requirement for sialic acid on neutrophils in a gmp-140 (padgem) mediated adhesive interaction with...

Vol. 172, No. 3, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

November 15, 1990 Pages 1349-1356

REQUIREMENT FOR SIALIC ACID ON NEUTROPHILS IN A GMP- 140 (PADGEM)

MEDIATED ADHESIVE INTERACTION WITH ACTIVATED PLATELETS

Laura Corral, Mark S. Singer, *Bruce A. Macher, and Steven D. Rosen

Department of Anatomy and Program in Immunology, University of California,

San Francisco, CA 94143-0452

*Departments of Chemistry and Biochemistry, San Francisco State University,

San Francisco, CA 94 132

Received October 4, 1990

Summarv. Platelet GMP- 140, along with ELAM- 1 and gp9OmL, comprise the l.,EC-C&l family of cell-cell adhesion proteins. The three proteins demonstrate a highly related domain organization, which includes an extracellular calcium-type lectin motif. gp90MEL, a lymphocyte homing receptor, mediates lymphocyte attachment to high endothelial venules of lymph nodes through recognition of a sialylated ligand on the endothelial cells. The rosetting of neutrophils or promyelocytic HL60 cells by activated platelets is mediated by GMP- 140 on the platelets. We show here that treatment of neutrophils or HL60 cells with 3 broad spectrum sialidases completely prevents rosetting. However, the Newcastle disease virus sialidase, an enzyme specific for a2,3 and a28 linkages of sialic acid does not affect rosetting of HL60 cells. These results indicate that the ligand for GMP- 140 requires sialic acid and suggest that an a2,6 linkage may be critical, ‘3 1990 Academic Press. IIIC.

Granular-membrane protein (GMP- 140), also known as PADGEM or CD62, is a

glycoprotein of 140,000 daltons found in the alpha granules of platelets and the Weibel-Palade

bodies of endothelial cells (l-5). Upon stimulation of these cells by agonists such as thrombin,

degmnulation leads to the rapid externalization of this membrane-bound protein to the plasma

membrane. Recent studies have established that GMP-140 mediates the calcium-dependent

adhesion of neutrophils or monocytes to activated platelets (6, 7) as well as the early binding of

neutrophils to histamine- or thrombin-stimulated endothelium (8). Possible functional

consequences of these adhesive events include the removal of activated platelets from the blood by

neutrophils (or monocytes) and the rapid adherence of neutrophils to endothelium at sites of acute

inflammation. The molecular cloning of GMP- 140 has revealed a modular protein consisting of an

amino terminal lectin-like domain (calcium-dependent), an EGF motif, 9 tandem consensus repeats

related to those found in complement-binding proteins, a transmembrane domain, and a short

cytosolic tail (9). Two other proteins, ELAM- 1 and gp90 MEL, share a high degree of sequence

homology with GMP-140 and the same basic domain structure, but are distinctive in the number of

consensus repeats (lo- 12). ELAM- 1 is a cytokine-inducible molecule on endothelial cells which

Abbreviations: Au, Arthrobacer ureafaciens ; C-type, calcium-type; Cp, Cfostridium perfringens HEV, high endothelial venule; NeuAc, 5 N-acetylneumminic acid; NDV, Newcastle disease virus; Vc, Vibrio choler .


mediates neutrophil attachment (10, 13). 1~ viva, this molecule may promote neutrophil adherence

to endothelium during the later stages of certain acute inflammatory reactions. gp90MEL, a

constitutive cell surface protein on murine lymphocytes identified by the MEL- 14 mAb, is

responsible for the organ-selective adherence (“homing”) of lymphocytes to the endothelial cells of

high endothelial venules (HEV) in lymph nodes, an adhesive event that initiates extravasation

during the process of lymphocyte recirculation (14- 16). These three proteins, each of which is

involved in a highly selective cellular interaction within the blood vascular compartment, comprise

a new family of cell-cell adhesion molecules designated as the LEC-CAMS (17) or selectins (8).

In the case of gp9OMEL, the lectin domain has been directly implicated in the adhesive

function of the receptor. Specific carbohydrates have been identified that interact with this domain

in a calcium-dependent manner and competitively inhibit lymphocyte attachment to HEV ( 18,22).

Furthermore, the HEV-ligand for gp’)@ELdemonstrates a functional requirement for sialic acid as

follows: 1) lymphocyte attachment to HEV is blocked by pretreatment of the HEV with sialidases

or with the sialic acid specific lectin, Limaxflavus agglutinin (23-25); 2) the binding of a

recombinant form of gp90MELto HEV in tissue sections is abrogated by exposure of the HEV to

either sialidases or Limax agglutinin (25); and 3) desialylation of the isolated HEV-ligand for

gp90MELprevents its binding to gp90 MEL (Imai, Singer, Fennie, Lasky, and Rosen, submitted).

In view of the strong structural similarities between gp90 MELand the other members of the LEC-

CAM family, we investigated whether a sialic acid requirement also exists for the binding of GMP-

I40 to its ligand.

MATERIALS AND METHODS

Cell Culture and Cell Preparation. The HL60 human promyelocytic cell line was cultured in RPMI- 1640, supplemented with 10% fetal bovine serum, 100 U/ml penicillin and 100 ug/ml streptomycin. All culture reagents were obtained from the UCSF Cell Culture Facility. The cells were washed 3 times by centrifugation (150 x g) and resuspended in Dulbecco’s PBS at a concentration of 3 x 107 cells/ml. The cells were fixed by the addition of an equal volume of 1% parafotmaldehyde (EM Sciences, Cherry Hill, NJ) in 0.1 M sodium cacodylate, pH 7.3. After 1 hr at 40 C, the cells were washed 3 times and suspended in PB buffer (0.0 1 M Na,H;P04,0.1 SM NaCl, 5mM CaC12, pH 6.5) or PBS. Polymorphonuclear neutrophils were isolated from fresh heparinized human blood by density gradient centrifugation (Mono-Poly Resolving Media, Flow Laboratories, McLean, VA) following the manufacturer’s instructions. The neutrophils were fixed according to the procedure of Hamburger and McEver (7) and washed 3 times.

Enzyme Treatment. Fixed HL60 cells (150 ul at 3 x 107 cells/ml in PB) were treated with either Arthrobacter ureafaciens neuraminidase (Calbiochem, Inc., San Diego, CA) or Clostridium perftingens neuraminidase (Type X, affinity purified, Sigma Chemical Co., St. Louis, MO) at a final concentration of 0.1 U/ml (international units) or with Vibrio cholera neuraminidase (Gibco, Grand Island, NY) at a final concentration of 250 U/ml (according to the manufacturer’s definition of a unit) and incubated at 37O C for 1 hr. Fixed neutrophils were treated with C. perfringens neuraminidase under the same conditions used for HL60 cells. In some experiments, HL60 cells were exposed to Arthrobacterureafaciens neumminidase (0.1 U/ml final ) in PBS instead of PB. Newcastle disease virus (NDV) sialidase was prepared and assayed as described by Paulson et al. (26). Fixed HL60 cells at 4 x 107/ml in PBS were mixed with NDV sialidase (0.2 U/ml final) and incubated at 370 C for 1 hr. After exposure to an enzyme, the cells were washed 3 times with PBS before use in the platelet binding assay. Control samples were treated in parallel without exposure to enzymes.

Platelet isolation. Blood from normal human donors was anticoagulated with ACD (85 mM sodium citrate, 7 1 mM citric acid, 111 mM D-glucose) at a 6: 1 (v/v) mtio. After centrifugation at

1350


200 x g for 20 min at 22O C, the platelet-rich plasma was pooled and fractionated by gel filtration at 220 C with Sepharose CL-2B (Sigma Chemical Co.), equilibrated with HEPES buffer (5.5 mM D-glucose, 137 mM NaHC03, 2.68 mM KCl, 20.0 mM HEPES, 0.35% human albumin), pH 7.4. Platelets were stimulated with human thrombin (T-6759, Sigma Chemical Co.) at a final concentration of 0.15 U/ml for a minimum of 10 min and were used in the adhesion assays within 30 min of activation. In some cases, the isolated platelets were already activated as determined morphologically and were used without thrombin treatment. Similar results were obtained with thrombin-activated platelets and those activated during isolation.

Platelet Rosetting Assay. The procedure for platelet rosetting was slightly modified from the protocol of Larsen et al. (6). For each sample, 50 ul of cells (3 x 107 cells/ml in PBS) were mixed with an equal volume of activated platelets (2 x IO* cells/ml) in a microfuge tube and allowed to react for 20 min at 220 C without agitation. Each treatment was carried out in 3-4 replicates. The cell suspensions were fixed by the addition of glutaraldehyde to a final concentration of 1.25%. Platelet rosetting of cells was determined by Hoffman modulation contrast microscopy. Cells with two or more adherent platelets were scored as rosetted. Samples (l-2) of cells (SO-loo) were evaluated from each replicate tube. Results are indicated as percent rosetted cells based on the means of the 3-4 replicates. SEMs denote the statistical variability. The effect of calcium chelation was determined by exposing activated platelets to 10 mM EGTA for 10 min prior to mixing with the HL60 cells or neutrophils

Flow Cytometry Analysis. Fixed HL60 cells prepared as above (3.3 x 107/ml) were exposed to NDV sialidase (0.2 U/ml) in PBS, Arthrobacter ureafaciens neuraminidase (0.1 U/ml final) in PBS, or to PBS alone and incubated for 1 hr at 37O C. The cells were washed 3 times in PBS containing 1% fetal bovine serum (FBS). The washed cells were exposed to CSLEX- 1 mAb (27) or to VIM-2 mAb (28), each diluted as an ascites fluid l/50 in PBS-l% FBS. Incubation was for 30 min at 4O C. The cells were sequentially exposed to to biotinylated-horse anti-mouse IgG (Vector Laboratories, Burlingame, CA) and FITC-extrAvidin (Sigma Chemical Co.) according to the manufacturers’ recommendations. Fluorescence histogram profiles were determined with a Becton-Dickinson FACScan II flow cytometer. The background level of fluorescence was defined with no primary antibody added,

RESULTS

A well-described phenomenon is the binding of activated platelets to neutrophils or the

HL60 promyelocytic cell line (6, 7,29). Two groups (6, 7) have shown that the rosetting is

calcium-dependent and appears to be mediated almost entirely by GMP- 140. We found that fixed

HL60 cells or neutrophils were rosetted by thrombin-activated platelets at levels comparable to

those previously reported. In further correspondence, the interaction was completely inhibited by

the inclusion of EGTA in the rosetting step (Figs. l-3), and rosetting by resting platelets was

greatly diminished (lo- 15% rosetting, not shown). To determine the role of sialic acid on HL60

cells in rosetting by activated platelets, we treated fixed HL60 cells with broad-spectrum bacterial

sialidases prior to the rosetting step. Each of the three sialidases tested, including an affinity-

purified sialidase from Clostridium perfringens, abolished the capacity of HL60 cells to be rosetted

(Fig. 1). In the case of the Arthrobacterureahciens sialidase, inclusion of the substrate sialyl-

lactose during the enzyme incubation, substantially preserved the rosetting with HL60 cells, thus

demonstrating that the sialidase itself, rather than a contaminating activity, was responsible for the

diminished rosetting (data not shown). Sialidase treatment of viable HL60 cells resulted in a 60-

70% reduction in platelet resetting in contrast to the complete inhibition observed with fixed cells

(not shown). This incomplete effect may reflect the ability of metabolically active cells to restore

critical sialylated residues during the 20 min incubation step with the platelets. To determine

whether neutrophils were also susceptible to sialidase treatment, fared neutropbils were treated

1351


with the Closttidium perfringens enzyme. As was the case for HL60 cells, sialidase treatment of

neutrophils eliminated rosetting (Fig. 2).

The bacterial sialidases employed were all broad spectrum enzymes which can cleave a2,3,

a2,6, and a2,8 linked NeuAc from glycoconjugates (30). In an attempt to define a particular

linkage of sialic acid that might be important in platelet rosetting, we employed the Newcastle

disease virus (NDV) sialidase, which cleaves a2,3 and a2,8 but not a2,6 linked NeuAc (26, 31).

As shown in Figure 3, treatment of HL60 cells with a high concentration of NDV sialidase had no

effect on their rosetting by platelets, whereas the cells were still susceptible to the broad spectrum

Atihrobacter ureafaciens sialidase. To confirm that the NDV sialidase was, in fact, active upon

cell surface sialyloligosaccharides of HL60, we tested the effect of this enzyme on two independent

epitopes defined by the monoclonal antibodies, CSLEX-l(27) and VIM-2 (32). CSLEX- 1

recognizes the sialylated Lewis X structure:

NeuAca2,3Galp1,4[Fucal,3]GlcNAcp1,3Galp-

whereas VIM-2 binds to the heptasaccharide:

NeuAca2,3Gal~l,4GlcNAc~l,3Ga1[31,4[Fucal,3]GlcNAc~l,3Gal-

For both antibodies, binding shows an absolute dependence on the presence of a2,3 linked NeuAc

(27,28). Neither of these carbohydrates epitopes is itself the ligand for GMP- 140, since these

antibodies have no effect on rosetting of HL60 or neutrophils by activated platelets (data not

shown). As shown in Fig. 4, treatment of HL60 cells with either NDV sialidase or Arthrobacter

ureafaciens sialidase under the same conditions employed for the rosetting experiment (Fig. 3)

completely eliminated the epitopes defined by both CSLEX- 1 and VIM-2.

80

1

0 1

, Siatidnses 0

Control Sialidax (Cp) EGl-A

Treatment 2

Treatment

Figure 1. Effects of sialidases or EGTA on the resetting of HL60 cells by platelets. Fixed HL60 cells were treated with Arthrohxter ureafaciens (Au) sialidase. Closttidium verfrinens (CD) sialidase, Vibrio cholera (Vc) sialidase or control‘buffer and thek tested for r&.etting%y ac&ted platelets. HL60 cells were also tested for rosetting in the presence of EGTA (10 mM). The means and SEMs were computed on the basis of three independent replicates.

Fimre 2. Effects of Clostridium pcrfhgem sialidase or EGTA on the resetting of neutrophils by platelets. Fixed neutrophils were treated with Cbstkiium perftingens (Cp) sialidase or control buffer and then tested for rosetting with activated platelets. Neutrophils were also tested for rosetting in the presence of EGTA (10 mM). The means and SEMs were computed on the basis of three independent replicates.

1352


COlltd NDV Sialidase Au Sialidase

Treatment

EGTA

Figure 3. Effects of NDV sialidase and Arthrobacter ureafkiens sialidase on the rosetting of HL60 cells by platelets. Fixed HL60 cells were treated with NDV sialidase or Arthrobacterureaf (Au) sialidase in PBS as described in MATERIALS AND METHODS. A control sample was sham treated with PBS alone. Rosetting of control cells was also determined in the presence of EGTA (10 mM). The means and SEMs were computed on the basis of four independent replicates. The incomplete effect of the Arthrobacter is probably attributable to the higher pH of PBS vs. the PB buffer used for the data in Figure 1.

DISCUSSION

Our previous work has shown a functional requirement for sialic acid on the HEV-ligand

for gp9WEL, a lymphocyte homing receptor. It is not known in this system whether sialic acid

contributes directly to a recognition determinant of the ligand or modulates the conformation or

activity of a distal recognition element. However, the known capacity of different forms and

linkages of sialic acid to encode highly specific recognition elements (33-35) argues for the former

possibility. A notable feature of the lectin domain of gp9wEL (11) and that of its closely related

human homologue (22,36,37) is a very large number of lysine residues, i.e., 16 in the case of the

mouse molecule and 12 in the case of the human molecule. This high concentration of positive

charges within the lectin domain is compatible with the sialic acid requirement for the HEV-ligand

as well as with the finding that all of the sugars known to mimic the ligand are anionic (2 1). In

contrast, other receptors containing the C-type lectin domain (e.g., chicken, rat, and human hepatic

lectins, mannose-binding receptor, etc.) contain an average of 5 lysine residues and, insofar as is

known, recognize only neutral sugars (38). It is, thus, notable that the other two members of the

LEC-CAM family also exhibit a large number of lysine residues in their lectin domains (i.e., 14 for

GMP- 140 and 10 for ELAM- I), suggesting that their cognate ligands might be anionic, a

prediction which is supported for GMP- 140 by the finding that it can bind to several anionic

polysaccharides (39). The high degree of sequence homology among the lectin domains of the

LEC-CAMS (60-700/o) lends further support to the possibility of related sugar-binding specificities.

In the present study, we provide direct evidence for a similarity by showing that sialidase treatment

of HL60 cells and neutrophils prevents rosetting by activated platelets, an intelaction attributed to

1353


CSLEX- 1 VIM-2

D

Relative Fluorescence

Figure 4. Effects of NDV sialidase and Arthrobacter ureafkiens sialidase on the expression of the epitopes for CSLEX-1 and VIM-Z. Fixed HL60 cells were treated with NDV sialidase, Admbacter urediciens sialidase or PBS alone under the identical conditions used in Figure 3. The binding of CSLEX- 1 and VIM-2 mAbs was determined by indirect bnmunofluorescence using flow cytometry. The y-axis indicates the relative cell frequencies and the x-axis indicates relative fluorescence intensity. A) no primary antibody; B) control-treated cells; C) cells treated with Arthrohcter ureafaciens (Au) sialidase; and D) cells treated with NDV sialidase. The modal fluorescence values were: no primary antibody (3); CSLEX-control(328); CSLEX-Au (5); CSLEX-NDV (7); VIM-2 control (215); VIM-2 Au (3); VIM-2 NDV (6).

GMP- 140. As three different broad spectrum sialidases produce this effect (on HL.60 cells) and

competitive inhibition by sialyl-lactose blocks the inactivation, a sialic acid requirement is clearly

indicated. The use of the linkage-specific sialidase of NDV has provided information about the

13.54


possible nature of the linkage of the essential sialic acid on HL60 cells. Under conditions of

enzyme treatment that completely eliminated two epitopes dependent on a2,3 linked NeuAc

(CSLEX- 1 and VIM-2), there was no detectable effect on the ligand activity for GMP- 140.

NeuAca2,3Gal on isolated oligosaccharides or on cell surface associated glycoconjugates is

completely susceptible to hydrolyis by NDV sialidase, whereas the NeuAca2,6Gal and

NeuAca2,6GalNAc linkages are completely resistant (26). From the activity of NDV sialidase on

the NeuAca28NeuAc linkage in a milk oligosaccharide, it is strongly suspected that the enzyme

would also cleave a2,8 linked NeuAc at the cell surface (26). If it is assumed that the ligand for

GMP- 140 does not involve an unusual form or linkage of sialic acid with resistance to NDV

sialidase (e.g., 4-0-acetyl NeuAc [30]), our results argue for the existence of a critical a2,6 linked

NeuAc. Further biochemical chamcterization of the ligand will allow evaluation of this prediction.

A question of considerable additional interest is whether the sialic acid requirement extends to

ELAM- 1 and thus, is general for all members of the LEC-CAM family. If so, the ligands for

ELAM- 1 and GMP- 140 would be predicted to involve distinct sialyloligosaccharides, since

functional evidence suggests that the ligands are independent and nonoverlapping (40).

The results of this study may shed light on the interaction between certain tumor cells and

platelets. It has long been known that some tumor cells induce platelets to aggregate (4 1). Interest

in this phenomenon has been considerable, since the formation of tumor-platelet emboli in the

blood may be an important determinant of metastatic arrest and growth of the tumor cells at

secondary sites (41,42). For certain tumor cells, a correlation exists among the degree of cell

surface sialylation, their ability (or that of shed factors derived from them) to aggregate platelets,

and their metastatic potential in animals (42,43). Recently, it has been shown that sialic acid

moieties on the surface of a metastatic colon adenocarcinoma cell line (mouse) are directly involved

in platelet co-aggregation (44). Since, as reported here, GMP- 140 is likely to mediate platelet-

neutrophil binding through recognition of a sialylated structure, the participation of GMP- 140 in

tumor cell-induced platelet aggregation warrants investigation.

Acknowledgments- We thank Drs. David Brown and Israel Charo of COR Therapeutics for their advice on the isolation of platelets. We thank Dr. Walter Knapp (Institute of Immunology, University of Vienna, Vienna) and Dr. Paul Terasaki (Dept. of Surgery, UCLA School of Medicine, Los Angeles) for their gifts of VIM-2 and CSLEX- 1, respectively. We are grateful to Dr. David True and Supama Bhattacharya for culturing of NDV. The research was supported by NIH Grants: MAC P60 AR20684 and GM23547 to SDR and CA32826 to BAM.

1. 2.

3.

4.

5. 6.

7.

REFERENCES

McEver, R. P., and Martin, M. N. (1984) J. Biol. Chem. 259, 9799-9804. Stenberg, P. E., McEver, R. P., Shuman, M. A., Jacques, Y. V., and Bainton, D. F. (1985) J. Cell Biol. 101, 880-886. Berman, C. L., Yeo, E. L., Wencel-Drake, J. D., Furie, B. C., Ginsberg, M. H., and Furie, B. (1986) J. Clin. Invest. 78, 130-137. McEver, R. P., Be&stead, J. H., Moore, K. L., Marshall-Carlson, L., and Bainton, D. F. (1989) J. Clin. Invest. 84, 92-99. Bonfanti, R., Furie, B. C., Furie, B., and Wagner, D. D. (1989) Blood 73, 1109-l 112. Larsen, E., Celi, A., Gilbert, G. E., Furie, B. C., Erban, J. K., Bonfanti, R., Wagner, D., and Furie, B. (1989) Cell 59, 305-312. Hamburger, S. A., and McEver, R. P. (1990) Blood 75, 550-554.

1355


8.

9. 10.

11.

12. 13.

14. 1.5. 16.

17. 18. 19. 20.

21.

22. 23.

24.

25. 26.

27.

28.

29.

30.

:;:

33. 34.

2:

37.

38. 39.

40.

41.

1::

44.

Geng, J. G., Bevilacqua, M. P., Moore, K. L., McIntyre, T. M., Prescott, S. M., Kim, J. M., Bliss, G. A., Zimmerman, G. A., and McEver, R. P. (1990) Nature 343, 757-760. Johnston, G. I., Cook, R. G., and McEver, R. P. (1989) Cell 56, 1033- 1044. Bevilacqua, M. P., Spengeling, S., Gimbrone, M. A., Jr., and Seed, B. (1989) Science 243, 1160-l 165. Lanky, L. A., Singer, M. S., Yednock, T. A,, Dowbenko, D., Fennie, C., Rodriguez, H., Nguyen, T., Stachel, S., and Rosen, S. D. (1989) Cell 56, 1045-1055. Siegelman, M. H., Van, de, Rijn, M, and Weissman, I. L. (1989) Science 243, 1165 1172. Bevilacqua, M. P., Pober, J. S., Mendrick, D. L., Cotran, R. S., and Gimbrone, M. A. J. (1987) Proc. Natl. Acad. Sci. USA 84, 9238-9242. Gallatin, W. M., Weissman, I. L., and Butcher, E. C. (1983) Nature 303, 30-34. Geoffroy, J. S., and Rosen, S. D. (1989) J. Cell Biol. 109, 2463-2469. Watson, S. R., Imai, Y., Fennie, C., Geoffroy, J. S., Rosen, S. D., and Lasky, L. A. (1990) J. Cell Biol. 110, 2221-2229. Stoolman, L. M. (1989) Cell 56, 907-910. Stoolman, L. M., Tenforde, T. S., and Rosen, S. D. (1984) J. Cell Biol. 99, 15351540. Yednock, T. A., Stoolman, L. M., and Rosen, S. D. (1987) J. Cell Biol. 104, 713-723. Yednock, T. A., Butcher, E. C., Stoolman, L. M., and Rosen, S. D. (1987) J. Cell Biol. 104, 725-731. Imai, Y., True, D. D., Singer, M. S., and Rosen, S. D. (1990) J. Cell Biol. 111, 1225- 1232. Bowen, B. R., Nguyen, T., and Lasky, L. A. (1989) J. Cell Biol. 109, 421-427. Rosen, S. D., Singer, M. S., Yednock, T. A., and Stoolman, L. M. (1985) Science 228, 1005- 1007. Rosen, S. D., Chi, S.-I., True, D. D., Singer, M. S., and Yednock, T. A. (1989) J. Immunol. 142, 1895-1902. True, D. D., Singer, M. S., Lasky, L. A., and Rosen, S. D. J. Cell Biol., in press Paulson, J., Weinstein, J., Dorland, L., van Halbeek, H., and Vliegenthart, J. F. G.( 1982) J. Biol. Chem. 257, 12734-12738. Fukushima, K., Hirota, M., Terasaki, P. I., Wakisaka, A., Togashi, H., Chia, D., Suyama, N., Fukushi, Y., Nudelman, E., and Hakomori, S.-I. (1984) Can. Res. 44, 5279-5285. Majdic, O., Bettelheim, P., Stockinger, H., Aberer, W., Liszka, K., Lutz, D., and Knapp, W. (1984) Int. J. Can. 33, 617-623. Jungi, T. W., Spycher, M. O., Nydegger, U. E., and Barandun, S. (1986) Blood 67, 629- 636. Schauer, R. (1982) Adv. Carb. Chem. Biochem. 40, 131-233. Dtzeniek, R. (1973) Histochem. J. 5, 27 l-290. Macher, B. A., Buehler, J., Scudder, P., Knapp, W., and Feizi, T. (1988) J. Biol. Chem. 263, 10186-10191. Schauer, R. (1985) Trends Biochem. Sci. 10,357-360. Sharon, N., and Lis, H. (1989) Science 246, 227-234. Wiley, D. C., and Skehel, J. J. (1987) Ann. Rev. Biochem. 56, 365-394. Siegelman, M. H., and Weissman, I. L. (1989) Proc. Natl. Acad. Sci. USA 86,5562- 5566. Tedder, T. F., Isaacs, C. M., Ernst, T. J., Demetri, G. D., Adler, D. A., and Disteche, C. M. (1989) J. Exp. Med. 170, 123-133. Drickamer, K. (1988) J. Biol. Chem. 263, 9557-9560. Skinner, M. P., Foumier, D. J., Andrews, R. K., Gorman, J. J., Chesterman, C. N., and Bemdt, M. C. (1989) Biochem. Biophys. Res. Commun. 164, 1373-1379. Gamble, J. R., Skinner, M. P., Bemdt, M. C., and Vadas, M. A. (1990) Science 249, 414-417. Gasic, G. J., Gasic, T. B., Galanti, N., Johnson, T., and Murphy, S. (1973) Int. J. Cancer 11, 704-718. Karpatkin, S., and Pearlstein, E. (1981) Ann. Intern. Med. 95,636-641. Pearlstein, E., Salk, P. L., Yogeeswaran, G., and Karpatkin, S. (1980) Proc. Natl. Acad. Sci. USA 77, 4336-4339. Kijima-Suda, I., Miyazawa, T., Itoh, M., Toyoshima, S., and Osawa, T. (1988) Can.Res. 48, 3728-3732.

1356

requirement for sialic acid on neutrophils in a gmp-140 (padgem) mediated adhesive interaction with...

Documents