the dividing line between them is both blurry and somewhat
arbitrary. The modal use of the term “artificial cell” appears to refer to an entity with a liposome
membrane, whose physical dimensions are comparable to those of natural cells, that serves a function
such as enzyme delivery, drug delivery for cell therapy, and red blood cell substitutes.^8 ) However, if
synthetic cells are to be useful or controllable, it will be necessary to insert control points that can supply
external instructions or “reprogram” the cell for specialized tasks (e.g., a virus that injects DNA into the
cell to insert new pieces of code or instructions).
Researchers are interested in expanding the size and complexity of pathways for synthetic cells that
will do more interesting things. But there is little low-hanging fruit in this area, and today’s computa-
tional and mathematical ability to predict cellular behavior quantitatively is inadequate to do so, let
alone to select for the desired behavior. To bring about the development of synthetic cells from concept
to practical reality, numerous difficulties and obstacles must be overcome. Following is a list of major
challenges that have to be addressed:
- A framework for cellular simulation that can specify and model cellular function at different levels of
abstraction (as described in Section 9.2). Simulations will enable researchers to test their proposed de-
signs, minimizing (though not eliminating) the need for in vivo construction and experimentation. Note
that the availability of such a framework implies that the data used to support it are also available to
assist in the engineering development of synthetic cells.
(^6) J.W. Szostak, D.P. Bartel, and P.L. Luisi, “Synthesizing Life,” Nature 409(6818):387-390, 2001.
(^7) A. Pohorille, “Artificial Cells: Prospects for Biotechnology,” Trends in Biotechnology20(3):123-128, 2002.
(^8) See, for example, T.M.S. Chang, “Artificial Cell Biotechnology for Medical Applications,” Blood Purification 18:91-96, 2000,
available at http://www.medicine.mcgill.ca/artcell/isbp.pdf.