The objectives of the laboratory's research are:

First, to understand how the function of membrane-spanning channels is determined by their host bilayer. Though it is well established that membrane protein function is regulated by the host bilayer's lipid composition, the mechanistic basis for this regulation is poorly understood. In some cases, the regulation results from specific, stoichiometric binding of selected lipids to their "target" proteins. In many cases, membrane protein function is modulated by "general" bilayer properties: thickness, elastic moduli and lipid intrinsic curvature (or lipid "shape"). This general regulation remains enigmatic, as evident from the variety of descriptors that have been proposed for the underlying mechanisms, e.g.: bilayer fluidity; bilayer free volume; propensity to form non-bilayer phases; bilayer curvature stress; bilayer lateral pressure profile; bilayer stiffness; etc. Some of the difficulties arise from the complex organization of biological membranes, but the uncertainties identified above pertain even to "simple" model systems, which points to a serious gap in our current understanding. To obtain insight into the underlying mechanism(s) we examine the energetic (functional) consequences of the hydrophobic coupling between bilayer-spanning proteins and their host bilayer.

Second, to evaluate to what extent it is possible to use atomistic models (molecular dynamics, MD, and Brownian dynamics, BD) to gain in-depth mechanistic insight into ion permeation through narrow, cation-selective channels. To establish such a "proof-of-principle" requires emphasis on a family of particularly well-defined ion channels. For this purpose, we select the gramicidin channels because this family of ion channels allows us to combine information obtained from experiments (electrophysiology, thermodynamics and NMR) and computations based on a hierarchical implementation of MD and BD simulations. Gramicidin channels thus allow for a tight combination of experimental and computation studies because the channels are small enough that it is feasible to undertake extensive simulations, yet large enough that they have well-defined functional and structural properties that can be fine-tuned by discrete chemical modifications - to enable detailed comparisons of theory and experiment. Most importantly, gramicidins are non-trivial in the sense that their permeability properties have proven difficult to simulate quantitatively.

Contact us

Mailing address:
Olaf S. Andersen
Department of Physiology and Biophysics
Weill Medical College of Cornell University
C-501B, 1300 York Avenue
New York, NY 10065


Office phone: (212) 746-6350

Lab phone: (212) 746-6221

Fax: (212) 746-8690