By Stuart A. Kauffman
This article originally appeared in Nature 382
August 8, 1996.
Copyright 1996 by Nature.
On page 525 of this issue , David Lee and colleagues describe what
appears to be the first case of a self-replicating
peptide, a result that may prove to be either a mere chemical curiosity,
The authors show that a 32-amino-acid peptide, folded into an alpha-helix
and having a structure based on a region of
the yeast transcription factor GCN4, can autocatalyse its own synthesis by
accelerating the amino-bond condensation of
15- and 17-amino-acid fragments in solution (see Fig. 1 on page 525).
The design of this replicator was based on a protein found in nature, an
alpha-helical coiled coil. Reasoning that a given
alpha-helical subunit of the entire structure could be seen as a
complementary binding surface, acting cooperatively to
organize other participating peptide subunits in the coiling, the authors
hoped that a similar 'template' function could be
found in smaller fragments. The ligation, or joining, site was constructed
so as to lie on the solvent-exposed surface of
the alpha-helical structure of their 32-amino-acid sequence.
Lee et al. established autocatalysis by showing that by increasing initial
concentrations of the 32-amino-acid template,
with constant concentrations of the 15- and 17-amino-acid substrates, a
marked increase was produced in the initial
rates of template production. The increase correlates with the square root
of the initial template concentration, as seen in
self-ligating polynucleotide systems [2,3]. The reaction is region- and
chemically selective, yielding less that 15% side
products, and proceeds through the major autocatalytic pathway open to the
Do these results reflect a rare chemical quirk in the repertoire of
peptides and polypeptides, or might they hint at a route
to self-reproducing molecular systems on a basis for wider than
Watson-Crick base-pairing in polynucleotides? At this
stage, we cannot know, but the way is now open to investigate.
The first step, beyond independent verification of the reaction system, is
to construct self-reproducing cross-catalytic
systems of two peptides, A and B, here A catalyses the formation of B from
B's two fragment substrates, and B does
the same for A. Such a system would be collectively autocatalytic--no
molecule would catalyse its own formation, but
the system would collectively catalyse its own formation from 'food'
sources--where, the two A fragments and the
two B fragments. If collectively autocatalytic peptide sets with two
catalytic components can be constructed, can
systems with three, four or hundreds of cross-coupled catalysts be
Such experiments are important. A free-living cell, prokaryote or
eukaryote, is in fact a collectively autocatalytic
system--virtually no molecule, including DNA, catalyses its own formation.
Most of the cell's catalysts are proteins, so
if collective autocatalysis in complex peptide systems is possible, we
would have a new model for self-reproducing
A host of experimental and theoretical questions about such systems
present themselves. If such systems can be
created, is it possible in general to constrain the side reactions enough
for the autocatalytic system to increase in
concentration as the fragment 'fuel' is added? Would such an increase be
aided by confining reactions to a surface, or
within a small volume? Are there means other than thioester promotion to
drive the synthesis of peptide bonds? Even if
such systems can be designed using clever synthetic chemistry, is the
spontaneous formation of collective autocatalytic
sets of peptides rare or common as a function of the diversity of the
peptides, and of the regions of sequence space they
are derived from? Can such autocatalytic systems be constructed or
spontaneously assembled from mixed polymer
systems consisting of DNA, RNA, peptides and perhaps other polymers and
their building blocks? Can such systems
evolve to 'neighbouring' autocatalytic systems while retaining 'catalytic
closure', and could current life have evolved
The new autocatalytic ligation-reaction system is merely exergonic: left
to its own devices, the system will simply run
to equilibrium. Can an autocatalytic system be created that carries out
thermodynamic work cycles whereby the system
sustains displacement from equilibrium, performs coordinated work and
achieves such coordination by controlling,
constraining and 'correcting' unwanted side reactions (as in DNA editing
and repair; P. W. Anderson, personal
communication and ref. 4) to enhance its own rate of reproduction?
The dominant view of life assumes that self-replication must be based on
something akin to Watson-Crick base pairing.
The 'RNA world' model of the origins of life conforms to this view. But
years of careful effort to find an enzyme-free
polynucleotide system able to undergo replication cycles by sequentially
and correctly adding the proper nucleotide to
the newly synthesized strand have not yet succeeded [5,6].
A polynucleotide system based on a ribozyme polymerase able sequentially
to add the correct nucleotides (and thus
copy itself) might work. In contrast, the simple and successful
reproducing molecular systems described by Lee et al.
 and by von Kiedrowski  which uses a single-stranded DNA hexamer and
its two trimer fragments) are based on
a polymer catalysing its own formation from two fragments. Both show that
autocatalytic systems based on specific
ligation reactions are possible. Because a variety of polymers and small
molecules can catalyse such reactions, these
results may prove seminal: the creation or spontaneous formation of simple
or collectively autocatalytic sets may occur
far more readily than we thought. Given the emerging field of 'molecular
diversity', with its capacity to synthesize
high-diversity DNA, RNA and peptide libraries , these questions are now
open to detailed scrutiny.
[1.] Lee, D. H., Granja, J. R., Martinez, J. A., Severin K., and Ghadiri,
M. R. Nature 382 525-528 (1996). [return
[2.] von Kiedrowski, G. Agnew. Chem. 25 932-935 (1986).
[3.] von Kiedrowski, G., Wiotzka, B., Hielbing, J., Matzan, M. and Jordan,
S. Agnew, Chem. 30 423-426 (1991).
[4.] Hopfield, J. J. Proc. Natl Acad. Sci. USA 71 4135-4139 (1974).
[5.] Joyce, G. F., and Orgel, L. E. J. Mol. Biol. 188 433-437 (1986).
[6.] Joyce, G. F. Cold Spring Harb. Symp. Quant. Biol. 52 (1987).
[7.] Scott, J. K., and Smith, G. P. Science 249 386-389 (1990).