THE PROTEOLIPID NATURE OF THE NEUROKERATIN NETWORK OF MYELIN *

EWALD and K ~ ~ H N E (1 877) described the presence of a substance in the central nervous system, retina and peripheral nerves which, l i e keratin of epidermis, resisted solution in lipid solvents, dilute acids and dilute bases, and was not digested by gastric and pancreatic juices. In transverse sections of peripheral nerve and of white matter, this material, which they designated neurokeratin, appeared as pairs of concentric hollow tubes joined by bridges of the same material. The irregular network of stainable substance which persists in myelin sheaths following fixation and extraction of lipids has since been called ‘neurokeratin’, although in a histological rather than a biochemical sense. The investigations of H o f i ~ ~ (1936). (1950) and others (reviewed by STOECKENIUS and ZQGER. 1956) indicate clearly that the neurokeratin network of myelin sheaths seen in ordinary histological preparations is a precipitation artifact derived evidently from protein present in the myelin sheath. Because various biochemical preparations of neurokeratin have shown considerabk differences in elementary composition, some doubt has existed as’ to whether new+ keratin is a real constituent of the nervous system or an artifact derived from other proteins during its preparation (reviewed by BLOCK, 1951). Studies carried out by FOLCH and his co-workers in recent years have clarified this problem. The neurokeratin prepared by the classical method is evidently a mixture derived from the protein moiety of proteolipid (FOLCH and Lam, 1951) and a trypsin-resistant protein residue (TRPR) which contains inositol phosphatide (LEBARON and FOLCH, 1956); both of these proteins resist tryptic and peptic digestion. Proteolipid protein comprises 2 per cent of white matter and 0-6 per cent of grey matter; the TRPR is evenly distributed between white and grey matter and constitutes 0.38 per cent of brain (LEBARON and FOLCH, 1956; FOLCH and LEBARON, 1957). These data suggest that proteolipid protein would contribute more to neurokeratin as prepared by the classical method than the TRPR. Proteolipid first a p p m in the developing brain when myelination commences and increases in amount as myelination proceeds (FOLCH, 1955). This and the high concentration of proteolipid in white matter as compared with grey matter imply that proteolipid is a constituent of the myelin sheath. The TRPR has been suggested as