Sulfated Glycolipids Are the Platelet Autoantigens for Human Platelet ...

THEJOURNALOF

BIOLOGICAL CHEMISTRY

Vol. 266, No. 23, Issue of August 15, pp. 15414-15419,1991 Printed in CJ.S. A.

Q 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

Sulfated Glycolipids Arethe Platelet Autoantigens for Human Plateletbinding Monoclonal Anti-DNA Autoantibodies* (Received for publication, January 14, 1991)

Hiroshi MurakamiS, Zamas Lam§, Barbara C. FurieS, VernonN. Reinhold§, Taro AsanoS, and Bruce FurieS From the $Center for Hemostasis and Thrombosis Research, Division of Hematology-Oncology, New England Medical Center and the Department of Medicine and Department of Biochemistry, Tufts University Schoolof Medicine, Boston Massachusetts 021 021 1I and the §Department of Nutrition, Harvard School of Public Health, Boston Massachusetts 15

Thehumanmonoclonal autoantibody HF2-1/17, produced bya human-human hybridomaderived from lymphocytes of a lupus patient with thrombocytopenia, reacts with singlestranded DNA and platelets. To determine the chemical nature of the autoantigen against which thisantibody is directed on platelets, this platelet antigen was purified by the lipid extraction of sonicated platelets, DEAE-Sephadex chromatography, and high performance liquid chromatography. The purified glycolipids, a trace component in platelets,demonstrated high reactivity with the HF2-1/17 antibody using a competition enzyme-linked immunosorbent assay system or immunostaining of thin layer chromatograms. The purified glycolipids co-migrated with bovine sulfatides by thin layer chromatography. The purified glycolipids contain sulfate and galactose but not sialic acid or phosphate.Fast atom bombardment-mass spectrometry revealed these sulfatides to be sulfated monohexyl ceramides. The dominant species has a molecular weight of 794 while a minor form has a molecular weight of 812 due to an extrahydroxyl group and loss of a double bond. These results indicate that the platelet autoantigen against which the human monoclonal anti-DNA antibody is directed represents a family of novel monogalactosyl sulfatides.

Autoimmune thrombocytopenia is a common disorder associated with systemic lupus erythematosus and idiopathic thrombocytopenia purpura (ITP).’ In these syndromes, autoantibodies bind to a platelet surface antigen leading to the rapid clearance of the platelet-autoantibody complex. In addition to the immunoregulatory pathology that is the basis for excess autoantibody production, the chemical nature of the platelet autoantigen hasyet to be clarified. While the sera of some patients with ITP contain autoantibodies reactive with the major platelet glycoproteins (1, 2), including glycoprotein Ib (3),glycoprotein IIb/IIIa (4), and glycoprotein IIIa (5),proteins were not identified as autoantigens in the sera of many ITP patients studied (6). Indeed, both neutral and acidic glycolipids havebeen implicated in autoimmune thromof platebocytopenia (7-9). To characterize further the nature let autoantigens in this disorder, we have studied platelet-

* This work was supported by Grants A119794 and HL42442 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. The abbreviations used are: ITP, idiopathic thrombocytopenic purpura; ELISA, enzyme-linked immunosorbent assay; HPLC, high performance liquid chromatography.

binding monoclonal lupus autoantibodies derived from human hybridomas. Human and murinemonoclonal anti-DNA autoantibodies are polyspecific in as much as they bind to DNA, synthetic nucleotides (10, l l ) , phospholipids (lo), the cytoskeletal protein vimentin (12), rheumatoid factor (13), cellsurface proteins of lymphoblastoid cells and erythrocytes (14), and platelets (15). These observations indicate that the autoantigen in autoimmune diseases could be a common structural motif which is broadly distributed in nature. We have previously partially characterized the platelet binding properties of platelet-binding lupus anti-DNA autoantibodies(16). These antibodies are apparently specific for a platelet antigen that does not involve DNA, protein, phospholipid, or sialic acid. The purpose of the current study was to purify and structurally define the plateletautoantigen against which these autoantibodies are directed. In this article, we demonstrate thatHF2-1/17, a prototype of the 16/6 idiotype family of monoclonal anti-DNA autoantibodies, binds to human platelet-sulfated glycolipids. A preliminary account of this work has appeared (7). MATERIALS AND METHODS’

RESULTS

Extraction of the Platelet Antigen from Platelet Membranes-Platelets (200 units) were sonicated and the lipids extracted with chloroform/methanol/waterand with chloroform/methanol/sodium acetate as described under “Materials and Methods.” The lipid preparation was analyzed for the presence of the platelet antigen using a solid phase ELISA system. As shown in Fig. 1,HF2-1/17, an IgM, bound to lipid derived from the platelet lipid extract that was used to coat the ELISA plate. In contrast, normal polyclonal IgM demonstrated minimal binding to thislipid fraction. These results suggest that the antigen against which HF2-1/17 is directed is present in the platelet lipid extract. neutral and Binding of HF2-1/17 toGlycolipids-Both acidic glycolipids extracted from platelets and separated by DEAE-Sephadex chromatography were evaluated for their ability tointeract with HF2-1/17 autoantibodies using a competitive ELISA. Platelet lipid extracts were coated onto wells in a 96-well plate. As shown in Fig. 2, the acidic glycolipids in the solution phase inhibited HF2-1/17 binding to the solid phase; 50% inhibition was observed at about 50 pg/ ml of the acidic glycolipids. In contrast, neutral glycolipids

* Portions of this paper (including the “Materials and Methods” and Figs. 4,5,9, and10) are presented in miniprint at theend of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photocopies are included in the microfilm edition of the Journal thatis available from Waverly Press.

15414

Sulfatides Autoantigens as Platelet

1 S

0.6

d

1I

/d

I

/

Antlbody

[pg/ml]

Wn-1

FIG. 1. Binding of HF2-1/17 to platelet lipid extract. Platelet lipids were extracted using a mixture of chloroform/methanol/ water. After drying, the lipid was redissolved in methanol and 100 pg applied to each microtiter well. HF2-1/17 ( 0 )or polyclonal human IgM ( O ) ,100 pl at the indicat.ed concentration, was incubated in the well. Bound antibodywas detected using goat peroxidase-labeled antihuman IgM as a second antibody and o-phenylenediamine as substrate.

E

0.4

2

0.3

15415

P U ~

FIG. 3. Binding of IgM antibodiesto acidic glycolipids. Acidic glycolipids ( 2 pg/ml) were coated onto 96-well plastic plates. Antibody (100 pl) was added to thewell, and allowed to incubate for 90 min at 23 "C. HF2-1/17,0 HF3-16/6,0, HF2-18/2,0; HF2-1/ 13b, . ;HF6-21/28, A; HF9-11/3, A; polyclonal IgM, X.

1

2 3 4 5 6213 4 5 6

FIG. 6. Thin layer chromatography of the HPLC-purified fractions. Lane 1, glycolipid standards (top t o hottorn: bovine sulfatides, GM:,, asialo-GM,, GM,, GD,,, and G T d ; lane 2; PI; lane 3, 1'2; lane 4 , P:,; lane 5, P,; lane 6, P,. Lipids were visualized with: A, nnaphthol; R, Phospray.

The acidic glycolipids were furtherfractionated by thin layer chromatography (Fig. 4). The glycolipids, visualized with FIG. 2. Binding of HF2-1/17 to the acidic glycolipid frac- a-naphthol, contained multiple components with RF values tion. The platelet lipid extracts were attached to the solid phase. In ranging from 0.52 to 0.67. By comparison, a bovine hematoside the experiment illustrated, the aqueous phase of the crude platelet standard had an R p of 0.60 and a bovine sulfatide standard extract (0) or the acidic glycolipid fraction that bound to the DEAE- had an R p of 0.71. A parallel lane of the thin layer chromaSephadex column (0)inhibited the bindingof HF2-1/17 to the solid phase containing acidic glycolipids. Neither the organic phaseof the togram was stained with HF2-1/17 and developed with peroxidase-labeled anti-goat human IgM antibodies to identify crude lipid extract (0)nor the neutral glycolipids that failed to bind to DEAE-Sephadex (B)inhibited the binding of HF2-1/17 to the the glycolipid fractions that were reactivewith HF2-1/17. solid phase. Two bands ( R p 0.71 and 0.63-0.67) were reactivewith the antibody (Lane B ) . The component with the highest R p coshowed no evidence of antibody interaction using the commigrated with a bovine sulfatide standard. petitive ELISA. Furthermore, only the aqueous phase and notThe acidic glycolipids were separated by HPLC usinga the organic phase of the original lipid extract demonstrated silica gel column derivatized with -NH2. The chromatogram reactivity with the HF2-1/17. Theseresultsindicatethat (Fig. 5A), monitored a t 214 nm, contained a major peak, Po, acidic glycolipids in the aqueous phase of the platelet lipid which eluted a t 3 min, followed by a series of small peaks extract contain the platelet autoantigen. eluting from 9 to 12 min, designated Ps (Fig. 5B, inset), P1, Wehavepreviously shownthat16/6 idiotype-positive P2, P3, P,, and Ps. To correlate these peaks with the compomonoclonalIgMs, HF2-1/17, HF2-18/2, HF2-1/13b, and nents visualized by thin layer chromatography, each of the HF3-16/6 bind to platelets, but the 16/6 idiotype-negative peaks was chromatographed as before by thin layer chromaHF9-11/3 did notbind to platelets (16). To evaluate the tography and the chromatogram visualized with a-naphthol specificity of the interaction of these antibodies with the (Fig. 6A). PI corresponded to the component with an R p of acidic glycolipids, a direct binding ELISA was employed in 0.67; Pe to the component with an R p of 0.64; P:,, 0.60; P1, which the acidic glycolipid fraction was coated on the solid 0.60; P5,0.60. Staining of the chromatogram to detectphosphase. The platelet-binding IgM autoantibodies HF2-1/17, phorus indicated the presence of phospholipids in PI and P, HF2-18/2, HF2-1/13b, and HF3-16/6 bound to the acidic (Fig. 6B). glycolipids (Fig. 3). These interactions were saturable, with Characterization of the Platelet Antigen-The interaction half-maximal binding a t about 10 pg/ml of antibody. In con- of HF2-1/17 with thepurified glycolipids was analyzed using trast, HF6-21/28, a platelet-binding IgM that does not exa competition ELISA. HF2-1/17 was incubated with each of press the dominant 16/6-idiotype, did not bind significantly the glycolipids, then the antibody was allowed to react with t o acidic glycolipids. In addition, HF9-11/3 and polyclonal acidic glycolipids coated on thesolid phase. As shown in Fig. IgM, which do not bind to platelets, did not bindsignificantly 7, Ps was the only significant inhibitor. Inhibition of 50% was t o acidic glycolipids. These results indicate that the antibodies observed a t 5 pg/ml. In contrast, P:),P4,and Ps did not bind expressing the 16/6idiotype bind specifically to acidic glyco- to HF2-1/17. PI and Pr showed some inhibition, but demonlipids in contrast to otherIgMs. strated significantly lower antigenicity. These results allow 0.1

1

lnhlbltor

10

[pg/rnl]

100

1m

Sulfatides as Platelet Autoantigens

15416

phorus demonstrated that neither the Ps fraction, thebovine sulfatide standard, nor hematoside standard containedphosphate. Using the thiobarbituric acid assay, no sialic acid was detectable in Ps. However, sulfate was detected in Ps using the Azure-A assay. Carbohydrate analysis using sodium bisg ulfide-impregnated TLC confirmed thepresence of galactose i in the hydrolyzed sample of Ps (Fig. 9). These results indicate that Ps is a sulfated glycolipid containing galactose. Structural Analysis-Ps was directly analyzed by negative ion fast atom bombardment-mass spectrometry using triethanolamine matrix. The mass spectrum showed a prominent 0.0 ! 1 10 100 loo0 pseudo-molecular ion (M-H)-, m/z 794 (Fig. lOA), which on Inhibitor [pglrni] acetylation was found to have the addition of 5 acetate units to give m/z 1004 (Fig. 10B). A minor ion m/z 812 (30% compared to m/z 794) on acetylation was converted to m/z 1064, indicating six sites of acetyl attachment. These data suggested that the m/z 812 species differs from the m/z 794 species by one extra hydroxyl group and the absence of a double bond. The spectrum of the acetylated material also showed a number of minor ions centereda t m/z 1004 with 14 atomic mass unitdifferences. Similarly, the mass spectrum of the methylated sampleshowed a prominent ion m/z 878 (Fig. lOC) which corresponded to 6 units of methyl addition to the O .. m/z 794 species while the m/z 812 species was shifted up by 1 10 1w 7 methyl groups to m/z 910. Inhibitor [unillml] The mass spectrumof the methylated product treated with FIG. 7 . Binding of purified lipid fractions to HF2-1/17.The binding of purified glycolipids to HF2-1/17 was studied in a compe- mild acid showed a small molecular ion m/z 878 with a new tition ELISA. Acidic glycolipids were coated on the solid phase. The ion m/z 798. The loss of 80 atomic mass units from m/z 878 inhibition of the bindingof HF2-1/17 to thesolid phase was measured was consistent with the loss of an acid-labile sulfate group, by preincubation of HF2-1/17 with the purified glycolipids isolated further confirming that thisglycolipid is a sulfatide. by HPLC fractionation. A, Ps, 0. I?, Ps,0 PI,0 P,, . ;Ps, & P,, A; The differences between the number of methyl and acetyl P:,, 0. Because of insufficient quantities of' purified lipids to allow the groups added is due to N-methylation of the amide nitrogen determination of mass, a unit was defined as the amount of purified procedure. As neither methylationnor lipid obtained from 100 pcg of the acidic glycolipids purified by DEAE- during the methylation acetylation derivatize the sulfatemoiety, the nativesulfatides Sephadex chromatography. Given the similar extinction coefficients of glycolipids a t 214 nm, the specific activity of Ps is greater than contain 5 hydroxyl groups. 100-fold that of the other lipids tested. Fast atom-bombardment-mass spectrometry analysis of bovine sulfatide standardsshowed a molecular ion m/z 888 (Fig. 1OD) with the associated sphingosineheterogeneity. PeraceA B tylation of this material shifted the molecular ionto m/z 1056, which indicated a n addition of 4 acetyl groups. Permethylation of this material shifted themolecular ion by 5 additions of methyl groups to m/z 958. These data suggested that the bovine sulfatides possess4hydroxyl sitesand 2 units of unsaturation. A

I

'1

DISCUSSION

Autoantibodies found in the serumof patients with autoimmune disorders areknown to be reactive with a diversegroup 1 2 3 1 2 3 of molecular structures. With the preparationof monoclonal FIG. 8. Immunostaining of the purified sulfated glycolipids. mouse and human autoantibodies, it has been demonstrated A hematoside standard, a bovine sulfatide standard, and thepurified sulfated glycolipids were analyzed by thin layer chromatography using that the broad serologic specificity of serum from patients with autoimmunediseases is not necessarily due to thediverchloroform/methanol/water.The plate was developed using ""I-labroad beled HF2-1/17, then autoradiographed (Panel A ) . The plate was sity of antibodies in the sera but rather due to the then visualized with e-naphthol (Panel B ) . I, hematoside standard; distribution of antigens against which a structurally related 2, purified platelet autoantigen; 3, bovine sulfatide standard. family of antibodies are directed. The human monoclonal lupus anti-DNA autoantibody HF2-1/17 is a member of the The sequence of this antibody, us toconclude that theglycolipid Ps is the dominant platelet 1616-idiotypefamily(28). known from direct protein analysis and in part from predicantigen that binds toHF2-1/17. The thin layer chromatogram of glycolipid Ps yielded a tions from thecDNA sequence (29-31), is identical or nearly major doublet, with several bands of low intensity apparent identical with other antibodies in this family. Although this at slightly greater mobility. Ps co-migrated with the bovine antibody was initially examinedbecause of its single stranded sulfatide standard and had a mobility in excess of the hema- DNA binding properties, we have focused upon its platelet toside standard (Fig. 8). Development of the chromatogram binding properties.HF2-1/17 binds to unstimulated and stimwith ""I-labeled HF2-1/17 detected Ps under the conditions ulated platelets, and the interaction is inhibited by single employed, but the bovine sulfatide or hematoside standards stranded DNA (16). The interactionof antibody and platelets were not detected. Staining of the chromatogram for phos- is not alteredby pretreatment of the platelets with nuclease,

Sulfatides as Platelet Autoantigens

15417

atomic mass units lower than brain sulfatide; when the number of hydroxyl sites and unsaturation units are taken into account, this may correlate to 7 CH, groups. We suspect that the geometry of sulfate groups among adjacent sulfatides in the platelet plasma membrane defines the epitope for HF21/17. As the repeating phosphate structure on single stranded DNA is critical to theinteraction of this antibody with single stranded DNA (lo), the repeating sulfate structure on the plasma membrane is critical to platelet-antibody interaction. Glycolipids have been demonstratedasautoantigens in B several autoimmune diseases. Autoantibodies to aneutral glycolipid, asialo-GM, have been observed in systemic lupus erythematosis, especially in those patients with neurological symptoms (36). Antibodies against acidic glycolipids, ganglioside GM1 and asialo-GM1, have also been observed in the serum of patients with multiple sclerosis and lupus erytemaFIG. 11. Structure of the sulfatides derived from human tosus (37). Monoclonal IgM antibodies against gangliosides platelets. The platelet autoantigen includes a sulfated monogalac- (38) and sulfatides (39) of peripheral nerves have been implitose. The linkage stereochemistry, precise length of the aliphatic chains, and position of the sulfate hasnot been determined. Assuming cated in the polyneuropathy and demyelinating processes a C,, sphingosine base and a-hydroxylation of the fatty acyl group, associated with certain monoclonal gammopathies. In autoimthe structure of the platelet antigen (m/z 794) could be represented mune thrombocytopenia, however, lipid components of the by the structure B. The structure of bovine brain sulfatide is shown platelet surface have not been previously identified as targets for comparison (A); x is 2 except for those carbons adjacent to a of autoantibodies, although anti-sulfatide antibodies have single double bond. been identified in ITP sera (9, 40). The autoantigen of polyclonal anti-platelet autoantibodies trypsin, or neuraminidase. In order to determine the chemical has been elusive. In some patients, anti-platelet antibodies nature of the platelet autoantigen, we evaluated platelet frac- can be detected that are reactive with glycoprotein IIb-IIIa, tions and established that this antigen is defined by glyco- the fibrinogen receptor on theplatelet surface (4,5,40).These lipids (16). antibodies are unreactive with platelets lacking glycoprotein In the current series of experiments we have established IIb-IIIa that were isolated from patients with Glanzmanns that the platelet antigen is a platelet lipid with chemical thrombasthenia (9,40). Thisantibody specificity, found in 35 properties characteristic of acidic glycolipids. Chemical and of42 patientsinthis study, has suggested that in some physical analyses of the purified antigen revealed that this instances glycoprotein IIb-IIIa is the autoantigen. The studies antigen is a sulfatide. Sulfatides have, until recently, not been of Kunicki et al. (41) failed to demonstrate the binding of identified inhumanplatelets despite extensive studies of platelet autoantibodies to Glanzmanns platelets. Other studplatelet lipid composition and metabolism (32, 33). However, ies have implicated glycoprotein Ib, the von Willebrand factor indirect evidence for the presence of human plateletsulfatides receptor, as theplatelet autoantigen (3).Nonetheless, approxhas accrued based upon demonstration of the binding of von imately 80% of autoimmune sera areunreactive with platelet Willebrand factor and thrombospondin to acidic glycolipid membrane proteins in some studies (4). extracts from platelets (34, 35). Sulfatides are in low abunAlthough we have isolated and determined the structure of dance in platelets, and the current study represents the first the platelet component that is the antigentarget of the purification and structural characterization of platelet sulfa- monoclonal autoantibody HF2-1/17, it remains to be demtides. Despite their low abundance, a physiologicrole as onstrated thatthese sulfatides are thedominant autoantigens membrane receptors has been suggested. for all or most autoantibodies that bind to platelets. PrelimiPlatelets were fractionated and the glycolipid components nary dataindicate that polyclonal anti-platelet autoantibodies further purified and analyzed. The platelet autoantigen was in sera of various patients with autoimmune thrombocytoassociated with the acidic glycolipid extract, which was sub- penia bind to the isolated sulfatide and that this sulfatide sequently fractionated to yield homogeneous or nearly ho- preparation inhibits these antibodies from binding to platemogeneous components by HPLC. Although two phospholipid l e t ~ Further .~ study will establish whether this family of peaks retained some platelet antigen activity, on a weight sulfatides is the dominant autoantigen targetof platelet autobasis thesecomponents had minimal activity compared to the antibodies. sulfatide fraction, Ps. The microheterogeneity of Ps by thin layer chromatography likely is due to ceramide heterogeneity. Acknowledgments-We are grateful to Dr. Paul Schick for many Mass spectral analyses suggest the putative sulfatide struc- helpful discussions and toDrs. Robert S. Schwartz and David Stollar tures shown in Fig. 11. Salient features of the platelet sulfa- for their continued interest. tide, with a molecular weight of 794, include the monogalactose, a single sulfate group likely to be located at position 3 REFERENCES (based upon other human sulfatides), and aceramide moiety 1. Nugent, D. J. (1989) Blut 59,52-58 containing 2 units of unsaturation, an hydroxyl group, and 2. Beardsley, D. S. (1989) Blut 59,47-51 saturated hydrocarbon chains that are, in the aggregate, 3. Woods, V. L., Jr., Kurata, Y., Montgomery, R.R., Tani, P., Mason, D., Oh, E. H., and McMillan, R. (1984) Blood, 64,156shorter in the human plateletsulfatides than thebovine brain 160 sulfatides. Several differences distinguish the human platelet 4. Woods, V. L., Oh, E. H., Mason, D., and McMillan, R. (1984) sulfatides and the bovine brain sulfatides. First, the platelet Blood 63,368-375 sulfatide has one more site of hydroxylation than the brain 5. Beardsley, D. S., Spiegel, J. E., Jacobs, M. M., Handin, R. I., and sulfatide (5 versus 4). Second, the platelet sulfatide possesses Lux, S. E., IV (1984) J. Clin. Inuest. 7 4 , 1701-1707 1unit of unsaturation while the brain sulfatides have 2 units. Third, the molecular weight of the platelet sulfatide is 94 H. Murakami, B. C. Furie, B. Furie, unpublished results. A

15418


6. Devine, D. V., and Rosse, W. (1984) Blood 6 4 , 1240-1245 7. Asano, T., Furie, B. C., and Furie, B. (1986) Clin. Res. 3 4 , 654A 8. Koerner, T. A. W., Weinfeld, H. M., Bullard, L. S. B., and Williams, L. C. J. (1989) Blood 7 4 , 274-284 9. Van Nliet, H. H. D. M., Kappers-Klunne, M. C., van der Hel, J. W. B., and Abels, J. (1987) Br. J. Haematol. 6 7 , 103-108 10. Lafer, E. M., Rauch, J., Andrzejewski, C., Jr., Mudd, D., Furie, B., Furie, B., Schwartz, R. S., and Stollar, B. D. (1981) J. Exp. Med. 153,897-909 11. Shoenfeld, Y., Rauch, J., Massicotte, H., Datta, S., AndreSchwartz, J., Stollar, B. D., and Schwartz, R. S. (1983) N. Engl. J . Med. 308,414-420 12. Andre-Schwartz, J., Datta, S. K., Shoenfeld, Y., Isenberg, D. A., Stollar, B. D., and Schwartz, R. S. (1984) Clin.Immunol. Immunopathol. 3 1 , 261-271 13. Rubin, R.L., Balderas, R. S., Tan, E. M., Dixon, F. J., and Theofilopoulos, A. N. (1984) J. Exp. Med. 1 5 9 , 1429-1440 14. Jacob, L., Tron, F., Bach, J. F., and Louvard, D. A. (1984) Proc. Natl. Acad. Sci. U. S. A. 8 1 , 3843-3845 15. Shoenfeld, Y., Hsu-Lin, S. C., Gabriels, J . E., Silberstein, L. E., Furie, B. C., Furie, B., Stollar, B. D., and Schwartz, R. S.(1982) J. Clin. Invest. 7 0 , 205-208 16. Asano, T., Furie, B. C., and Furie, B. (1985) Blood 66,1254-1260 17. Svennerholm, L., and Fredman, P. (1980) Biochim. Biophys. Acta 617,97-109 18. Roberts, D. D., Haverstick, D. M., Dixit, V. M., Frazier, W. A., Santoro, S. A,, and Ginsburg, V. (1985) J. Biol. Chem. 2 6 0 , 9405-941 1 19. Leeden, R. W., and Yu, R. K. (1982) Methods Enzymol. 8 3 , 139191 20. Gazzotti, G., Sonnino, S., and Ghidoni, R. (1985) J . Chromatogr. 348,371-378 21. Hirabayashi, Y., Hamaoka, A., Matsumoto, M., Matsubara, T., Tagawa, M., Wakabayashi, S., and Taniguchi, M. (1985) J. Biol. Chem. 260,13328-13333 22. Magnani, J. L., Smith, D.F., and Ginsberg, V. (1980) Anal. Biochem. 1 0 9 , 399-402

23. Magnani, J. L., Spitalnik, S. L., and Ginsberg, V. (1987) Methods Enzymol. 1 3 8 , 195-212 24. Warren, L. (1959) J. B i d . Chem. 2 3 4 , 1971-1975 25. Kean, E. D. (1968) J. Lipid Res. 9 , 319-327 26. Lewis, B. A., and Smith, F. (1969) in Thin Layer Chromatography (Stalhl, E., ed) 2nd Ed, pp. 807-837, Springer-Verlag, New

York 27. Ciucanu, I., and Kerek, F. (1984) Carbohydr. Res. 1 3 1 , 209-217 28. Shoenfeld, Y., Isenberg, D. A., Rauch, J., Madaio, M. P., Stollar, B. D., and Schwartz, R. S . (1983) J. Exp. Med. 1 5 8 , 718-730 29. Atkinson, P. M., Lampman, G. W., Furie, B. C., Naparstek, Y., Schwartz, R. S.,Stollar, B.D., and Furie, B. (1985) J. Clin. Invest. 7 5 , 1138-1143 30. Lampman, G. W., Furie, B., Schwartz, R. S., Stollar, B. D., and Furie, B. C. (1989) Blood 7 4 , 262-269 31. Dersimonian, H., Schwartz, R. S., Barrett, K. J., and Stollar, B. D. (1987) J. Immunol. 1 3 9 , 2496-2501 32. Marcus, A. J., Ullman, H. L., and Safier, L. B. (1972) J. Clin. Invest. 5 1 , 2602-2612 33. Tao, R.V., Sweeley, C. C., and Jamieson, G. A. (1973) J. Lipid Res. 1 4 , 16-25 34. Roberts, D. D., Haverstick, D. M., Dixit, V.M., Frazier, W. A., Santoro, S. A., and Ginsburg, V. (1985) J. Biol. Chem. 2 6 0 , 9405-9411 35. Roberts, D.D., Williams, S. B., Gralnick, H.R., and Ginsberg, V. (1986) J . Biol. Chem. 2 6 1 , 3306-3309 36. Hirano, T., Hashimoto, H., Shiokawa, Y., Iwamori, M., Nagai, Y., Kasai, M., Ochiai, Y., and Okumura, Y. (1980) J . Clin. Invest. 6 6 , 1437-1440 37. Endo, T., Scott, D. D., Stewart, S. S., Kundu, S. K., and Marcus, D. M.(1984) J. Immunol. 1 3 2 , 1793-1797 38. Ilyas, A. A,, Quarles, R. H., Dalakas, M.C., and Brady, R. 0. (1985) Proc. Natl. Acad. Sci. U. S. A. 8 2 , 6697-6700 39. Chou, K. H., Ilyas, A.A., Evans, J. E., Quarles, R.H., and Jungalwara, F. B. (1985) Biochem. Biophys. Res. Commun. 128, 383-388 40. Van Leeuwen, E. F., van der Ven, J. T. M., Engelfriet, C. P., and von dem Borne, A. E. G. K. (1982) Blood 59,23-26 41. Kunicki, T. J., and Aster, R. H. (1978) J. Clin. Inuest. 6 1 , 12251231

SUPPLEMENTARY MATERIAL FOR SULFATED GLYCOLIPIDS ARE THE PLATELET AUTOANTICENS FOR HUMAN PLATELETBINDING MONOCLONAL ANTI-DNA AUTOANTIBODIES Hiroshi Muratam,, Zamas Llm, Barbar. C. Furie, Vernon N Reinhold. Taro Asano and Bruce Furic


15419 794

, 812

1004

878

do

888

1 2 3 4 5 6 7 8