Membrane Electrostatics

The gene product Src has a central history in cancer biology, as it was the first proto-oncogene to be discovered. In addition, illuminating experiments on the membrane targeting of this tyrosine kinase had been instrumental for our understanding of the organelle distribution of many cytosolic proteins (McLaughlin and Aderem, 1995).

Src is attached to the cyto­solic leaflet of the plasma membrane through an N-terminal sequence, which led to the hypothesis that a membrane pro­tein recognizes this sequence. Because experiments aimed at identifying this putative Src receptor failed, another hypothesis emerged: the N-terminal sequence of Src, which is very rich in positively charged residues, interacts directly with negatively charged lipids.

This second hypothesis turned out to be true: the membrane targeting of Src, both in vitro and in vivo, as well as its transforming activity correlate with the density of posi­tively charged residues in the N-terminus, regardless of its exact amino acid sequence (McLaughlin and Aderem, 1995).

As rudimentary as it may seem, membrane electrostatics is a powerful mean to target cytosolic proteins to the plasma membrane. The cytosolic leaflet of this membrane is indeed very rich in negatively charged lipids, notably PS (net charge = - 1) and PI(₄,₅)P₂ (net charge = - 4), as compared to other organelles.

Thus, artificial constructs bearing 2, 4, 6, or 8 accessible positively charged residues show a gradual enrich­ment at the plasma membrane at the expense of internal or­ganelles (Yeung et al., 2006). In many cases, close variants of the same protein (e.g., small G proteins of the Ras or Rho subfamilies) are targeted to different organelles owing to dif­ferences in the number of positively charged residues of their membrane targeting sequence.

Nonspecific electrostatic interactions between peripheral proteins and membranes are quite weak. Consequently, they generally combine with other elementary mechanisms of membrane anchorage. In the case of Src, a myristate (a 14- carbon acyl chain) is covalently attached to a glycine upstream of the N-terminal-basic sequence.

Strong binding of Src to the membrane results both from myrisate insertion in the hydro­phobic core of the membrane and from electrostatics inter­actions of the N-terminal sequence with the membrane surface (McLaughlin and Aderem, 1995). The combination of weak interactions is a recurrent theme in membrane targeting mechanisms and has two advantages.

First, it allows cycles of membrane association-dissociation because disrupting one elementary membrane interaction (e.g., phosphorylation of a basic sequence) is generally sufficient to promote protein detachment.

Second, it improves the specificity of membrane targeting: binding occurs only when the membrane displays features that are compatible with the association of each elementary membrane-anchoring motif (e.g., negatively charged lipids and curvature).

Overall, membrane electrostatics governs a membrane ter­ritory that starts at the trans Golgi, includes various endosomal membranes and culminates at the plasma membrane. Not surprisingly, protein machineries that work in or on these membranes take advantage of this feature. A good example is cortical actin, whose nucleators at the plasma membrane interact with the phosphoinositide PI(4,5)P₂ and PS (Papayannopoulos et al., 2005).