The 9th ISSOL Meeting and the 12th International Conference on the Origin of Life,
San Diego, July 11-17, 1999

Molecular Simulations of Protocellular Membrane Functions

Andrew Pohorille, Michael A. Wilson, Karl Schweighofer, Christophe Chipot, and Michael H. New

Exobiology Branch, NASA Ames Research Center

Department of Pharmaceutical Chemistry, University of California, San Francisco

Laboratoire de Chimie Theorique, U.M.R. C.N.R.S. No 7565, Universite Henri Poincare-Nancy I

The emergence of the earliest forms of cellular life -- protocells -- was a central step in the evolution of simple organic matter into the present-day diversity of life. In this fundamental step, organic material assembled into membrane-bounded structures that acquired the capabilities necessary for self-maintenance, growth and replication. Several of these capabilities, such as acquisition of organics from the environment and bioenergetics, are mediated by membranes. The assumption of the continuity of evolution implies that, at some stage, membrane-mediated, protocellular functions must have been performed by peptides, which eventually evolved into the membrane-integral proteins of modern cells. This assumption, however, does not imply that peptides were the first or only functional molecules in protocells. Thus, the significance of membrane-related protein functions does not depend on a specific scenario for the origin of life.

It is well known that short, isolated peptides are typically disordered in water. Thus, to play a role in protocellular evolution, simple membrane peptides must have fulfilled three conditions. First, they must have been able to adopt an ordered structure in contact with the membrane. Second, they must have become inserted into the membrane and associated to form higher-order structures, such as transmembrane channels. Finally, such structures must have been capable of performing functions. Simple peptide models of membrane-related functions have been developed and their structure, stability and mechanism of action have been examined using computer simulations. These simulations have been performed on detailed, realistic models acting in appropriate cellular environments, so predictions based on these simulations are directly testable experimentally.

The mechanism by which peptides fold into ordered structures at the water-membrane interface was studied through the example of an 11-mer of  poly-L-leucine. Initially placed as a random coil on the water side of the interface, the peptide folded into an -helix in 36 ns. Simultaneously, the peptide translocated into the membrane-mimetic side of the interface. The folded peptide was preferentially oriented parallel to the interface but occasionally rotated to adopt a transmembrane orientation.

The aggregation of folded peptides into larger, transmembrane structures was investigated using peptides formed from hydrophobic leucine (L) and polar serine (S) with sequence (LSLLLSL). Four such peptides, once folded into amphiphatic -helices, can associate in the membrane to form ion channels lined by the hydrophilic serine side chains. Single helices remain adsorbed at the water-membrane interface but can adopt a transmembrane orientation in the presence of a transmembrane electric field. Once in a transmembrane orientation, the monomers associate to form dimers which, subsequently, further aggregate into tetrameric channels. An alternative mechanism, wherein dimers are first formed parallel to the interface and then insert into the membrane, was found to be unlikely.

Channels formed from tetramers of -helices can transport protons as well as larger ions. A protobiologically relevant model of such a channel was constructed of 25 residue-long peptides based on a transmembrane fragment of the Influenza A M protein. Despite its simplicity, this channel transports protons with remarkable efficiency and selectivity. The channel pore is occluded by a ``gate'' of four histidines located near its center. Two mechanisms of gating have been proposed. In one mechanism, all four histidines become protonated and the gate opens due to repulsion between their positive charges. In the alternative mechanism, a proton is captured on one side of a histidine in the gate while another proton is released from the opposite side. The gate returns to its initial state through tautomerization. The results of simulations of the channel embedded in a hydrated phospholipid membrane are consistent only with this latter mechanism.

In summary, the simulations demonstrate that, in the presence of membranes, simple peptides with proper sequences of hydrophobic and hydrophilic residues can adopt ordered conformations and aggregate to form functional structures. These results provide clues to the design of such structures that can be used to construct laboratory models of protocells.

This work was supported by the NASA Exobiology Program.