Published in Evolutionary Systems: Biological and Epistemological Perspectives on Selection and Self-Organization,   G. Van de Vijver, S.N. Salthe and M. Delpos (editors), Dordrecth: Kluwer, 1998, pp. 59-66.

EMERGENCE OF LIFE AND BIOLOGICAL SELECTION 
FROM THE PERSPECTIVE OF COMPLEX SYSTEMS DYNAMICS

Bruce H. Weber

Department of Chemistry and Biochemistry

California State University, Fullerton

Fullerton, CA 92834-6866   USA

IN THE BEGINNING...

All cultures, including the scientific, have creation stories. These reflect the knowledge and values of the tellers of such narratives. Until relatively recently, most such scenarios posited a divine creator or engineer-like god responsible for the creation of the cosmos and the origin of life..

Modern naturalistic explanations for the origin of life could be said to begin with the "active materialism" of early nineteenth-century French biologist such as Lamarck and Geoffroy, which was introduced into England by "philosophical anatomists" Robert Grant (one of Darwin's early teachers) and taken up avidly by English socialists of the 1820s and 1830s (Desmond 1989). This view assumed that matter was capable of spontaneous self-organization. The firestorm that greeted this view formed the background to the development of Darwin's thinking (Desmond and Moore 1991). In his public writings Darwin ducked the issue of the origin of life and spoke of life being "breathed into a few forms or into one" (Darwin 1859, 490). Privately, however, he let his materialistic inclinations fuel speculation on a possible scenario for the origin or life in a letter to Joseph Hooker in 1871 in which he wrote, "But if (and oh what a big if) we could conceive in some warm little pond with all sorts of ammonia and phosphoric salts, light, heat, electricity and etc., present, that a protein compound was chemically formed, ready to undergo still more complex changes" (DAR 94, 188-9).

It would be some time before further speculation about the origin of proteins and metabolism would occur and indeed could even be reasonably undertaken. In the period after World War I, biochemistry was in a period of rapid growth, which allowed for more informed models of life's origin. J.B.S. Haldane and Alexander Oparin

suggested that there could be an abiotic synthesis and accumulation of organic compounds of increasing complexity from which would emerge a proto-metabolism (Haldane 1929; Oparin 1938). Within this approach, J.D. Bernal suggested that clays could have provided the catalytic and template properties that would have facilitated the emergence of such a proto-metabolism (Bernal 1951). Concurrent with the solution of the structure of DNA, the abiotic synthesis of amino acids was demonstrated (Miller 1953). The demonstration of the abiotic synthesis of proteinoids suggested that not only might polypeptides be catalysts of proto-metabolism, but that their formation might even be prior (Fox 1965). The demonstration of the abiotic synthesis of purine and pyrimidine bases led to the suggestion that RNA might be the original catalytic agent in the origin of life (Woese 1967; Orgel 1968). This suggestion became more plausible with the demonstration of catalytic activity of present-day RNA (Cech 1986, 1987; Doudna and Szostak 1989; Wang, Downs, and Cech 1993; Lorsch and Szostak 1994). Recently the emphasis has been on an RNA-first model for the origin of life in which RNA becomes template,replicator, and ribozymic catalyst (Orgel 1992; Joyce and Orgel 1993). Indeed, it has become quite popular among biochemists and molecular biologists to speak of an "RNA world" that preceded the emergence of true life (Gesteland and Atkins 1993). This view is particularly attractive if the fundamental fact about life is taken to be the self-replication of information-bearing nucleic acids. Indeed, in Dawkins' vision, life began with the highly improbable accident of the first replicating macromolecule that subsequently decorated itself with proteins, lipids, and carbohydrates, which provided protection for the genes and exploited catalysis and metabolism for their benefit, such that first cells and then later organisms emerged as "gene machines" (Dawkins 1976, 1989).

THIS (SYSTEMS) VIEW OF LIFE

Each of the approaches to thinking about the origin of life emphasizes some aspect of contemporary living systems (metabolism, enzymic catalysis, nucleic acid replication) as the primary feature that brings life into being. Some proponents of the RNA world are well aware that such a simple picture of RNA replication and catalysis is incomplete and that life emerged, and continues to exists, far from thermodynamic equilibrium with the consequence that resultant autocatalytic cycles allow the emergence of information and systems complexity (Eigen 1992). Also, there are some serious problems in the RNA-first model, such as the fact that no convincing chemical scenario has been advanced as to how the purine and pyrimidine bases might be attached to ribose abiotically, even under the catalytic activity of clays (Joyce and Orgel 1993). Unfortunately, in a more general sense, while RNA can do some limited catalysis of reactions on itself, and even has been induced in vitro to have an kinase activity (Lorsch and Szostak 1994), it is highly unlikely that it has the catalytic capacity of a set of polypeptides.

Alternatively, life can be viewed as emerging with its general organization of catalytic and metabolic functions prior to, or at least along with the appearance of genetic replication (Fox 1980; Cairns-Smith 1982; Wicken 1987; Weber et al. 1989, Salthe 1991; Morowitz 1992). J.D. Bernal commented that DNA did not wash up on the beach (Bernal 1967). Fox has argued for life emerging within proteinoid microspheres that would give significantly greater concentration than the dilute solution chemistry assumed in many accounts (Fox 1965, 1988). More recently, Morowitz has argued that amphiphile bilayers could have formed prior to other events and than within such membranes the emergent chemistry could occur (Morowitz 1992). This idea that life arose within a proto-cellular context is particularly attractive since this also suggests that primitive chemiosmotic functions could have been involved in providing energy transduction to keep such proto-cellular systems far from equilibrium (Mitchell 1961, 1979; Harold 1986).

Morowitz postulates that vesicular protocells spontaneously form from amphiphilic bilayers of primitive phospholipids (formed by terrestrial processes or of meteoric origin). This bilayer is a barrier to the diffusion of polar solutes, thus allowing molecules once brought into the protocell to keep higher concetrations than in the ambient fluid and for concentration gradients to form. The lipid phase in between the aqueous phases can partition nonpolar solutes, some of which could be chromophores capable of capturing and transducing light energy thus permitting chemical, electrical and osmotic work. Given such a source of free energy the protocellular system can be driven (or constrained) further from equilibrium by such work. "Nutrients" could accumulate and more amphiphilic molecules would be made allowing growth of the protocell. Mechanical and thermodynamic constraints would provide physical selection for optimal sizes and drive the breakdown of larger vesicles or the fusion of smaller vesicles, just as has been observed for Bernard convection cells under similar dissipative conditions (Swenson 1989, 1998). This gives a model of: protocell + nutrients + energy -----> protocells + waste products + entropy (Morowitz, Deamer, and Smith 1991).

There is empirical evidence to support such a model. Amphiphiles have been isolated from a number of carbonaceous chondrites and repeated drying and wetting of such amphiles (mimicing the action of a tide pool) give rise to membrane-like structures, including bilayered micelles and vesicles (Deamer and Pahsley 1989). Such structures show autocatalytic self-replication (Bachmann, Luisi and Lang 1992). Alternative lipid-like molecules of fully terrestrial origin, such as terpenoids, could also have formed the membranes of primitive protocells (Ourisson and Nakatani, 1994). Polycyclic aromatic hydrocarbons derived from carbonaceous chondrites will "dissolve" in the non-polar phase of amphiphile bilayer vesicles and upon illumination translocate protons across the membrane generating a pH shift and hence a proton-motive force that could energize chemical and osmotic work (Deamer 1992).

Such phase-separated vesicles would not yet be living entities, but they would provide good cradles within which life could emerge. With a free-energy source producing a proton gradient, for example, polyphosphate could be made (Morowitz 1992; Deamer and Harang 1990). The formation of polyphosphate is important because such compounds have a unique mix of thermodynamic instability and kinetic stability (Westheimer 1987) and will participate in phosphorylation of amino acids and nucleotides that were accumulated by the cell also through the use of the electrochemical potential of the proton gradient. The phosphorylation of monomers in the protocell would help drive thermodynamically their polymerization to polypeptides and polynucleotides. At some point polyphosphate was replaced by ATP. Interactions of such macromolecules would not be random, even if their sequences at this point were random (Carter and Kraut 1974) and at the very least their interaction would provide some mutual stability against hydrolysis (Wicken 1987). Both polypeptides and polynucleotides could catalyze reactions, though the potential for polypeptides is much larger. Indeed, Kauffman estimates that a random set of catalytic peptides could "cover" the catalytic task space of the hundred billion or so possible chemical reactions that could be catalyzed by such catalysts (Kauffman 1993). Poised away from equilibrium, nonlinear interactions and autocatalytic cycles will spontaneously arise. As Kauffman has shown by his modeling, such systems will develop toward catalytic closure in which there is a tightening of the interactions and synergism that gives rise to a protometabolism. Even though there is no information in the genetic sense in such a system, there will be competition with other such autocatalytic systems for energy fluxes. As successful patterns of dissipation emerge under these competitive conditions, we would expect progressive tightening of the nucleic acid-protein relationship as the protocells evolve into true cells. It is thermodynamic or chemical selection of the efficient, as David Depew and I have argued (Depew and Weber 1995; Weber and Depew 1996), rather than physical selection of the stable or biological selection of the fit where the relevant units of selection are energy-capturing and energy-utilizing cycles. Self-organization under thermodynamic constraints provides much order without selection in such autocatalytic cycles; chemical selection can act upon these cycles to produce increased efficiency.

The emerging ensemble of proteins that catalyzed the reactions constituting the primitive metabolism of the protocell would have been "generic" proteins (Wicken 1987). Over time they would have acquired more specific catalytic functions. There would have been selection pressure for any entity that could increase the efficiencies of its autocatalytic cycles by storing the information needed for storing the information needed for autocatalysis and for expanding autocatalytic prowess by using these information-storing capacities in new ways. The close coupling of replicating nucleic acids (with some residual catalytic activity) and catalytic proteins involved in autocatalytic processes would have been highly advantageous in this context. It would have been of enormous competitive advantage to such catalytic units if they were able to "remember" information that enhanced autocatalytic activity by encoding it in the polymers of nucleic acids formed by chemical selection itself. Genetic information would have accumulated under thermodynamic selection for stable patterns of entropy production. There is no reason to think that the properties that were required for the emergence of life would not be maintained by living systems thereafter. Thus living things can be viewed as bounded, self-replicating, informed autocatalytic dissipative systems that sustain themselves by efficient environmental energy exhanges (Wicken 1987; Depew and Weber 1995), yet vary under the drive to configurational randomness (Brooks and Wiley 1986,1988) in such a way that new information, guiding new catalytic functions, can be selected from the variation.

Natural selection of the reproductively fit is emergent from chemical selection of the autocatalytically efficient; it is a process that can be ascribed only to the autocatalytic dissipative structures that capture information within strongly defined boundaries and use it to guide efficient autocatalysis. Such entities would have to be able to pass the information they have to successor entities, for without that additional property, the whole point of internalizing information would be lost. Among such entities themselves, therefore, what would have been even more crucially contested than storing and deploying metabolic information would have been the ability to reproduce themselves and the information they possess. Fitness is a measure of that ability. It cannot be reduced to the efficiency of chemical selection, any more than the efficiency of chemical selection can be reduced to the stability of physical selection, for the relevant processes and entities capable of engaging in them do not exist at those levels. To say that these processes cannot be reduced to lower-level ones is not, however, to say that they are not part of a single, coherent process. It is to say rather that evolution exhibits emergent levels and emergent properties and phenomena.

From this perspective, biological information would emerge as a probable, indeed expected, result of deep physical and chemical principles governing self-organization in complex webs of catalysts rather than as a ³fortuitous, frozen accident² (Kauffman, 1993; Salthe, 1991). These principles, however construed or modeled, represent an interplay of self-organizational and selective processes. The physical imperatives of self-organization and dissipation require that the particular sort of selection process leading to the emergence of living systems was at first the selection of the stable (physical selection) and of the efficient (chemical selection) rather than of the reproductively fit (biological selection). So construed, the problem of the origin of life is the emergence of the phenomenon of natural selection out of these more basic forms of selection (Weber and Depew 1996). From this perspective, it may be more fruitful to regard primitive protocellular systems as the sites of the dynamics leading to life involving a coevolution of proteins and replicating nucleic acids over a "replicators first" strategy. The more basic forms of selection that obtain in such sites are inseparable from the amplification of stochastic events by the self-organizational tendencies and propensities of open systems and dissipative structures.

Kauffman's NK Boolean models allow exploration of the interplay of selective and self-organizational principles at a more abstract and level-independent manner that emphasizes the relational rather than causal aspects of complex systems. In the ordered regeme of NK space, there is either no or too little variation in regions described by simple point or limit-cycle attractors, not a very interesting dynamics. In the chaotic regions, described by large numbers of strange attractors, there is insufficient stability or coherence; over time such a system will settle onto and cycle through attractors with vast numbers of states, taking billions of times longer that the history of the universe to traverse these enormous attractors. Such systems do not do anything interesting either. But, under the right conditions (supplied by the programmer in virtual systems or by the nonlinear interactions of real, autocatalytic systems), at or near the membrane-like edge of order and chaos, there is a region where interesting things can and do happen. The behavior of systems in this range is complex. It might move toward one or several different attractors. Transient islands of ordered structure can arise in a sea of chaos, only to melt away as new order appears elsewhere. In this region is also seen the distinctive dynamics of self-organization and adaptability (Langton 1986, 1992; Kauffman 1993; Ulanowicz 1994). Adaptation, considered as a process, becomes an exploration of parameter space near the edge of chaos in search of a better fit to a given fitness landscape onto which Kauffman maps his NK model, or even to find better attractors. The structure of this region of phase space and its attractors are properties of the self-organization of the system itself. Kauffman goes on to speculate that it is selection that pulls such complex adaptive systems into this fruitful region. This suggest that complex adaptive behavior, at any hierarchical level (molecules, cells, organisms, ecosystems, phylogenies), develops only in systems whose range of elements and connections is such that wide variation due to stochastic processes, a feed-back-driven selection process, and self-organization are integral aspects of a single process. The interplay of chance with physical and chemical self-organization and selection led ultimately to the emergence of natural selection concomitant with the emergence of life itself, but life and natural selection remain embedded in this context from which it arose, even as they exhibit novel, emergent behavior.

Rather than cut natural selection off from its physical and chemical roots, as biological autonomists would, it may be more useful to consider that there is an on-going intimate tie between chance, self-organization and selection, including natural, or biological, selection. Such a conception, along with the use of dynamical models based upon complex systems more generally, could vitalize the Darwinian tradition and facilitate dialogue between its adherents and the proponents of the developmentalist tradition (Depew and Weber, 1995; Weber and Depew 1996).

---

REFERENCES

Bachmann, P.A., P.L.Luisi, and J. Lang, 1992. Autocatalytic self-replication micelles as modes for prebiotic structures. Science 357: 57-58.

Bernal, J.D., 1951. The Physical Basis of Life. London: Routledge and Kegan Paul.

Bernal, J.D., 1967. The Origin of Life. Cleveland: World.

Brooks, D.R. and E.O. Wiley, 1986. Evolution as Entropy: Toward a Unified Theory of Biology. Chicago: University of Chicago Press.

Brooks, D.R. and E.O Wiley, 1988. Evolution as Entropy: Toward a Unified Theory of Biology, Second Edition. Chicago: University of Chicago Press.

Cairns-Smith, A.G., 1982. Genetic Takeover and the Mineral Origins of Life. Cambridge: Cambridge University Press.

Carter, C.W. Jr. and J. Kraut, 1974. A proposed model for interaction of polypeptides with RNA. Proc. Nat. Acad. Sci. USA 71: 283-287.

Cech, T.R., 1986. A model for the RNA-catalyzed replication of RNA. Proc. Nat. Acad. Sci. USA 83: 4360-4363.

Cech, T.R., 1987. The chemistry of self-splicing RNA and RNA enzymes. Science 236: 1532-1539.

Darwin, C.R., 1859. On the Origin of Species by Means of Natural Selection or the Preservation of Favored Races in the Struggle for Life. London: John Murray.

Dawkins, R., 1976. The Selfish Gene. Oxford: Oxford University Press.

Dawkins, R., 1989. The Selfish Gene: New Edition. Oxford: Oxford University Press.

Deamer, D.W., 1992. Polycyclic aromatic hydrocarbons: primitive pigment systems in the prebiotic environment. Adv. Space Res. 12: 183-189.

Deamer, D.W. and E. Harang, 1990. Light-dependent pH gradients are generated in liposomes containing ferrocyanide. BioSystems 24: 1-4.

Deamer, D.W., and R.M. Pashley, 1989. Amphiphilic components of the Murchison carbonaceious chondrite: surface properties and membrane formation. Origin of Life and Evolution of the Bioshpere 19: 21-38.

Depew, D.J. and B.H. Weber, 1995. Darwinism Evolving: Systems Dynamics and the Genealogy of Natural Selection. Cambridge MA: Bradford/The MIT Press.

Desmond, A., 1989. The Politics of Evolution. Chicago: University of Chicago Press.

Desmond, A., and J. Moore, 1991. Darwin. London: Michael Joseph.

Doudna, J.A. and J.W. Szostak, 1989. RNA-catalysed synthesis of complementary-strand RNA. Nature 339: 519-522.

Eigen, M., 1992. Steps Towards Life: A Perspective on Evolution. Oxford: Oxford University Press.

Fox, S.W., 1965. Simulated natural experiments in spontaneous organization of morphological units from protenoid. In The Origins of Prebiological Systems and Their Molecular Matrices, S.W. Fox (ed), New York: Academic Press, pp. 361-382.

Fox, S.W., 1980. The origins of behavior in macromolecules and protocells. Comparative Biochemistry and Physiology 67B: 423-436.

Fox, S.W., 1988. The Emergence of Life: Davwinian Evolution from the Inside. New York: Basic Books.

Gesteland, R.F. and J.F. Atkins, 1993. The RNA World: The Nature of Modern RNA Suggests a Prebiotic RNA World. Cold Spring Harbor: Cold Spring Harbor Laboratory Press.

Haldane, J.B.S., 1929. The origin of life. Rationalist Animal. Reprinted in The Origin of Life, J.D. Bernal (ed) 1967, Cleveland: World, pp. 242-249.

Harold, F.M., 1986. The Vital Force: A Study of Bioenergetics. New York: Freeman.

Joyce, G.F. and L.E. Orgel, 1993. Propsects for understanding the origin of the RNA world. In The RNA World, R.F. Gesteland and J.F. Atkins (eds), Cold Spring Harbor: Cold Spring Harbor Laboratory Press, pp. 1-25.

Kauffman, S.A., 1993. The Origins of Order: Self-Organization and Selection in Evolution. New York: Oxford University Press.

Krüppers, B.-O., 1990. Information and the Origin of Life. Cambridge MA: The MIT Press.

Langton, C.G., 1986. Studying artificial life with cellular automata. Physica (D) 22: 120-149.

Langton, C.G., 1992. Life at the edge of chaos. In Artificial Life II, C.G. Langton, C.Taylor, J.D. Farmer, and S. Rasmussen (eds), Reading MA:Addison-Wesley, pp. 41-91.

Lorsch, J.R. and J.W. Szostak, 1994. In vitro evolution of new ribozymes with polynucleotide kinase activity. Nature 371: 31-36.

Miller, S., 1953. A production of amino acids under possible primitive earth conditions. Science 117: 528-531.

Mitchell, P.D., 1961. Coupling of phosphorylation to electron and hydrogen transfer by a chemiosmotic type of mechanism. Nature 191: 144-148.

Mitchell, P.D., 1979. Keilin's respiratory chain concept and its chemiosmotic consequences. Science 206: 1148-1159.

Morowitz, H., 1992. Beginnings of Cellular Life: Metabolism Recapitulates Biogenesis. New Haven: Yale University Press.

Morowitz, H, D.W.Deamer, and T. Smith, 1991. Biogensis as an evolutionary process. Journal of Molecular Evolution 33: 207-208.

.5

Oparin, A.I., 1924. Proiskhozhdenie zhizy. Moscow: Moskovski Rabochii.

Oparin, A.I., 1938. The Origin of Life. London: Macmillan.

Orgel, L.E., 1968. Evolution of the genetic apparatus. Journal of Molecular Biology 38: 381-393.

Orgel, L.E., 1992. Molecular replication. Nature 358: 203-209.

Ourisson, G., and Y. Nakatani, 1994. The terpenoid theory of the origin of cellular life: the evolution of terpenoids to cholesterol. Current Biology

1: 11-23.

Salthe, S.N., 1991. Formal consideration on the origin of life. Uroboros 1: 45-65.

Swenson, R., 1989. Emergent attractors and the law of maximum entropy production: Foundation to a theory of general evolution. Systems Research 6: 187-197.

Swenson, R., 1998. Spontaneous Order, Evolution and Natural Law: An Introduction to the Physical Basis for an Ecological Psychology. Hillsdale, NJ: Lawrence Erlbaum Associates.

Ulanowicz, R.E., 1998. Boundaries on the complexity of evolving networks: A window of vitality. In Between Instruction and Selection: Studies in Nonequilibrium Biology, J.D. Collier (ed), Baltimore: Johns Hopkins University Press.

Wang, J.-F., W.D. Downs, and T.R. Cech, 1993. Movement of the guide sequence during RNA catalysis by a groupI ribozyme. Science 260: 504-508.

Weber, B.H., D.J. Depew, C.Dyke, S.N. Salthe, E.D. Schneider, R.E. Ulanowiczand J.S. Wicken, 1989. Evolution in thermodynamic perspective: An ecological approach. Biology and Philosophy 4: 373-405.

Weber, B.H. and D.J. Depew, 1996. Natural selection and self-organization: Dynamical models as clues to a new evolutionary synthesis. Biology and Philosophy 11: 33-65.

Westheimer, F.H., 1987. Why nature chose phosphates. Science 235: 1173-1178.

Wicken, J.S., 1987. Evolution, Information and Thermodynamics: Extending the Darwinian Program. New York: Oxford University Press.

Woese, C., 1967. The evolution of the genetic code. In The Genetic Code, New York: Harper and Row, pp. 179-195.