The Evolution & Medicine Review

Biological evolution is, obviously, a historical (i.e., time-dependent) process.  However, the importance to evolution of dynamics occurring on multiple time scales is still being delineated. I recently attended a symposium sponsored by the Case Department of Physiology and Biophysics.  The last speaker was Martin Karplus, an eminent chemist affiliated with both Harvard and the Universite Louis Pasteur in Strasbourg, France. Professor Karplus has been a pioneer in the use of computer simulations to investigate and understand macromolecular dynamics, i.e. structural changes over time.  Such studies have facilitated the appreciation of the extent to which protein function depends on time-dependent structural transitions and not merely on the static average structure [e.g., Mobley and Dill. Structure. 2009 Apr 15;17(4):489-98].  He began his lecture by expressing the view that many biomedical scientists using the sophisticated techniques of modern genetics to identify disease-associated genes and gene products do not necessarily have a deep understanding of the physical mechanisms by which proteins actually mediate their functions.  This assertion is consistent with a recent essay in Nature [2009 Apr 23;458(7241):969] by Mark Isalan and Matthew Morrison which suggests that much conventional thinking about interactions among proteins, and the cellular consequences of those interactions, may be somewhat simplistic.

It is common for molecular and cellular biologists to summarize their findings about interactions between proteins with schematic diagrams employing arrows labeled by plus and minus signs, thereby fostering the view that functions are intrinsic (and time-independent) properties of gene products or other biomolecules.  As Isalan and Morrison point out, however, the concentrations of molecules, and their variations over time, as well as their intrinsic biophysical and chemical properties can greatly influence the measurable consequences of the interactions being studied.

One example used by the authors involves the interactions between the cellular proteins p53 and Mdm2.  A study by Lev Bar-Or et al. [Proc Natl Acad Sci U S A. 2000 Oct 10;97(21):11250-5] suggested that oscillations in the concentrations of these two gene products can occur in response to external stimuli (e.g. ionizing radiation) and these oscillations are plausibly necessary for at least one critical function associated with p53, namely DNA repair.  Constant, high-level expression of p53 would be expected to lead to destruction of the cell (as opposed to repair) through apoptosis.

It is more than plausible to suspect that selection will be extremely sensitive to effects that require a quantitative and multidimensional perspective to fully describe (or understand).  In other words, the evolution of various genotypes and the phenotypes with which they are associated can probably not be fully elucidated if the patterns of thought employed by investigators are limited to the sorts of simplifications (such as ignoring dynamics) that can be heuristically valuable in investigating a phenomenon but that also can pose a risk of constraining the ultimate conceptualization of that phenomenon.


Mobley DL, Dill KA. Binding of small-molecule ligands to proteins: “what you see” is not always “what you get”. Structure. 2009 Apr 15;17(4):489-98.

Isalan M, Morrison M. This title is false. Nature 2009 Apr 23;458(7241):969.

Lev Bar-Or R, Maya R, Segel LA, Alon U, Levine AJ, Oren M. Generation of oscillations by the p53-Mdm2 feedback loop: a theoretical and experimental study. Proc Natl Acad Sci U S A. 2000 Oct 10;97(21):11250-5.