[44] Gangliosides that modulate membrane protein function

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

[44] G a n g l i o s i d e s T h a t M o d u l a t e Protein Function

521

Membrane

B y Y A S U Y U K I IGARASHI, H I S A O NOJIRI, N O B U O H A N A I , and SEN-ITIROH H A K O M O R I

Introduction Two classes of glycosphingolipid function can be distinguished: one as receptors for bacterial toxins as well as for cell-microbial and cell-cell interactions, the other as regulators of cell proliferation through modulation of functional membrane proteins such as receptors, transducers, and transporters. This chapter describes experimental procedures providing evidence for the latter view, that is, the effect of gangliosides on cell growth and the associated modulatory effects of gangliosides on receptor-associated tyrosine kinase, protein kinase C, and other kinases. The effect of gangliosides as differentiation inducers for hematopoietic cells is also described. Evidence That Gangliosides May Regulate Cell Proliferation and Induce Differentiation: Brief Overview Several lines of evidence that gangliosides may regulate cell proliferation have accumulated in the past decade (Table I). Results from these experimental systems have been interpreted to suggest that gangliosides by themselves may transduce signals stimulating or inhibiting cell growth, or may modulate cell growth through influencing the function of receptors, transducers, or transporters which are essential for cell growth regulation. The effect of endogenous gangliosides stimulated by ligand has been best demonstrated through utilization of cholera toxin subunit B, which binds to GM1 and stimulates 3T3 cell growth in the quiescent state through an unknown mechanism associated with Ca 2+ influx) ,2 The same ligands inhibited growth of transformed cells) Cell growth is also modulated by antiganglioside antibodies3,4 and sialidase5,6 (Table I), which may also affect organization, mobility,and quantity of membrane gangliosides, as a consequence triggering a cascade reaction leading to cell growth inhibition S. Spiegel and P. H. Fishman, Proc. Natl. Acad. Sci. U.S.A. 84, 141 (1987). z A. Vaheri, E. Ruoslahti, and S. Nordling, Nature (London, New Biol.) 238, 211, (i972). 3 C. Lingwood and S. Hakomori, Exp. CellRes. 108, 385 (1977). 4 W. G. Dippold, A. Knuth, and K.-H. M. zurn Buschenfelde, Cancer Res. 44, 806 (1984). 5 S. Usuki, S.-C. Lyn, and C. C. Sweeley, J. Biol. Chem. 263, 6847 (1988). 6 S. Usuki, P. Hoops, and C. C. Sweeley, J. Biol. Chem. 263, 10595 (1988). METHODS IN ENZYMOLOGY, VOL. 179

Copyright© 1989by AcademicPress,Inc. All rightsof reproductionin any formreserved.

522

MISCELLANEOUS

[44]

TABLE I EVIDENCE THAT GLYCOLIPIDS MAY REGULATE CELL PROLIFERATION

Contact inhibition of cell growth accompanies change of glycolipid synthesis ("cell-contact response" of glycolipids)~'b Cell cycle-dependent change of glycolipid organization: exposure at G~ or Go phase~'d Butyrate induces cell growth inhibition and enhances Gm synthesise Retinoids induce contact inhibition of cell growth and enhance Gm synthesis and glycolipid responsel Antibodies to Gm but not to globoside inhibit 3T3 and NIL cell growth and enhance Gm synthesisc; MAb to Gr), inhibits melanoma cell growths Cholera toxin subunit B, which binds Gin, induces DNA synthesis in quiescent 3T3 cells, while the same ligand induces stimulation of transformed 3T3 cellsh Sialidase induces and sialidase inhibitor reduces cell proliferation i Membrane-associated sialidase increases in various tumor cell lines/and decreases in association with contact inhibition;b sialidase inhibitor for Gm induces cell growth behavior similar to that of normal ceilsk Exogenous addition of glycolipids incorporated into cell membranes inhibits cell growth through extension of the G~ phaset a S. Hakomori, Proc. NatL Acad. Sci. U.S.A. 67, 1741 (1970); H. Sakiyama, S. K. Gross, and P. W. Robbins, Proc. Natl. Acad. Sci. U.S.A. 69, 872 D. R. Critchley and I. A. Macpherson, Biochim. Biophys. Acta 296, 145 (1973); S. Kijimoto and S. Hakomori, Biochem. Biophys. Res. Commun. 44, 557 (1971); R. Langenbach and S. Kennedy, Exp. CellRes. 112, 361 (1978). b G. Yogeeswaran and S. Hakomori, Biochemistry 14, 215 i (1975). c C. Lingwood and S. Hakomori, Exp. CellRes. 108, 385 (1977). a C. G. Gahmberg and S. Hakomori, J. Biol. Chem. 250, 2438 (1975). e p. H. Fishman, J. L. Simmons, R. O. Brady, and E. Freese, Biochem. Biophys. Res. Commun. 59, 292 (1974). I L. Patt, K. Itaya, and S. Hakomori, Nature (London) 273, 379 (1978). W. G. Dippold, A. Knuth, and K.-H. M. zum Buschenfelde, Cancer Res. 44, 806 (1984). h S. Spiegel and P. H. Fishman. Proc. Natl. Acad. Sci. U.S.A. 84, 141 (1987); S. Spiegel and C. Panagiotopoulos, Exp. CellRes. 177, 414 (1988). J A. Vaheri, E. Ruoslahti, and S. Nordling, Nature (London) New Biol. 238, 211 (1972); S. Usauki, S.-C. Lyn, and C. C. Sweeley, J. Biol. Chem. 263, 6847 (1988); S. Usuki, P. Hoops, and C. C. Sweeley, J. Biol. Chem. 263, 10595 (1988). J C.-L. Schengrund, R. N. Lansch, and A. Rosenberg, J. Biol. Chem. 248, 4424 (1973). k S. Hakomori, W. W. Young, Jr., U M. Patt, T. Yoshino, L. Halfpap, and C. A. Lingwood, in "Structure and Function of Gangliosides" (U Svennerholm, H. Dreyfus, and P.-F. Urban, eds.), p. 247. Plenum, New York, 1980. t R. A. Laine and S. Hakomori, Biochem. Biophys. Res. Commun. 54, 1039 (1973). T. W. Keenan, E. Schmid, W. W. Franke, and H. Wiegandt, Exp. CellRes. 92, 259 (1975).

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

523

or stimulation. Ganglioside-dependent cell growth modulation may well operate by modified signal transduction through the receptor/transducer kinases, since gangliosides are modulators for receptor-associated tyrosine kinases,7- to protein kinase C, ~,~2 and other kinases ~3-~5 (see Appendix). On the other hand, ganglioside-dependent induction of differentiation has been observed, and the inductive phenomenon is increasingly apparent in hematopoietic cells 16,17 and in neuroblastoma, ~s although the mechanism of the phenomenon is still unknown. The presence of GQ~b-sensitive cell surface protein kinase and its activation could be the cause of GQtb-induced neurite formation in neuroblastoma? 9 This research area is still in a stage of ferment, and our knowledge is highly fragmentary. Various aspects of phenomena involving bioactive gangliosides are summarized in the Appendix. The following sections of this chapter describe only the experimental procedures carried out and results obtained in our laboratory. Effect of G M I and GM3 o n P D G F - D e p e n d e n t Swiss 3T3 Cell Growth and on P D G F R e c e p t o r Tyrosine Kinase Cell Culture. The medium used for cell culture was a mixture of Dulbecco's modified Eagle's medium (DME) and Ham's FI2 medium, in the weight ratio of 3 : 1, and supplemented with 0.4 g/liter of L-histidine and 15 m M HEPES (DME-F12). Cells were grown in the above basal medium supplemented with 5% fetal calf serum (FCS) and penicillinstreptomycin in a humidified atmosphere of 95% air-5% CO2. Serum-free 7 E. G. Bremer and S. Hakomori, Biochem. Biophys. Res. Commun. 106, 711 (1982). 8 E. Bremer, J. Schlessinger, and S. Hakomori, J. Biol. Chem. 261, 2434 (1986). 9 N. Hanai, T. Dohi, G. A. Nores, and S. Hakomori, J. Biol. Chem. 263, 6296 (1988). 10N. Hanai, G. A. Nores, C. MacLeod, C.-R. Torres-Mendez, and S. Hakomori, J. Biol. Chem. 263, 10915 (1988). 11 D. Kreutter, J. Y. H. Kim, J. R. Goldenring, H. Rasmussen, C. Ukomadu, R. J. DeLorenzo, and R. K. Yu, J. Biol. Chem. 262, 1633 (1987). 12 y. A. Hannun and R. M. Bell, Science 235, 670 (1987). ta j. R. Goldenring, L. C. Otis, R. K. Yu, and R. J. DeLorenzo, J. Neurochem. 44, 1229 (1985). 14 K.-F. J. Chan, J. Biol. Chem. 262, 5248 (1987). 15 K.-F. J. Chan, J. Biol. Chem. 263, 568 (1988). ~6H. Nojiri, F. Takaku, Y. Terui, Y. Miura, and Y. Saito, Proe. Natl. Acad. Sci. U.S.A. 83, 782 (1986). 17 H. Nojiri, S. Kitagawa, M. Nakamura, K. Kirito, Y. Enomoto, and M. Saito, J. Biol. Chem. 263, 7443 (1988). 18 S. Tsuji, M. Arita, and Y. Nagai, J. Biochem. (Tokyo) 94, 303 (1983). 19 S. Tsuji, T. Yamashita, and Y. Nagai, J. Biochem. (Tokyo) 104, 498 (1988).

524

MISCELLANEOUS

[44]

cell growth was in the same basal medium, supplemented with 1/tg/ml insulin, 1/tg/ml transferrin, 1 #g/ml hydrocortisone, 100/~g/ml fatty acidfree bovine serum albumin (BSA) plus linoleic acid, 2/zg/cm 2 fibronectin, 10 ng/ml platelet-derived growth factor (PDGF), and 10 ng/ml epidermal growth factor (EGF). 2° For assays, cells were seeded into 24-well plates (Costar, Cambridge, MA) at a density of 1 - 2 × 104 cells/well. Purified gangliosides were added to the cell culture medium as follows. Ganglioside in chloroform-methanol (C/M) solution was transferred to glass screw-cap test tubes, evaporated under a stream of N2, and dissolved in 0.5 ml of distilled water with the aid of sonication. This solution was then diluted with 5 ml of DME-F12, passed through a 0.2-/1m sterilizing filter (Gelman, Ann Arbor, MI), and diluted with sterile medium to a final ganglioside concentration of 50 #M. Growth Factors. Highly purified PDGF was isolated from outdated human platelet-rich plasma as described by Raines and Ross) ~ EGF was purchased from Collaborative Research, Waltham, MA.

PDGF-Dependent Mitogenesis and Inhibition by Exogenous Addition of GM3. Swiss 3T3 cells were grown in D M E - F 1 2 in the presence or absence of gangliosides, as described above. The confluent cell cultures were preincubated for 1- 2 days in 1 ml of the medium plus 1% human plasma-derived serum, followed by incubation for 3 days in D M E - F I 2 plus 5% FCS, with or without addition of 50 nmol/ml ofganglioside. This medium was then replaced by D M E - F 12 without serum but with various concentrations of PDGF (2-12 ng/ml), with or without ganglioside, for 48 hr. Each well was supplemented with 1/ICi/ml of [3H]thymidine (60 Ci/mmol; ICN, Irvine, CA). Cells were harvested 18 hr later, washed 3 times with ice-cold 5% trichloroacetic acid (TCA), and the TCA-insoluble material was solubilized with 0.5 ml of I N NaOH. The base was neutralized with 0.1 ml glacial acetic acid, 0.5 ml of the solution was added to 5 ml of scintillation mixture (e.g., Formula 963, New England Nuclear, Boston, MA), and activity was determined in a scintillation counter. Membrane Preparation. Membranes were prepared by scraping the cells, grown in 30 to 50 150-mm dishes, in the presence of phosphate-buffered saline (PBS) with a rubber policeman. The scraped cells were pelleted and resuspended in 6 ml of 5 m M HEPES (pH 7.4), 5 m M MgC12, and 5 m M 2-mercaptoethanol. The ceils were then homogenized with 30-50 strokes in a Dounce homogenizer using a tight-fitting pestle (Wheaton Scientific, Millerville, NJ), and 2.1 ml of 1 M sucrose was added to the homogenate, which was then transferred to polycarbonate tubes and centrifuged at 100,000 g for 1 hr. The pellet was resuspended for assay in 20 D. Barnes and G. Sato, Anal Biochem. 102, 255 (1980). 2~ E. W. Raines and R. Ross, J. Biol. Chem. 257, 5161 (1982).

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

525

20 m M HEPES (pH 7.4) and 100 mM NaCI. Membrane protein was determined by fluorescamine protein assay. Phosphorylation Assays" and Effect of Gangliosides. Phosphorylation of membrane proteins by membrane kinase was carried out in a total volume of 40pl containing (final concentration): 20 m M HEPES (pH 7.4), 100 mM NaCl, 0.2% Triton X-100, 10mM MnC12, 40/zM adenosine 5'-(fl,y-imino)triphosphate, 5 mM p-nitrophenyl phosphate, 10 nM [y-32p]ATP (1000-3000 Ci/mmol), and approximately 30-40/tg of membrane protein. In addition, the assay system contained 70 nM PDGF plus 5 pg of carrier BSA or 5 pg of BSA for controls. These conditions were similar to those described by Pike et al. 22 In order to observe the effect of gangliosides, various quantities of gangliosides (GM1, GM3, NeuAcnLc4) and Gb4 in C/M solution were placed in individual test tubes and evaporated to dryness. To each tube, the buffer solution containing Triton X-100 was added and sonicated. The concentration of glycolipid in each test tube varied from 5 to 30 pM. The whole assay system, containing [7-32P]ATP, PDGF, carrier BSA, and many other components as described above, was then added to the solution in each test tube. Assays were started by addition of the membrane suspension and incubated at 30 ° for 30 min. The reactions were terminated by addition of 40/A Laemmli's sample buffer and subsequently boiled for 3 min prior to application to an 8% polyacrylamide gel. 23 In order to hydrolyze serine or threonine O-phosphate, the gels were fixed and then treated with 1 N NaOH at 50 ° for 1 hr according to the method of Cheng and Chen,24 dried, and autoradiographed for 2 - 8 hr. After visualization of the gel by autoradiography, the region containing the Mr 170,000 component was cut out of the gel, and the 32p activity was determined by a liquid scintillation counter. Results. Inhibition of cell growth and PDGF-dcpendent DNA synthesis in 3T3 cells in the presence of gangliosides GM1 and G m is shown in Fig. 1A,B. The effect of these GM~ and GM3 on tyrosine kinase activity associated with the PDGF receptor is shown in Fig. 2. Effect of GM3 o n EGF-Dependent Mitogenesis and E G F Receptor Phosphorylation

Cell Culture. Human oral epidermoid carcinoma KB cells and ovarian epidermoid carcinoma A431 cells were used throughout this study. KB cells have been characterized as having a moderate concentration of a 22 L. J. Pike, D. F. Bowen-Pope, R. Ross, and E. G. Krebs, J. Biol. Chem. 258, 9383 (1983). 23 U. K. Laemmli, Nature (London) 227, 680 (1970). 24 y. S. E. Cheng and L. B. Chen, ProcNatl. Acad. Sci. U.S.A. 78, 2388 (1981).

526

MISCELLANEOUS

[44]

A

60 tO

40 E CO

-o 2O

0

0

I

I

I

I

I

I

2

4

6

8

10

12

PDGF Concentrofion ( n g / m l ) 12

B

10

Qx s E 7 --

4

2 0

i

I

I

1

2

3

5

Days in Culture Flo. 1. Inhibitory effect on PIN}F-dependent cell growth by exogenous addition of gangliosides. (A) 3T3 cells were seeded into 24-well plates at a density of 2 X 104 cells/well and grown in serum-free chemically defined D M E - F 1 2 containing no ganglioside (@), 50 nmol/ml GMI (E]), 50 nmol/ml GM3 (A), or 50 nmol/ml NeuAcnLc4 (A). The medium was changed on days 2 and 4. Each day, cells were harvested with trypsin and numbers were counted. Each data point represents the average of six determinations -+ S.D. (B) 3T3 cells were cultured for 3 days in D M E - F 1 2 supplemented with 5% FCS and containing no ganglioside (@), 50 nmol/ml Gm (D), 50 nmol/ml GM3 (A), or 50 nmol/ml of NeuAcnLc4 (A). The medium was then replaced with D M E - - F I 2 (without serum) containing various concentrations of PDGF (as indicated on the abscissa) for 48 hr, and incorporation of [3H]thymidine in TCA-insoluble material was measured 18 hr later. The results arc plotted as -fold stimulation (ratio of [3H]thymidine incorporation in the presence versus absence of added PDGF) as compared with initial PDGF concentration. Each data point represents the average of four determinations - S.D.

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

527

100

(_9

"*6 40'

T

2O

0

5

I0 15 20 2[5 3C) Glycolipid Concentration (.uM)

35

FIG. 2. Effect of gangliosides on PDGF-dependent protein phosphorylation. Swiss 3T3 membranes were incubated with l0 nM [7-nP]ATP and 60 nmol PDGF with glycolipid at the concentrations indicated on the abscissa for 30 rain at 30", and proteins were separated by polyacrylamide slab gel electrophoresis. 23 After visualization of 7a2p-labeled protein by autoradiography, the M, 170,000 protein was cut from the gel and counted in a liquid scintillation counter. Results are expressed as percentage of maximum response (glycolipid concentration 0) versus concentration of glycolipid added. Data points represent the average _ S.E. of at least three determinations. O, Gut; E], G~3; O, Gb4; A, NeuAcnLc 4.

high-affinity EGF receptor at the cell surface and show clear EGF-dependent growth stimulation. 2s A431 cells contain an unusually high content of the EGF receptor, but only a very small quantity of EGF can stimulate cell g r o w t h . 26 These cells were grown in DME supplemented with 10% FCS. For experiments that required addition of purified glycolipids to the culture medium, 5% FCS was used. 27,2a Purified glycolipids were added to the cell culture medium as follows. The glycolipid in C/M solution was transferred to a glass screw-cap test tube, evaporated under a stream of N2 dissolved in 0.5 ml of distilled water with the aid of sonication. The glycolipid was then diluted with 5 ml of DME, passed through a 0.2-#m sterilizing filter (Gelman), and diluted with sterile medium to a final concentration of 50 nmol/ml.

EGF-Dependent Mitogenesis and Inhibition by Exogenous Addition of 25 A. C. King and P. Cuatrecasas, J. Biol. Chem. 257, 3053 (1982). 26 T. Kawamoto, J. D. Sato, A. Le, J. Polikoff, G. H. Sato, and J. Mendelsohn, Proc. Natl. Acad. Sci. U.S.A. 80, 1337 (1983). 27 R. Callies, G. Schwarzmann, K. Radsak, R. Siegert, and H. Wiegandt, Eur. J. Biochem. 80, 425 (1977). 2~ G. Schwarzmann, P. Hoffmann-Bleihauser, J. Schubert, K. Sandhoff, and D. Marsh, Biochemistry 22, 5041 (1983).

528

MISCELLANEOUS

[44]

GM3. The experiment was performed with KB cells, since mitogenic stimulation by EGF can be observed more clearly over a wider range of EGF concentrations than found in A431 cells. KB cells were grown in Falcon 24-well plates (15 mm diameter/well; Becton Dickinson, Oxnard, CA) in DME supplemented with 5% FCS with or without addition of ganglioside. Twenty-four hours prior to determination of mitogenesis, the medium was replaced with DME containing I% human plasma-derived s e r u m 29 to which different quantities of EGF were added, and the cells were cultured for 18 hr, followed by labeling with 1/~Ci/ml [3H]thymidine for 2 hr. Cells were washed with PBS and extracted with 5% TCA. The insoluble residue collected on a Millipore filter was counted. Phosphorylation Assays. Phosphorylation of membrane proteins by membrane kinases was carded out in a total volume of 40/11 containing (final concentration): 20 mM HEPES (pH 7.4), 100 m M NaC1, 0.2% Triton X-100, l0 m M MnCI2, 40/~M adenosine 5'-(fl,7-imino)triphosphate, 5 m M p-nitrophenyl phosphate, l0 nM [7-32p]ATP (1000-3000 Ci/ mmol), and approximately 30-40/tg of membrane protein. In addition, the assay system contained 100 ng of EGF plus 5/tg of carrier BSA or 5 #g of BSA alone for controls. In order to observe the effect of gangliosides and other lipids, various quantities (5-40 nmol) ofgangliosides (GMI, GM3, NeuAcnLc4), globoside (Gb4), or 17-20/tg of phospholipids in C/M solution were placed in individual test tubes and evaporated to dryness under a stream of N2. The lipids were solubilized in 25/tl of the buffer containing 0.2% Triton X-100 as above with slight warming (37°). To each 25/d solution of glycolipid or phospholipid in the above buffer solution was added l0/11 of a membrane preparation containing 30-40/~g of membrane protein, and this was incubated with or without 100 ng EGF for l0 rain at room temperature prior to the addition of [7-32p]ATP.3° After 20 min, the assay tubes were cooled to 0 ° in an ice bath. [7-32p]ATP in 5/~1 of buffer solution was then added, and the reaction was allowed to proceed for 10 min at 0 °. The reactions were terminated by addition of 40/~l Laemmli's sample buffer and subsequently boiled for 3 min prior to application to an 8% polyacrylamide slab gel.23 For some experiments, in which phosphorylated products were purified by immunoadsorption, the reaction was terminated by addition of 0.5 ml of solubilization buffer followed by immunoadsorption as described below. For experiments in which phosphorylation was performed on the isolated EGF receptor by immunoadsorption, the phosphorylation assay was performed identically to that with membrane preparations except that 29 D. F. Bowen-Pope and R. Ross, J. Biol. Chem. 257, 5161 (1982). ao G. Cohen, G. Carpenter, and L. King, J. BioL Chem. 255, 4834 (1980).

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

529

addition of the anti-EGF receptor antibody-protein A-Sepharose complex (an amount equivalent to 30-40/~g of total membrane protein) was used instead of addition of cell membranes. The beads were centrifuged and treated with 40/tl of Laemmli's sample buffer, boiled, and subjected to slab gel electrophoresis23 as above (see subsequent section). In order to reduce serine or threonine O-phosphate, the gels were fixed and then treated with 1 N NaOH at 50 o for 1 hr according to the method of Cheng and Chen. 24 The treated gel was then dried, and phosphoproteins were visualized by autoradiography. After visualization of the gel by autoradiography, the region containing the EGF receptor (Mr 150,000170,000) was excised from the gel, and the 32p activity was determined in a liquid scintillation counter. Immunoadsorption of the EGF Receptor. Cell membranes were solubilized with 1 ml of solubilization buffer, consisting of 20 m M HEPES (pH 7.4), 150 m M NaC1, 1% Triton X-100, 10% glycerol, 0.03% NAN3, and 0.1 unit/ml aprotinin (protease inhibitor, Sigma, St. Louis, MO) for I hr on ice. Unsolubilized material was removed by centrifugation. The solubilized material was then incubated with 100 #l of washed Blue Sepharose (Sigma) for l hr at 4 ° on a rocking table in order to select for adenosine-binding proteins. The Blue Sepharose beads were then washed 4 times with 1 ml of solubilization buffer. The Blue Sepharose-bound proteins were eluted with l M NaC1 in a solubilization buffer for l hr at 4 °. After centrifugation, the eluate was used for isolation of the EGF receptor by immunoadsorption as described below. The EGF receptor was immunoadsorbed from the above partially purified preparation by anti-EGF receptor MAb 29.13~ linked to a protein A-Sepharose column. 32 Approximately 25/lg of the purified antibody was incubated with 2 mg of protein A-Sepharose 4B (Sigma) in 100/A of PBS at room temperature for 30 min. The beads were then centrifuged for 1 min in an Eppendorf centrifuge and washed with 0.5 ml of PBS. Aliquots of the Blue Sepharose eluate were diluted 5 times with PBS containing 0.1% BSA and 10% glycerol and incubated with the antibody-protein A-Sepharose complex on a rocking table for 2 hr at 4 °. The beads were then washed 4 times with 1 ml of a solution containing 20 m M HEPES (pH 7.4), 150 m M NaC1, 0.2% Triton X-100, 10% glycerol, and 0.03% NAN3. The washed beads were used for phosphorylation assay. Results. The effects of gangliosides on the EGF receptor tyrosine kinase activity of KB and A431 cells are shown in Fig. 3. 31 y. Yarden, 1. Harari, and J. Schlessinger, J. Biol. Chem. 260, 313 (1985). 32 C. Schneider, R. A. Newman, D. R. Sutherland, U. Asser, and M. F. Greaves, J. Biol. Chem. 257, 10766 (1982).

5 30

MISCELLANEOUS

[44]

100

75 0

25

I

I

I0

20

I~-

I

40

nmoles gongliosides added I00

cO

L~ 5O

~6 o~ 25

0

o

,3

J

2'o 2'5

nmoles cjangliosidesodded FIG. 3. Reduction in EGF receptor phosphorylation by Gin3 ganglioside. (A) A431 membranes were incubated with 100 ng EGF for 20 rain at room temperature and 10 n M [7:2p]ATP for 10 rain at 0 ° in the presence or absence of glycolipids as described in the text. After visualization of 32p-labeled protein by autoradiography, the EGF receptor (Mr 170,000) was cut from the gel and radioactivity detected in a liquid scintillation counter. Results are expressed as percentage of maximum response (glycolipid concentration 0) versus concentration of glycolipid added. Data points represent the average _+ S.E. of at least three determinations. (B) KB cell membranes were incubated with EGF followed by [7-s2p]ATP in the presence or absence of various amounts of glycolipids (Gin, G u , , and NeuAcnLc4(SPG). After visualization of 3:P-labeled protein by autoradiography, the activity of the EGF receptor area was determined as described for A.

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

531

Inhibitory Effect of Lyso-GMa and Stimulating Effect of De-Nacetyl-GM3 on E G F R e c e p t o r Phosphorylation

Cells and Materials. A431 cells were cultured under the same conditions as described in the preceding section. Confluent cell cultures in 150-mm plastic dishes were scraped, pelleted in PBS (800 g), and resuspended in 7 ml of 20 m M HEPES buffer, pH 7.5, 1 m M EGTA, 0.5 m M phenylmethylsulfonyl fluoride (PMSF) in 250 m M sucrose, homogenized in a Dounce homogenizer with a tight-fitting pestle (Wheaton Scientific), and centrifuged (3,000 g, 10 min). The supernatant fraction was centrifuged (100,000 g, 1 hr), and the pellet was resuspended in 300 pl of 20 m M HEPES buffer, pH 7.4. In order to keep the kinase activity stable, the membrane fraction thus prepared was aliquoted and stored in liquid nitrogen until use. Lyso-GM3 was prepared by N-acetylation of neuraminyllactosylsphingosine when the amino group of sphingosine was protected by inclusion of the compound in liposomes. 33 NeuroaminyUactosylsphingosine was prepared by alkaline hydrolysis of GM3 with 1 N KOH in 90% 1-n-butanol as described by Taketomi and Kawamura. 34 Lactosylsphingosine was prepared by hydrolysis of lactosylceramide under the same conditions as described above. 34 Phosphorylation Assay of EGF Receptor. In order to observe the biphasic effect of GM3 and Iyso-GM3 and the stimulatory effect of de-N-acetylG~3 on EGF receptor phosphorylation, a high concentration of ATP (10 #M) was employed under various detergent concentrations, as follows. Cell membranes were incubated in the buffer (20 m M HEPES, pH 7.4, 1 m M MnC12, 10 # M ZnC12, 30/zM NaVO3) including 0.33 pA//EGF (receptor grade; Collaborative Research) plus 1.5 # M carrier BSA and various concentrations of Triton X-100 (high-purity detergent, Pierce Chemical Co., Rockford, IL) in the presence or absence of gangliosides for 10 min at 25 °. The reaction was initiated by addition of 1.0 or 10 # M [7-32p]ATP (10 pCi) for 10 min at 0 °. The reaction volume was 50 #1, and the amount of membrane protein was 25 #g. Reactions were terminated by addition of 50 #1 of Laemmli's sample buffer.23 Aliquots of the incubation mixture were subjected to 8% polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate. The gel was washed with 1 N NaOH for 15 min at 25 °, then treated with 1 N NaOH for 1 hr at 40 ° to reduce serine or threonine O-phosphate 24 and dried, followed by visualization by autoradiography. The region containing the EGF receptor (Mr 170,000) 33 G. A. Nores, N. Hanai, S. B. Levery, H. L. Eaton, M. E. K. Salyan, and S. Hakomori, Carbohydr. Res. 179, 393 (1988). 34 T. Taketomi and N. Kawamure, J. Biochern. (Tokyo) 68, 475 (1970).

532

MISCELLANEOUS

[44]

200-

1169268-

45-

12

345

6

FIG. 4. EGF-dependent phosphorylation of immunoprecipitated EGF receptor. The EGF receptor was immunoprecipitated from solubilized A431 cell membranes, and the immunoprecipitated material was assayed for EGF-stimulated phosphorylation as described in the text, with the addition of 0.65 mM phosphatidylethanolamine. Glycolipids were included in the reaction mixture as indicated below. Aiter phosphorylation, the samples were subjected to gel electrophoresis, dried, and exposed to X-ray film. The resulting autoradiogram is shown above. The arrow indicates the location of the EGF receptor. Molecular weight standards are shown (X 10-3). Lane 1, no EGF, no glycolipid; lane 2, 100 ng EGF, no glycolipids; lane 3, 100 ng EGF, 7.0 nmol Gin; lane 4, 100 ng EGF, 14 nmol GMI; lane 5, 100 ng EGF, 7 nmol Gin; lane 6, 100 ng EGF, 14 nmol Gm.

was excised from the gel and the 32p activity was determined in a liquid scintillation counter (see Fig. 4). In a separate experiment, reaction mixtures containing various concentrations of [y-32p]ATP (i.e., 0.1, 0.5, 2.0, and 10.0 ~ / ) were tested with a constant detergent concentration (0.05% Triton X-100). Since the specific

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

533

activity of [y-32p]ATP at the four ATP concentrations was 3300, 516, 119, and 25/~Ci/mmol, respectively, decreased kinase activity was attained at the higher concentration of ATP. Although the physiological concentration of ATP is greater than 10/~M, performing experiments with ATP concentrations above 10/13,/was difficult in practice because of the great decrease in specific activity. Therefore, we used 1.0 or l0 #MATP concentrations throughout these experiments. Results. The inhibitory effect Of GM3 on EGF receptor kinase activity of A431 cell membranes (Fig. 5, filled triangle with solid line) showed a clear biphasic transition depending on mixed micellar conditions, i.e., inhibitory at low detergent concentration (up to 0.075%) but stimulatory at higher detergent concentration (over 0.1%). Such a biphasic effect was not seen with lyso-GM3 (Fig. 5, open triangle with dotted line), which showed a monophasic continuous inhibitory effect regardless of detergent concen-

.•

5-

4 O

fY

s

o

2

L

o

L

I

L

i

o.o5 o.~ o.~5 02 Triton X - l O 0 (%)

FIG. 5. EGF receptor kinase activity of A431 cell membranes: dependence on Triton X-100 concentration and on GM3, de-N-acetyl-Gu3, and lysO-GM3.In vitro phosphorylation assay was performed using membranes from A431 cells as described in the text. O, Control without ganglioside addition; f-l, with 500#M lyso-CDH (lactosylsphingosine); A, with 500/134 GM3;0, with 250 ~ de-N-acetyl-GM3;A, with 500/zM lysO-GM3.

534

MISCELLANEOUS

::::-

[44]

rh, .-.

%,

C

:':~'.'.Tyr ,.

c-i

:

pHI.9

~,

pill.9

FIe. 6. Phosphoamino acid pattern in hydrolyzate of EGF receptor band: effect of lyso-Gm and de-N-acetyl-GM3. The EGF receptor (M r 170,000) bands were excised from the slab gel (examples as shown in Fig. 4). [32p]Phosphoamino acids from the EGF receptor band were analyzed by two-dimensional thin-layer electrophoresis as described previously (J. A. Cooper, B. M. Sefton, and T. Hunter, this series, Vol. 99, p. 387). (I) A, Without EGF; B, with EGF; C, with EGF plus lyso-GM3. (II) A, With EGF; B, with EGF and de-N-acetyl-GM3.

tration. In striking contrast, de-N-acetyl-Gm (Fig. 5, filled circle with solid line) showed a strong stimulatory effect regardless of detergent concentration. The inhibitory effect of lyso-Gm and stimulatory effect of de-N-acetyl-Gra3 were observed only with tyrosine phosphate (Fig. 6). Effect of Gangliosides on Insulin R e c e p t o r Kinase Activity

Cells and Cell Culture. Human myeloma cell line IM-9 cells35 were grown in Falcon 3045 tissue culture flasks (Becton Dickinson Labware, Lincoln Park, NJ) in RPMI 1640 medium supplemented with 10% heatinactivated (56 ° for 30 min) FCS at 37 ° in a humidified atmosphere of 5% CO2. Solubilization and Partial Purification of Insulin Receptor. Approximately 5 × 109 IM-9 cells were collected and washed with PBS. The cells were solubilized with 40 ml of 1% (v/v) Triton X-100 in 50 m M HEPES buffer (pH 7.4) supplemented with 0.5 trypsin inhibitor units/ml aprotinin and 2 m M PMSF. The suspension was mixed by vortexing and allowed to 35j. L. Fahey, D. N. Buell, and H. C. Sox, Ann. N.Y. Acad. Sci. 190, 221 (1972).

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

535

stand at 4 ° for 40 min with intermittent mixing. The detergent-insoluble material was sedimented by centrifugation at 200,000 g for 40 min. The insulin receptors were further enriched by chromatography on a wheat germ agglutinin-agarose column and elution with 0.3 M N-acetylglucosamine in 50 m M HEPES buffer (pH 7.4) containing 0.1% (v/v) Triton X-100 as described by Kasuga et aL36 Protein concentration was determined by the Bradford reagent (Bio-Rad, Richmond, CA) with BSA as a standard. Phosphorylation Assay. An appropriate amount of ganglioside was taken in microcentrifuge tubes and evaporated under a stream of N2. Solubilized insulin receptor (-10 #g protein) was added to the tubes and incubated with 1 a M insulin at 22 ° for 1 hr in 55/tl of the buffer (50 m M HEPES, pH 7.4, 5 m M MnCI2) in the presence or absence of gangliosides. Phosphorylation was initiated by adding 5/d of 100aM [?-32p]ATP (25 pCi). The final concentration of Triton X-100 was 0.05%, which was reported to be optimal. 37 The reaction mixtures were vortexed and incubated for 10 min at 22 °. The reaction was terminated by adding 15/tl of 5-fold concentrated Laemmli's sample buffer23 supplemented with 75 mg/ ml dithiothreitol and heating this mixture in a boiling water bath for 3 min. The phosphoproteins were separated by 7.5% polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate according to the method of Laemmli 23 and visualized by autoradiography. The phosphorylation corresponding to the fl subunit of the insulin receptor (95 kDa) was excised from the dried gel, and radioactivity was quantified in a liquid scintillation counter. Immunoprecipitation of Phosphorylated Insulin Receptor. The phosphoprotein of 95 kDa was confirmed to be the p subunit of the insulin receptor by immunoprecipitation with antiinsulin receptor antibodies. The phosphorylation reaction was stopped by cooling the mixture to 4 ° and adding 1 ml of 50 m M HEPES buffer (pH 7.4) containing 0.1% Triton X-100, 10 m M sodium fluoride, 4 m M EDTA, and 1 m M sodium vanadate. Antiinsulin receptor antibody designated B-10, which was obtained from a patient with an autoimmune form of insulin-resistant diabetes, was added to this mixture at a dilution of 1 : 300 and incubated at 4 o for 2 hr. The immune complex was precipitated from the solution by incubation with 200/~1 of 10% Pansorbin (Calbiochem-Behring Diagnostics, La Jolla, CA) for at least 1 hr. The immunoadsorbed protein was solubilized in Laemmli's sample buffer containing dithiothreitol and resolved by 7.5% polyacrylamide slab gels in the presence of 0.1% SDS as described by 36 M. Kasuga, M. F. White, and R. C. Kahn, this series, Vol. 109, p. 609. 37 D. t . Hwang, Y.-C. Tay, G. Barseghian, A. Roitman, and A. Lev-Ram, J. Recept. Res. 5, 367 (1985).

536 A Insulin Gongliosides(50NM)

MISCELLANEOUS -

+

+ Gg3

+

-

B

2,3SPG

[44]

(50pM)

+

-tGM3

52qSPG

kDo 200-

-,-95kDo

]16.392.5-

~95kDo

66.2-

45.0-

FIG. 7. Effectof gangliosides on autophosphorylation of insulin receptor. Autoradiogram of 32p-labeledinsulin receptor fraction partially purified on a wheat germ agglutinin column (A) and immunoprecipitate by antiinsulin receptor antibody B-10 (B). Note that only the 95-kDa band (/Ysubunit of insulin receptor) showedenhanced phosphorylation on addition of insulin. This autophosphorylation was inhibited slightly in the presence of 50/tM Gm and completely in the presence of 50 #M 2 --* 3SPG. Kasuga et al. 36 Resulting phosphoproteins were visualized by autoradiography. Results. Insulin receptor kinase associated with the 95-kDa subunit was inhibited strongly by IV3NeuAcnLc4 (2 ~ 3SPG), and to a lesser extent by G m (Fig. 7). In striking contrast, IVrNeuAcnLc4 (2 ---, 6SPG) and disialosyl I (VI3NeuAclV6NeuAca2 ~ 3Galfll --o 4GlcNAcnLc6) had no significant effect (Fig. 8). The effect o f 2 ~ 3sialylnorhexaosylceramide was m u c h higher in comparison to 2 ---, 6SPG or disialosyl I. The effects o f GM2, Gma, G i n , and Gx~b gangliosides were m u c h less in comparison to 2 ~ 3SPG or 2 ~ 3sialylnorhexaosylceramide (Fig. 8). Furthermore, 2 ~ 3SPG inhibited insulin-dependent cell growth o f h u m a n myelogenous leukemia cell lines HL6038 and K56239 in a dose-dependent manner. E f f e c t o f GM3a n d D e r i v a t i v e s on P r o t e i n K i n a s e C A c t i v i t y o f A431 Cells A431 cells were grown in a mixture o f D M E and H a m ' s F12 m e d i u m (weight ratio 1:1) supplemented with 10% FCS. Cells harvested from 50 150-cm diameter dishes were treated simultaneously for partial purifica3s s. J. Collins, R. C. Gallo, and R. E. Gallagher, Nature (London) 270, 347 (1977). 39C. B. Lozzio and B. B. Lozzio, Blood4$, 321 (1975).

[44]

537

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

A. Gongh'osen'es

B. Loc/o senes

100

0

'..= 6 0 (..) <~

~ 4o cr 20

nHC ......

Znsulm

I0

5

I

(-) ..........

I

150

....

I

Znsulm ( - ) . . . . .

I

I

250

50

i

Gangliosides (pM)

~

I 0

~.7~--

I

I

250

FIG. 8. Concentration-dependent effect of various gangliosides on autophosphorylation of the insulin receptor. Aliquots of insulin receptor fraction were incubated with [)mP]ATP, 1 ltM insulin, and various concentrations of several kinds of gangliosides. Incubation mixtures were subjected to polyacrylamide gel electrophoresis under reducing conditions, and the 95-kDa band was excised and counted. Relative activity was expressed as percentage of the same band in the absence of ganglioside.

tion of C-kinase, by the method of Kreutter et aL H Briefly, cells were scraped with a rubber policeman, suspended in 50 ml of 20 m M Tris-HC1 (pH 7.5), 2 m M EDTA, 0.5 m M EGTA, 0.15 U/ml aprotinin, and 0.25 M sucrose, and homogenized by 50 strokes at 4* in a Dounce homogenizer with a tight-fitting pestle (Wheaton Scientific). The homogenized cells were ultracentrifuged at 100,000 g for 60 min. The supernatant was purified on a DE-52 column equilibrated with 20 m M Tris-HC1 (pH 7.5), 2 m M EDTA, and 0.5 m M EGTA (buffer B), and washed well with this buffer. The protein kinase C activity was eluted with buffer B containing 0.1 M NaC1. The activity in this fraction was 200-500 pmol P/min/mg protein. The fraction, which was free of A-kinase and other kinases, was aliquoted and kept at - 8 0 °. In view of the extremely variable results obtained in mixed micelle systems, as described by Hannun and BelU 2 the standard liposome method described by Kraft and Anderson4° was slightly modified, and the effect of gangliosides was studied under these conditions. In conical tubes (1.5 ml, Sarstedt), phosphatidylserine (5 #g/tube) and 1,2-diolein (50 ng/tube), with or without an appropriate quantity of ganglioside or its derivatives, were added in organic solvent (ethanol or C/M), and the mixture was evaporated under a stream of N2. The lipid mixture was sonicated in 30 #1 of 40 A. S. Kraft and W. B. Anderson, Nature (London) 301,621 (1983).

538

MISCELLANEOUS

[44]

20 m M Tris-HCl (pH 7.5) for 30 min. The liposomes in the tube were supplemented with the reaction mixture, consisting of 25 mM Tris-HC1 (pH 7.5), 10 mM MgC12, 400/~M EDTA, 50/tM EGTA, 500/tM CaCl2, 200/tg/ml histone III-S, and 20 pM [y-32p]ATP (2 × 10 6 cpm); final volume was 90 pl. The reaction was initiated by addition of l0/A of the protein kinase C fraction (containing 1-2/tg protein) prepared as described above, and the reaction mixture was incubated for l0 min at 30 °. The reaction was terminated by addition of 1 ml of 25% TCA with 200 pl of 1% BSA in 1 mM ATP solution (pH 7.5). The precipitate was centrifuged, washed twice with l ml of 25% TCA, dissolved in l ml of 1 N NaOH containing 0.1% deoxycholate with slight heating (80 ° for l0 min), and counted in a scintillation counter. The value without phosphatidylserine, 1,2-diolein, or Ca2+ was used as a reference blank. The effects of various gangliosides and lysoglycolipids are shown in Fig. 9. Differentiation Induction by Gangliosides of Human Promyelocytic Leukemia Cell Line I-IL60 Cells into Monocytes/Macrophages or Neutrophilic Mature Granulocytes Cells of the human promyelocytic leukemia cell line HL603s can differentiate in two directions, one leading to monocytes and macrophage cells, the other to neutrophils. The former type of differentiation has been observed to be induced by 12-O-tetradecanoylphorbol 13-acetate (TPA), the latter by dimethyl sulfoxide or retinoic acid. 41 A similar differentiation is caused by exogenous addition of particular gangliosides, in association with growth inhibition. Induction of differentiation into monocytes/macrophages resulted from culturing HL60 cells in chemically defined medium in the presence of Gin; differentiation into neutrophilic mature granulocytes was induced by culturing cells in the presence of neolacto series gangliosides (nLc-Gs) prepared from human neutrophils. It is noteworthy that GM3, which increases characteristically during macrophagelike cell differentiation of HL60 cells induced by TPA, 42 can induce monocytic differentiation of HL60 cells, whereas neolacto series gangliosides, which increase characteristically during neutrophil differentiation of HL60 cells induced by dimethyl sulfoxide or retinoic acid, 42 can induce neutrophil differentiation of HL60 cells. Preparation of Culture Media Containing Ganglioside. A C/M solution of gangliosides was aliquoted in suitable quantity into glass screw-cap test 41 S. J. Collins, Blood 70, 1233 (1987). 42 H. Nojiri, F. Takaku, T. Tetsuka, K. Motoyoshi, Y. Miura, and M. Saito, Blood 64, 534 (1984).

[44]

GANGLIOSIDES IN MEMBRANE PROTEIN FUNCTION

539

C-Kinase Activity (incorporotion of 52p into Histone ]]I-S, cpmxlO -5 ] 5 I0 15 20 None CDH Lyso CDH D-erythro- sphingosine GM3 D5 D] D2 D3

~\\\\\\\\\\\\\\\\\\\~l~ ~\\\\\\\\\\\\\\\\\\\'.~--~ ~\\\\\\\\\\\\\\\\\\\%--~ ~\\\\\\\\\\\\\\\\xk~ ~\\\\\\\\\\\\\\\\\\\'q~ ~\\\\\\\\\\\\\\\\\\\'x"]-~ ~\\\\\\\\\\\\\\\\\'q~

FIG. 9. Effects of various sphingolipids and derivatives on protein kinase C activity. Protein kinase C activity was measured using histone III-S substrate under the standard assay conditions described in the test, in the presence (50 #M) or absence of various glycolipids. Values represent mean - S.E. of three separate experiments. Abbreviations on the ordinate are as follows: CDH, lactosylceramide; lyso-CDH, laetosylsphingosine; D-erythro-splfingosine, D-erythro-1,3-dihydroxy-2-amino-4,5-trans-octadecene (synthetic compound). D1, D2, D3, and D5 are derivatives of GM3, identified as follows: Dl, de-N-acetyl-Gu3 (amino group of sialic acid is free in GM3); D2, de-N-acetyllyso-GMa (amino groups of both sialic acid and sphingosine are free through complete hydrolysis of the N-acetyl group on sialic acid and the N-fatty acyl group on sphingosine); D3, lyso-GM3 (amino group of sphingosine is free); D5, sialyilactosyl-N-acetylsphingosine.

tubes, evaporated under N 2 , and the residue dissolved in distilled water with sonication. The resulting solution was sterilized through a 0.22-#m Millipore filter, mixed with an equal volume of 2 × concentrated serumfree DME-F12, and subsequently used as a stock solution. The ganglioside concentration of this stock solution was determined via the quantity of sialic acid. Conditions of Cell Culture. HL60 cells were cultured in serum-free medium 43 containing insulin (5 #l/ml), transferrin (5 /lg/ml), selenium dioxide (30 nM), sodium bicarbonate (1.2 g/liter), and HEPES (l 5 mM, pH 7.2). The basic medium was D M E - F I 2 (l : 1 by weight). Induction of Differentiation. HL60 cells were seeded at an initial concentration of 2 × 105 cells/ml, and differentiation was induced by replacing the initial culture medium with one containing the appropriate ganglioside. The optimal concentration of GM3 to induce monocyte/macrophage differentiation was 50/IM; the optimal concentration of neolacto series ganglioside to induce neutrophil differentiation was 2.0 #M. Cell 43 T. R. Breitman, S. J. Collins, and B. R. Keene, Exp. Cell Res. 126, 494 (1980).

540

MISCELLANEOUS

[44]

TABLE II ASSESSMENT OF DIFFERENTIATIONOF HE60 CELLS INDUCED BY GANGLIOSIDESa Positive cells (%) Cells

CAE

NBE

OKM 1

OKB2

OKM5

Phagocytosis

Control GM3-treated (50 pM) nLc-Gs-treated (2/zM)

81.5 11.7 81.0 b

10.5 84.8 13.5

3.2 35.7 79.7

3.2 1.3 67.3

1.9 11.7 0.3

16.5 52.0 47.0

a CAE, Naphthol AS-D chloroacetate esterase; NBE, naphthyl butyrate esterase. OKM1, OKB2, and OKM5 are cell surface markers for, respectively, macrophage/granulocyte, granulocyte/B lymphocyte, and monocyte/platelet cell lines; each is defined by a MAb of the same name. Cells were cultured for 6 days in the presence of individual gangliosides. b The number of cells strongly positive for CAE was increased in nLc-Gs-treated cells.

counts were made with a hemocytometer, and viability of cells was assessed by dye exclusion with erythrosine B. Evaluation o f Differentiation. Morphological changes were observed in aliquots of cultured cells under a light microscope after Wright-Giemsa staining. The activity of naphthol AS-D chloroacetate esterase, which is specific to granulocytic lineages, or ot-naphthyl butyrate esterase, which is specific to monocytic lineages, was determined by the esterase doublestaining method of Li et al. 44 Surface membrane antigens were assessed by cytofluorometry in an Ortho Spectrum III (Ortho Diagnostic Systems, Westwood, MA) using the authorized MAbs, including OKM1 (CD11), which reacts with human monocytes and granulocytes, OKB2 (CD24), which reacts with human granulocytes and B lymphocytes, and OKMS, which reacts with human monocytes and platelets. Functional differentiation was assessed on the basis of phagocytic activity of the cells toward polystyrene latex particles (Dow Chemical, Indianapolis, IN; 1.09 g m diameter). Evaluation criteria are shown in Table II. 44 C. Y. Li, K. W. Lam, and L. T. Yam, J. Histochem. Cytochem. 21, 1 (1973). 43 p. H. Fishman and R. O. Brady, Science 194, 906 (1976). 46 Review: S. Hakomori, Annu. Rev. Biochem. 50, 733 (1981). 47 E. Bremer, S. Hakomori, D. F. Bowen-Pope, E. Raines, and R. Ross, J. Biol. Chem. 259, 6818 (1984). 4s S. Spiegel, J. S. Handler, and P. H. Fishman, J. Biol. Chem. 261, 15755 (1986). 49 M. Chorev, A. Feigenbaum, A. K. Keenan, C. Gilon, and A. Levitzki, Eur. J. Biochem. 146, 9 (1985). 5o A. Levitzki, Science 241, 800 (1988). 5, Igarashi et al., unpublished. 52 Nojiri et al., unpublished. 53 S. Tsuji, J. Nakajima, T. Sasaki, and Y. Nagai, J. Biochem. (Tokyo) 97, 969 (1985).

Appendix: BIOACTIVEGANGL1OSIDES Phenomenon a Receptor for bacterial toxins (cholera toxin B subunit) Receptors for other bioactive substances (toxins and hormones) (evidence is ambiguous) Growth modulation EGF-dependent BHK cell growth (NIL cells, growth ~ ) PDGF-dependent 3T3 cell growth (Swiss 3T3, growth ~ ) EGF-dependent A431 and EB cell growth (A431 human epidermoid carcinoma, growth ~ ) B subunit-dependent DNA synthesis (quiescent 3T3, growth T, transformed 3T3 and rapidly dividing normal 3T3, growth J, ) EGF-dependent A431 cell growth (A431, growth 1' ) EGF-dependent A431 cell growth (A431, growth J, ) Ion-transport modulation (Na÷ transport) fl-Adrenergic receptor Differentiation induction (receptors and involved mechanism unknown) Neurite formation in human neuroblastoma cell lines Monocytic differentiation in human leukemia HL60, U937 Granulocytic differentiation in HL60 cells Modulation of protein kinascs protein kinase C HL60 protein kinase C ( J, ) Brain protein kinase C ( ~ ) A431 protein kinase C ( ~ ) Tyrosine kinase EGF-R autophosphorylation ( ~ ) EGF-R autophosphorylaton ( ~ ) EGF-R autophosphorylation ( ~ ) Insulin-R autophosphorylation ( J, ) Other kinases Ca2+/CaM-dependent kinase (1') CaE+/GQ~bactivated kinase (1') Brain kinase PKJ ( 1' ) Brain kinase PKL ( ~ ) a T, increase; ~, decrease.

Ganglioside

Ref.

GMI

45

Various gangliosides

46

GM3

7

GMI

47

GM3

8, 9

GMI

1

De-N-acetyI-GM3

9

Lyso-GM3

10

Gm

48

Unknown ganglioside coupling with receptor molecule

49, 50

GQIb

18

GM3

16

Neolactogangliosides (nLc-Gs)

17

Polysialogangliosides "Lysoganglioside" GM3,IysO-GM3

11 12 51

GM3 Lyso-GM3

8

2---~3 SPG

10 9 52

Gin, GQIb Polysialogangliosides Polysialogangliosides

13 53 14 15

De-N-acetylGM3