Protein Engineering

Volume 22 of the series Nucleic Acids and Molecular Biology pp 315-337

Understanding Membrane Proteins. How to Design Inhibitors of Transmembrane Protein—Protein Interactions

  • J. S. SluskyAffiliated withDepartment of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania
  • , H. YinAffiliated withDepartment of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania
  • , W. F. DeGradoAffiliated withDepartment of Biochemistry and Biophysics, School of Medicine, University of Pennsylvania

* Final gross prices may vary according to local VAT.

Get Access

Recent experiments demonstrated that buried membrane-protein hydrogen-bonding is less energetically favorable than the values that are discussed in section 3.3 of this chapter 1. Double mutant cycle analysis allows for the analysis of the interaction of two particular amino acids to the exclusion of their interaction with the rest of the protein. Such analysis of bacteriarhodopsin showed that amino acids in the correct orientation for hydrogen bonding contribute stabilization energies between −1.7 and +0.4 kcal mol−1 with an average over 8 pairs of −0.6 kcal mol−1. This is approximately the same value as that of a hydrogen bond in a soluble protein, and is consistent with the notion that the core of membrane proteins and soluble proteins are biochemically similar. However, it should be noted that the burial of polar amino acids exerts a stablizing effect on membrane proteins beyond hydrogen bonding as well. Though hydrogen bonds may contribute less to the stability of a membrane protein than previously understood, this does not necessarily argue against the importance of the membrane protein hydrogen bond. The membrane protein's unfolded state is more similar-and therefore closer in energy-to its folded state than the unfolded soluble protein is to its folded state. Therefore, although the energy of the hydrogen bond is the same, the relative contribution of the hydrogen bond to the stability of the fold is greater in membrane proteins.

This same study also showed a disparity between hydrogen bonding at lipid-exposed positions and at positions in the core of the protein. Distances between atoms and the strength of the bond between those atoms are tightly correlated. Analysis of the distance of hydrogen bond donors and acceptors in membrane proteins and in soluble proteins revealed that hydrogen bonding in buried positions of membrane proteins have a similar distance distribution to that of solvent exposed positions on soluble proteins. However, lipid-exposed hydrogen bonds are shorter, and thus likely exert a greater stabilizing effect than buried hydrogen bonds in the membrane. This is consistent with the change in dielectric constant for the two environments.

Protein—protein interactions in the membrane are just beginning to be explored. Recently, significant advances have been made in disrupting protein—protein interactions in the membrane through protein design. These advances have allowed for the manipulation of biological processes in vivo, and have been shown to be useful probes for understanding the features that stabilize protein—protein interaction in the membrane. By bringing together information on how individual amino acids modulate transmembrane structure, what forces are responsible for oligomerization in the membrane, and how to computationally encode those concepts, a method has been established to create and disrupt protein—protein interactions in the membrane. This review aims to describe the necessity and utility of such probes, as well as provide a “how-to manual” for the design of such probes.