Field of Science

Water and Amyloid Self-Assembly
One of the most significant, interesting and yet poorly understood effect in chemistry and biology is the hydrophobic effect. An important manifestation of its role in biology is invoked by the picture of two parallel hydrophobic plates with water between them that are slowly brought close to each other. At a certain distance called the 'dewetting distance', dewetting is observed and the water suddenly gets expelled out from them. Since nature abhors the resulting vacuum, the two hydrophobic surfaces collapse and 'stick' to each other.

Bruce Berne and his colleagues at Columbia had hypothesized that such dewetting could be observed in biological systems. A remarkable dewetting transition in the protein mellitin and in other proteins in the PDB provided evidence. Now they and others investigate such possible roles of water in the formation of amyloid by molecular dynamics simulations. They focus on the central core peptide of the Alzheimer's Aß (1-42) peptide consisting of the 16-22 region. Since one of my projects involves investigation of the self-assembly of this segment, it is of particular interest to me. This oligopeptide is interesting because it seemingly is the smallest stretch of the bigger parent that also forms the characteristic cross-beta sheet structure of amyloid. Importantly, it is also soluble which means it is more amenable to structural characterization compared to the bigger peptide.

Berne's group investigates the role that water plays between two sheets of 16-22 consisting of nine peptides each. The MD simulations were run for 1 ns, a typical time for such systems. The dewetting critical distance was found to be 12.8 A. Compare this with the classic repeat distance between amyloid beta-sheets which is 9.9 A. Basically two types of phenomena were observed depending on different trajectories and conditions; dewetting, which means that water expulsion was followed by hydrophobic 'collapse', and the simultaneous occurence of expulsion and collapse. The latter phenomena can be interpreted as a kind of 'lubrication' effect that water has been hypothesized to play in the folding of proteins. They haven't been able to put a finger on which is the phenomenon in the 'real' system, but it at least seems plausible that dewetting could be observed in such systems. In the cases where dewetting is not the driving force for assembly, they dissect the interaction energies of different amino acid residues that contribute to the total energy, and find that the Phe-Phe energy contributes to the most.

From a methodological standpoint, the investigation is complicated by the fact that, as in other such studies, the results seem to depend on the methods used. This is a common characteristic of modeling and theoretical investigations. In this case, turning off the protein-water Van der Waals forces causes dewetting in every instant, perhaps not a surprising conclusion since there is now no attraction at all between the water and protein. Also, turning off the electrostatic forces caused no changes, which means that electrostatic attraction plays a small role in the system. Again probably not too surprising. A third observation is that the choice between the two phenomena depends on rather small changes in temperature.

The interpretations are also rendered ambiguous by the fact that amyloid surfaces and indeed most 'hydrophobic' protein surfaces are a poor approximation to two parallel hydrophobic plates. It should be noted that the attractive energy of interaction between two flat plates separated by a distance R is inversely proportional to the square of R. Compare this with the case of two spheres of radius R whose classical Van der Waals attractive energy is proportional to the sixth power of R. That means that in the case of the plates the attraction falls off much more slowly which can lead to significant dispersion forces even at relatively long separation.

This paper, while not leading to very novel or practical results, sheds light through detailed investigation of the possible role water can play between amyloid sheets. It again demonstrates the important effect that choice of methods and parameters can have on observations.

Krone, M.G., Hua, L., Soto, P., Zhou, R., Berne, B.J., Shea, J. (2008). Role of Water in Mediating the Assembly of Alzheimer Amyloid Aß (16-22) Protofilaments. Journal of the American Chemical Society, 130(33), 11066-11072. DOI: 10.1021/ja8017303


  1. two parallel hydrophobic plates with water between them that are slowly brought close to each other

    Isn't this referred to as depletion force?

  2. yes i believe it's called that

  3. Welcome back Ashutosh !

    There is also something called the Casimir force [ Nature vol. 447 pp. 772 - 774 '07 ]which brings two plates together (but in a vacuum).

    Which of the many models of the amyloid fibril do you prefer? [ Proc. Natl. Acad. Sci. vol. 105 pp. 7462 - 7466 '08 ] There are zillions of them, including one of Max Perutz's last papers -- [ Proc. Natl. Acad. Sci. vol. 103 pp. 1546 - 1550 '06 ].

    I found [ Proc. Natl. Acad. Sci. vol. 105 pp. 7720 - 7725 '08 ] and the technique it uses particularly fascinating (and possibly conclusive). What do you think?


  4. Thanks Retread; I had to be away because of a family crisis.
    As for the amyloid models, I believe Perutz's model has now been discounted. The one I know that is the most frequently referred to is Tycko's (Biochemistry; 2006; 45(2); 498-512. DOI: 10.1021/bi051952q)
    I haven't seen the 2D IR paper you referred to; will check it out.
    The Casimir force is indeed one of the most mysterious manifestations of the quantum.


Markup Key:
- <b>bold</b> = bold
- <i>italic</i> = italic
- <a href="">FoS</a> = FoS