Jordi Folch

THE ISOLATION FROM BRAIN TISSUE OF A TRYPSIN-RESISTANT PROTEIN FRACTION CONTAINING COMBINED INOSITOL, AND ITS RELATION TO NEUROKERATIN*

INTRODUCTION IT has been well established that meso-inositol exists in brain in both free and combined states. More than fifty years ago, THUDICHUM (1901) reported the presence of free inositol, and this observation has since been amply confirmed. More recent work (FOLCH and WOOLLEY, 1942) demonstrated the OcCuITence of inositol combined in brain lipides, and resulted in the isolation from brain of diphosphoinositide (FOLCH, 1949), a phosphatide which contains inositol metadiphosphate as a constituent. Results obtained in this laboratory in the course of work on the nature of neurokeratin have established the presence in brain tissue of a second form of combined inositol, namely, that of a phosphoinositide combined to protein. This paper reports these findings and some other observations bearing on the nature of neurokeratin. Preliminary reports of this work have already appeared (FOLCH and LEBARON, 195 1 , 1953). Neurokeratin is a fraction of brain proteins first isolated by EWALD and K u m (1877) which can be summarily described as a proteolytic enzyme-resistant protein residue obtained primarily from nerve tissue myelin, and which has traditionally been considered to be associated with the protein network of the myelin sheath. The classical procedure for preparing neurokeratin consists essentially of two major operations; namely, the quantitative removal of lipides with organic solvents, and the subsequent treatment of the proteins thus obtained, with digestive enzymes. The procedure takes several months, and involves Such drastic steps as continuous extraction with hot ethanol for four weeks, digestion with pepsin in dilute hydrochloric acid for three weeks, and, in some of its versions, treatment with hot alkali. The product is a protein residue insoluble in water and in dilute acids and bases, rich in sulphur, and resistant to the action of proteolytic enzymes. The preparations obtained by various workers differed somewhat in composition; and BLOCK (1951) concluded that neurokeratin is most likely an artifact, derived from a definite group of substances specific to nervous tissue. The work which led to the isolation of the protein fraction containing combined inositol was an attempt to develop a method of preparation of neurokeratin less

THE EFFECT OF pH AND SALT CONCENTRATION ON AQUEOUS EXTRACTION OF BRAIN PROTEINS AND LIPOPROTEINS

A NUMBER of papers have recently appeared describing studies on various types of aqueous extracts of brain proteins, particularly with relation to fractionation of such extracts by electrophoresis. In these studies, the rate of extraction of proteins by physiological saline has been investigated (NAKAMURA, HAYASHI and TANAKA, 1954; HAYASHI, 1953) and the effect of the pH of the extracting solution on the subsequent electrophoretic patterns has been determined (KEUP, 1955; POLYAKOVA and GOTOVTSEVA, 1957). However, to our knowledge, no systematic study has been undertaken to determine the conditions necessary for extraction of the maximum amount of protein material from brain in aqueous medium. In connexion with our work on brain proteins and lipoproteins, these data were requisite for the development of satisfactory fractionation methods, so a survey of the effect of p H and ionic strength on the amount of material extracted in aqueous solution was carried out. To this end, we prepared extracts at pH’s from 4 to 11-6 and ionic strengths from 0 to 5 and determined the amounts of nitrogen and phosphorus which could be precipitated by trichloracetic acid from these extracts. The results of these experiments are reported in this paper as well as some chemical analyses and the results of fractionation studies on the extracts thus obtained. A preliminary report of this work has been published previously (LEBARON and FOLCH, 1958). The main reason why these data had not been obtained previously for brain tissue is that the high proportion and variety of lipids found in this tissue are bound to each other and to proteins in various types of complexes which easily form emulsions and suspensions in aqueous media that are difficult to resolve. We have found that these mixtures can be separated by the use of a relatively high centrifugal force, and waterclear extracts can be thus obtained. Using the method detailed below, maximum extraction occurs when the extracting solution has an ionic strength of from 2.5 to 3 and when the pH of the homogenate is between 6 and 9. The p H of homogenates made with unbuffered salt solutions is within the optimum range. About 20 per cent of white matter protein and 50 per cent of grey matter protein can thus be obtained in aqueous solution. Such extracts have been separated by dialysis into globulin and albumin fractions, the former containing about 25 per cent lipid, of which about onefifth is cholesterol.

Preparation of Blood Lipid Extracts Free from Non-Lipid Extractives:

Deficiencies in Purification of Lipids by Petrol Ether. Resolution in petrol ether has been a classical procedure for analytical purification of extracted fats. Thus the blood fats extracted with Bloors 1 efficient alcohol-ether mixture are, for certain analyses, dried and purified by resolution in petrol ether. 1 , 2 , 3 The non-lipid extractives, such as urea, glucose, amino acids, and inorganic salts, dissolve in varying amounts in the alcohol-ether, but they do not by themselves dissolve in petrol ether. It has been recognized, however, that the petrol ether solutions show higher N:P ratios than could be expected from any of the known phosphatides. Several attempts have been made to identify the extra nitrogen. 4 , 5 , 6 The present writers have been able to identify most of it as urea, determinable with urease and other urea regeants. Urea by itself is insoluble in petrol ether, but dissolves measurably in it when the blood lipids are present. Measurable amounts of amino acids, determinable by the specific amino acid carboxyl method of Van Slyke and Dillon, 7 are also present. On the other hand, petrol ether fails to redissolve the phosphatides completely. A fraction of them remains in the undissolved residue. It is slight in normal plasmas, but in certain pathological ones it may represent 40% of the phosphatides. It has the following properties suggestive of sphingomyelin: soluble in alcohol, insoluble in petrol ether, N/P ratio of 2, C/P ratio of about 45. All the other lipids seem to be completely redissolved by the petrol ether. Proposed Extraction. The proteins and lipids are precipitated together by colloidal iron, and the water-soluble extractives are washed away. The lipids are then extracted by stirring up the wet precipitate with alcohol and ether.

Protection of Animals against Cl. welchii (Type A) Toxin by Injection of certain Purified Lipids.

Summary Purified total lipids obtained from erythrocytes, plasma, and liver have been found to have a protective effect against Cl. welchii (type A) toxin in mice and dogs. The protection appears to be due to a substrate partition effect in which the lecithinase of the Cl. welchii toxin hydrolyzes the relatively large added source of lecithin and in part spares the lecithin of the animal cells from destruction.