- Email: [email protected]
Glycosyltransferases with endogenous acceptor activity in plasma membranes isolated from rat liver
820
Biochimica et Biophysica Acta, 4 9 7 ( 1 9 7 7 ) 8 2 0 - - 8 2 4 © E l s e v i e r / N o r t h - H o l l a n d B i o m e d i c a l Press
BBA Report BBA 2 1 4 5 0
GLYCOSYLTRANSFERASES WITH ENDOGENOUS ACCEPTOR ACTIVITY IN PLASMA MEMBRANES ISOLATED FROM RAT LIVER
W.D. M E R R I T T * , D.J. M O R R E , W.W. F R A N K E a n d T.W. K E E N A N
Departments o f Biological Sciences, Medicinal Chemistry and Pharmacognosy, and Animal Sciences, Purdue University, West Lafayette, Ind. 47907 (U.S.A.) and Division o f Membrane Biology and Biochemistry, Institute o f Experimental Pathology, German Cancer Research Center, 69 Heidelberg (G.F.R.) (Received J a n u a r y 4 t h , 1 9 7 7 )
Summary Plasma membrane fractions from rat liver exhibited glycosyltransferase activity with endogenous membrane-associated acceptors and either UDPgalactose, UDPglucose, UDP-N-acetylglucosamine, or GDPmannose donors. Of these, incorporation into non-lipid acceptors was most active with UDPgalactose and only with UDPgalactose and UDPmannose was there incorporation into endogenous lipid acceptors. CMP-N-acetylneuraminic acid was inactive as a donor with the isolated plasma membranes. In order to demonstrate transferase activity, low concentrations of substrate sugar nucleotides and short incubation times were used as well as sulfhydryl protectants and a phosphatase inhibitor (NaF) in the reaction mixtures. The findings support the concept of surface localization of at least a galactosyl transferase in cells of rat liver.
The subcellular localization of glycoprotein glycosyltransferases, which were earlier t h o u g h t to be contained primarily or exclusively in the Golgi apparatus [1--4], is currently disputed. Evidence suggests that certain glycosyltransferases, especially those which add the core region carbohydrates to secretory glycoproteins, are localized in the endoplasmic reticulum [ 5--10]. In addition, numerous investigators have reported ectoglycosyltransferases, glycosyltransferases on the cell surface (for review, see ref. 11). Evidence for surface glycosyltransferases is basgd largely on indirect experiments in which sugar nucleotides are included in the medium surrounding suspensions or monolayers of presumably intact cells. This procedure has been criticised *Present address: Sterling Hall of Medicine, Yale University, New Haven, Conn. 06510, U.S.A.
821
[ 11--14], and reports of glycosyltransferase activities based on this method must be considered equivocal [13]. Previous studies failed to demonstrate significant glycosyltransferase activities in isolated plasma membranes of r a t liver with UDP-N-acetylglucosamine as donor for endogenous glycoprotein acceptors [1] or with a variety of glycosphingolipid acceptor and sugar nucleotide donor combinations [15]. As part of a systematic study on distribution of glycosyltransferases among rat liver membranes, we report here the presence in plasma membranes of a galactosyltransferase active with endogenous lipid-soluble and -insoluble membrane-associated acceptors. Plasma membranes were isolated b y the method of Yunghans and Morre [16] from livers of 200-g male rats of the Holtzman strain. Purity of these fractions was judged to be greater than 90% from marker enzyme and morphometric analysis following methods detailed previously [ 17]. Before removal from animals, livers were perfused via the portal vein with at least 200 ml of 1 mM NaHCO3 to minimize the possibility of contamination of plasma membrane fractions with erythrocyte ghosts and of serum protein absorption onto membranes. The homogenization medium, sucrose gradient solutions, and resuspension solutions all contained 5 mM 2-mercaptoethanol, and all assays were with freshly isolated membranes. Based on preliminary studies with plasma membranes and more extensive studies with endoplasmic reticulum and Golgi apparatus fractions (ref. 18; and Merritt, W.D. and Franke, W.W., in preparation), the following conditions were chosen as b o t h necessary and sufficient to demonstrate glycosyltransferase activities in isolated plasma membranes. Reaction mixtures contained, in a final volume of 0.1 ml, 40 to 60 pg of plasma membrane protein, 2.5 t~M sugar nucleotide (either UDPgalactose (UDPGal), 290 Ci/mol; UDPglucose (UDPGlc), 254 Ci/mol; UDP-N-acetylglucosamine (UDPGlcNAc), 122 Ci/mol; GDPmannose (GDPMan), 240 Ci/mol; or CMP-N-acetylneuraminic acid (CMPNAcNeu), 6.6 Ci/mol), 15 mM MnC12, 5 mM MgC12, 3 mM NaF, 5 mM 2-mercaptoethanol, and 25 mM Tris .HC1, pH 7.0. All sugar nucleotides were labelled with 14C and were from New England Nuclear, Boston, Mass. Reactions were terminated after 5 min by addition of 2 ml of chloroform/methanol (2 : 1, v/v), and the labelled endogenous acceptors were separated into three fractions as modified from Waechter et al. [19]. After collection of precipitates b y centrifugation, pellets were washed once with 2 ml of chloroform/methanol (2 : 1, v/v) and the supernatants were combined. After partitioning with 0.9% NaC1 (1 ml), lower phases were collected a.~d washed twice with 0.~% NaCi/methanol (2 : 1, v/v; 1.5 ml). The final lower phases ("lipid phase" hereafter) were dried under a stream of nitrogen and radioactivity was determined. Pellets remaining after chloroform-methanol extraction were resuspended in 1 ml of distilled water and treated with ultrasound. After centrifugation, pellets were washed 3 times with water and 2 times with chloroform/methanol/water (1 : 1:0.3, v/v). These latter two washes were devoid of radioactivity. Radioactivity was measured in the final pellet, which was presumed to contain only plasma membrane proteins and glycoproteins. Fractions were dissolved in 1 ml of hydroxide of hyamine and radioactivity was determined in a toluene-based scintillation fluid. Proteins were determined according to Lowry et al. [20].
822 Results obtained are summarized in Table !. All data were corrected for zero time blanks in which all reaction constituents were present but the reaco tion was stopped immediately by addition of chloroform/methanol. Of the sugar nucleotides tested, incorporation using UDPGal was highest into both protein and lipid fractions. Sugars were also transferred to a low but significant extent into proteins from UDPGlc, UDPGlcNAc, and GDPMan, but, except possibly for GDPMan, little or no incorporation was observed into lipids with these sugar nucleotides. With CMPNAcNeu as donor, we observed no incorporation above zero time values into either proteins or lipids. An identical result was obtained when membranes were pretreated with neuraminidase (0.1 unit/mt Clostridium perfringens neuraminidase, pH 5.5, 20°C for 30 min). These results were obtained with carrier-free concentrations of radioactive donor. Parallel experiments indicated that incorporation levels were not increased by addition of carrier sugar nucleotides. Presumably this was because the levels of endogenous acceptor were rate limiting. Due to low concentrations of donor, reproducible results were obtained only when NaF was included as a phosphatase inhibitor [21]. The presence of NaF appears to be particularly critical in studies with isolated plasma membranes, since high nucleotide pyrophosphatase and phosphatase activities are associated with these membrane fractions [22--25]. Finally, even in the presence of NaF, incubation times must be relatively short to ensure linearity of incorporation. These considerations help to distinguish plasma membrane activities from those measured previously for Golgi apparatus or endoplasmic reticulum [ 1, 15]. Assays for the latter have relied on relatively high substrate concentrations and reaction times adapted for exogenous acceptor activity and special precautions to prevent hydrolysis of substrate have been unnecessary. In contrast to plasma membranes which incorporated mannose poorly, Golgi apparatus fractions, when tested in the system outlined herein, incorporated mannose nearly as readily as galactose. Endoplasmic reticulum fractions incorporated mannose more readily than galactose (ref. 18; and Merritt, W.D. and Franke, W.W., in preparation). Based on measurements of thiamine pyrophosphatase activity, and in agreement with morphometric estimates, the m a x i m u m contamination of the plasma membrane fractions by Golgi apparatus would be 10%. The specific activity of Golgi apparatus galactosyl transferase determined in parallel for endogenous acceptors was 7.5 pmol galactose incorporated from UDPgalactose/min/mg protein. Therefore, the
TABLE
I
INCORPORATION OF RADIOACTIVE SUGARS FROM SUGAR NUCLEOTIDES INTO ENDOGENOUS ACCEPTORS OF PLASMA MEMBRANES ISOLATED FROM RAT LIVER Units of specific activity are pmol sugar incorporated/rain/rag e x p e r i m e n t s -+ s t a n d a r d d e v i a t i o n . S e e t e x t f o r d e t a i l s . Sugar nucleotide
P r o t e i n residue
Lipid phase
UDPgalactose UDPglucose UDP-N-acetylglucosamine GDPmannose CMP-N-acetylneuraminie acid
8.8 2.3 1.4 1.3 0
1 6 . 4 +- 9 . 0 0 0 0.2 + 0.1 0
-+ 4 . 1 +- 1 . 4 +- 0 . 1 + 0.9
protein. Results are averages from four
823
plasma membrane activity of Table I with UDPGal as donor could not be explained on the basis of Golgi apparatus contamination. These findings provide evidence for surface glycosyltransferases for UDPGal, UDPGlc, UDPGlcNAc, and GDPMan, and broaden our concept of the role of different cellular membrane systems in glycosylation. Certain sugars, for example galactose, fucose, N-acetylneuraminic acid, and possibly distal N-acetylglucosamine residues, are added to secretory (and membrane) glycoproteins at the Golgi apparatus [1,5--8,26--30]. Other evidence shows the site of addition of the eore region carbohydrates of secretory glyeoproteins to be the endoplasmic reticulum [ 5--10]. Glycosyltransferases which use exogenous N-acetylglucosamine [1,2,31] or gangliosides [15] as acceptors are highly concentrated in Golgi apparatus. In contrast, experiments with either isolated plasma membranes [1,15] or whole cells [32] indicate that sugars cannot be incorporated into exogenously supplied gangliosides or simple sugars by the cell surface membrane. With intact cells, only glyeopeptides from embryonic liver cells were active as exogenous acceptors [32]. In addition, these glycopeptides inhibited incorporation into endogenous acceptors [32]. Glycosylation of certain macromolecules may require the cooperation of several membrane systems, including the plasma membrane. While Golgi apparatus and endoplasmic reticulum probably play the dominant roles in glycosylation of membrane proteins, the presence of incomplete carbohydrate chains in glycoproteins of Golgi apparatus membranes [33] could be interpreted as evidence that either secretory vesicles or the plasma membrane function in completion of the biosynthetic process. With endogenous acceptors and donors other than UDPgalaetose, the levels of glyeosyltransferase activity in plasma membrane fractions are low compared to levels found in Golgi apparatus and endoplasmic reticulum. The significance of plasma membrane galactosyltransferase reactions to completion of carbohydrate chains of membrane glycoproteins remains to be assessed. We thank Keri Safranski, Kristine Hess, and Beth Weber for excellent technical assistance and Dorothy Morr~ and Dorothy Werderitsch for assistance with the electron microscopy. This work was supported by NSF PCM7511908 and Phi Beta Psi; T.W.K. is supported by research career development award GM 70596 from the National Institute of General Medical Science. Portions of this study were made possible by awards from the Alexander Von Humboldt Foundation (W.D.M.) and the Deutsches Krebsforschungszentrum (D.J.M.). References 1 M o r r d , D . J . , Merlin, L.M. and Keenan, T.W. ( 1 9 6 9 ) B i o c h e m . B i o p h y s . R e s . C o m m u n . 37, 8 1 3 - - g 1 9 2 S c h a c h t e r , H., J a b b a l , I., H u d g i n , R . L . , P i n t e r i c , L., M c G u i r e , E.J. a n d R o s e m a n , S. ( 1 9 7 0 ) J. Biol. Chem. 245, 1090--1100 3 S c h a c h t e r , H. ( 1 9 7 4 ) B i o c h e m . S o c . S y m p . 4 0 , 5 7 - - 7 1 4 S c h a c h t e r , H. ( 1 9 7 4 ) in A d v a n c e s in C y t o p h a r m a e o l o g y ( C e c c a r e l l i , B., C l e m e n t i , F. a n d Meldolesi, J., eds.), Vol. 2, p p . 2 0 7 - - 2 1 8 , R a v e n Press, N e w Y o r k 5 W h u r , P., H e r s c o v i c s , A. a n d L e B l o n d , C.P. ( 1 9 6 9 ) J. Cell Biol. 4 3 , 2 8 9 - - 3 1 1 6 Oliver, G . J . A . , H a r r i s o n , J. a n d H e m m i n g , F.W. ( 1 9 7 5 ) E u r . J. B i o c h e m . 58, 2 2 3 - - 2 2 9 7 B o u c h i U o u x , S., C h a b a u d , O., M i c h e l - B $ c h e t , M., F e r r a n d , M. a n d A t h o u e ' l - H a o n , A.M. ( 1 9 7 0 ) B i o c h e m . B i o p h y s . Res. C o m m u n . 4 0 , 3 1 4 - - - 3 2 0 8 Z a g u r y , D., U h r , J.W., J a m i e s o n , J . D . a n d P a l a d e , G . E . ( 1 9 7 0 ) J. Cell Biol. 4 6 , 52---63
824
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
29 30 31 32 33
Lawford, G.R. and Schachter, H. (1966) J. Biol. Chem. 241, 5408--5418 Lennaxz, W.J. (1975) Science 188, 986--991 Shur, B.D. and Roth, S. (1975) Biochim. Biophys. Acta 415, 473--512 Deppert, W., Werchau, H. and Walter, G. (1974) Proc. Natl. Acad. Sci. U.S. 71, 3 0 6 8 - - 3 0 7 2 Keenan, T.W. and MorrO, D.J. (1975) FEBS Lett. 55, 8--13 Harwood, R., Grant, M.E. and Jackson, D.S. (1975) Biochem. J. 152, 291--302 Keenan, T.W., MorrO, D.J. and Basu, S. (1974) J. Biol, Chem. 249, 310--315 Yunghans, W.N. and MorrO, D.J. (1973) Prep. Bioehem. 3, 301--312 MorrO, D.J., Yunghans, W.N., Vigil, E.L. and Keenan, T.W. (1974) in Methodological Developments in BiochemistrY (Reid, E., ed.), Vol. 4, pp. 195--236, Longman, London Merritt, W.D. and Franke, W.W. (1976) J. Cell Biol. 70, 304a Waechter, C.J., Lueas, J.J. and Lennarz, W.J. (1973) J. Biol. Chem. 248, 7 5 7 0 - 7 5 7 9 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265--275 Hiibscher, G. and West, G.R. (1965) Nature 205, 799--800 Fleiseher, B. and Fleischer, S. (1969) Bioehim. Biophys. Acta 183, 265--275 Bischoff, E., Liersch, M., Keppler, D. and Decker, K. (1970) Hoppe-Seyler's Z. Physiol. Chem. 351, 729--736 Touster, O., Aronson, N.N., Dulaney, J.T. and Hendriekson, H. (1970) J. Ceil Biol. 47, 604--618 Evans, W.H. (1974) Nature 250, 391--394 Wagner, R.R. and Cynkin, M.A. (1971) J. Biol. Chem. 246, 143--151 Haddad, A., Smith, M.D., Herseovics, A., Nadler, N.J. and LeBlond, C.P. (1971) J. Cell Biol. 49, 856--882 Sturgess, J.M., Mitranic, M. and Moscarello, M.A. (1972) Biochem. Biophys. Res. Commun. 46, 1270--1277 Sturgess, J.M., Minaker, E., Mitranic, M.M. and Moscarello, M.A. (1973) Biochim. Biophys. Acta 320, 123--132 LeBlond, C.P. and Bennett, G. (1974) in The Cell Surface in Development (Moscona, A.A., ed.), pp. 29--49, Jo hn Wiley, New Y o r k Merritt, W.D. and MorrO, D.J. (1973) Biochim. Biophys. Acta 3 0 4 , 3 9 7 - - 4 0 7 Arnold, D., Hommel, E. and Risse, H. (1976) Mol. Cell. Biochem. 10, 81--95 Elder, J.H., MorrO, D.J. and Keenan, T.W. (1976) Cytobiologie 13, 279--284
Biochimica et Biophysica Acta, 4 9 7 ( 1 9 7 7 ) 8 2 0 - - 8 2 4 © E l s e v i e r / N o r t h - H o l l a n d B i o m e d i c a l Press
BBA Report BBA 2 1 4 5 0
GLYCOSYLTRANSFERASES WITH ENDOGENOUS ACCEPTOR ACTIVITY IN PLASMA MEMBRANES ISOLATED FROM RAT LIVER
W.D. M E R R I T T * , D.J. M O R R E , W.W. F R A N K E a n d T.W. K E E N A N
Departments o f Biological Sciences, Medicinal Chemistry and Pharmacognosy, and Animal Sciences, Purdue University, West Lafayette, Ind. 47907 (U.S.A.) and Division o f Membrane Biology and Biochemistry, Institute o f Experimental Pathology, German Cancer Research Center, 69 Heidelberg (G.F.R.) (Received J a n u a r y 4 t h , 1 9 7 7 )
Summary Plasma membrane fractions from rat liver exhibited glycosyltransferase activity with endogenous membrane-associated acceptors and either UDPgalactose, UDPglucose, UDP-N-acetylglucosamine, or GDPmannose donors. Of these, incorporation into non-lipid acceptors was most active with UDPgalactose and only with UDPgalactose and UDPmannose was there incorporation into endogenous lipid acceptors. CMP-N-acetylneuraminic acid was inactive as a donor with the isolated plasma membranes. In order to demonstrate transferase activity, low concentrations of substrate sugar nucleotides and short incubation times were used as well as sulfhydryl protectants and a phosphatase inhibitor (NaF) in the reaction mixtures. The findings support the concept of surface localization of at least a galactosyl transferase in cells of rat liver.
The subcellular localization of glycoprotein glycosyltransferases, which were earlier t h o u g h t to be contained primarily or exclusively in the Golgi apparatus [1--4], is currently disputed. Evidence suggests that certain glycosyltransferases, especially those which add the core region carbohydrates to secretory glycoproteins, are localized in the endoplasmic reticulum [ 5--10]. In addition, numerous investigators have reported ectoglycosyltransferases, glycosyltransferases on the cell surface (for review, see ref. 11). Evidence for surface glycosyltransferases is basgd largely on indirect experiments in which sugar nucleotides are included in the medium surrounding suspensions or monolayers of presumably intact cells. This procedure has been criticised *Present address: Sterling Hall of Medicine, Yale University, New Haven, Conn. 06510, U.S.A.
821
[ 11--14], and reports of glycosyltransferase activities based on this method must be considered equivocal [13]. Previous studies failed to demonstrate significant glycosyltransferase activities in isolated plasma membranes of r a t liver with UDP-N-acetylglucosamine as donor for endogenous glycoprotein acceptors [1] or with a variety of glycosphingolipid acceptor and sugar nucleotide donor combinations [15]. As part of a systematic study on distribution of glycosyltransferases among rat liver membranes, we report here the presence in plasma membranes of a galactosyltransferase active with endogenous lipid-soluble and -insoluble membrane-associated acceptors. Plasma membranes were isolated b y the method of Yunghans and Morre [16] from livers of 200-g male rats of the Holtzman strain. Purity of these fractions was judged to be greater than 90% from marker enzyme and morphometric analysis following methods detailed previously [ 17]. Before removal from animals, livers were perfused via the portal vein with at least 200 ml of 1 mM NaHCO3 to minimize the possibility of contamination of plasma membrane fractions with erythrocyte ghosts and of serum protein absorption onto membranes. The homogenization medium, sucrose gradient solutions, and resuspension solutions all contained 5 mM 2-mercaptoethanol, and all assays were with freshly isolated membranes. Based on preliminary studies with plasma membranes and more extensive studies with endoplasmic reticulum and Golgi apparatus fractions (ref. 18; and Merritt, W.D. and Franke, W.W., in preparation), the following conditions were chosen as b o t h necessary and sufficient to demonstrate glycosyltransferase activities in isolated plasma membranes. Reaction mixtures contained, in a final volume of 0.1 ml, 40 to 60 pg of plasma membrane protein, 2.5 t~M sugar nucleotide (either UDPgalactose (UDPGal), 290 Ci/mol; UDPglucose (UDPGlc), 254 Ci/mol; UDP-N-acetylglucosamine (UDPGlcNAc), 122 Ci/mol; GDPmannose (GDPMan), 240 Ci/mol; or CMP-N-acetylneuraminic acid (CMPNAcNeu), 6.6 Ci/mol), 15 mM MnC12, 5 mM MgC12, 3 mM NaF, 5 mM 2-mercaptoethanol, and 25 mM Tris .HC1, pH 7.0. All sugar nucleotides were labelled with 14C and were from New England Nuclear, Boston, Mass. Reactions were terminated after 5 min by addition of 2 ml of chloroform/methanol (2 : 1, v/v), and the labelled endogenous acceptors were separated into three fractions as modified from Waechter et al. [19]. After collection of precipitates b y centrifugation, pellets were washed once with 2 ml of chloroform/methanol (2 : 1, v/v) and the supernatants were combined. After partitioning with 0.9% NaC1 (1 ml), lower phases were collected a.~d washed twice with 0.~% NaCi/methanol (2 : 1, v/v; 1.5 ml). The final lower phases ("lipid phase" hereafter) were dried under a stream of nitrogen and radioactivity was determined. Pellets remaining after chloroform-methanol extraction were resuspended in 1 ml of distilled water and treated with ultrasound. After centrifugation, pellets were washed 3 times with water and 2 times with chloroform/methanol/water (1 : 1:0.3, v/v). These latter two washes were devoid of radioactivity. Radioactivity was measured in the final pellet, which was presumed to contain only plasma membrane proteins and glycoproteins. Fractions were dissolved in 1 ml of hydroxide of hyamine and radioactivity was determined in a toluene-based scintillation fluid. Proteins were determined according to Lowry et al. [20].
822 Results obtained are summarized in Table !. All data were corrected for zero time blanks in which all reaction constituents were present but the reaco tion was stopped immediately by addition of chloroform/methanol. Of the sugar nucleotides tested, incorporation using UDPGal was highest into both protein and lipid fractions. Sugars were also transferred to a low but significant extent into proteins from UDPGlc, UDPGlcNAc, and GDPMan, but, except possibly for GDPMan, little or no incorporation was observed into lipids with these sugar nucleotides. With CMPNAcNeu as donor, we observed no incorporation above zero time values into either proteins or lipids. An identical result was obtained when membranes were pretreated with neuraminidase (0.1 unit/mt Clostridium perfringens neuraminidase, pH 5.5, 20°C for 30 min). These results were obtained with carrier-free concentrations of radioactive donor. Parallel experiments indicated that incorporation levels were not increased by addition of carrier sugar nucleotides. Presumably this was because the levels of endogenous acceptor were rate limiting. Due to low concentrations of donor, reproducible results were obtained only when NaF was included as a phosphatase inhibitor [21]. The presence of NaF appears to be particularly critical in studies with isolated plasma membranes, since high nucleotide pyrophosphatase and phosphatase activities are associated with these membrane fractions [22--25]. Finally, even in the presence of NaF, incubation times must be relatively short to ensure linearity of incorporation. These considerations help to distinguish plasma membrane activities from those measured previously for Golgi apparatus or endoplasmic reticulum [ 1, 15]. Assays for the latter have relied on relatively high substrate concentrations and reaction times adapted for exogenous acceptor activity and special precautions to prevent hydrolysis of substrate have been unnecessary. In contrast to plasma membranes which incorporated mannose poorly, Golgi apparatus fractions, when tested in the system outlined herein, incorporated mannose nearly as readily as galactose. Endoplasmic reticulum fractions incorporated mannose more readily than galactose (ref. 18; and Merritt, W.D. and Franke, W.W., in preparation). Based on measurements of thiamine pyrophosphatase activity, and in agreement with morphometric estimates, the m a x i m u m contamination of the plasma membrane fractions by Golgi apparatus would be 10%. The specific activity of Golgi apparatus galactosyl transferase determined in parallel for endogenous acceptors was 7.5 pmol galactose incorporated from UDPgalactose/min/mg protein. Therefore, the
TABLE
I
INCORPORATION OF RADIOACTIVE SUGARS FROM SUGAR NUCLEOTIDES INTO ENDOGENOUS ACCEPTORS OF PLASMA MEMBRANES ISOLATED FROM RAT LIVER Units of specific activity are pmol sugar incorporated/rain/rag e x p e r i m e n t s -+ s t a n d a r d d e v i a t i o n . S e e t e x t f o r d e t a i l s . Sugar nucleotide
P r o t e i n residue
Lipid phase
UDPgalactose UDPglucose UDP-N-acetylglucosamine GDPmannose CMP-N-acetylneuraminie acid
8.8 2.3 1.4 1.3 0
1 6 . 4 +- 9 . 0 0 0 0.2 + 0.1 0
-+ 4 . 1 +- 1 . 4 +- 0 . 1 + 0.9
protein. Results are averages from four
823
plasma membrane activity of Table I with UDPGal as donor could not be explained on the basis of Golgi apparatus contamination. These findings provide evidence for surface glycosyltransferases for UDPGal, UDPGlc, UDPGlcNAc, and GDPMan, and broaden our concept of the role of different cellular membrane systems in glycosylation. Certain sugars, for example galactose, fucose, N-acetylneuraminic acid, and possibly distal N-acetylglucosamine residues, are added to secretory (and membrane) glycoproteins at the Golgi apparatus [1,5--8,26--30]. Other evidence shows the site of addition of the eore region carbohydrates of secretory glyeoproteins to be the endoplasmic reticulum [ 5--10]. Glycosyltransferases which use exogenous N-acetylglucosamine [1,2,31] or gangliosides [15] as acceptors are highly concentrated in Golgi apparatus. In contrast, experiments with either isolated plasma membranes [1,15] or whole cells [32] indicate that sugars cannot be incorporated into exogenously supplied gangliosides or simple sugars by the cell surface membrane. With intact cells, only glyeopeptides from embryonic liver cells were active as exogenous acceptors [32]. In addition, these glycopeptides inhibited incorporation into endogenous acceptors [32]. Glycosylation of certain macromolecules may require the cooperation of several membrane systems, including the plasma membrane. While Golgi apparatus and endoplasmic reticulum probably play the dominant roles in glycosylation of membrane proteins, the presence of incomplete carbohydrate chains in glycoproteins of Golgi apparatus membranes [33] could be interpreted as evidence that either secretory vesicles or the plasma membrane function in completion of the biosynthetic process. With endogenous acceptors and donors other than UDPgalaetose, the levels of glyeosyltransferase activity in plasma membrane fractions are low compared to levels found in Golgi apparatus and endoplasmic reticulum. The significance of plasma membrane galactosyltransferase reactions to completion of carbohydrate chains of membrane glycoproteins remains to be assessed. We thank Keri Safranski, Kristine Hess, and Beth Weber for excellent technical assistance and Dorothy Morr~ and Dorothy Werderitsch for assistance with the electron microscopy. This work was supported by NSF PCM7511908 and Phi Beta Psi; T.W.K. is supported by research career development award GM 70596 from the National Institute of General Medical Science. Portions of this study were made possible by awards from the Alexander Von Humboldt Foundation (W.D.M.) and the Deutsches Krebsforschungszentrum (D.J.M.). References 1 M o r r d , D . J . , Merlin, L.M. and Keenan, T.W. ( 1 9 6 9 ) B i o c h e m . B i o p h y s . R e s . C o m m u n . 37, 8 1 3 - - g 1 9 2 S c h a c h t e r , H., J a b b a l , I., H u d g i n , R . L . , P i n t e r i c , L., M c G u i r e , E.J. a n d R o s e m a n , S. ( 1 9 7 0 ) J. Biol. Chem. 245, 1090--1100 3 S c h a c h t e r , H. ( 1 9 7 4 ) B i o c h e m . S o c . S y m p . 4 0 , 5 7 - - 7 1 4 S c h a c h t e r , H. ( 1 9 7 4 ) in A d v a n c e s in C y t o p h a r m a e o l o g y ( C e c c a r e l l i , B., C l e m e n t i , F. a n d Meldolesi, J., eds.), Vol. 2, p p . 2 0 7 - - 2 1 8 , R a v e n Press, N e w Y o r k 5 W h u r , P., H e r s c o v i c s , A. a n d L e B l o n d , C.P. ( 1 9 6 9 ) J. Cell Biol. 4 3 , 2 8 9 - - 3 1 1 6 Oliver, G . J . A . , H a r r i s o n , J. a n d H e m m i n g , F.W. ( 1 9 7 5 ) E u r . J. B i o c h e m . 58, 2 2 3 - - 2 2 9 7 B o u c h i U o u x , S., C h a b a u d , O., M i c h e l - B $ c h e t , M., F e r r a n d , M. a n d A t h o u e ' l - H a o n , A.M. ( 1 9 7 0 ) B i o c h e m . B i o p h y s . Res. C o m m u n . 4 0 , 3 1 4 - - - 3 2 0 8 Z a g u r y , D., U h r , J.W., J a m i e s o n , J . D . a n d P a l a d e , G . E . ( 1 9 7 0 ) J. Cell Biol. 4 6 , 52---63
824
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
29 30 31 32 33
Lawford, G.R. and Schachter, H. (1966) J. Biol. Chem. 241, 5408--5418 Lennaxz, W.J. (1975) Science 188, 986--991 Shur, B.D. and Roth, S. (1975) Biochim. Biophys. Acta 415, 473--512 Deppert, W., Werchau, H. and Walter, G. (1974) Proc. Natl. Acad. Sci. U.S. 71, 3 0 6 8 - - 3 0 7 2 Keenan, T.W. and MorrO, D.J. (1975) FEBS Lett. 55, 8--13 Harwood, R., Grant, M.E. and Jackson, D.S. (1975) Biochem. J. 152, 291--302 Keenan, T.W., MorrO, D.J. and Basu, S. (1974) J. Biol, Chem. 249, 310--315 Yunghans, W.N. and MorrO, D.J. (1973) Prep. Bioehem. 3, 301--312 MorrO, D.J., Yunghans, W.N., Vigil, E.L. and Keenan, T.W. (1974) in Methodological Developments in BiochemistrY (Reid, E., ed.), Vol. 4, pp. 195--236, Longman, London Merritt, W.D. and Franke, W.W. (1976) J. Cell Biol. 70, 304a Waechter, C.J., Lueas, J.J. and Lennarz, W.J. (1973) J. Biol. Chem. 248, 7 5 7 0 - 7 5 7 9 Lowry, O.H., Rosebrough, N.J., Farr, A.L. and Randall, R.J. (1951) J. Biol. Chem. 193, 265--275 Hiibscher, G. and West, G.R. (1965) Nature 205, 799--800 Fleiseher, B. and Fleischer, S. (1969) Bioehim. Biophys. Acta 183, 265--275 Bischoff, E., Liersch, M., Keppler, D. and Decker, K. (1970) Hoppe-Seyler's Z. Physiol. Chem. 351, 729--736 Touster, O., Aronson, N.N., Dulaney, J.T. and Hendriekson, H. (1970) J. Ceil Biol. 47, 604--618 Evans, W.H. (1974) Nature 250, 391--394 Wagner, R.R. and Cynkin, M.A. (1971) J. Biol. Chem. 246, 143--151 Haddad, A., Smith, M.D., Herseovics, A., Nadler, N.J. and LeBlond, C.P. (1971) J. Cell Biol. 49, 856--882 Sturgess, J.M., Mitranic, M. and Moscarello, M.A. (1972) Biochem. Biophys. Res. Commun. 46, 1270--1277 Sturgess, J.M., Minaker, E., Mitranic, M.M. and Moscarello, M.A. (1973) Biochim. Biophys. Acta 320, 123--132 LeBlond, C.P. and Bennett, G. (1974) in The Cell Surface in Development (Moscona, A.A., ed.), pp. 29--49, Jo hn Wiley, New Y o r k Merritt, W.D. and MorrO, D.J. (1973) Biochim. Biophys. Acta 3 0 4 , 3 9 7 - - 4 0 7 Arnold, D., Hommel, E. and Risse, H. (1976) Mol. Cell. Biochem. 10, 81--95 Elder, J.H., MorrO, D.J. and Keenan, T.W. (1976) Cytobiologie 13, 279--284