Mutations of the Proteolipid Protein Gene

Abstract

Proteolipid protein (PLP) is the most abundant integral membrane protein of myelin in central nervous system (CNS), but the normal function of this myelin component has been difficult to define. The primary structure of PLP has been determined by direct protein sequencing and through cDNA cloning: PLP comprises 276 residues and its primary translation product lacks a N-terminal signal peptide (Milner et al., 1985 and references herein). Four highly hydrophobic stretches are likely to constitute membrane-spanning domains and the most widely accepted topological model of PLP assumes that both the Nand C-terminus are located at the cytoplasmic membrane surface. Such a model of PLP as a “4 helix bundle” protein is supported by both theoretical considerations and experimental data (Popot et al., 1991; Weimbs and Stoffel, 1992). By alternative mRNA splicing a second PLP isoform, termed DM20, is generated which lacks 35 residues from the intracellular loop region (Nave et al., 1987a). DM20 is less abundant than PLP in compacted myelin but constitutes the more prevalent isoform in early CNS development (Dickinson et al., 1996) and in some non-glial cells (the function of proteolipids outside the nervous system is not known). The PLP/DM20 primary structure is highly conserved in evolution and is 100% identical in mouse and man, suggesting that PLP engages in multiple protein-protein interactions with little tolerance to evolutionary changes. More recently, DM20 has emerged as the prototype of a new protein family that includes two neuronal membrane proteins, M6A (EMA) and M6B (Yan et al., 1993).