Glycosyltransferases of rat brain that make cerebrosides: substrate specificity, inhibitors, and abnormal products.

Brain homogenates from young rats were assayed for their ability to synthesize cerebrosides from radioactive UDP-galactose or UDP-glucose and ceramide. A comparison of galactose transfer with ceramides made from different 2-hydroxy acids showed that the shortest one tested (C,) was by far the best acceptor, while the poorest contained 18 carbon atoms; longer fatty acids were better than CIS. Glucosyltransferase, on the other hand, showed rather little chain length specificity or discrimination against hydroxy acid ceramides. Synthetic compounds analogous in structure to ceramides were tested as inhibitors of the sugar transferases. Some were found to act as sugar acceptors themselves, particularly amides of DL-erythro-1phenyl-2-amino-1,3-propanediol. Some amides were good inhibitors of glucosyltransferase, particularly decanoyl norephedrine, decanoyl threo-l-phenyl-2-amino-1,3-propanediol and decenoyl phenylalaninol. The secondary amine analogous to the first of these, N-decyl norephedrine, was also very effective. No strong inhibitors of galactosyl transferase were found, although octanoyl o-threo-p-nitrophenylaminopropanediol showed promise (42% inhibition at 0.3 mM). Octanoyl phenylalaninol was nearly as good an inhibitor; the inhibition appeared only after a lag period. It is suggested that the glucosyltransferase inhibitors might he useful in therapy of Gaucher's disease, by reducing the degradative load normally falling on glucocerebrosidase. BRAIN contains glycosyltransferases which have the ability to form glucoand galactocerebroside from the corresponding uridine nucleotide sugars and ceramide (MORELL & RADIN, 1969; BASU et al., 1968, 1971). The galactosyltransferase shows a marked preference for ceramides made from 2-hydroxy fatty acids, while the glucosyltransferase utilizes the hydroxy and nonhydroxy ceramides with roughly equal ability (SHAH, 1971; MORELL et al., 1970). This report presents details of the relative effectiveness of the enzymes with ceramides made from fatty 'acids of different chain lengths, with and without a Dor This work was supported by grants NS-03192 and GB 36735 from the United States Public Health Service and National Science Foundation. Postdoctoral trainee, Mental Health Research Institute, supported by grant MH-07417 from the National Institute of Mental Health. Present address: Department of Cellular Physiology, Walter Reed Army Institute of Research, Washington, DC 20012, U.S.A. 'Present address: Meyer L. Prentis Cancer Center, Detroit, MI 48201, U.S.A. Present address: Department of Biochemistry, The Chicago Medical School, Chicago, IL 60612, U.S.A. Abbreviations used: Fatty acids are described digitally by chain length and number of double bonds (e.g.: 10:0 is decanoic acid). The letter h in front of the abbreviation indicates the presence of a 2-hydroxy group. PAPD is l-phenyl-2-amino-1,3-propanediol. Ceramide is the fatty acid amide of sphingenine. L-hydroxyl group. (The D-configuration is the natural one in hydroxy cerebrosides.) The primary purpose of this study was to search for effective inhibitors of the enzymes in an effort to control the metabolism of the glycolipids. In particular, it seemed possible that one might be able to ameliorate Krabbe's and Gaucher's diseases by slowing cerebroside biosynthesis in patients by means of suitable inhibitors. In these genetic disorders, hydrolysis of the two cerebrosides is seriously impaired and it is plausible to attempt reducing the load placed on the hydrolases by inhibiting cerebroside synthesis. This hypothesis could be examined by producing a model form of the genetic disorders in rats by means of hydrolase inhibitors-an attempt now under trial-and then correcting the disorders with synthe-