°r~t ~~ Ab L1 1 r i Michigan bra U: E W This is to certify that the thesis entitled CHARACTERIZATION OF A GLYCOLIPID N-ACETYLGALACTOSAMINYL- TRANSFERASE ACTIVITY IN NIL AND 3T3 CELL LINES. VIRAL TRANSFORMATION presented by EFFECT OF Michael Warren Lockney has been accepted towards fulfillment of the requirements for Ph.D. degree in Biochemistry Major professor Date July 25, 1980 0-7639 w: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove charge from circulation records CHARACTERIZATION OF A GLYCOLIPID N-ACETYLGALACTOSAMINYLTRANSFERASE ACTIVITY IN NIL AND 3T3 CELL LINES. EFFECT OF VIRAL TRANSFORMATION By Michael Warren Lockney A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1980 Game-- ABSTRACT CHARACTERIZATION OF A GLYCOLIPID N- ACETYLGALACTOSAMINYLTRANSFERASE ACTIVITY IN CULTURED NIL AND 3T3 CELL LINES. EFFECT OF VIRAL TRANSFORMATION. ' by Michael Warren Lockney Changes in glycolipid content and distribution in the plasma mem- branes of cells upon transformation suggest a role for these membrane components in cell growth control. The pathways for glycolipid biosyn- thesis involve the sequential addition of carbohydrate residues to glyco- lipid molecules catalyzed by glycosyltransferases which show specificity for both glycolipid acceptors and sugar nucleotide donors as substrates. While glycolipid glycosyltransferases are believed to exist as multien- zyme complexes which catalyze the addition of several sugars to a glyco— lipid acceptor, the means of regulation of these transferases or multien- zyme complexes is not clear. In the present studies, an assay for an fleacetylgalactosaminyltrans- ferase activity in cultured NIL-8 cells was developed. The activity of this enzyme in embryonic chicken brain, NIL-8 cells, and BALB/c 3T3 cells and their Kirsten murine sarcoma virus transformants was determined and the enzyme activity in NIL-8 cell homogenates was characterized. The UDP-GalNAc sugar nucleotide donor (unlabeled) was synthesized by the method of Carlson gt_al, (1964) and the product characterized by Michael Warren Lockney cellulose TLC, descending paper chromatography, and 13C-NMR analysis. GalNAc transferase activity in homogenates of embryonic chicken brain was maximum at UDP-GalNAc and GL-3 concentrations of 0.4 mM and 1.0 mM, respectively, as reported by Chien 33 31, (1973). The Km,app values obtained for UDP-GalNAc and GL-3 in NIL-8 cells were 0.137 mM and 0.415 mM, respectively. The enzyme required Mn+2 for maximum activity with maximum activation at 4 mM. Mg+2 was not able to replace Mn+2. Detergent was required in the assay mixture for enzyme activity. Sodium taurodeoxycholate gave the greatest enzyme activation at 50 ug/assay. A broad pH optimum (pH 4.5 - 8.0) was obtained with maximum enzyme activity at pH 6.0 using MES buffer. While the NIL-8 homogenates did not show GalNAc transferase activity with any other glycolipid acceptors than GL-3, GL-4 and GMB ganglioside in- hibited GalNAc transferase activity with GL-3 acceptor while GL-Z had little effect. The product obtained in the assay was shown to be GL-4 by its co-migration with GL-4 on TLC plates and by cleavage of the labeled GalNAc residue with jack bean a-hexosaminidase. GalNAc transferase activity with GL-3 acceptor was found to be about twice as high in 3T3 cell homogenates as in NIL-8 homogenates. Activity in 3T3 KiMSV homogenates was about 1/10 that of the normal 3T3 cells. A much lower level of activity was obtained with GL-Z acceptor, which was also decreased in the transformed cells. GalNAc transferase activity with 6M3 acceptor in the transformed 3T3 cells was about 10 times the level of the normal cells. The GalNAc transferase specific activity in the transformed cells with 6M3 acceptor was about the same as GalNAc transferase activity in the NIL-8 cells using GL-3 acceptor. In 3T3 and 3T3-KiMSV mixed homogenates, GalNAc transferase activity with GL-3, GL-Z, Michael Warren Lockney and 6M3 as acceptors was proportional to the ratio of normal vs. transformed cell homogenates, indicating that no soluble GalNAc transferase inhibitor was present. The Km,app for GL-3 in normal 3T3 cells was 0.249 mM and that for 6M3 in 3T3-KiMSV was 0.288 mM. These results are discussed in terms of the role of glycolipids and their glycosyltransferases in cell growth control and transformation. In addition to these studies, the effect of sodium butyrate on KB cell cycling was also examined. Since butyrate treatment causes pro- nounced changes in cell morphology and specifically induces the activity of a glycolipid sialyltransferase by the effect of this drug on cell growth was of interest. Butyrate was found to block cell growth, giving a peak of DNA synthesis in cells upon release from butyrate treatment. The position of this peak relative to that obtained with thymidine-block- ed cells indicated butyrate was blocking cells at some point in 61 phase. Further studies demonstrated that butyrate prevented M-phase cells (mitotically selected) from entering S phase while thymidine block- ed cells were able to progress through S phase in the presence of butyr- ate. In addition, butyrate pre-treatment before release from the thymi- dine block, which caused the butyrate-induced morphological change, did not prevent cells from progressing through S phase. These results indi- cate that the morphological changes and cell cycle block by butyrate are not directly related. To Cindy, for her love and support throughout my graduate studies, and to my parents, who always encouraged me to strive for the best. ii ACKNOWLEDGEMENTS I would like to thank Dr. Charles C. Sweeley for his guidance and support during my graduate career. I would also like to express my appreciation to Dr. John Wang and Dr. Clarence Suelter for many helpful discussions, along with Dr. Bruce Macher, Dr. Joseph Moskal, Dr. Fumito Matsuura, and the many graduate and undergraduate students in Dr. Sweeley's lab for their friendship and encouragement. TABLE OF CONTENTS LIST OF TABLES.................................................... LIST OF FIGURES................................................... ABBREVIATIONS..................................................... INTRODUCTION...................................................... Glycolipids In the Plasma Membrane Methods of Glycolipid Analysis............................... Glycolipids of Animal Tissues................................ Glycolipid Biosynthesis...................................... Glycolipids In Normal and Virally Transformed Cultured Cells. Glycosyltransferases In Normal and Virally Transformed Cultured Cells............................................. Localization of Glycosyltransferases......................... Effect of Sodium Butyrate 0n Cultured Cells.................. Glycolipid Synthesis During the Cell Cycle................... Statement of the Problem..................................... MATERIALS AND METHODS Materials....................................................... mthOds..00....0......OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO Preparation of DeGalactosamine-l-phosphate................... Yeast galactokinase..................................... Morgan-Elson Method..................................... Modified Fiske-SubbaRow Method.......................... Acetylation of GalN-l-P...................................... Preparation of UDP-GalNAc.................................... Biological Activity of UDP-GalNAc............................ Embryonic Chicken Brain................................. Cell Cultures................................................ .NfAcetylgalactosaminyltransferase Assay...................... iv Page vi vii mm-FHHX 19 22 24 29 31 32 33 37 37 37 38 4O 41 41 44 45 45 47 Product Analysis........ ..... .......... .......... . ........... Analysis of Cell Growth...................................... DNA Pulse-Label......................................... Synchronization of K8 Cells............................. Double Thymidine Block............................. Mitotic Selection.................................. Galactose Transport.......................................... RESULTS........................................................... Synthesis of UDPeflfAcetylgalactosamine....................... Biological Activity of Synthetic UDPfiflgAcetylgalactosamine... NfAcetylgalactosaminyltransferase Activity In Embryonic Chicken Brain......................................... 'NrAcetylgalactosaminyltransferase Activity In NIL-8 Cell Homogenates...................................... Characterization of NIL-8‘N7Acetylgalactosaminyltransferase.. Substrate Concentration................................. Metal Ion Requirements.................................. Detergent Activation.................................... pH Optimum.............................................. Acceptor Specificity.................................... Inhibition by 6M3, GL-4................................. Product Identification.................................. 'NrAcetylgalactosaminyltransferase Activities In Other Cell Lines...................................................... 3T3/BALB vs. 3T3/B-Kirsten MSV Transformants............ CHO and CHOHGA"°'""°°°°"°""‘"°°"""°°'°°"""° Effect of Sodium Butyrate 0n Cell Cycling of K8 Cells........ DISCUSSION........................................................ Sugar Nucleotide Substrates.................................. NfAcetylgalactosaminyltransferase Assay...................... Effects of Sodium Butyrate Dn KB Cell Cycling................ Regulation of Glycosyltransferase Activities................. BIBLIOGRAPHY............................................... ....... 48 52 52 53 53 53 54 55 55 75 76 85 85 85 100 100 100 109 109 109 117 120 135 135 149 149 150 158 162 168 Table 2. 3. 4. 5. 6. LIST OF TABLES Page Synthesis of UDPfiNfAcetylgalactosamine...................... 58 GalNAc Transferase Assay. Analysis of Counts............... 77 Kinetic Constants for N- -Acetylgalactosaminyltransferase ACtIVIty In NIL-8 and 3T3 CEII HomogenateSOOO0.0.0.0....O. 99 N-Acetylgalactosaminyltransferase Acceptor Specificity In 3T3 and 3T3-K1 CEIISOOOOOOOOOOOOOO0.00.0.0000...0.0.0.0... 121 N- -Acetylgalactosaminyltransferase Acceptor Specificity In CHO and CHMA CEI'ISO000.0000.00.0000000000000000.0000000.136 Butyrate-Induced Morphological Changes In Synchronized KB cells.ooooooooooooooooooooooooooooooooooeooooooeoaeooooaoo14S vi Figure 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. I7. 18. 19. LIST OF FIGURES Example of Glycosphingolipid Structure...................... Pathways for Glycolipid Biosynthesis........................ Pathway for Globo Type Glycolipid Synthesis................. Example of Ganglioside Synthetic Pathways................... Blood Group Active Glycolipid Biosynthetic Pathways......... Paper Chromatographic Technique for Glycosyltransferase Assays.................................................... Synthesis of UDPgNrAcetylgalactosamine...................... Elution Profile of Galactosamine-I-P from Dowex 50X-H+...... NMR Spectrum of Galactosamine-l-P........................... Thin Layer Chromatography of Sugar Nucleotide Synthesis ProductS.................................................. NMR Spectrum of NeAcetylgalactosamine-l-P................... Paper Electrophoresis of Coupling Reaction Mixture.......... NMR Spectrum of UDPgNgAcetylgalactosamine................... Descending Paper Chromatography of UDP-fl-Acet‘y‘ga‘actosanine.0..000.00.000.00.00.00.000.000. Embryonic Chicken Brain NrAcetylgalactosaminyltransferase... Effect of UDP-N-Acetylgalactosamine Concentration of Embryonic Cthken Brain GalNAc Transferase Activity....... Effect of Globotriaosylceramide (CL-3) Concentration on Embryonic Chicken Brain GalNAc Transferase................ Effect of Protein Concentration on NIL-8 and NIL-8HSV GaINAc TranSferase ActiVitYOOOOOOOOOOOOOOOOOOOICOIOOOOOOO. Effect of UDP-N-Acetylgalactosamine Concentration on NIL-8 GaINAc TranS?erase ACtiVitYOOOOOOOOOOOOOOO0.00.00.00.00... vii Page 10 12 14 18 SO 57 61 63 65 67 70 72 74 8O 82 84 87 89 Figure Page 20. Lineweaver-Burke Plot of GalNAc Transferase: UDPngAcetylgalactosamine Concentration Curve in NIL-8 HomogenateSOIOOOOOOOOO0.000000000000000000000000000IOOOOO. 92 21. Eadie-Hofstee Plot of GalNAc Transferase:UDP-N-Acetyl- galactosamine Concentration Curve in NIL-8 Homogenates.... 94 22. Effect of GL-3 Concentration on GalNAc Transferase Activity 1" NIL-8 Gel] HomogenateSOOOOOOOOOOO0.0.00.000.00.00....0. 96 23. Eadie-Hofstee Plot of GalNAc TransferasezGL-3 Concentration Curve in NIL-8 Homogenates.................. 98 24. Effect of Divalent Metal Cation Concentration on NIL-8 GalNAc TraHSferase Act1Vity00000000000.000.00.00....00...O 102 25. Detergent Requirement for NIL-8 GalNAc Transferase Activity. 104 26. Activation of NIL-8 GalNAc Transferase by Sodium TaUI’OChOIate V50 SOdIUfl TaUT‘OdEOKYChOIateo00-00000-0000000 106 27. Optimum pH of NIL-8 GalNAc Transferase...................... 108 28. Inhibition of NIL-8 GalNAc Transferase by Potential Glyc01ipid AcceptorSOO0.0.00......OOOOOOOOOOOOOOOOOOOOOOOO 111 29. Inhibition of NIL-8 GalNAc Transferase by 6M3 Gang]1051de00000000000000.0.00......0.00000000000000000000 113 30. Thin Layer Chromatography of GalNAc Transferase Assay PrOdUCt from Assay PaperSOOOOO...OOOOOOOOOOOOOOOOOOOOO0..O 116 31. Thin Layer Chromatography of GalNAc Transferase Product Treated with a-Hexosaminidase............................. 119 32. GL-3:GalNAc Transferase Activity in Mixed 3T3 and 3T3-KiMSV HomogenatESOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0.0.0.0000....0... 124 33. 6M3:GalNAc Transferase Activity in Mixed 3T3 and 3T3- KIMSV HomogenateSOO000.000.000.000000000000000000000000000 126 34. Effect of GL-3 Concentration on 3T3 GalNAc Transferase ActiVityOOOOOOOOOOOOOO0.0.00.0...OOOOOOOOOOOOIOOO ......... 128 35. Eadie-Hofstee Plot for GalNAc Transferase:GL-3 Concentration Curve in 3T3 Cell Homogenates............... 130 36. Effect of 6M3 Concentration on 3T3-KiMSV GalNAc TranSferase ACtiVityooooooooooooooooooooooooooo ooooooooooo 132 37. Eadie-Hofstee Plot of GalNAc Transferase:GM3 Concentration Curve in 3T3-KiMSV Homogenates.............. 134 viii Figure Page 38. DNA Synthesis in KB Cells Synchronized by Thymidine or Butyrate BIOCkS.0.00.0.0.0...OOOOOOOOOOOOOOOOOO0.00.0.0... 139 39. Effect of Butyrate on Cycling of Mitotically-Selected Cells. 142 40. Effect of Butyrate on Thymidine-Blocked KB Cells............ 145 41. Transport of Galactose in Butyrate-Treated KB Cells. ........ 147 42. Possible Pathways for Synthesis of Asialo-GMZ in 3T3-KIMSV CEIIS.OCOOOOOOOOOOOOOOO0.000000000000000000000.0 156 430 cell cyc1e Mada].ooooooooooooooooooooooooooooooooooooooooooo 161 44. Phosphorylation-Dependent Regulatory Mechanism for Glycosyltransferase Activity.............................. 165 ix ABBREVIATIONS N-acetyl -_D__-gal actos ami ne, Gal NAc; N-acetyl -a all-gal actosami ne-l-phos- phate, GalNAc-I-P; =l__)=--galactosanine-l-phosphate, GalN-l-P; uridine diphos- phate-a-N-acetyl-galactosamine, UDP-GalNAc; galactose, gal; glucose, glc; N-acetylglucosamine, GlcNAc; glucosylceranide (CL-1), glc-cer; lactosyl- ceranide (CL-2), lac-cer; globotriaosylceramide (CL-3) , Gste3Cer; globotetraglycosylceranide (CL-4, globoside) , GbOse4Cer; globopenta- glycosylceramide (CL-5, Forssman antigen), Gste5Cer; cyclohexylcar- bodiimide, DCC; acetic anhydride, AcZO; adenosine 5'-triph05phate, ATP; phosphoenolpyruvate, PEP; N(-)3-phosphoglyceric acid, PGA; sodiun tauro- cholate, TC; sodiun taurodeoxycholate, TDC; embryonic chicken brain, ECB; N-acetylgalactosaminyltransferase, GalNAc transferase; thymidine, thy; 2(N-morpholino)ethane sulfonic acid, MES; Kirsten murine sarcoma virus, KiMSV; N-Z-hydroxyethylpiperazine-N‘-2-ethanesulfonic acid, HEPES; tris(hydroxymethyl)aminomethane, TRIS; chloroform:mthanolzwater, C:M:W. INTRODUCTION According to the fluid dynamic model of the animal plasma membrane proposed by Singer and Nicolson (1972), integral membrane proteins exist in a “sea" of lipids. The lipids are arranged in a bilayer in which individual components are able to diffuse laterally (Zwaal, 1973). In addition, some lipids may segregate to form more structured regions, par- ticularly around membrane proteins, and are separate from the freely dif- fusable lipids (Nicolson, 1976; Singer and Nicolson, 1972; Edidin, M., 1973; Zwaal, 1973; McConnell, 1975). There are basically two types of membrane proteins: (1) proteins integrated into the membrane (integral membrane proteins), and (2) proteins bound to the membrane surface by hydrophillic interactions (or peripheral membrane proteins) (Singer and Nicolson, 1972). Of the various classes of lipids in the plasma membrane, the glyco- lipids are one of the most interesting. Glycosphingolipids have the gen- eral structure shown in Figure 1, consisting of a sphingosine residue linked through its amino group by an amide linkage to a long saturated or mono-unsaturated fatty acid. This glycolipid backbone structure is call- ed ceramide. A variety of polar head groups can be attached to the hydroxyl group at the 1 position of the sphingosine base. In the case of glycosphingolipids, the polar group is a carbohydrate moiety which can contain from 1 to 20 or 30 glucose units. Thus, glycolipids contain a hydrophobic tail region, as do phospholipids, but they have relatively large, complex polar regions which can participate in cell surface events. In addition, it has been shown that the carbohydrate moieties of .copamuuu nude warm: ccpuuawg emacmmmcmcp u Dihydrosphingosine (2) Sphingosine can be derived from dihydrosphingosine through a ketosphingo- sine intermediate (Braun and Snell, 1968), through a desaturase reaction (Brady gt_al., 1958), or independently from hexadecanoyl-CoA and serine, similar to the reactions above (DiMari gt 31., 1971; Stoffel gt 21., 1967; Nakano and Fujino, 1973). The synthesis of galactosyl- or gluco- syl-ceramide probably occurs by acylation of the sphingosine and transfer of the sugar residue to the resulting ceramide moiety (Morell and Radin, 197D; Morell g; 21., 1970; Radin g; 31., 1972; Hammarstrom, 1970). A minor pathway for ceramide biosynthesis may also occur involving the transfer of galactose or glucose to sphingosine to give psychosine or glucosylsphingosine, respectively (Cleland and and Kennedy, 1960; Brady, 1962; Hildebrand g; 11., 1970; Basu g; 31., 1973). The resulting glycosylsphingosine is then acylated to give galactosyl- or glucosyl- ceramide. The biosynthesis of gluco- and galactosyl-ceramide has been reported by several groups using 1g litgg_assay techniques (Basu gt_al,, 1968; Basu 93 31., 1973; Cleland and Kennedy, 1960; Morell and Radin, 1969; Basu 3511., 1971). While invi—vg studies have been carried out by adding labeled glycolipid precursors to anhnals, there are several diffi- culties in interpreting these data (Burton, 1969). These include the variations due to the age and species of the animals being used and esti- mation of the molar size of the body's pool of precursor. However, lg Igiyg studies have provided a great deal of useful information. For instance, when glycolipid metabolism in fetal and newborn tissues is examined, the differing rates of development among species show up as different ages of glycolipid synthesis inductions (Moser and Karnovsky, 1959; Burton, 1969). In another in vivo study using cultured neuroblas- toma strains, Kemp and Stoolmiller were able to show a precursor-product relationship supporting the biosynthetic sequence 6M3 —>GM2 —->- 6M1 (1976a). This agreed with the sequence established using lg 31359 transferase assay systems (Kaufman g£_al., 1966; Roseman, 1970). Other 15.1139 studies have shown a greater incorporation of DL-[3-14C] neuraminic acid into gangliosides of neuron-rich sections of rat brain than in glial-rich fractions (Jones g§_al., 1972). Because Of the difficulties inherent in studying glycolipid synthe- sis lg giyg,‘ig 11359 assays were developed. In these assays the trans- fer of labeled sugar nucleotides to exogenous glycolipid acceptors are measured. The particulate enzyme and glycolipid substrates are solubil- ized with detergents for maximum activity. Using these assay systems several groups have reported glycolipid biosynthesis in vitro in a vari- ety of tissues and cultured cells. Basu gt_al. (1971) reported synthesis of lactosylceramide (CL-2) in embryonic chicken brain, and Hildebrand and Hauser (1969) demonstrated synthesis of lactosylceramide and triglycosyl- cermnide in rat brain. In the latter case the enzyme activities for the synthesis of lactosylceramide (a-galactosyltransferase) and globotria- osylceramide (CL-3) (a-galactosyltransferase) were shown to be different by heat inactivation and substrate inhibition studies. Globoside synthe- sis was demonstrated in embryonic chicken brain by Chien £5 31. (1973) and Forssman synthesis was reported in guinea pig (Kijimoto gt al,, 1974; Ishibashi g; 21., 1974), in Y-1 adrenal tumor cells (Yeung st 31., 1974), and in porcine erythrocyte bone marrow (Dawson and Sweeley, 1972). The synthesis of gangliosides, beginning with the sialylation of lactosylcer- amide, was initially reported by Basu gt al. (Basu, 1966; Basu and Kaufman, 1965; Kaufman gt 31., 1966). The pathways of ganglioside bio- synthesis were more completely identified by Basu, Kaufman, Steigerwald, and Roseman (Basu £3 31., 1965; Steigerwald g5 11., 1975; Kaufman gt 31., 1967; Kaufman gt 11., 1968; Roseman, 1970). Based on these studies the pathways for neutral glycolipid and ganglioside biosynthesis shown in Figures 2-4 were presented. These pathways have been corroborated in frog brain (Yip and Dain, 1970), rat brain (Yip and Dain, 1970; Hildebrand £3 31., 1970; Arce 33 31., 1970), in neuroblastoma clones (Moskal ggual., 1974) and 13 vivo with neuroblastoma cells (Kemp and Stoolmiller, 1976a). .mxazgama mmmgu co $.23 c8 «58 9.22.95 a mm 32.3 “cacao—oipag 635283833 .muEZoubm a6 quxu so mazosm uvmmn sac; cusp uaup>pu m. m—mmnuzzmo'n cpappouxpm .mcrsacmu seem .mpmmgucxmovm evappouxpu go; mzazsumm .N weaned [lllll I, TO u¢>h <4<¢ macaw cooam ma>h o~awzh cmcac ) ..8_i§§+_££-m8-m 32.78.9433 omucmmmcacuocpzm emacmwmcacupxmouum_mw ma<3zhmo_m c—mudou>4c ll .Acouaas cmammsomv mubw o» a: czczm m? mcvnrpouzpm camp onopw .Pacuaw: cow mocmzawm use .mvmmguczm vvnppouxpc camp onopo see zazguma .m beamed 12 emu—122cc +2 33 .1298 ipmu=ou on cog» :mu gupzz .mzouopmvmm m>pm on Nubw on umuuu my macvmmc o :mzu< .8233 4~3o2u< E . .. .._ M m . . L8.iaa§i§§izaGola —>Gna- In a study of ganglioside biosynthesis in rat brain, Arce gt 31. (1971) demonstrated that, in a non-detergent system, the glycosyltrans- ferases do not interact freely with all the substrates available. Thus, the actual glycosyltransferase activities 1_.vivo may be regulated, in part, by substrate or enzyme distribution. Because of this, it is clear that enzyme activities measured in viggg_may not reflect actual activi- ties _i_n_v_i_vg_. Evidence for an alternate pathway of ganglioside synthesis was ob- tained by Basu gt a1. (1974), where a s-N-acetylgalactosaminyltransferase transfers a GalNAc residue to lactosylceramide to give asialo-GMZ. However, this activity was found in the supernatant fraction after a 12,000 x g centrifugation. Since most glycolipid glycosyltransferases are believed to be membrane bound, this activity may have been due to very high glycoprotein glycosyltransferase activity. Synthesis of blood group active glycolipids has also been studied. The synthesis of blood group glycolipid precursor triglycosylceramide 16 (GlcNAc81.—-> 3Gale —->4Glc —>Cer) was reported by Kijimoto e; a_l_. (1974). Several other blood group glycosyltransferase activities have also been studied (Basu and Basu, 1972; Basu and Basu, 1973; Moskal et. .31., 1974; Presper gt_gl., 1976). Recent studies have demonstrated blood group related glycolipids in a variety of sources, including neuronal tissue (Chien £3 31., 1978; Uemura egg” 1978) and hunan kidney (Rauvala _e£_a_l_. , 1973) as well as human erythrocytes (Natanabe £2 31,, 1978; Watanabe 33 31., 1979). The .12.!1322 biosynthesis of blood group-related glycolipids has been report- ed in rat Ascites hepatoma cells (Taki e_t 31. , 1979a), hunan serun (Pacuszka and Koscielak, 1976), and several mouse and human cell lines (Basu gt al,, 1979; Basu gt 31., in press). From these studies the bio- synthetic pathways for blood group active glycolipids shown in Figure 5 have been proposed. The acceptor specificities of the glycosyltransferases catalyzing glycolipid biosynthesis have been demonstrated in only a few cases. Since many of the carbohydrate transfers involve addition of the same sugar residues, the possibility exists that many different glycolipids could be synthesized by the same enzymes. Thus, the question of enzyme specificity is an important one. As mentioned previously the 5- and a-galactosyltransferases which are involved in the synthesis of lactosyl- ceramide and globotriaosylceramide, respectively, appear to be separate enzyme activities. This was demonstrated by heat inactivation studies in which the galactosyltransferase which utilized glucosylceramide acceptor was inactivated by heating to 50°C for 5 min while the lactosylceramide galactosyltransferase was not (Hildebrand and Hauser, 1969). Galactosyl- transferase activities which catalyze the synthesis of l7 .mvmogucxm cvavpoozpm m>yaum m aaogw voopm o» mummp mozzuma «nah open; och .mrmmgucxm pum use < azocu voopm on muamp xmzsuma agar cocouuma mgh .mscwmms mmouumpoo uxo: on» mo mmm3:?_ one an cm:Fngumv .mmazh cocouuob new onus; on“ .mmgucmsn o3» ope? mmuw>+u Amzsuma ecu “grog mvsu u< .mibw oa mauvmmg opuum azogm coopm .mxuzguam ovuozucamo?m uvavpouzpo m>Pau< azogw coopm .m mesmwm 18 um: PBS m .595 :83 9 of P6 m EST—mu pceaupupmum au—mu4wuomhasialo Gm —>GM1, but were unable to make 6M2 ganglioside. One of the problems associated with measuring "cell surface“ glyco- syltransferase activities with intact cells was demonstrated by Deppert .22.21- (1974) and Hirshberg gt 31. (1976). These groups measured the hydrolysis of several sugar nucleotides in vitro with whole cells. Deppert gt 31. (1974) showed that, within 1 hr of incubation with intact BHK cells, 95% of the added UDP-Gal was hydrolyzed. Neither UDP-Gal nor Gal-l-P were able to permiate the membrane. Hirshberg gt 31. (1976), using GNP-[14C] sialic acid and [3HJ-CMP-sialic acid, showed that over 75% of the sialic acid incorporation by intact NIL, BHK, and 3T3 cells from CMP-sialic acid was due to uptake of free sialic acid and 28 utilization within the cell. They also pointed out that the concentra- tion of unlabeled sugars used by other groups to effectively "wash out" incorporation of hydrolyzed, labeled sugar was too low, since the apparent Km for free sugars was around 10 mM (vs. 1 mM unlabeled sugar added). Merritt gt 31. (1977) found that glycosyltransferase activity in plasma membrane fractions of rat liver only used UDP-Gal and GDP-Man as donors for lipid acceptors. More recently, a plasma membrane-enriched fraction from BALB/c 3T12 cells was shown to contain a galactosyltrans- ferase activity, which was partially characterized (Cunnings gt a_l_., 1979). Possibly the most persuasive studies that have been done to show cell surface glycosyltransferase activities have been those utilizing exogenous acceptors which were attached to solid supports. Yogeeswaren 33 al, (1974) demonstrated the transfer of galactose from exogenously added UDP-[14C]-Gal to glycolipids attached to glass beads upon con- tact with NIL or BHK cells. Furthermore, glycosylation was not as signi- ficant with polyoma transformants. They proposed that retinal could serve as an in 3112 sugar donor, since glycosylation of the glass parti- cles was enhanced by adding retinal to the medium. Unfortunately, these results could not be reproduced by Hakomori's group so their validity is questionable. Verbert gt 31. (1976) demonstrated the transfer of galac- tose from UDP-Gal to a non-phagocytosable exogenous acceptor and to endo— genous membrane receptors by spleen lymphocyte cells. The newly glyco- sylated cells were shown to have different agglutination properties with soybean agglutinin. Schnaar gt 51. (1978) have also shown a carbohy- drate-specific binding of chicken hepatocytes to polyacrylamide gels derivatized with Neacetyglucosamine but not e-glucose, s-galactose, 29 s—mannose, s-maltose, a-malibiose, a-cellibiose, or a- or a-lactose. This binding was dependent upon calcium ion levels and was decreased at low temperatures and could be inhibited by Nracetylglucosamine or GlcNAc-glycosides. While a considerable amount of data has been obtained by various techniques supporting and refuting the concept of cell surface glycosyl- transferases and mechanisms of intercellular adhesion, the existence of such molecules at the cell surface or their function is still not clear. With the increasing awareness of the importance of cell surface glycoconjugates in cell growth processes and oncogenesis, there arose a need for model systems in which the regulation of glycoconjugate meta- bolism could be studied. One such system was reported by Simmons 35 al, (1975) and Fishman £5 al. (1974b) in which Hela cells treated with 2 mM sodium butyrate for 18-24 hr showed a specific increase in CMP-sialic acid:lactosylcermmide sialyltransferase which was 15-20 times that of untreated cells. This change in sialyltransferase activity was accom- panied by a distinct morphological change in which the cells became long and spindle- shaped and extended processes. Similar results were obtain- ed by Macher gt 31. (1978) using KB cells. When HeLa or KB cells were cultured in butyrate-containing media for 24 hr and then placed in fresh medium (without butyrate), the levels of sialyltransferase returned to near normal levels within 24 hr (Simmons 33 al,, 1975). When butyrate-treated cells were trypsinized and replated in butyrate-free medium they transiently retained the "butyrate" morphology. When cycloheximide was added to the butyrate treatment median this return to butyrate morphology after trypsinization was prevented (Henneberry and Fishman, 1976). Cycloheximide treatment after removal of butyrate from 30 the cells prevented reversion of the cells to the normal morphology as well as the decline in 6M3 levels. The sialyltransferase activity returned to normal levels even in the presence of cyclohexamide. Addi- tion of cycloheximide or Actinomycin D with butyrate inhibited morpho- logical changes (Altenberg gt 31., 1976). Sialidase activities for 6M3 and 601 substrates were normal during and after butyrate treatment (Tallman gt 11., 1977). In addition to the changes in sialyltransferase activity and cell morphology, several other, seemingly unrelated biochemical changes occur during butyrate treatment. These include an inhibition of histone deacetylation (Candida gt 91., 1978), increase in acetylcholinesterase activity in cultured neuroblastoma cells (Schneider, 1976), the activa- tion of tyrosine hydroxylase in (Prasad and Sinha, 1976), and the induc- tion of alkaline phosphatase in 86 and CHL cells (Koyama and Ono, 1976). Altenberg and Steiner (1976) reported that butyrate caused the appearance of cyto-actin fibers in MSV-MLV transformed NRK cells resembling those of non-transformed cells. They also demonstrated the presence of microfila- ments in the long, butyrate-induced processes and showed that, while transformed NRK cells had random patterns of microtubules, butyrate- treated cells had parallel arrays radiating from the nucleus. While differentiation of some cells can also be caused by addition of dibu- tyryl-cyclic AMP (db-cAMP), and cAMP levels are altered by butyrate treatment, induction of sialyltransferase does not occur upon addition of db-cAMP to the culture medium. Furthermore, butyrate is a better inducer of alkaline phOSphatase than db-cAMP, and the effects of these two com- pounds added together is synergistic. Thus, alterations in cell 31 morphology and biochemistry by butryate are probably due to a separate process than the CAMP-mediated changes. Butryate has also been shown by several groups to alter cell growth. Ginsberg gt 31. (1973) reported the growth inhibition of Hela, Chang liver, L-132, intestine, rat sarcoma, chicken embryo, and SV/3T3 cells by butyrate as well as morphological changes in all but the primary epithe- lial and fibroblast cells. Hagopian gt 21. (1977) found that chick em- bryo fibroblasts and Hela cells stop dividing in media containing 2 mM and 5 mM butyrate, respectively. Protein and RNA synthesis in these butyrate-treated cells was not affected. In addition, the cytosol from butyrate-treated cells was able to inhibit DNA synthesis in control nuclei while nuclei from butyrate-treated cells remained inactive in con- trol cytosol. More recently, Macher gt 31. (1978) and Fallon and Cox (1979) have shown that butyrate acts as a cell cycle blocking agent, arresting cells in the 61 phase of the cell cycle. Wolf and Robbins (1974) and Chatterjee §t_al. (1975) have shown that glycoconjugate synthesis is cell cycle dependent. These studies showed that glycolipid biosynthesis occurs primarily in M and early G1 phase of the cell cycle, while glycoprotein synthesis takes place largely in S and early 62 phase. The fact that butyrate specifically alters glyco- lipid metabolism and also affects cell cycling implies that these two phenomena are related, particularly since both the butyrate block and glycolipid biosynthesis occur during G1 phase. Thus, the butyrate sys- tem may provide a means to study cell cycling behavior as well as a tool for the study of the regulation of glycolipid metabolism. 32 Statement of the Problem The studies described above demonstrate a clear correlation between glycolipid metabolism and transformation in cell culture systems. How- ever, studies showing, directly, a role for glycolipids in cell growth control and/or viral transformation are lacking. One approach to this problem is through the study of the enzymes involved in glycolipid metab- olism with the goals of 1) demonstrating regulatory changes in these enzymes upon transformation and 2) developing methods to selectively regulate these enzyme activities in in vivo systems. The present study is designed to characterize a glycolipid Nracetylgalactosaminyltransfer- ase activity in a cultured cell line which may be subject to regulation during cell growth and transformation. The goals of this study were as follows: 1. Preparation of UDPtfl-acetylgalactosamine for use as a substrate in enzyme assays. 2. Assessment of synthetic UDP-GalNAc using previously characteriz- ed embryonic chicken brain GalNAc transferase. 3. Characterization of GalNAc transferase activity in normal NIL-8 cells, including substrate specificity, substrate concentration curves and binding constants, pH optimum, requirements for metal ions and deter- gents, and possible inhibitory factors. 4. Assessment of GalNAc transferase activities in normal vs. virally transformed cells. These studies will lead to a better understanding of the regulation of a specific glycosyltransferase in normal and transformed cell lines, and enable future investigators to obtain more direct evidence for the role(s) of glycolipids and their metabolic enzymes in cell growth regula- tion and transformation. MATERIALS AND METHODS MATERIALS SOLVENTS CHEMICALS lngalactosamine°HCl Adenosine—5'-triphosphate Phosphoenolpyruvate 'Q(-)3-Phosphoglyceric acid QfN-Acetylgalactosamine Amino-2-naphthol-4- sulfuric acid Dicyclohexylcarbodiimide Morpholine DETERGENTS Sodium Taurocholate Grade A Sodiun Taurodeoxycholate Grade A Triton X-100 33 Solvents were redistilled by constant-flow rotary evaporation. Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Pierce Chemical Co. St. Louis, MO Kodak Chemical Co. Rochester, NY Aldrich Chemical Co. Milwaukee, MN Aldrich Chemical Co. Milwaukee, MN Calbiochem La Jolla, CA Calbiochem La Jolla, CA Research Products International Corp. Elk Grove Village, IL 34 ATLAS G-3634-A Cutscum Tween 20 CHROMATOGRAPHY SUPPLIES Dowex SOW-XB (100-200 mesh) Dowex 1X-8 (IOO-ZOO mesh) Bio-Gel P-2 Silica Gel G Thin-Layer Chromatography Plates Silica Gel 60 High Performance Thin-Layer Chromatography Plates Cellulose Thin-Layer Chromatography Plates CELL CULTURE SUPPLIES Plastic Tissue Cultures Flasks 25, 75, 150 cm2 Minimal Essential Medium (Earle's Salts) Dulbecco's Modified Eagle Medium Alpha Modified Minimal Essential Medi an ICI United States, Inc. Wilmington, DL Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Bio-Rad Laboratories Richmond, CA Analtech, Inc. Newark, DE EM Reagents Oarmstadt, Germany Analabs, Inc North Haven, CT Falcon Plastics Oxnard, Ca Corning Glass Works Corning, NY GIBCO Laboratories Grand Island, NY GIBCO Laboratories Grand Island, NY GIBCO Laboratories Grand Island, NY 35 Trypsin (1:250) Bovine and Fetal Bovin Serun Membrane Filters (0.22 n) RADIO-LABELED MATERIALS Uridine diphosphate-[14CJ1flf AcetylfgfiGalactosamine (50 mCi/mmol) [14CJ-Thymidine [3H]-Thymidine [14CJ7Q:Galactose MISCELLANEOUS REAGENTS Yeast Hexokinase a-Glucosaminidase from Jack Bean Bio-Solv BBS-3 PPO (2,5-inhenyloxasole) Dimethyl-POPOP (1,4-bis-2-[4-methyl-5- phenyloxazolle-benzene) GIBCO Laboratories Grand Island, NY GIBCO Laboratories Grand Island, NY Kansas City Biological Kansas City, KS Millipore Corp. Bedford, MA New England Nuclear Boston, MA New England Nuclear Boston, MA ICN Pharmaceuticals, Inc. Irvine, CA ICN Pharmaceuticals, Inc. Irvine, CA Sigma Chemical Co. St. Louis, MO Sigma Chemical Co. St. Louis, MO Beckman Instruments, Inc. Fullerton, CA Research Products International Corp. Elk Grove Village, IL Research Products International Corp. Elk Grove Village, IL 36 Porcine and canine intesting glycolipids were previously isolated and purified in this laboratory by Dr. Kenneth Dean and Dr. Joseph Sung. All other reagents used in these studies were reagent grade. METHODS 1. Synthesis of Uridine Diphosphategflfacetylgalactosamine. A. Preparation of D-Galactosamine-l-phosphate. Yeast galactokinase. The prepartion of GalN-l-P was carried out accord- . ing to the method of Carlson gt 31. (1964). Galactokinase was extracted from 10 g of lyophilized, galactose-adapted yeast (Sigma, St. Louis, M0) by autolyzing the yeast overnight in 30 ml of 0.1 M NaHC03 at 25°C. This suspension was centrifuged at 32,000 x g in a Sorval RCZ-B centri- fuge equipped with an 55-34 rotor (Sorval, Newtown, CN) for 30 min. The pellet was washed with 30 ml of 0.1 M NaH003 and centrifuged again. The combined extracts contained 40-50 mg protein per ml. Galactokinase assay. The assay mixture contained, in wholes, phosphate buffer (pH 7.8) 200; MgClz, 2; adenosine-5‘-triphosphate (ATP) (from equine muscle), 40; galactosamine-HCl (GalN-HCl), 40; phosphoenolpyruvate (PEP), 10; 0(-)3-phosphoglyceric acid (PGA), 10; and 0.01 ml of galacto- kinase preparation, in a final volume of 0.14 ml. The pH of the reaction mixture was adjusted to 7.8 before addition of galactokinase. After the reaction mixture was incubated for 30 min it was heated to 100°C for 2 min and centrifuged. An aliquot (0.05 ml) of the supernatant was assayed for GalN-l-P as follows: (1) Reduction of remaining substrate. One drop of capryl alcohol was added to each sample, followed by 0.025 ml of 1.0 M NaBH4. This solution was vortexed and allowed to stand at room temperature for 5 min with occasional shaking. The borohydride addition was repeated and, after 5 min, 0.025 ml of acetone was added. When this solution had been 37 38 allowed to stand 5 min the excess borohydride was removed by heating the mixture to 100°C for 3 min. (2) Acetylation. Aqueous 0.5 M acetic anhydride (0.1 ml) was added to the reaction mixture while the pH was maintained between 7 and 9 (pH paper) with saturated NaHCO3. The reaction was complete within 10 min. (3) Hydrolysis. The GalNAc-l-P in the reaction mixture was hydro- lyzed by adding 0.2 ml of 2 N HCl and heating to 100°C for 10 min. After the reaction mixture had cooled it was neutralized with 2 N NaOH using phenolphthalein indicator and water was added to give a final volume of 1 ml. An aliquot (0.5 ml) of this fraction was assayed for free GalNAc by the modified Morgan-Elson procedure (Reissig 35:31,, 1955). The Morgan-Elson Method for detection of acetylated sugars. To the sam- ple, blank, and GalNAc standards (0.01-0.1 umole/assay), each in a total volume of 0.5 ml, was added 0.1 ml of 0.8 M potassium tetraborate (pH 9.1). This mixture was heated in a boiling water bath for 3 min after which the tubes were cooled in tap water. Dimethylaminobenzaldehyde (0MBA) reagent* (3 ml) was added to each tube and the solutions were allowed to incubate in a water bath at 36-38°C for 20 min. Tubes were cooled and the absorbances read immediately at 585 (or 540) nm on a Gilford 2400 Spectrophotometer (Gilford Instrument Laboratories, Oberlin, OH). Large scale preparation of GalNAc-l-P. The reaction mixture for large scale preparation included the following components, in mmoles: GalN-HCl, 12; PEP, 20; ATP, 12; PGA, 20; potassiun phosphate buffer (pH *DMBA reagent was prepared by dissolving 1 g of DMBA in 10 ml of glacial acetic acid containing 12.5% (V/V) 10 N HCl. This stock solution can be stored at 2°C for one month. When used for an assay the working solution was made up fresh by diluting the stock solution with glacial acetic acid, 9:1. 39 7.8), 60; MgClz, 6; and 60 ml of the crude galactokinase preparation in a total volume of 280 ml. The pH was maintained by addition of 1 N NaOH. After 6 hr at 30°C, 0.2 ml of toluene was added and the mixture was kept at 30°C for an additional 12 hr. The reaction was terminated by heating the mixture for 5 min at 100°C. After cooling, the mixture was centrifuged for 30 min at 32,000 x g. Dowex 50, H+ column chromatography. The supernatant fraction from the reaction mixture was applied to a Dowex 50, H+ column (400 ml, 100-200 mesh) (Sigma, St. Louis, MO) and eluted with water. Fractions (10-15 ml) were collected and analyzed for total and inorganic phosphate by the Fiske-SubbaRow method (1925) as well as for GalN-l-P by the procedure described above. These analyses showed that the product had been separa- ted from most of the GalN-l-P-containing reactants and by-products. However, yellow coloration of the GalN-l-P-containing fractions indicated that they contained PEP, so a second column was run. This time the pool- ed fractions from the first column were lyophilized and redissolved in 50 ml of water. The second column separated the yellow component from the GalN-l-P-containing fractions, which were pooled and lyophilyzed. The residue was dissolved in 20 ml H20 and the product was precipitated out of solution by the addition of 20 ml 95% ethanol. Crystals were allowed to grow for 3-4 days and the supernatant fraction was decanted. The supernatant fraction was allowed to stand for an additional 7 days and the second crop of crystals were collected. The crystals were dried to a constant weight over P205.1n.vagug. A 13C-NMR spectrum of this product was obtained, after which the product was chromatographed on 250 micron cellulose TLC plates (Analabs, Inc., North Haven, CT) developed with ethanol: 1 M ammonium acetate (pH 7.2)/7.5:3 and with the same 40 solvent at pH 3.8. Chromatograms were visualized by spraying with a solution containing 5 ml 60% perchloric acid, 25 ml 4% ammonium molyb- date, 10 ml 1 N HCl, and 60 ml H20. After spraying, plates were dried in air and exposed to short wave-length UV light with a hand-held Mineralight (Ultraviolet Products, Inc., San Gabriel, CA). Modified Fiske-Subbarow'method for phosphate determination. For deter- mination of total phosphate, samples containing 0.1-1.0 umoles of phos- phate were treated with an equal volume of 2 N HCl and placed in a boil- ing water bath for 7 min. These were cooled and neutralized with an equal volume of 2 N NaOH (ml HCl/ml NaOH . 1). All samples were then diluted to 3 ml. Acid molybdate reagent was prepared by adding 27.2 ml of H2504 to 70 ml of H20. Ammonium molybdate (5 g) was dissolved in 100 ml H20 and was added to the H2504 and H20 was added to give a final volune of 200 ml. The reducing reagent was prepared by dissolving 29 g of NaHSO3 and 1 g of Na2503 in 200 ml of H20. Amino-Z-naphthol-4-sulfuric acid (0.5 9) (Kodak, Rochester, NY) was added to the sulfite solution and the mixture was shaken vigorously for 10-15 min. This solution was stored in a dark bottle at 4°. Assay procedure. For each analysis a water blank, unhydrolyzed samples, and phosphate standards (0.4-0.8 wmoles KH2P04) were prepared. The volumes of all tubes were adjusted to 3 ml with H20 and reagents were added in the following order: 1 ml molybdate reagent, 1 ml reducing reagent, and 5 ml H20. The tubes were vortexed and allowed to stand at room temperature for 20 min before reading the absorbance at 660 nm. 41 B. Acetylation of GalN-l-P. Crystalline GalN-l-P (1.5 g) was dissolved in a solution containing 0.5 g NaHCO3, 15 ml methanol, and 100 ml water. This mixture was cool- ed on ice and 2.5 ml of acetic anhydride was added in 0.5 ml increments over 30 min. During this time the solution was maintained between pH 6.5 and 8.0 by adding solid NaHCO3. After 1 hr this solution was stirred with 250 ml of Dowex 50, HT, 200-400 mesh for 30 min and filtered. The resin was washed with water and this wash was combined with the filtrate. Residual acetic acid was removed by extracting the aqueous solution six times with 1 liter portions of diethyl ether. The product was converted to the potassiun salt by passage over a 250 ml Dowex 50, H+ column. The colunn was washed with 100 ml water and the combined colunn effluents were adjusted to pH 10.0 with 1 N KOH. This solution was dried to a light-brown syrup in .!3222. The syrup was dissolved in a small amount (5-10 ml) of methanol and an ethanol-acetone (1:1) mixture was added until a faint turbidity persisted. This addition was repeated several times over 2-3 weeks. Crystals were harvested by centrifugation and washed twice with 95% ethanol, twice with absolute ethanol, and once with diethyl ether. The crystals were then dried in vagug over P205. This product was analyzed by cellulose TLC as described above and 13C-NMR. These analyses indicated that the acetylation was complete and the product was free of contamination. C. Preparation of UDP-GalNAc. Preparation of UDP-GalNAc was carried out according to the procedure of Moffatt for sugar nucleotide synthesis (Moffatt, 1966). This proce- dure involved the coupling of the nucleotide, uridine-5'-monophosphate, 42 (UMP), with the 1-phospho-sugar. The coupling reagent dicyclohexylcar- bodiimide (OCC) was used to prepare the morpholidate derivative of UMP. Trioctylamine salt of GalNAc-l-P. 1 mmole (300 mg) of GalNAc-I-P was converted to the trioctylamine salt by passage, in aqueous solution, over a Dowex 50, H+ column (1 x 50 cm), eluting directly into 20 ml of pyri- dine, followed by the addition of 1 mmole of trioctylamine. The column effluent mixture was dried to a syrup in vagug_and dry pyridine was added and evaporated similarly. Pyridine was added and evaporated again, and this was repeated a total of 5 times, followed by drying the sample in 32239 over CaH overnight. UMP-morpholidate preparation. A solution of DCC (4.5 g in 75 ml) was added dropwise to a refluxing solution containing 2 g of UMP (acid form, from a Dowex 50, H+ column), 54 ml H20, 54 ml t-butyl alcohol, and 1.84 ml morpholine. After a few hours of refluxing the mixture was exam- ined by paper electrophoresis on Whatman #3 chromatography paper with a triethylamine bicarbonate buffer (14 ml triethylamine/2 l water, adjusted to pH 7.2 with C02). The paper was dipped in buffer and placed in an electrophoresis apparatus (Savant flat plate electrophoresis, Savant Instruments, Inc., Hicksville, NY). Excess buffer was blotted off and samples were applied 3 cm apart on a line 1/3 of the total length from the (-) side reservoir. Picric acid spots were applied to each chromato- gram as markers. The electrophoresis was run for 1 1/2 hr at 2000 V. Papers were dried and spots visualized by UV light (for nucleotides) or with the phosphate spray reagent described earlier (for phosphorylated sugars). The initial sampling showed that the reaction was only 20-30% complete so additional amounts of DCC and morpholine were added. 43 Refluxing for an additional 2 hr gave nearly total conversion to the nucleoside phOSphomorpholidate. The mixture was cooled to room tempera- ture, filtered, and the filtrate was washed with t-butyl alcohol followed by water. The filtrate was evaporated in vagug_to about half the origin- al volume and extracted twice with diethyl ether. The aqueous solution was evaporated to dryness in 32239 and the residue was dissolved in a small volume of methanol and transferred to a 40 ml centrifuge tube where it was dried to 3-4 ml. Dry diethyl ether was added to give a gum which, upon addition of fresh ether, gave a white powder. This product was dried in vagug_with repeated addition of dry pyridine and placed over CaH in M overnight. Coupling of UMP-morpholidate with GalNAc-l-P-trioctylamine. The morphol- idate, in 20 ml dry pyridine, was transferred to the trioctylamine-Gal- NAc-1-P flask in a dry box with a N2 atmosphere and P205 dessicant. The reaction mixture was stirred for 3-4 hr and allowed to stand. The reaction was followed by paper electrophoresis as described above. After five days at room temperature the reaction mixture was applied to a Dowex 1, Cl' column (1 x 20 cm) and eluted with a 0.01 to 0.1 M gradient of LiCl in 0.003 M HCl (1 ml/4 l). The eluate was monitored with an ISCO UA-5 UV monitor at 254 nm (Instrumentation Specialties Co., Lincoln, N8) and 20 ml fractions were collected. The fractions Which contained the product were pooled and evaporated to a syrup in gaggg_at 30°C. This product was tested for purity by paper electrophoresis as described previously. The product was desalted by passing over a Bio-Gel P-Z column (Bio-Rad Laboratories, Richmond, CA) (2 cm x 1 m, ZOO-400 mesh). The sample was applied as a 20 ml aqueous solution and eluted with H20. The eluent was monitored with the UV monitor as before and 10 ml 44 fractions were collected. The fractions containing the product were pooled, lyophilized, and the residue redissolved in 2 ml H20. A drop of this solution was tested for Cl' by adding a drop of AgNO3 (1% in 1 N HN03). A white precipitate indicated Cl“ still remained in the sample. It was reapplied to the P-2 column and eluted as before except the flow rate was decreased from 20 ml/hr to 5-10 ml/hr. Only the last two UDP-GalNAc-containing fractions showed traces of Cl'. Those frac- tions which did not contian Cl' were pooled and dried to a clear brown syrup in vaggg at 30°C. The product was dried to a constant weight over P205 and portions were dissolved in H20 and stored at -20°C for analysis or use in GalNAc transferase assays. Synthetic products were analyzed by 13C-NMR by Dr. H. Nunez and Dr. J. O'Connor. Shown for these analysis are 15.01 MHz proton-decoupled spectra obtained with a NIT-360 spectrometer at 35°C using a 0.2 M 020 solution in a 5 mm tube. Transients (3000) were accumulated at a sweep width of 3000 Hz at 0.733 Hz/computer point and 75° pulse. Chemical shifts are expressed in ppm downfield from tetramethylsilane. II. Biological Activity of Synthetic UDP-GalNAc. The biological activity of the UDP-GalNAc prepared as described above was tested by using it as a sugar nucleotide donor for UDP-GalNAc: globotriaOSylceramide GalNAc transferase in homogenates of embryonic chicken brain (ECB) and normal and virally transformed hamster fibro- blasts (NIL-8, NIL-8HSV). 45 A. Embryonic chicken brains were obtianed from 15 day old fertilized chicken eggs. Brains were homogenized in a ground glass homogenizer in an ice bath (15 strokes) in 2 voltmes of 0.25 M sucrose containing 0.05% ethylenediamine-tetraacetic acid (EDTA) and 1% dithiothreitol (DTT) and frozen in 0.5 ml aliquots. A crude microsomal preparation was obtained by centrifuging the homogenate at 110,000 x g for 1 hr (Bechnan L3-50 Ultracentrifuge, SW 50.1 rotor, Beckman Instrunents, Palo Alto, CA) and rehomogenizing the light part of the pellet with 2 volumes of 0.25 M sucrose as before. 8. Cell Cultures. NIL-8 cells. A NIL-8 hamster fibroblast line and a hamster sarcoma virus (HSV) transformant of this line were obtained from Dr. Phillips Robbins at the Center for Cancer Research, Massachussetts Institute of Technology, Cambridge, MA. The cells were grown in monolayer on Falcon (Falcon, Oxnard, CA) or Corning (Corning Glass Works, Corning, NY) tissue culture flasks or in glass 950 cm2 roller bottles (Bellco Glass Inc., Vineland, NY). The medium used for these cells was Dulbecco's Modified Eagle Medium supplemented with 5% fetal calf serun (DMEM + 5% FCS) (Gibco Laboratories, Grand Island, NY). Cells were seeded into flasks at ap- proximately 104 cells/cm2 and placed in a hunidified incubator at 37°C with an atmosphere of 5% C02. Cells reached confluency within 3-4 days. Roller bottles were seeded with one near-confluent 75 cm2 flask per bottle and bottles were gassed for 30 seconds with air containing 5% C02. Tightly capped bottles were placed on a Bellco roller bottle ap- paratus and rotated in a 37°C room at 0.25-0.5 rpm. 46 Cells were routinely subcultured as follows: Media was removed from flasks by pouring or aspiration and a solution of trypsin (1:250, GIBCO), 0.5 g/l, containing 0.02% EDTA was added. After rinsing the cell layer, the trypsin was removed and cells were washed a second time with trypsin. This wash was removed and cells were allowed to incubate 5-10 min at 37°C. Trypsinization was stopped by the addition of fresh serum-contain- ing media and cells were suspended by repeated trituration. Aliquots of this suspension were diluted appropriately and used to seed new cul- tures. Flasks containing cells to be used for transferase assays were wash- ed twice with warm phosphate buffered saline (PBS) (0.01 M phosphate buffer, pH 7.2, 0.85% NaCl), followed by the addition of PBS containing 0.05% EDTA. Flasks were allowed to incubate with PBS-EDTA for 10 min at 37°C followed by vigorous shaking to remove the cells. Cells were pel- leted by centrifugation (International Equipnent Co. PR-6000, Needham Hts., MA) at 1000 rpn for 7 min. The pellet was washed with PBS, and cells were frozen in one volune of 0.32 M sucrose. KB, CHO, and BALB[c 3T3 cells. Procedures for other cell lines were essentially the same as those used for NIL-8 cells except different growth media were used. KB cells were cultured in Minimum Essential Medium with Earle‘s salts supplemented with 10% calf serun (MEM + 10% CS). Chinese hamster ovary (CHO) cells and a wheat germ agglutinin (WGA)-resistant mutant (CHOWGA) were gifts from Dr. Pamela Stanley (University of Toronto). They were cultured in MEM Alpha Median (a-MEM) with 10% CS. Mouse BALB/c 3T3 cells and their Kirsten murine sarcoma virus transformants (B/3T3-Ki) were obtained from Dr. Sen Hakomori University of Washington, Seattle, WA), and were grown in DMEM + 10% 47 FCS. All media were purchased as powders (GIBCO) and sterilized by fil- tration through 0.22 H membrane filters (Millipore Corp., Bedford, MA) and contained 100 units/ml penicillin and 100 ug/ml streptomycin. Bovine serun was obtained from 61800 or from Kansas City Biological, Inc. (Lenexa, KS). Cultures were routinely tested for mycoplasma contamina- tion and lines were discarded after 30 passages in our laboratory. C. Glycolipid GalNAc Transferase Assay System. The initital assay system used for UDP-GalNAc:globotriaosylceramide Nfacetylgalactosaminyltransferase activity (GalNAc transferase) was that described by Yeung gt 31. (1974), and included the following components: in a final volune of 0.05 ml, (in micromoles) acceptor lipid, 0.05; sodium taurocholate (TC) (Grade A, Cal Biochem, San Diego, CA), 125 ug; 2(N-morpholino)ethane sulfonic acid (MES) (Sigma), pH 6.38, 10; MnClz, 0.5; UDP-[14CJ-GalNAc (2.5 x 105 cpm/umol) (New England Nuclear, UDP-GalNAc [GalN-1-14C] so mCi/nlnol, specific activity reduced with synthetic, unlabeled UDP-GalNAc), 0.02; and 0.2-0.5 mg protein. This assay mixture was later modified by using 50 pg of sodium taurodeoxycho- late (TDC) (Cal Biochem) instead of TC, and MES buffer at pH 6.0. ECB homogenates were prepared as described previously. Cultured cells (5 x 105) were homogenized using a Dounce homogenizer with the B pestle (30 strokes). Microscopic analysis of this preparation showed only a few intact cells remaining. Protein determinations were done by the method of Lowry (1951). Assay mixtures were prepared by drying the acceptor lipid (in chloroform:methanol (C:M)l 2:1) and the detergent (in theoretical upper phase, C:M:W/3:48:47) in 6 x 50 mm disposable glass culture tubes 48 (Kimble). A pre-mix was then added which contained buffer, MnClz, and UDP-GalNAc. If one of these components was being varied it was added to the assay tubes separately. Finally, the homogenate was added and the tubes were thoroughly vortexed. Samples were incubated for 1 hr (unless specified) at 37°C and the reactions were terminated by adding 0.01 ml of 0.25 M EDTA, pH 7.1. Double chromatographic analysis of products. Samples were applied in 25 mm lanes 26 cm from the top of a 46 x 57 cm sheet of Whatman #3 chromato- grapy paper. After the aqueous samples were applied, tubes were rinsed twice with 0.05 ml of C:M/2:1 and these rinses were applied to the same spots. Papers were developed (descending) with 1% NaB407 overnight. After drying, the chromatograms were cut off 3-4 cm below the origin and the origin-containing sections were ascendingly developed with either C:M:W/60:35:8 (for neutral glycolipid synthesis) or propanol:water/7:3 (for ganglioside synthesis). Each lane of these chromatograms was cut into 1" sections and the sections were counted in scintillation cocktail (4 g PPO, 0.15 g POPOP, 1 liter toluene, Beckman LS-100 Liquid Scintillation Counter, Fullerton, CA as shown in Fig. 6). Counts of all sections (excluding the origin) for each lane were totaled. The counts for an endogenous acceptor control (no lipid acceptor added) were sub- tracted to give a net cpm for each assay condition. These counts could then be converted to nmoles from the known specific activity of the UDP-GalNAc. 0. Product Analysis. Extraction of Chromatography Papers. Papers from several assays were collected after individual counting and extracted three times with 49 .m «Fame c. exogm ma ummmmooga men mucaoo ucvupzmmg map can emacaoo ems» m, mmompq omega we seem .mmomva cue? use one. use m. oca— some ecu czosm me «so uzo use mmcap .Aznzuov mummumm c? xgnmgmopmsogzu mcpucouma ace oumeonaeuou guy: xgamemoumsoegu mcvvcwummc Lmuw< .maumm< omagmmmcugu—xmoua—w toe oaavccumh urgaesmouasoegu some; .w mezmwm 50 #2300 A”— .ao uLv it; ----4 7---- pao 51 toluene to remove fluorophores. After drying, the papers were extracted twice with C:M:W/100:50:10. This extract was dried on a Rotevap and redissolved in C:M/2:1. The large amount of insoluble material was partially removed by passage through glass wool. The resulting material (about 25,000 cpm) was passed over a silicic acid column (10 g) as described by Vance and Sweeley (1967). The neutral glycolipid fraction (acetone:methanol/9:1) contained very low counts. The following elution (methanol) contained most of the counts applied to the column. This was probably due to incomplete washing of the column with acetone:methanol. The methanol eluent was dried and redissolved in C:M/2:1. An aliquot of this solution was counted and found to contain 7250 cpm compared to 1000 cpn for the acetone-methanol fraction. The methanol elution product was chromatographed on high performance thin layer chromatography (HP-TLC) plates (Silica Gel 60, EM Reagents, Darmstadt, Germany) in C:M:W/65:35:8 and on cellulose TLC plates (Analabs, Inc., North Haven, CT) in butanolzacetic acid:water/3:3:2. Plates were scanned for radioactivity using a Berthold Radio Scanner (Varian Aerograph, Walnut Creek, CA) and radioactive peaks were compared to standard globoside and free sugars. Large scale assay. Since the recovery of the product from the assay papers was poor, a large scale assay was run in which the specific activity of the UDP-GalNAc was increased three times and the quantities of all the assay components were increased five times. After 2 hr at 37°C, 2 ml of C:M/2:1 and 0.1 ml of H20 were added to give a Folch partition (Folch £3 11., 1957). After vigorous mixing, the lower phase was removed and the upper phase was extracted twice with theoretical lower phase (C:M:W/3:48:47). The combined lower phase washes were dried. About 20,000 cpn were obtained. Duplicate aliquots (1500 cpm) were 52 spotted on HP-TLC plates and developed with C:M:W/65:35:8. The sample-containing lanes were scanned for radioactivity and showed only one radioactive spot, corresponding to the standard GL-4 migration. These spots were scraped and eluted from the gel with C:M:W/100:50:10. After drying, the duplicate samples were treated with jack bean a-glucos- aminidase (EC 3.2.1.30, Sigma, St. Louis, MD) as described below. After hexosaminidase treatment the samples were partitioned as described above. The lower phase extracts were dried, spotted on HP-TLC plates in a small volume of C:M/2:1, and developed as before. Plates were again scanned for radioactivity. e-Hexosaminidase Treatment. To the duplicate dried extracts from the TLC plate described above were added .075 ml of 0.1 M sodium citrate buffer, pH 5.0, and 0.025 ml of Jack Bean B-glucosaminidase. One of the samples was placed in a boiling water bath for 3 min. Both samples were incubat- ed overnight at 37°C partitioned and chromatographed as described above. III. Analysis of Cell Growth. A. DNA Pulse-label. To measure DNA synthesis in cell cultures the incorporation of 14C-thymidine into acid precipitable material was determined during short (0.5-1 hr) pulses. Cultures were labeled by adding 0.25 uCi of 14C-thymidine (New England Nuclear, Boston, MA) in 0.5 ml of serum- free MEM to each 25 cm2 flask or 25 cm2 petri dish. These cultures were incubated for 60 min at 37°C. Labeling was stopped by aspirating the media and washing the monolayers twice with ice-cold PBS, followed by incubation for 20 min at 4°C with a 5% trichloroacetic acid (TCA) 53 solution. After this incubation monolayers were washed with cold TCA and scraped into TCA with a rubber policeman. Cells were pelleted by centri- fugation at 1500 rpm at 4°C. The pellets were washed once with cold PBS and solubilized overnight in 0.1 N NaOH. Aliquots were assayed for pro- tein (Lowry gt_al,, 1951) and for radioactivity (scintillation fluid: 1 liter toluene, 100 ml Biosolv, 7 g PPO and 0.6 g POPOP) using a Beckman LS-150 liquid scintillation counter. B. Synchronization of K8 cells. KB cell cultures (monolayers) were synchronized by the double thymi- dine block method (5 phase) or a single thymidine block followed by mito- tic selection (M phase). Double thymidine block. Sparse toimedium density cultures were blocked at S phase by the addition of thymidine to the culture medium (final con- centration of 2 mM). After 20 hr, cultures were washed and fresh media added for 12 hr. The addition of thymidine was repeated, and after 20 hr cells were released from the second thymidine block to give a synchronous population of growing cells at the Gl/S interface. Mitotic selection. To obtain well-synchronized M phase cultures of K8 cells, cells were single-thymidine-blocked as described above. About 10 hr after release from the thymidine block, selection of mitotic cells by gentle shaking was begun. Shaking was repeated every 15 min for 2 hr. Upon collection, mitotic cells were centrifuged at 1000 rpm and stored at 4°C until collection was completed (not more than 3 hr). Mitotic cells were seeded at 20,000 cells/cm2 and allowed 1 hr for attachment to occur before any experimental treatments were performed. 54 C. Galactose Transport. Measurement of galactose transport in control and butyrate-treated KB cells was accomplished by pulsing cultures in 25 cm2 petri dishes with D-galactose-1-14C (Gal) (ICN Pharmaceuticals, Inc., Irvine, CA). Ten dishes were set up for each experimental point. At a given time 1-2 uCi of the labeled compound was added in serun-free MEM to all of the dishes and quickly mixed with the media (2 ml total vol une). The dishes were immediately placed in an incubator at 37°C and duplicate dishes were removed at 0, 3, 6, 9, and 12 min after addition. As soon as the dishes were removed from the incubator the labeled media was aspirated and the dishes were washed twice with cold PBS. Cells were scraped into PBS and centrifuged at 1500 rpn (IEC centrifuge, 4°C) for 10 min. Pellets were dissolved in 0.1 N NaOH and aliquots were counted in scintillation fluid containing Biosolv (7 g PPO, 0.6 g POPOP, 100 ml Biosolv, 1 l toluene) and protein determinations were made by the method of Lowry 35 31. (1951) (see thymidine pulse-label section). RESULTS I. Synthesis of Uridine Diphosphate NrAcetyl-a:Q-Galactosamine. The synthesis of UDPfiflfacetylgalactosanine (UDP-GalNAc) was accom- plished by a combination of enzymatic and chemical techniques. The star- ting material for this sequence of reactions was GalN—HCl, which was con- verted to a-D-GalN-l-P enzymatically. This intermediate was N-acetylated with acetic anhydride and the resulting compound was converted to the tri-n-octylamine salt for coupling with the UMP-nucleoside phosphor- amidate, which occurs directly as shown in Figure 7. The use of galacto- kinase for synthesis of GalN-l-P gives only the a anomer, rather than a mixture of a and a anomers (Carlson gt 31., 1964). Thus, while the total yield of GalN-l-P is somewhat less than that possible by direct chemical phosphorylation (O'Brien, 1964), the yield of usable (a) product by the enzyme synthesis is relatively good. The yields obtained at each step of the reaction and the methods used for identification are listed in Table 1. A. a-D-Galactosamine-l-Phosphoric Acid. The only modifications to the procedure described by Carlson £3 21. (1964) were: 1) the pH of the galactokinase assay mixture was adjusted before addition of the yeast extract, and 2) the quantities listed for the large-scale GalN-I-P synthesis were doubled. The pH of the assay buffer was lowered by the addition of the cofactors (ATP, PEP, PGA), making adjustment of the pH before enzyme addition necessary. 55 56 .uusuoea :owuumms peeve mgu m>pm ow mamuppognsosamz: gu—z capaaoo can» m. Hostage mpg» .mcwsuzgca dramas maomzca saw: umumpzuoum m_ mnpuzpaw mcpapamms men can appeuwaus>~em umumpxyocnmosa m? o:?5emoaoopaw .ocvsamouuupampaumu .3. 3:: 94230.83: A: mi.io<2_oo Sin: mi_no<2_oo any mu.iz_oo 3.9 m:...z_oo A8 .o:-z_u@ 32255 .2695 33255 .3332 9535 333432.33? w a 59 The initial Dowex 50, H+ column failed to adequately separate the GalN-l-P from the reactants and reaction by-products, so a second colunn was run on the pooled GalN-I-P-containing fractions. The pooled GalN-l-P fractions from the first column were lyophilized and redissolved in 5 ml H20 to give a smaller application volune. Two phosphate peaks were eluted from the second column (fractions 2-24 and 28-46) (Figure 8). The second peak contained the GalN-l-P. The pooled GalN-I-P-containing frac- tions were lyophilized and redissolved in water. A 13C-NMR spectrum of this product is shown in Figure 9. This spectrum indicated that the desired product was obtained and was essentially pure. Additional proof of the product purity was demonstrated by cellulose TLC using ethanol/1 M ammoniun acetate, 7.5:3 (Figure 10). B. Acetylation. The acetylation of GalN-l-P was carried out as des- cribed by Carlson 35 31. (1964) with quantities scaled up five times. About 30-35% of the product was obtained in the first two harvests of crystals. The yield was lower than that obtained by Carlson, probably because a shortened precipitation time was used (1 week vs 3 weeks) and some product was lost during filtration. A 13C-NMR spectrun (Figure 11) demonstrated that the acetylated product was indeed obtained, and cellulose-TLC chromatography showed that the product was free of non-acetylated reactant (Figure 10). C. UDP-N-Acetylgalactosamine Synthesis. The coupling of the dicyclohexylcarbodiimide salt of GalNAc-l-P with the UMP-morpholidate was carried out with great precautions against get- ting moisture in the reaction mixture. Presence of water during the 60 983 2232... .313 “772:5 one AOIOV ouogomogo 33 so“? carapace 638:8 mew: mcozuot E 2.2 new 55:... Nap an oouzpo was 5:38 o+zlxcm xgoo EC}... QIFIwaEmmouuapmw $0 wpvnvea cosh—b .m 0.257. mmmSDZ zo_._.o2 sup: ooNVPoamr> ego: o oooN be c.5 om ace :3; mo: mvmogogooguoopu .goooo xzoogmouosoggo m .oz ensues: co meauxvs :owuuoos m:r_o:oo q. we» uo mommcogooguoo—m goueo omcropoo Eoeuvaogogoosuoopo co mucomogoos Eogaopo a_sh .ogauxpz :o.uooom mapposou to mvmogozoogpompm swoon .Np ogamwd 70 0.04 UEUE O wm:...x.2 me O anjoxdmos. - ass 0 0.04 QED—a O 1T 2.03.0 71 new pusooeo poo?» mg» yo oesuoacum .oeo; czozm use mucoscmvmmo goon mzz .m ogsmru so; ownesomoo mo ooocoomg we: Eoguooom .o:.Eomouuopompzpoo «a ooxommo mew: apnea :oxopgo ovcozeoso mo monocomoso; moose .coeuosuooocoo cwouogo eo uooweu .omogoemcegupzc.somouoe_ompxuoopuoo owmvooom o .>u_>¢ao< amusemmcagh om:wu.=z .010 35:: .01. .moezuxvs zommm one o» cocoa mew: mopocomoeo; —poo mooeo mo mu==oEm mzovgo> .covuoom moosumz new m—o—eopoz oz» op ooovgummo mcovuwocoo zommo on» one eouowuoo mugw anew: .ae.>_ee< ameeaemeaee emxo-st eee o-e~z eo eeeeeeeeaeeoo eeaeaea ea eoaeem .o_ aeemea 87 38833 20.23.2328 255%. oom ooa oom ooN oo. o o l u . \ o l O 0 lo l o oo. 0 O ooN a... m / N 8» 00¢ 88 .oopgo> we: opuoo uewvumom acoumcou .au*>vuo< omosmemcoeh dFuowomwe .egxms\pos: mm~.o wow :5 emp.o mg on ocaoe mew: mmope> on> :o ooo.e o x ooguws are» ween: .mpmooeovome covuoeucwocoo wuoeumnam use zupoopw> we» eo mpmzpoco mweozom umow— eow=VF owucmpm: w an owcwsemawo mew: mw=Fo> on> new oom.sx .mmuocwaoao: muwmz :H w>e=u covueeucwocou weesomouowPomquwoom3mcpm .om mesmee 92 "" 0 a a (on X) ,-(,-Ju-aiowd) A/l 0.2 0.3 0.4 0.5 US (UDP GalNAc, mM) OJ ‘_L‘ I. . 93 we goo.a .e2\oe:8e 83 u x > “:5 empac u x .om wezmvm cw ::o2m om—o wee mew; omega—o omen wee .om—poeucoo apwaoeaooo ma coo coppmeucwocoo wuoeumoam we» use eoeew op pownaam wee mucwsmeomows xuvoopw> ems: omeeweweo we use mw=~e> on> use 52 eo coeuosrumw peamr> ewuuwo o mw>wm coeoozow :opcmzamepwosovz we» eo seoe mesh .mwuocwmosoz mnmuz =H w>e=u =o_pmeu=wo2ou mewsomouuopompauwowuu< wmmewwmcmeh o~m>_uomomme .e2\me\mposc umm.o mac :5 mfie.o mew: mime eom x oom.s mmsFe> eE> use 2 .om wezmwe eoe mo omNAPmco mew: emoo m>ezo cowmeeucmocou .mmuecwmoeo: m-mez :H m>ezu cowumeucmocou mimoummmememcoep ovuo< wmmememcmeh eeo ee moweem .eN aeemee 102 F \, l N .1 a e 9. ° (30: X) Ju/fiw/salowu 20 12 16 (mM) M2+ 103 .Ameeo m mcpmo omeomoms mo: auv>vpoo mmoewwmcoeu oH.>Fuo< mmoememcoeh o mumpozooeooh Eamoom an wmmewemcmeh opuo< .oN meomee 106 23:: 2: ezmommemo 00m 09 00. on I d d M ON 0? 00 '_Jq ..,61u. selouau (am we 107 .835. <14 a E. :1. “wee 3088. 010 52.010 858. :5 2.8.010 .Aemeeon zoo.o mo 20mm P: mpv mmnou momma op apwumewomm owooo mew: mooruopom eweezm .mwmcoe mopeweeoa m>_uomomwe eewzu cvzup: mwopo> 2o mooreo> mo o2 woos mew: memwwoo owuoopo:_ .mmoememcmeh o.mfl mmmczeo on cmxmu ocm mama :2 p ;u_: meopo mwooa sommo o» omooo wem: mopmovpmcom mzw oco .oima .Nimw eo mucoosm owuoovo=_ .meeeaaoe< eeae_eee_o Peeeeaeee 2: emaewemeeee e mo: covpmeucwocoo mugs .aoeaeepmeao mzo e2 ameeaeaeaee ewu0< ammomemcmafi. U_uomomme .e2\mE\mwpos: mm.m one :5 mo~.o mew: om=_muoo mmopm> xms> ocm oom.sx m2» .mmumcwmoaoz —pmu mhm cm m>e=u coeuoeacwocou miduuwmmememcoeh ovuowomme Q a .ecxme\mmposc emm.o ocm :2 mm~.o mew: omcvmmoo mm2Fo> xos> ozo o o 22 mop .maeeeaoeeaz >m2:x-mem =0 m>e2u =o_umeucmo:ou mzanmmoememcoe» om2§inhn 0.0 A (Pm. selowu) 135 same enzyme is acting on both acceptors, the fact that one activity is much higher in normal cells while the other activity is higher in the transformants suggests that separately regulated enzymes are involved. GalNAc Transferase Activity In CHO and CHO Cell . Since GalNAc WG transferase activity is greatly affected by cell transformation it was of interest to see if similar changes might occur in non-Virally transformed cells after cell surface alterations. For these experiments, CHO cells selected for resistance to wheat germ agglutinin (CHOWGA) (Stanley and Carver, 1977) were used for GalNAc transferase assays. As shown in Table 5, GL-3 was the most active acceptor for the GalNAc transferase in both wild type and mutant cells. Some activity was also seen with GL-Z and GM3 acceptors. A problem with measuring transferase activities in these cells was that, at lower protein concentrations (less than 250 ug/assay), activity was very low. At 150 ug/assay, for example, the activity was increased four-fold by doubling the protein concentration. This was possibly due to instability of the GalNAc transferase at low protein concentrations. For this reason, care was taken to run all com- parative CHO or CHOWGA GalNAc transferase studies at the same pro- tein concentrations, which were, where possible, greater than 250 ug/assay. V. Effects of Sodium Butyrate on Cell Cycling of K8 cells. The initial observation by Fishman and coworkers (Simmons gt 21., 1975; Fishman gt 31., 1974; Henneberry and Fishman, 1976) that the addi- tion of butyrate to HeLa cell cultures caused a 15-20 fold induction of a 136 8.8.8 «o: . oz 8.2:. 2.2.8 Q .325... ._..E._-m=_.mm.o=a c a of oz 3 a. eN. N2o-e_e_a e: 8. NNe Re Nam 8v Nam ”we Io Nam on 1 a. .88... a eozo ozo 3:23 1 no: N 22o N _ N3 o m....u :2: .oz 5.8:. £8 ozo E... ozo e. 2858.; sawed. @4282 8:»: .Eomoeoo.om.ibwo;h xm uo~wcosgu=>m mppou mx ca mpmogucxm eg mppoe we new seem new: mxmepw Pecueee one use ppe ea eeeee some em; euewxuee wonky xeepe ecwewsxsu eeeeom ecu seew eemee—ec ace: appee we» can: .xee—e eewewexgu eceeem on» Eeww emeepee ocewee moswa mzewce> um mxmepw we meow ea eeeee we: A25 ev eueczuem .eexeepe eewewsxguuepeaee ewe: mp—ee mg .mppee mg eexeepm-ecwewewgw ce memexezm we eeewwm .ee ecsewe 144 4.0 7 l0 0.5 150 L O Q In ugetmd b‘ri/uado 00 6.0 1.5 2.0 25 3.0 3.5 Time after release (hrs) 145 + ++ + 33.9. cote 25... 1.11...»on 9. 325233 e. 3034;... 30112231111132 eeeeee Togo. exam 0 w...m<._. m. V O. o :5 22:38:93. .e 583 146 Figure 41. Transport Of Galactose In Butyrate-Treated KB Cells. KB cells were treated for 24 hr with 4 mM butyrate. At T=0 1-2 uCi of [14CJ-galactose was added to duplicate 25 cm2 petri dishes containing 2 ml of media. At the indicated times after addition of the label, cells were washed with ice-cold PBS, scraped from the dish, and counted as for thymidine pulse-label experiments. g—o, Control; 0— o, Butyrate. 147 60 45 TIME (MIN) 30 15 250 * 200 - 50 - / ,_...O--’ 150 - 100 " 5295 03:30 148 or not this phenomenon is related to other changes induced by butyrate is not clear. DISCUSSION Sugar Nucleotide Substrates. The use of sugar nucleotides as substrates for glycosyltransferase reactions is required for enzymatic activity in in 11352 assay systems. In this form the sugar is activated and can be easily transferred to an appropriate acceptor. Part of the problem with using sugar nucleotides in glycosyltransferase assays, however has been the lhnited availability of these compounds in unlabeled form. While several unlabeled sugar nucleotides are commercially available, UDP1Nracetylgalactosamine (UDP-GalNAc) is not. Only high specific acti- vity UDP-GalNAc (approximately 50 uCi/mmole) is commercially available and the use of this preparation at substrate-saturating levels (0.4 mM) would be impossible. Therefore, many studies have been carried out using extremely low concentrations of UDP-GalNAc. Certainly kinetic character- ization of UDP—GalNAc transferase activity is out of the question under these conditions, and significant problems with non-kinetically oriented studies can also be encountered. For instance, in comparing glycosyl- transferase activities from several cell lines (or a single cell line with several different drug treatments), differing levels of pyrophos- phatase activities could easily cause alterations in the observed trans- ferase activities by hydrolyzing a portion of the sugar nucleotide sub- strate. At higher sugar nucleotide concentrations, i.e., above saturat- ing levels, such hydrolysis would not have as great an impact on trans- ferase activities. Other problems might include substrate competition of low, sub-saturating concentrations of sugar nucleotide substrates with other compounds, such as nucleoside triphOSphates or sugar phosphates. 149 150 Thus, it is clear that unlabeled sugar nucleotide substrates must be available for reliable studies of glycosyltransferases. There are two primary methods for the synthesis of sugar nucleo- tides. One involves the direct chemical phosphorylation of the sugar with ortho-phosphoric acid (O'Brien, 1964), while the other utilizes the enzymatic addition of phosphate to the free sugar (Carlson ££.El- 1964). The advantage of the latter method is that only the useable (a) anomeric form of the sugar-l-phosphate is obtained, while both anomers, in nearly equal proportions, are obtained by the chemical phosphorylation. The coupling of the sugar-l-phosphate to the nucleotide can also be done by both chemical (Roseman £3 31., 1961) and enzymatic (Maley, 1970) methods. In this case the chemical method is more straight forward and may provide a better yield. The only problems encountered in the preparation of UDP-GalNAc, as described in the Materials and Methods section, were in the recrystallization and washing steps of the final product. A rela- tively large loss of product occurred at this stage due to lack of come plete recrystallization during the washes. The use of gel filtration (BioGel P-2) column chromatography is better for the removal of reactants and reaction by-products than recrystallization because of this problem. N-Acetylgalactosaminyltransferase Assay. To determine the effectiveness of the synthetic UDP-GalNAc as a glycosyltransferase substrate, IS-day- old embryonic chicken brain was used as an enzyme source, since this has been shown by Basu and coworkers (Chien gt 31., 1973) to have relatively high glycosyltransferase activities. Time course and protein concentra- tion curves were similar using either crude homogenates or high speed (100,000 x g) membrane preparations, so crude homogenates were used for 151 the remainder of the assays due to the relatively large amounts of mater- ial required for membrane preparations. Activities in NIL-8 cell homogenates were significantly lower than in embryonic chicken brain, but the results show that the assay is con- sistent and reliable, even at relatively low incorporated counts (SO-1000 cpm/assay). It should be mentioned here that higher counts could be ob- tained simply by increasing the specific activity of the UDP-GalNAc or even by counting the samples at a wider window setting on the scintilla- tion counter, but the results obtained under the present conditions pro- vided reproducible results in all the systems tested. Throughout these assays no endogenous acceptor activity was observed, suggesting that only transfer to the added acceptor was being measured. There was also a high substrate specificity under the assay conditions employed, as is demon- strated by NIL-8 cells, which only utilized GL-3 as an acceptor of the various substrates tested. Interestingly, in the large scale assay, labeled glycolipids were formed which had Rf values similar to GL—S and asialo-6M2, indicating some endogenous acceptor participation in these reactions. While no endogenous acceptor activity was detected using the paper chromatographic assay, the low level of counts incorpor- ated with the usual assay conditions may be below the level of detection. It would be interesting, therefore, to run a large-scale assay with no added glycolipid acceptor to determine the relative endogenous levels of synthesis of the glycolipids observed here. Subsequent studies with other cell lines, particularly CHO cells, demonstrated occasional lack of consistency in reproducing specific acti- vity values. The reason for this is not clear, but may be the result of incomplete cell homogenization or variations in cell growth, harvest, or 152 stbrage conditions. In the cases where this was a problem, data points were compared only within the experhnent and representative results from those of several experiments were reported. It was also noted that, in CHO cells at low protein concentrations, a doubling of protein concentra- tion increased GalNAc transferase activities by four fold. Though the reason for this effect was not clear, it was found to occur only when the lower protein values were below 250 ug/assay. Fbr these experiments, then, all assays were run at identical protein values of 300 ug/assay or greater. While Vmax values varied, to some extent, with all the cell lines tested, Km values were quite consistent. These values, ,a P obtained for UDP-GleAc in NIL-8 cells and for several glycolipid accep- tors in NIL-8 and 3T3 cells, were similar to those reported for glycolip- id glycosyltransferase activities in other systems (Chien gt 31., 1973; Chandrabose and MacPherson, 1975). Comparison of NIL-8 GalNAc transfer- ase activity with that of NIL-BHSV cells demonstrated a significantly lowered synthesis of globoside in the transformed cells. This follows the general trend reported in the literature of reduced glycolipid com- plexity upon transformation (Gahnberg and Hakomori, 1975; Itaya gt 31., 1976). The BALB/c 3T3 cells and their Kirsten MSV transformants differed from the NIL-8 cells in that they demonstrated GalNAc transferase acti- vity with 6M3 ganglioside acceptor as well as with GL-3. As can be seen from the glycolipid biosynthetic pathways shown in Figure 3 and 4, the linkage involved in the synthesis of globoside (CL-3 acceptor) is GalNAcB1.-—e-3 Gal, while that involved in 6M2 ganglioside synthesis (6M3 or GL-Z acceptor) is GalNAcsl ——-4 Gal. The GalNAc transferase 153 activities utilizing 6M3 or GL-2 acceptors for synthesis of GMZ are presuned to be due to the same enzyme (Basu gill. , 1974). The data obtained for the 3T3 vs 3T3-KiMSV GalNAc transferase activities indicated that both GL-3 and GL-Z acceptor activities were significantly lower in transformed cells, while the GM3 acceptor activity was much higher in the transformed cells. While this suggests tht the GalNAc transferase activities using GM3 ganglioside and GL-Z as acceptors are different, it is possible that the low level of GL-2 acceptor activity in both nor- mal and transformed cells is due to the GalNAc31.-—a»3 Gal activity which normally utilizes GL-3 acceptor. Linkage studies could be used to ascer- tain if this were the case, but would require considerably more product. Alternatively, labeled GL-Z could be used as the acceptor and competition with GL-3 and GM3 could be measured. The results obtained in this study suggest that the high level of asialo 6M2 found in the 3T3-KiMSV is due to increased 5M2 synthe- sis followed by loss of sialic acid. Another possibility is that synthe- sis of more complex gangliosides is blocked by a reduced galactosyltrans- ferase activity leading to 6M1 synthesis. This, along with normal neuraminidase activity could cause increased asialo GMZ content. This is in agreement with previous studies showing reduction of 5M1 synthesis upon transformation of 3T3 cells by Kirsten-MSV (Fishman 33 .21., 1974a) using labeled glucosamine to measure jghvixg ganglioside syn- thesis, and the demonstration of the lack of asialo-6M2 synthesis from GL-Z (Kemp and Stoolmiller, 1976b). These pathways for asialo 6M2 synthesis are shown in Figure 42. The mixing experiments indi- cate that a soluble GalNAc transferase inhibitor is not present under these conditions. This lack of a demonstrable glycosyltransferase 154 Figure 42. Possible Pathways For Synthesis Of Asialo-GM2 In 3T3-KiMSV Cells. Elevated levels of asialo-6M2 could be the result of increased synthesis from GL-2 or due to increased synthesis of 6M3 followed by renoval of the sialic acid residue. A block in synthesis of 6M1 from 5M2: caused by viral transformation, could also cause an increase in asialo-GNZ synthesis by increasing GMZ concentrations. 155 Gem->4 Glam->1 Cer GL~2 \ \ \ \ \ \ asialo-G EM; \ u: Gglfil-“l Glc 131->1 Cer [ GalNAc 131-*460131-‘4610 131->1 Car 1 , _ a2 AcNeu ”(\b ._J GalNAc 131->4 G§|Bl~4Gchl->l Cer 6M2 aé AcNeu =-_-. 4 ”5" 1 Gal 131->4GolNAcBl->4Gzol 131->4Glc Bl ---I Cor 1 a2 AcNeu 3.4.4.1. 156 inhibitor in these cells is important, since a possible mechanism of regulation in cells having lowered activity of one of the GalNAc trans- ferases could be the inhibition of transferase activity by a soluble or membranebound inhibitor or by compartmentalization of enzymes and/or sub— strates. However, the data obtained here suggest two different possibil- ities: (1) that regulation occurs at the level of gg £219 synthesis or processing of the enzyme itself, or (2) that some modification of the enzyme has occurred (e.g., phosphorylation by a protein kinase) which causes alterations in transferase activity. Indeed, some evidence for the latter possibility has been found and will be discussed later. There have been many studies in recent years which strongly suggest that cell surface molecules regulate cell growth and metabolism. For instance, Whittenberg and Glaser (1977) showed that plasma membrane prep- arations from 3T3 cells inhibited DNA synthesis in non-confluent 3T3 cell cultures. Furthermore, the membrane preparations did not significantly affect the growth rate of SV40-transformed 3T3 cells, and membrane prep- arations from transformed cells did not alter the growth of normal cells. From these and earlier studies showing that succinylated concanavalin A (Con A) inhibited cell division by binding to cell surface receptors (Mannino and Burger, 1975), it was proposed that growth inhibition may resemble hormone action (Frazier and Glaser, 1979). The binding of specific receptors by the "hormone", which could be a surface molecule of 'an adjacent cell, would signal the cell to stop dividing. Similarly, Lingwood and Hakomari (1977) have demonstrated the inhibition of 3T3 and NIL cell growth by specific antibodies (Fab) to GM3 ganglioside while little effect was seen with tranformed counterparts (Kirsten MSV and 157 polyoma, respectively). Antibodies to globoside did not affect cell growth. Gahmberg and Hakomori (1975) have shown alterations in galactose oxidase-sodium borotritiide labeling of galactoprotein (LETS) and glyco- lipids after treatment with Ricinus conmunis lectin and Con A. After lectin treatment, glycolipids of normal NIL cells but not NIL-polyoma cells became more exposed, which may have been the result of glycoprotein clustering. Lingwood gt_gl. (1978) have shown that treatment of cells transformed with a temperature sensitive (ts) virus with antibodies (Fab) to gangliosides or galactoprotein inhibited expression of the transformed phenotype at the permissive temperature. This effect was possibly due to prevention of the reduction in cell surface GM3 upon transformation by antibody binding. These studies and those by several other groups (Natraj and Datta, 1978; Podolsky e_til_., 1978; Holley siel- , 1977, 1978) demonstrate the importance of cell surface glycolipids and glycoproteins in cell growth regulation and the transformation process. The dependence of cell surface structure on the cell cycle was demonstrated by Gahnberg and Hakomori (1974). They found that ceramide tri-, tetra-, and penta-saccharides were maximally labeled on NIL cells during G1 phase and minimally labeled during 5 phase using galactose oxidase-sodium borotritiide while the actual chemical amount of these lipids remained relatively constant. They also found a constant cell cycle labeling of these compounds in NIL-polyoma transformants. Similar studies have also been carried out by Fox 35 51. (1971) and Noonan and Burger (1973) using plant lectins. Chatterjee gt 31. (1975) demonstrated that glycolipid and glycoprotein synthesis were cell cycle specific in human epithelial (KB) cells, with maximum incorporation of 158 [14CJ-galactose into glycolipids in M/early G1 phase and into gly- coproteins in late S/early 62 phase. Delaat gt al. (1977) have shown that the membrane microviscosity of neuroblastoma cells changed during the cell cycle, with microviscosity decreasing during 61 phase, at a constant low level during 5 phase, increasing during G2 phase, and at a maximum during M phase. More recently, Basu gt 91. (1980) have shown a cell cycle dependence of a specific glycolipid glycosyltransferase (UDP-GalNAc:globotetraosylceramide‘Nracetylgalactosaminyltransfrease) during the cell cycle of synchronized guinea pig bone marrow cells. These results suggest that glycolipid composition and metabolism during the cell cycle may be important. Effects of Sodium Butyrate On KB Cell Cycling. Fishnan SEQ. (1974) denonstrated the specific induction of a glycolipid sialyltransferase during treatment of Hela cells with sodium butyrate. Since butyrate had also been shown to cause decreased cell growth (Ginsburg, gt 31., 1973) it seemed possible that the effects on the sialyltransferase may be related to growth regulation. The studies described in this thesis were designed to determine if butyrate was act- ing as a cell cycle blocking agent and if so, at what point during the cell cycle it was arresting cells. The results, along with those of Fallon and Cox (1979), demonstrated that butyrate arrested cells in the 61 phase. Furthermore, some evidence for the existence of a “window“ or a restriction point in the cell cycle beyond which butyrate did not affect cell cycling was obtained by Moskal gt 31. (1979). This possibil- ity is supported by the data presented here, which shows that cells at the G1/S interface (double thymidine blocked) continued to cycle in the presence of butyrate while early 61 phase cells (mitotically selected) 159 were prevented from entering S phase by butyrate. Whether or not this arrest point resembles (or is the same as) the GO differentiated state remains to be seen. The morphological changes brought about by butyrate were not direct- ly linked to the cell cyle block, since a morphological change could be induced by butyrate without stopping cell cycle arrest (in S and Gz phase cells). The possibility existed that the butyrate effect was merely a time-dependent phenomenon in which the majority of cells were arrested within 16 hr after butyrate addition. However, the ability of 61/5 phase cells to enter 5 phase and to reach mitosis even after 16 hr of butyrate pre-treatment shows that this is not the case. The fact that these cells displayed the “butyrate" morphology before release from the thymidine block and during S and Gz phases demonstrates that the morphological change does not necessarily accompany a cell cycle arrest. One possibility is that the morphological changes induced by butyrate cause a differentiation-like state to occur once the cell reaches the restriction point in 61 phase. Indeed, this membrane modification may be a general requirement for cell differentiation to occur. The commonly used representation of the cell cycle is shown in Figure 43. Differentiation is thought to occur at some point during 61 phase. The idea of “restriction points" for cell growth control was pre- sented by Pardee (1974). According to this model there is a distinct point at which cells are arrested by various nutritional deficiencies. Malignant cells have lost this restriction point control and, under adverse growth conditions, stop at random points in the cell cycle. Support for this hypothesis was provided by Gunther 33 51. (1974) who demonstrated a commitment of splenic lymphocytes to mitogenesis after 160 Figure 43. Cell Cycle Model. The cell cycle consists of four basic phases: M phase, during which mitosis occurs, G1 phase, S phase, during which the majority of DNA synthesis takes place, and 62 phase. Differentiated cells are believed to be in a non-cycling Go state, which probably occurs during 61 phase. Maximum glycolipid and glycoprotein synthesis occur in different points in the cell cyle. 161 .5255 . ee m_mmr._.z>m U o.a300>.._o EDS—x52 m.mm:._.z>m 2_m._.omn_oo>._w $5.2.sz 0 Ne ' ee_e_...e.eo. 162 addition of Con A. After this commitment point, but before mitogenesis, addition of o-methylmannoside to cultures failed to inhibit mitogenesis. This restriction point theory was later elaborated upon by Gelfant (1977) to include essentially all metabolic blocks that have been observed experimentally. The possibility that extra-nuclear changes may be involved in cell cycling and regulation of cell growth was shown by Yanishevsky and Prescott (1978), who demonstrated the induction of early S-phase DNA labeling in G1 phase nuclei by late S phase CHO cells. Thus, it is possible that glycolipid metabolism is important in and strictly regulated by cell cycling. If glycolipids act as cell surface receptors or modulate the chemical or physical properties of cell mem- branes, their function in regulation by cell cycling could be to provide appropriate signals for differentiation or continued cell growth. Glycolipids have been unanbiguously shown to function as cell sur- face receptors only in the case of the cholera toxin receptor. Early work by Cuatrecasas (1973a-c), Holmgren gt 21. (1973) and King and van Menigen (1973) clearly demonstrated the specific cholera toxin receptor activity of GM1 ganglioside. The toxin is a two subunit protein con- sisting of an A (activating) subunit and a B (receptor binding) subunit. Once the toxin is bound to the receptor, the A subunit enters the mem- brane and activates an adenylate cyclase. Once the adenylate cyclase has been activated the elevation of the level of cAMP within the cell modu- lates the observed biochemical responses to toxin treatment. The actual mechanism by which the activating subunit activates the adenylate cyclase system is currently under study by many groups. Binding of cholera toxin has been shown to redistribute ganglioside receptors on lymphocytes (Revesz and Greaves, 1975; Craig and 163 Cuatrecasas, 1975; Sedlacek gt 31., 1976) and may provide a model system for the interaction of'glycolipids with other extracellular ligands. This also suggests a mechanism for transmembrane communication by glyco- lipid receptors. The mechanisms of glycosyltransferase regulation are not well under- stood. One possible model for regulation has been suggested by Dawson gt .21- (1979; ibid., 1980) in which specific glycosyltransferases are regu- lated by phosphorylation-dephosphorylation via cAMP-dependent protein kinases (Figure 44). The support for this theory is largely correlative evidence from measurements of glycosyltransferase activities and cAMP levels in neuroblastoma cells possessing PGE1 and opiate receptors (Dawson, 1980). Thus,ganglioside synthesis was enhanced in cell lines where cAMP levels have been increased by treatment with phosphodiesterase inhibitors, prostaglandin E1 (PGE1), cholera toxin, 8-bromo-cAMP or dibutyryl-cAMP. When enkephalins were added to block PGE1 or cholera toxin stimulation of adenylcyclase, induction of ganglioside synthesis was also inhibited. It has also been shown that reduced levels of UDP-GalNAc:G 3 Neacetylgalactosaminyltransferase activity in polyoma or SV40 transformed NIL or 3T3 cells are accompanied by an inhibition of cAMP synthesis (Brady and Fishnan, 1976). Certainly the adenylcyclase activities and cAMP levels in KiMSV-transformed 3T3 cells used in the studies described in this thesis should be of interest. It was also found in these studies (Dawson, 1979) that a tolerance to opiates (characterized by loss of inhibition of PGE1 stimulation) was reflected by a loss of glycosyltransferase response to given levels of opiates. Thus, it is possible that alterations in membrane glycocon- jugates due to opiate treatment result in a lowered sensitivity to a 164 .eeum emewexo euepzceee we» we pe>e~ one we :ewue>wuee mwzu exeepe meeueeeee euewee ea meweewe eueweo .eewue>wuee ee mcwmeee emeeewmeewupzmeezpm e meuepaeeeemege emeewx mwew .emeewx eweueee e meue>weee wmwe an meeueeeee —mua we neweewe nee: mwe>ew ¢zwuee emeweXe euepaeeee we :ewue>ewe as» .Amwmpv uflm MM eemzea we eemeeeee .peeee mwgu ep meweeeee< .xuw>wue< emeeewmeegupameexpw sew Emweegeez zgeuepaoem veneeeeeoieewuePxeecemege .ee eeemwm 165 emeeeweeeezwmeezo 333.2 a 325. £22m 23.39?ng (0 3226 32233 Jr £240 nF< < .283: e830 . .2381 Jon. 166 particular level of opiates. Furthermore, they found that removal of the opiates from the growth medium resulted in a greatly increased level of adenylate cyclase activity as well as a corresponding increase in speci- fic glycosyltransferase activities. This could resemble the in £122 "tolerance-withdraw“ response, which could be due to alterations in mem- brane composition upon chronic opiate treabment. A similar situation, where membrane alterations are induced to change content or accesibility of membrane receptors may occur in viral transformation. Glycoconjugate changes in virally transformed cells could be mediated by the viral genome in order to escape immune surveil- lance or to inhibit further viral infection. It has been shown for RNA tunor viruses that once viral tranformation takes place, infection by another virus particle does not usually occur (Rubin, 1960). This so-called "interference" has been explained by the binding of the newly synthsized viral coat proteins or particles to virus particle receptors on the cell surface. Alterations in cell surface composition, particu- larly the glycoconjugates, could be another interference mechanism. In addition, since viral particles use specific portions of the plasma meme brane for the viral coat (Pessin and Glaser, 1980), modifications in gly- coconjugate metabolism by viral infection could be to provide newly form- ed virus particles with the appropriate viral coat. The selectivity of viral budding sites could also be due to specific glycolipid or glyc0pro- tein coding. Alternatively, since viral transformation causes a loss of growth control (implying loss of cell cycle control), the changes in glycoconju- gates which occur as a function of cell cycling may be lost. In this event glycoconjugate composition and glycosyltransferase activities would 167 appear to be altered. A further implication of this is that glycoconju- gates which may be responsible for growth control regulation may no long- er be present or available on the cell surface. This would explain why transformed cells often fail to show contact inhibition and why a single transformation-sensitive molecule has not been found which functions in all systems. The studies carried out in this thesis have been designed to address the broad question of how glycolipids are involved in the regulation of cell growth. The various approaches used, i.e., glycosyltransferase measurements in normal and transformed cells as well as in cell surface mutants and the cell biology of a known glycosyltransferase activator, demonstrate the methods available for such studies. These results, when used in conjunction with a variety of other cell surface studies (e.g., lectin and antibody effects on cell growth, biosynthetic pathways for glycolipid synthesis and their alterations upon transformation) begin to show the importance of glycolipids in cell growth control. They appear to have direct as well as indirect roles in the transformation process, in intercellular communication, and in cell morphological changes. By extending these studies to ascertain mechanisms of glycosyltransferase regulation and dependence of specific glycosyltransferase activities on cell cycling, a great deal can be learned about how glycolipids function in the plasma membrane and how the plasma membrane functions to regulate cell growth and communication. BIBLIOGRAPHY BIBLIOGRAPHY Abrahamsson, S., Dahlen, B., Lofgren, H., Pascher, I., and Sundell, S. (1977) in Structure of Biological Membranes (Abrahamsson, S., Pascher, I., eds.) Plenum Press (New York) p 1. Altenberg, B.C., Via, D.P., and Steiner, S.H. (1976) Exp. Cell Res. 192, 223. Arce, A., Maccioni, H.J., and Caputto, R. (1970) Fed. Proc. 22, 925. Arce, A., Maccioni, H.J., and Caputto, R. (1971) Biochem. J. 121, 483. Basu, M., and Basu, S. (1972) J. Biol. Chem. 241, 1489. Basu, M., and Basu, S. (1973) J. Biol. Chem. 248, 1700. Basu, M., Chien, J.-L., and Basu, S. (1974) Biochem. Biophys. Res. Commun. 69, 1097. Basu, M., Higashi, H., Basu, S., and Evans, C. (1980) Fed. Proc., New Orleans, LA. Basu, S. (1966) Ph.D. Thesis, Univ. of Michigan, Ann Arbor, Michigan. Basu, S., Basu, M., Chien, J.-L., and Presper, K.A., "Ganglioside Structure and Function“, Plenum Press (in press). Basu, S., Basu, M., and Potter, M. (1979) Biochemistry of Cell Surface Glycolipids Symposium, ACS, Washington, D.C. Basu, S., and Kaufman, B. (1965) Fed. Proc. 24, 479. Basu, S., Kaufiman, B., and Roseman, S. (1965) J. Biol. Chem. ggg, PC41. Basu, S., Kaufman, B., and Roseman, S. (1968) J. Biol. Chem. 243, 5802. Basu, S., Schultz, A., Basu, M., and Roseman, S. (1971) J. Biol. Chem. 246, 42. Basu, S., Kaufman, B., and Roseman, S. (1973) J. Biol. Chem. 248, 1388. Bennett, 6., LeBlond, C.D., and Haddad, A.J. (1974) J. Cell Biol. 69, 258. Bjorndal, H., Hellerqvist, C.G., Lindberg, B., Svensson, S. (1970) Angew. Chem. Internat. Edit. 3, 610. Bosmann, H.B. (1972) Biochem. Biophys. Res. Commun. 48, 523. Brady, R.O. (1962) J. Biol. Chem. 231, PC2416. 168 169 Brady, R.O., and Fishman, P.H. (1976) Biochim. Biophys. Acta 335, 12. Brady, R.O., Formica, J.V., and Koval, G.J. (1958) J. Biol. Chem. 233, 107. Brady, R.O., and Mora, P.T. (1970) Biochim. Biophys. Acta 218, 308. Braun, P.E., and Snell, E.E. (1968) J. Biol. Chem. 243, 3775. Bremer, E.G., Gross, S.K., and McCluer, R.H. (1979) J. Lipid Res. go, 1028. Brunngraber, E.G., Tettamanti, G., and Berra, B. (1976) in Glycolipid Methodolo (Witting, L.A., ed.) Amer. Oil Chemists' Society Champaign p 159. Burton, R.M. (1969) Lipids 5, 475. Candida, E.P.M., Reeves, R., and Davie, J.R. (1978) Cell 14, 105. Carlson, D.M., Swanson, A.L., and Roseman, S. (1964) Biochemistry 3, 402. Chambers, R.E., and Clamp, J.R. (1971) Biochem. J. 125, 1009. Chandrabose, K.A., and MacPherson, I. (1975a) Biochim. Biophys. Acta 422, 90. Chandrabose, K.A., and MacPherson, I. (1975b) Biochim. Biophys. Acta 422, 112. Chatterjee, S., Sweeley, C.G., and Velicer, L.F. (1975) J. Biol. Chem. 250, 61. Chien, J.-L., Li, S.-C., Laine, R.A., and Li, Y.-T. (1978) J. Biol. Chem. 253, 4031. Chien, J.-L., Williams, I., and Basu, s. (1973) J. Biol. Chem. 248, 1788. Cleland, R.H., and Kennedy, E.P. (1960) J. Biol. Chem. 235, 45. Craig, S.W., and Cuatrecasas, P. (1975) Proc. Natl. Acad. Sci. USA. 1;, 3844. Critchley, D.R., and MacPherson, I. (1973) Biochim. Biophys. Acta 326, 14. Cuatrecasas, P. (1973a) Biochemistry 12, 3547. Cuatrecasas, P. (1973b) Biochemistry 12, 3558. Cuatrecasas, P. (1973c) Biochemistry 12, 3577. 170 Cumar, F.A., Brady, R.O., Kolodny, E.H., McFarland, V.W., and Mora, P.T. (1970) Proc. Natl. Acad. Sci. USA. 61, 757. Cummings, R.O., Cebula, T.A., and Roth, 5. (1979) J. Biol. Chem. 254, 1233. Dawson, 6. (1978) in The Gl coconjugates, Vol. II (Horowitz, M.I., and Pigman, W., eds.) Academ1c Press,1nc. (New York) p 256. Dawson, G., McLawhon, R.W., and Miller, R.J. (1979) Biochemistry of Cell Surface Glycolipids Symposium, ACS, Washington, D.C. Dawson, G., McLawhon, R., and Miller, R.J. (1980) J. Biol. Chem. 255, 129. Dawson, G., and Sweeley, C.C. (1972) J. Biol. Chem. 241, 410. DeLaat, S.W., van der Saag, P.T., and Shinitzky, M. (1977) Proc. Natl. Acad. Sci. USA. 14, 4458. Den, H., Sela, B.-A., Roseman, S., and Sach, L. (1974) J. Biol. Chem. 242, 659. Deppert, W., and Walter, G. (1978) J. Supramol. Struct. 8, 19. Deppert, W., Werchau, H., and Walter, G. (1974) Proc. Natl. Acad. Sci. USA. 11, 3068. DiMari, S.J., Brady, R.N., and Snell, E.E. (1971) Arch. Biochem. Biophys. 143, 553. ' Dowd, J.E., and Riggs, D.S. (1965) J. Biol. Chem. gig, 863. Edidin, M. (1973) Ann. Rev. Biophys. Bioeng. 3, 179. Fallon, R.J., and Cox, R.P. (1979) J. Cell Physiol. 190, 251. Fishman, P.H., and Brady, R.O. (1976) Science 124, 906. Fishman, P.H., Brady, R.O., Bradley, R.M., Aaronson, S.A., and Todaro, G.J. (1974a) Proc. Natl. Acad. Sci. USA. 11, 298. Fishnan, P.H., Simmons, J.L., Brady, R.O., and Freese, E. (1974b) Biochem. Biophys. Res. Commun. 52, 292. Fishman, P.H., McFarland, V.W., Mora, P.T., and Brady, R.O. (1972) Biochem. Biophys. Res. Commun. 48, 48. Fiske, C.H., and SubbaRow, Y. (1925) J. Biol. Chem. 66, 375. Folch, J., Lees, M., and Sloane-Stanley, G.H. (1957) J. Biol. Chem. ggg, 497. 171 Fox, T.0., Sheppard, J.R., and Burger, M.M. (1971) Proc. Natl. Acad. Sci. USA. 55, 244. Frazier, W., and Glaser, L. (1979) Annu. Rev. Biochem. 55, 491. Gahmberg, C.G., and Hakomori, S. (1973) J. Biol. Chem. 535, 4311. Gahmberg, C.G., and Hakomori, S. (1975a) J. Biol. Chem. 559, 2438. Gahmberg, C.G., and Hakomori, S. (1975b) J. Biol. Chem. 559, 2447. Gelfant, S. (1977) Cancer Res. 51, 3845. Ghosh, S., Blumenthal, H.J., Davidson, E., Roseman, S. (1960) J. Biol. Chem. 555, 1265. Ginsberg, E., Solomon, D., Sreevalsan, T., and Freese, E. (1973) Proc. Natl. Acad. Sci. USA. 19, 2457. Gross, S.K., and McCluer, R.H. (1980) Anal. Biochem. 19g, 429. Gugther, G.R., Wang, J.L., and Edelman, G.M. (1974) J. Cell Biol. 5;, 66. Hagopian, H.K., Riggs, M.G., Swartz, L.A., and Ingram, V.M. (1977) Cell 12, 855. Hakomori, S. (1964) J. Biochem. 55, 205. Hakomori, S. (1970) Proc. Natl. Acad. Sci. USA. 51, 1741. Hakomori, S., and Jeanloz, R.W. (1964) J. Biol. Chem. 555, PC3606. Hakomori, S., Koscielak, J., Bloch, K.J., and Jeanloz, R.W. (1967) J. Immunol. 55, 31. Hakomori, S., and Murakami, W.T. (1968) Proc. Natl. Acad. Sci. USA. ‘53, 254. Hakomori, S., Teather, C., and Andrews, H. (1968) Biochem. Biophys. Res. Commun. 55, 563. Hammarstrom, S., Samuelsson, B., and Samuelsson, K. (1970) J. Lipid Res. 11, 150. Hansson, H.-A., Holmgren, J., and Svennerholm, L. (1977) Proc. Natl. Acad. Sci. USA. 15, 3782. Henneberry, R.C., and Fishman, P.H. (1976) Exp. Cell Res. 595, 55. Hildebrand, J., and Hauser, G. (1969) J. Biol. Chem. 531, 5170. Hildebrand, J., Stoffyn, P., and Hauser, G. (1970) J. Neurochem. 11, 403. 172 Hirokawa, N., and Kitamura, M. (1975) Naunyn-Schmiedebergs' Arch. Pharmacol. 551, 107. Hirshberg, C.B., Goodman, S.R., and Green, C. (1976) Biochemistry 55, 3591. Holley, R.W., Armour, R., Baldwin, J.H., Brown, K.D., and Yeh, Y.-C. (1977) Proc. Natl. Acad. Sci. USA. 14, 5046. Holley, R.W., Armour, R., and Baldwin, J.H. (1978) Proc. Natl. Acad. Sci. USA. 15, 339. Holm, M., Mansson, J.-E., Vanier, M.-T., and Svennerholm, L. (1972) Biochim. Biophys. Actaggg, 356. Holmgren, J., Lonnroth, I., and Svennerholm, L. (1973) Infect. Immunol. 8, 208. Ishibashi, T., Atsuta, T., and Makita, A. (1976) Biochem. Biophys. Acta 429, 759. Ishibashi, T., Kijimoto, S., and Makita, A. (1974) Biochem. Biophys. Acta 531, 92. Itaya, K., and Hakomori, S. (1976) FEBS Lett. 55, 65. Itaya, K., Hakomori, S., and Klein, G. (1976) Proc. Natl. Acad. Sci. USA. 15, 1568. Iwamori, M., and Nagai, Y. (1978) J. Biochem. 54, 1601. Jones, J.P., Ransey, R.B., Aexel, R.T., and Nicholas, H.J. (1972) Life Sci. 11, 309. Karlsson, K.-A. (1977) in Structure_of Biological Membranes (Abrahamsson, S. and Pascher, I., eds.)_Plenum Press (New York) p 245. Kaufiman, B., Basu, S., and Roseman, S. (1966) Methods in Enzymology, Vol. VIII, 365. Kaufiman, B., Basu, S., and Roseman, S. (1967) Proc. of the Third Int. Symposium on the Cerebral Shingolipodoses (Aronson, S.H., and Yolk, B.W., eds.) Pergamon Press (NewTYork) p 193. Kaufman, B., Basu, S., and Roseman, S. (1968) J. Biol. Chem. 535, 5804. Keenan, T.W., and Morre, D.J. (1975) FEBS Lett. 55, 8. Kemp, S.F., and Stoolmiller, A.C. (1976a) J. Biol. Chem. 55;, 7626. Kemp, S.F., and Stoolmiller, A.C. (1976b) J. Neurochem. ‘57, 723. Kijimoto, S., and Hakomori, S. (1971) Biochem. Biophys. Res. Commun. 43, 557. 173 Kijimoto, S., and Hakomori, S. (1972) FEBS Lett. 25, 38. Kijimoto, S., Ishibashi, T., and Makita, A. (1974) Biochem. Biophys. Res. Connun. 6_O, 177. King, C.A., and van Henigen, W.E. (1973) J. Infect. Dis. 557, 639. Koyoma, H., and Ono, T. (1976) J. Cell Phys. 55, 49. Langenbach, R., and Kennedy, S. (1978) Exp. Cell Res. 555, 361. Lindberg, B., and Lonngram, J. (1978) Methods in Enzymology, Vol. XXVIII, Academic Press (New York) p 3-42. Lingwood, C.A., and Hakomori, S. (1977) Exp. Cell Res. 595, 385. Lingwood, C.A., Ng, A., and Hakomori, S. (1978) Proc. Natl. Acad. Sci. USA. 15, 6049. Lockney, M.W., Moskal, J.R., Macher, B.A., and Sweeley, C.C. in preparation. Lowry, O.H., Rosebrough, N.J., Farr, A.L., and Randall, R.J. (1951) J. Biol. Chem. 125, 265. Macher, B.A., Lockney, M., Moskal, J.R., Fung, Y.K., and Sweeley, C.C. (1978) Exp. Cell Res. 117, 95. Maley, F. (1970) Biochem. Biophys. Res. Commun. 55, 371. Mannino, R.J., and Burger, M.M. (1975) Nature 555, 19. McConnell, M.M. (1975) in Functional Linkage in Bimolecular Systems. (Schmitt, F.O., Schneider, 0.M., andCrothers, E.H., eds.) Raven Press (New York) p 123. McGuire, E. (1972) in Membrane Research (Fox, C.F., ed.) Academic Press (New York) p 347. McKibbin, J.M. (1976) in Gl coli id Methodolo (Witting, L.A., ed.) Amer. Oil Chemists' Society (Champaign) p 77% Merritt, W.D., Morre, D.J., Franke, W.W., and Keenan, T.W. (1977) Biochim. Biophys. Acta 427, 820. Moffatt, J.G. (1966) Methods infgnzymology, (Vol VIII), (Neufeld, E.E., and Ginsberg, V., eds.)Academic Press (New York) p 136. Mora, P.T., Brady, R.O., Bradley, R.M. and Mcarland, V.W. (1969) Proc. Natl. Acad. Sci. USA. 55, 1290. Morell, P., and Braun, P. (1972) J. Lipid Res. 15, 293. 174 Morell, P., Constantino-Ceccarini, E.C., and Radin, N.S. (1970) Arch. Biochem. Biophys..l4l, 738. Morell, P., and Radin, N.S. (1969) Biochemistry 5, 506. Morell, P., and Radin, N.S. (1970) J. Biol. Chem. 545, 342. Morre, D.J., Merlin, L.M., and Keenan, T.W. (1969) Biochem. Biophys. Res. Commun. 51, 813. Moser, M.W., and Karnovsky, M.L. (1959) J. Biol. Chem. 554, 1990. Moskal, J.R., Gardner, D.A., and Basu, S. (1974) Biochem. Biophys. Res. Commun. 54, 751. Moskal, J.R., Lockney, M.W., Mason, P., and Sweeley, C.C. (1979) Fed. Proc., Dallas, Texas. - Nagai, Y., and Iwamori, M. (1980) Mal. and Cell. Biochem. 22. 81. Naiki, M., and Marcus, D.M. (1975) Biochemistry 44, 4837. Nakano, M., and Fujino, Y. (1973) Biochim. Biophys. Acta 555, 457. Natraj, C.V., and Datta, P. (1978) Proc. Natl. Acad. Sci. USA. 15, 3859. Neutra, M., and LeBlond, C.P. (1966) J. Cell Biol. 59, 137. Nicolson, G.L. (1976) Biochim. Biophys. Acta 451, 57. Noonan, K.D., and Burger, M.M. (1973) J. Biol. Chem. 545, 4286. O'Brien, P.J. (1964) Biochim. Biophys. Acta 55, 628. O'Connor, J., Nunez, H., and Barker, R. (1979) Biochemistry 15, 500. Oppenheimer, S.B., Edwin, M., Orr, C.W., and Roseman, S. (1969) Proc. Natl. Acad. Sci. USA. 55, 1395. Orr, C.W., and Roseman, S. (1969) J. Membrane Biol. 4, 109. Pacuszka, T., Duffard, R.O., Nishimura, R.N., Brady, R.O., and Fishman, P.H. (1978) J. Biol. Chem. 255, 5839. Pacuszka, T., and Koscielak, J. (1976) Eur. J. Biochem. 54, 499. Pardee, A.B. (1974) Proc. Natl. Acad. Sci. USA. 14, 1286. Patt, L., and Grimes, W.J. (1974) J. Biol. Chem. 542, 4157. Pessin, J.E., and Glaser, L. (1980) Fed. Proc., New Orleans, LA. Podolsky, D.K., Weisser, M.M., and Isselbacher, K.J. (1978) Proc. Natl. Acad. Sci. USA. 7_5, 4426. 175 Prasad, K.N., and Sinha, P.K. (1976) In Vitro 4;, 125. Presper, K.A., Basu, M., and Basu, S. (1976) Fed. Proc. 55, 1441. Price, D.L., Griffin, J.W., and Peck, K. (1977) Brain Res. 454, 379. Radin, N.S., Brenkert, A., Arora, R.C., Sellinger, 0.2., and Flangas, A.L. (1972) Brain Res. 52, 163. Rauvala, H., Krusius, T., and Finne, J. (1978) Biochim. Biophys. Acta 554, 266. Reissig, J.L., Strominger, J.L., and LeLoir, L.F. (1955) J. Biol. Chem. 541, 959. Revesz, T., and Greaves, M. (1975) Nature (London) 551, 103. Robbins, P.W., and MacPherson, I. (1971) Nature 555, 569. Roseman, S., Distler, J.J., Moffatt, J.G., and Khorana, H.G. (1961) J. Am. Chem. Soc. 55, 659. Roseman, S. (1970) Chem. Phys. Lipids 5, 270. Roth, S., McGuire, E.J., and Roseman, S. (1971) J. Cell Biol 54, 536. Roth, S., McGuire, E.J., and Roseman, S. (1971) J. Cell Biol. 5;, 536. Roth, S., and White, 0. (1972) Proc. Natl. Acad. Sci. USA. 52, 485. Rubin, H. (1960) Proc. Natl. Acad. Sci. USA. 45, 1105. Sakiyama, H., Gross, S.K., and Robbins, P.W. (1972) Proc. Natl. Acad. Sci. USA. 55, 872. Sakiyama, H., and Robbins, P.W. (1974) In Vitro 5, 331. Schnaar, R.L., Weigel, P.H., Kuhlenschmidt, M.S., Lee, Y.C., and Roseman, S. (1978) J. Biol. Chem. 555, 7940. Schneider, F.H. (1976) Biochem. Pharmacol. 55, 2309. Sedlacek, H.H., Staerk, J., Seiler, F.R., Zeigler, W., and Weigandt, H. (1976) FEBS Lett. 51, 272. Shur, 8.0., and Roth, 5. (1975) Biochim. Biophys. Acta 445, 473. Simmons, J.L., Fishman, P.H., Freese, E., and Brady, R.0. (1975) Cell Biol. 55, 414. Singer, S.J., and Nicolson, G.L. (1972) Science 415, 720. Slomiany, A., Slomiany, B.L., and Horowitz, M.I. (1976) in Glycolipid Methodology, Amer. Oil Chemists' Society (Champaign) p 49. 176 Stanley, P., and Carver, J.P. (1977) Proc. Natl. Acad. Sci. USA. 14, 5056. Steck, T.L., and Dawson, G. (1974) J. Biol. Chem. 545, 2135. Steigerwald, J.C., Basu, S., Kaufman, B., Roseman, S. (1975) J. Biol. Chem. 559, 6727. Stoffel, W. (1971) Ann. Rev. Biochem. 49, 57. Stoffel, W., Lekin, D., and Sticht, G. (1967) Hoppe-Seyler's Z. Physiol. Chem. 545, 1570. Sweeley, C.C., Moskal, J.R., Nunez, H., and Matsuura, F., in 27th Int. Con . of Pure and Appl. Chem. (Varmaruori, A., ed.) Pergamon Press (New 70??) p 233. Taki, T., Hirabayashi, Y., Matsumoto, M., and Kojima, K. (1979a) Biochim. Biophys. Acta 51;, 105. Taki, T., Hirabayashi, Y., Ishiwata, Y., Matsumoto, M., and Kojima, K., (1979b) Biochim. Biophys. Acta 515, 113. Taki, T., Hirabayashi, Y., Suzuki, Y., Matsumoto, M., and Kojima, K. (1978) J. Biochem. 55, 556. . Tallman, J.P., Fishman, P.H., and Henneberry, R.C. (1977) Arch. Biochem. Biophys. 455, 556. Thudichum, J.L.W. (1874) Rep. Med. Officer (London) 5, 113. Tjaden, V.R., Krol, J.H., Van Hoeven, R.P., Oomen-Meulemans, E.P.M., and Emmelot, P. (1977) J. Chromatog. 455, 233. Todaro, G.J., Habel, K., and Green, H. (1965) Virology 51, 179. Vemura, K., Yuzawa, M., and Taketomi, T. (1978) J. Biochem. 55, 463. van Heynigen, S.E. (1974) Nature 545, 415. Vance, D.E., and Sweeley, C.C. (1967) J. Lipid Res. 5, 621. Verbert, A., Cacan, R., and Montrevil, J. (1976) Eur. J. Biochem. 29, 49. Watanabe, K., Hakomori, S., Childs, R.A., and Feizi, T. (1979) J. Biol. Chem. 554, 3221. Watanabe, K., Powell, M., and Hakomori, S. (1978) J. Biol. Chem. 555, 8962. Whittenberg, B., and Glaser, L. (1977) Proc. Natl. Acad. Sci. USA. .14, 2251. 177 Ni1k1nson, G.M. (1969) Biochem. J. 55, 324. Wolf, A., and Robbins, P.W. (1974) J. Ce11 8101. 55, 676. Wong, C.G., Sung, S.-S.-J., Sweeley, C.C. (1979) in Methods in Carbohydrate Chemistry, Academic Press (New York). Wu, H.C., Meezan, E., Black, P.H., and Robbins, P.W. (1969) Biochemistry 8, 2509. Yamakawa, T., and Nagai, Y. (1978) Trends in Biol. Sci. 5, 128. Yang, H., and Hakomori, S. (1971) J. Biol. Chem. 545, 1192. Yanishevsky, R.M., and Prescott, D.M. (1978) Proc. Natl. Acad. Sci. USA. 75, 3307. Yeung, K., Moskal, J.R., Chien, J.-L., Gardner, D.A., and Basu, S. (1974) Biochem. Biophys. Res. Commun. 55, 252. Yip, M.C.M., and Dain, J.A. (1979) Biochem. J. ll§. 247. Yogeeswaren, G., Laine, R.A., and Hakomori, S. (1974) Biochem. Biophys. Res. Commun. 55, 591. Yogeeswaren, G., Sheinin, R., Wherrett, J.R., and Murray, R.K. (1972) J. Biol. Chem. 541, 5146. Yohe, H.C., and Yu, R.K. (1980) J. Biol. Chem. 555, 608. Yokosawa, N., Kijimoto-Ochiai, S., and Makita, A. (1978) Biochim. Biophys. Acta 555, 138. Yu, R.K., and Ledeen, R.W. (1972) J. Lipid Res. 55, 680. Zwaal, R.F.A. (1973) Biochim. Biophys. Acta 555, 159.