Orientation of proteolipid protein in myelin: comparison of models with X-ray diffraction measurements

Three models have been proposed for the arrangement of proteolipid protein (PLP) in the myelin membrane. We have tested these models by determining to what extent each is consistent with the membrane-membrane interactions and electron density profile of central nervous system myelin obtained from X-ray diffraction. Equilibrium periods and membrane separations were calculated from the proposed organization of lipids and proteins in the membrane, and compared with values obtained experimentally as a function of pH and ionic strength. The orientation of the proteins was also used to calculate electron density levels in the cytoplasmic and extracellular spaces. We found that the Stoffel and Hudson models for PLP were more consistent than the Laursen model with the range of pH over which the intermembrane separation at the extracellular apposition is a minimum. The Hudson model also fits better the swollen periods observed at alkaline pH. The Hudson PLP model has many more residues in the extracellular side of the membrane than does either of the other models, resulting in higher electron density in the extracellular space compared to the cytoplasmic space. Such an asymmetric distribution of electron density is offset by the electron density of myelin basic protein which is localized in the cytoplasmic space. The resulting similar levels of electron density at the two appositions are like those in profiles calculated from the X-ray data.