A RCHIVES OF BIOCHEMISTRY ANO BIOPHYSICS
`Vol. 233, No. 2, September, pp. 501-506, 1984
`
`Monoclonal Antibodies PMN 6, PMN 29, and PM-81 Bind Differently
`to Glycolipids Containing a Sugar Sequence Occurring
`in Lacto-N-Fucopentaose Ill
`
`JOHN L. MAGNANI,* EDWARD D. BALL,t MICHAEL W. FANGER,t
`SEN-ITIROH HAKOMORI,:j: A ND VIC'TOR GINSBURG"'
`
`"N ativ nal b1$t-it'Ute of A rthritis, DWhetes, and Digestive and Kidney Disease, National lnst·itutes of Ilealth,
`B ethesda, Mary/,and 20fl()5; fDepartments of MedUine and Microbi.ology, Dart:rrwut.h MedUxtl School,
`Hanover, N ew Ham TJShire OS756; and ;Divisimi of BiochemUxtl Oncol.ogy, Fred HutcMnsvn Cancer Research
`Cent.er and Department of Pat.hoUi,ology, University of Washingtvn, Seattle, Washingtvn 98104
`
`Received March 29, 1984, and in revised form May 7, 1984
`
`Three monoclonal antibodies, PMN 6, PMN 29, and PM-81, bind myeloid cells. An(cid:173)
`tibodies PMN 6 and PMN 29 bind specifically to granulocytes but differ in their ability
`to bind some other cell lines [E. D. Ball, R. F. Graziano, L. Shen, and M. W. Fanger
`(1982) Proc. Natl. Acad. Sci.. USA 79, 5374-5378]. Antibody PM-81, in addition to gran(cid:173)
`ulocytes, also binds to eosinophils, monocytes, and most acute myelocytic leukemia cells
`[E. D. Ball, R. F. Graziano, and M. W. Fanger (1983)J. Immunol.. 130, 2937-2941]. Despite
`these differences, the binding of all three antibodies to cells was inhibited by the
`oligosaccharide, lacto-N-fucopentaose III [ GaIP1-4(Fucal-3)GlcNAct11-3GalP1-4Glc].
`Solid-phase radioimmunoassays using purified glycolipids containing sugar sequences
`found in lacto-N-fucopentaose III demonstrated different binding characteristics for
`each antibody. PM-81 bound lower concentrations of glycolipids than PMN 29, while
`PMN 6 required the highest concentration of glycolipids for binding. Autoradiography
`of thin-layer chromatograms of glycolipid antigens supported these results. The binding
`of these monoclonal antibodies to cells probably depends on the density of antigens on
`the cell surface, each antibody requiring a different density. Thus, cells containing
`antigen below a certain threshold concentration may not bind low-affinity antibodies.
`
`Many monoclonal antibodies with ap(cid:173)
`parent specificities for various cells are di(cid:173)
`rected against the sugar sequence
`
`ual{H-4GlcNAcfll-3Gal · · ·
`3
`I
`Fucal
`which occurs in the human milk oligosac(cid:173)
`charide, lacto-N-fucopentaose III (1). This
`sequence is very immunogenic in mice, and
`is a marker for human adenocarcinoma of
`the colon, stomach (2; 3), and lung (4), as
`well as granulocytes and granulocyte pre(cid:173)
`cursors (5-8). It is also the murine em(cid:173)
`bryonic antigen known as SSEA-1 (9, 10).
`
`Although the antigen is restricted to
`myeloid cells among hemopoetic cells as
`evidenced by immunoftuorescence studies
`(5-8), small amounts of antigen were de(cid:173)
`tected in glycolipids from erythrocytes by
`immunostaining of thin-layer chromato(cid:173)
`grams (2). Recent ly, three monoclonal an(cid:173)
`tibodies, PMN 6, PMN 29, and PM-81, have
`been described which bind differently to
`cells of the myeloid series, including gran(cid:173)
`ulocytes, monocytes, and blasts from pa(cid:173)
`tients with acute myelogenous leukemia
`(11, 12). Despite these differences, t he
`binding of all three antibodies is inhibited
`by lacto-N-fucopentaose III. The data in
`the present paper suggest that differences
`
`501
`
`0003-9861/84 $3.00
`Copyright ~ 1984 by Academic Press, Jnc.
`All rights of reproduction in any Corm reserved.
`
`1 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`

`

`502
`
`MAGNANI ET AL.
`
`in the binding affinity of these antibodies
`for their antigen may explain their differ(cid:173)
`ential react ivities.
`
`EXPERIMENTAL PROCEDURES
`
`Materials. Monoclonal antibodies PMN 6 and PMN
`29 are produced by hybridomas prepared from spleen
`cells obtained from mice immunized with neutrophils
`from normal donors (11); PM-81 is produced by a
`hybridoma prepared from spleen cells obtained from
`a mouse immunized with the promyelocytic leukemia
`cell line, HL-60 (12). Monoclonal antibody AML-1-
`201 binds {J-2 microglobulin (12), and was used as a
`control antibody for these studies. All four antibodies
`are of the lgM isotype.
`Lacto-N-fucopentaosyl(III)ceramide (III'FucnLc(cid:173)
`Ose,Cer) was prepared from human colonic adeno(cid:173)
`carcinoma as previously described (3). The glycolipid
`was further purified by rechromatography on HPLC
`and was freed from Jacto-N-fueopentaosyl(II)ce(cid:173)
`ramide (Le• glycolipid; III' FucLcOse,Cer). The y2
`glycolipid (V'FucnLcOs~r) was prepared from hu(cid:173)
`man erythrocytes as previously described (13). Di(cid:173)
`fucosyl lacto-N-norhexaosylceramide (bands 4a-e;
`III'V'FueznLcOse~Cer) was prepared from a human
`colonic cancer metastasis in the liver (14).
`Globoside was purchased from Supelco Inc., Bel(cid:173)
`lefonte, Pennsylvania. Sialylated lacto-N-fucopen(cid:173)
`taosyl(III)ceramide was kindly provided by Dr. H.
`Rauvala (University of Helsinki, Helsinki, Finland).
`Oligosaccharides, lacto-N-fucopentaose III, and lato(cid:173)
`N-fucopentaose I were isolated from human milk as
`previously described (15).
`Affinity-purified goat anti-mouse IgM (Kirkegaard
`and Perry Laboratories, Inc., Gaithersburg, Md.) was
`iodinated with Naizl (ICN Biochemicals,Irvine, Calif.)
`to a specific activity of about 40 µCi/pg using Iodogen
`(16) (Pierce Chemical Co., Rockford, Ill.).
`Total lipid extracts were prepared by homogeni(cid:173)
`zation of cells in chloroform/methanol/H20 (30/60/
`4, final ratio) (17).
`Inhibition qf binding of antibodies tc cell.a by ol.igc>(cid:173)
`sa.ccharides. Monoclonal antibodies PMN 6, PMN 29,
`PM-81, and AML-1-201 (5 µg/ml) were preincubated
`with 5.4 mM lacto-N-fucopentaose I or lacto-N-fu(cid:173)
`cooentaose III for 30 min at room temoerature. This
`mixture was added to 10' neutrophils previowily
`washed with phosphate-buffered saline, pH 7.4, con(cid:173)
`tainingO.l % oovine serum a lhnmin anti O.Ofi% Andium
`azide, and incubated for 30 min at 4°C. After washing
`with the same buffer, fiuorescein isothiocyanate-con(cid:173)
`jugated goat F(11b')2 antibody directed to mouae im(cid:173)
`munoglobulin (Boehringer- Mannheim, Indianapolis,
`Ind.) was added and incubated for 30 min at 4°C.
`Controls in which each monoclonal antibody was in(cid:173)
`cubated with neutrophils in the absence of oligosac(cid:173)
`charides were run in parallel. Cells treated in this
`
`manner were analyzed for fluorescence on the Ortho
`Cytofluorograph System SOH.
`Solid-Pha,se ra.dioimmunoa8Say. The binding of an(cid:173)
`libody to purified glycolipids was measured by solid(cid:173)
`phase radioimmunoassay as previously described (18,
`19) with minor modifications. Glycolipids in 30 µI
`methanol were added to t he wells of a round-bottom
`Polyvinylchloride microtiter plate (Dynatech, Alex(cid:173)
`andria, Va.), and the solutions were dried by evap(cid:173)
`oration. The wells were then filled with 0.05 M Tris(cid:173)
`HCI, pH 7.8, containing 0.15 M NaCl, 1 % bovine serum
`albumin. and 0.1 % NaN8 (Buffer A). After 30 min, the
`wells were emptied and to each was added 30 µl buffer
`A containing5 µg/rnl monoclonal antibody. The wells
`were covered with parafihn, incubated for 3 hat 22°C,
`washed once with buffer A, and then to each was
`added 100,000 cpm of 1261-labeled goal anti-mouse IgM
`(40-50 µCiliig) in 30 µl buffer A. After 8 h, the wells
`were washed six times with cold phosphate-buffered
`saline (0.15 M NaCl, 0.01 M sodium phosphate, pH 7.4),
`cut from the plate, and assayed for 1261 in an Auto(cid:173)
`Gamma spectrometer.
`Autoradiography of glycolipid antigens. Glycolipid
`antigens were detected on thin-layer chromatograms
`by autoradiography as previously described (19) with
`minor modifications. Glycolipids were chromato(cid:173)
`graphed on aluminum-backed high-performance thin(cid:173)
`layer chromatography plates (silica gel 60, E. Merck,
`Darmstadt, West Germany) in chloroform/methanol/
`0.25% KCI (50/50/12, by volume). The dried chro(cid:173)
`matogram was soaked for 1 min in a 0.1 % solution
`of Polyisobutylmethacrylate beads (Polysciences, Inc.,
`Warrington, Pa.) dissolved in hexane. After drying
`in air, the chromatogram was sprayed with phos(cid:173)
`phate-buffered saline (0.15 M NaCl, O.Ql M sodium
`phosphate, pH 7.4) and immediately soaked in buffer
`A until all of the silica gel was wet (about 15 min).
`The plate was then removed and laid horizontally on
`a slightly smaller glass plate in a large Petri dish.
`Monoclonal antibody (5 µg/ml) diluted in buffer A
`was layered on the plate (60 µJ/cm2 chromatogram
`surface). After incubation at 22° for 2 h, the chro(cid:173)
`matogram was washed by dipping in four successive
`changes of cold phosphate-buffered saline at 1-min
`Intervals, and overlayed with buffer A containing 2
`X lOG cpm/ml 1261-1.abeled goat anti-mouse IgM. After
`1 h at 22°C, the chromatogram was washed as before
`in cold phosphate-buffered saline, dried, and exposed
`U> Xar-5 X-ray film (Eastman-Kodak, Roche.ster,
`N. Y.) for 10 h at 22°C.
`
`RESULTS
`Effects of Oligosaccharides on
`Cell Binding
`Monoc)onal antibodies PMN 6, PMN 29,
`and PM-81 bound to most neutrophils, and
`
`2 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`

`

`DIFFERENTIAL PMN 6, PMN 29, AND PM-~1 Rll'IDINC TO GLYCOLIPIDS
`
`503
`
`this binding was completely inhibited by
`5.4 mM lacto-N-fucopentaose III but not by
`5.4 mM Jacto-N-fucopentaose I (Fig. 1).
`Neither oligosaccharide
`inhibited
`the
`binding of monoclonal antibody AML-1-
`201, an lgM which binds /3-2 microglobu(cid:173)
`lin (12).
`
`A1,toradiography of Glycolipid Antigerl.$
`
`Glycolipid antigens were detected by au(cid:173)
`toradiography of thin-layer chromato(cid:173)
`grams as described under Experimental
`Procedures. Purified glycolipids Y2 and 4c,
`which contain a carbohydrate sequence
`found in lacto-N-fucopentaose III (see Ta(cid:173)
`ble I), bound PMN 6, PMN 29, and PM-81
`(Figs. 2A, B, C, lanes 1). The smaller
`glycolipid, lacto-N-fucopentaosyl(ffi)cer-
`
`amide, bound only PMN 29 and PM-81 un(cid:173)
`der these conditions.
`The reactivity of these antibodies, par(cid:173)
`ticularly PMN6, resembled that of SSEA-
`1 (9, 10); ZWG 13, ZWG 14, and ZWG 111
`(2); and FH-1 and FH-5 (20); which do not
`bind as well Lo Jacto-N-fucopentaosyl(cid:173)
`(IIl)ccramide as to glycolipids with longer
`carbohydrate chains, including di- and tri(cid:173)
`fucosylated derivatives.
`All three antibodies detected glycolipid
`antigens from total lipid extracts of gran(cid:173)
`ulocytes and HL-60 cells (Figs. 2A, B, and
`C; lanes 6 and 7). Both of these cell types
`have high concentrations of glycolipids
`containing a carbohydrate sequence found
`in lacto-N-fucopentaose III (5). Antigen
`comigrating with lacto-N-fucopentaosyl(cid:173)
`(III)ceramide, however, was not detected
`
`•!!tt1
`
`0
`
`, .. ,..n
`
`0
`
`, .. Jlij.
`
`Alll 120,
`
`I
`
`E
`
`,.,_,,
`
`+ lNF I
`
`PMN 21
`+ lNF I
`
`H
`
`,MN•
`
`+ l HF I
`
`AML- 1-201
`• LHF" I
`
`,..,.,
`
`+ l •F Ml
`
`F
`
`PllN2t
`• lNF Ill
`
`iltMN6
`• LHF tH
`
`... l -t-201
`.. lllf ..
`
`II
`
`c ... .. s ,,
`z ... ... ...
`
`0
`
`·-
`
`c
`
`l
`
`FLUORESCENCE INTENSITY
`
`FIG. 1. The effect of Jacto-N-fucopentaose I {LNF I) and Jacto-N-fucopentaose Ill (LNF 111) on
`the binding of monoclonal antibodies PM-81, PMN 29, PMN 6, and AML-1-201 was determined by
`cytoftuorography as described under Experimental Procedures. The ftuorescence of neutrophils
`stained with PM-81, PMN 29, PMN 6, and AML-1-201 is shown in panels A, D, G, and J , respectively.
`Tho eff~t of lacto-N-fucopentaose I and lacto-N-fucopentaose 111 on this fluorescence is shown in
`panels B, E, H, and K, and panels C, F, I, and L, respectively.
`
`3 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`

`

`504
`
`MAGNANI ET AL.
`
`TABLE I
`STRUCTURE OF CARBOHYDRATES
`
`Name
`
`Structure
`
`Glycolipids
`Lacto-N-fucopentaosyl(Hl)ceramide
`
`4c
`
`Globoside
`Oligosaccharides
`Lacto-N-fucopentaose III
`
`Lacto-N-fucopentaose I
`
`GaltJl-4GlcNActJ1 ·3GaltJl-4GlctJl-1 Cer
`3
`I
`Fucal
`GaltJl-4GleNAef11-3GaltJl-4GleNActJl-3GaltJ1-4Glct1l-1Cer
`3
`3
`I
`I
`Fueal
`Fueal
`GalP1-4GlcNAcPI-3GalPI-4GleNAcP1-3GalPI-4GlcPl-1Cer
`3
`I
`Fueal
`
`GalNAcPl-3Galal-4GaltJl-4GlctJ1-lCer
`
`GalP1-4GlcNAcP1-3GalPl-4Glc
`3
`I
`Fucal
`Fueal-2GaltJl-3GleNActJl-3GaltJl-4Glc
`
`by PMN 6. The same chromatographic pat(cid:173)
`tern was obtained by all three antibodies;
`however, the intensity of staining in(cid:173)
`creased from PMN 6 to PMN 29 to PM-81.
`No antigens were detected in the total
`
`lipid extracts of acute myelocytic leukemia
`cells or monocytes by antibodies PMN 6
`and PMN 29 (Figs. 2A and B, lanes 2, 3, 4,
`5). Under the same conditions PM-81 de(cid:173)
`tected low levels of antigen in both extracts
`
`A
`
`B
`
`c
`
`-
`
`LHF • cer-
`
`Origin-
`
`-
`
`123-4567
`
`123-4567
`
`2 3 • 5 6 7
`
`Fie. 2. Autoradiography of glycolipid antigens. Autoradiography of glycolipid antigens was per(cid:173)
`formed as described under Experimental Procedures. (A) was stained with antibody PMN 6, (B)
`with PMN l!9, and (C) with PM-81, each at 5 µg/ml. Purified glycolipids (30 ng) 4c, Y2 , and lacto(cid:173)
`N-fucopcntaosyl(III)ceramide (LNF III cer) were chromatographed in lane 1. The amount of extract
`chromatographed expressed as the volume of packed cells from which it was obtained is lane 2, 5
`µI AML blasta; lane 3, 2 µI AML blasta; lane 4, 5 µI monocytes; lane 5, 2 µI monocytes; lane 6, 2 µI
`granulocytes; and lane 7, 2 µI HL-60 cells. The positions of the purified glycolipids are shown on
`the left.
`
`4 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`

`

`DIFFERENTIAL PMN 6, PMN 29, AND PM-81 BfNDlNG TO CL YCOLIPrDS
`
`505
`
`(Fig. 2C, lanes 2, 3, 4, 5). These data support
`the previous findings that PMN 6 and PMN
`29 bind to neutrophils (11) while PM-81
`binds t-0 neutrophils, monocytes, and acute
`myelocytic leukemia cells (12).
`
`Sol.id-Phase Radi<Jimmunoassay
`Monoclonal antibodies PMN 6, PMN 29,
`and PM-81 were a8Saye<l for binding to pu(cid:173)
`rified glycolipids by solid-phase radioim(cid:173)
`munoassay as described under Experi(cid:173)
`mental Procedures. Differences in binding
`were found for each antibody as shown in
`Fig. 3. PM-81 bound to the lowest concen(cid:173)
`tration of glycolipids containing sugar se(cid:173)
`quences found in lacto-N-fucopentaose III.
`Higher cuncent.rations of glycolipids were
`required for binding antibody PMN 29.
`PMN 6 showed the least binding to high
`concentrations of glycolipids Y2 and 4c,
`and did not bind to lacto-N-fucopentao(cid:173)
`syl(Ill)ceramide at the concentrations
`tested. These results agree with the inten(cid:173)
`sity of staining of glycolipid antigens
`shown in Fig. 2.
`Differences in binding were also found
`for each purified glycolipid. All three an(cid:173)
`tibodies bound to glycolipids Y2 and 4c at
`lower concentrations than to lacto-N-fu(cid:173)
`copentaosyl(lll)ceramide (Fig. 3). These
`results also agree with the chromato(cid:173)
`graphic patterns of glycolipid antigens
`shown iu Figure 2. Noue uf the::ie ant.ibo<lies
`bound to a monosialoganglioside contain(cid:173)
`ing sialic acid linked a2-3 to the termi(cid:173)
`nal galactose of lacto-N-fucopentaosyl(cid:173)
`(III)ceramide (data not shown).
`
`DISCUSSION
`
`The carbohydrate sequence
`
`GalP1-4GlcNAcftl-3Gal · · ·
`3
`I
`Fucal
`
`is a. potent antigen to the mouse. Out of
`325 monoclonal antibodies from different
`laboratories that have been analyzed in our
`laboratory, 55 are directed against this se(cid:173)
`quence (21).
`Some antibodies directed against the
`same antigen, as judged by hapten binding
`or hapten inhibition studies, have different
`cell specificities. For example, a rabbit anti(cid:173)
`paragloboside antibody (22) and a Wal(cid:173)
`denstrom cold agglutinin (cold agglutinin
`McC) (23) both bind to paragloboside
`{Gal/H-4GlcNAcfJ1-3GalP1-4Glctn-1Cer),
`yet react differently with cells: the rabbit
`antibody reacts equally well with human
`cord and adult erythrocytes {22) while the
`cold agglutinin reacts strongly with cord
`cells but weakly or not at all with adult
`cells (23). This differential reactivity might
`be explained In some cases by the fact that
`some antibodies bind to different parts or
`to different sides of the same sugar chain
`(18, 20, 24, 25). If the adult erythrocyte
`antigen were actually substituted para(cid:173)
`globoside and the two antibodies bound to
`different parts of the paraglohoside sugar
`chain, the antibodies would react differ(cid:173)
`ently with the substituted paragloboside
`depending on where the substitution oc(cid:173)
`curred. This hypothesis, however, is not
`likely to explain the differential reactivity
`of antibodies PMN 6, PMN 29, and PM-81
`with various cell types, as the three an(cid:173)
`tibodies appear to bind to the same gly(cid:173)
`colipid antigens (Fig. 2). It is more likely
`that their differential reactivity is ex(cid:173)
`plained by their different affinities for an(cid:173)
`tigen {Fig. 3). PM-81 has the highest af(cid:173)
`finity and binds to more cell types than do
`PMN 6 or PMN 29. It is the only antibody
`that binds monocytes {11, 12) which contain
`little glycolipid antigen (Fig. 2C, lane 4).
`PMN 29 detects intermediate concentra(cid:173)
`tions of glycolipid antigen and bindi:J to
`some cell lines that PMN 6 does not (11).
`Thus, cells that contain antigen below a
`
`le
`
`I
`
`/;
`~/-
`.. ...
`
`j_
`.. -. ... ..
`
`t.
`) -_
`
`•
`
`,. ~ • tr •
`
`~ Ct A
`t •t
`~ .
`~ ,..
`........
`:••.
`
`GLYCOU'll>i....;
`
`FIG. 8. Binding of antibodies to purified glycolipids.
`Solid-phase radioimmunoassays were performed as
`described under Experimental Procedures. Antibody
`PMN 6 was used for assays in (A), PMN 29 for (B),
`and PM-81 for (C). Purified glycolipids tested we.re
`4c, 6; Ya, O; Jacto-N-fucopentaosyl(Ill)ceramide, O;
`and globoside, o. Structures of these glycolipids are
`depicted in Table I.
`
`5 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`

`

`506
`
`MAGNANI ET AL.
`
`certain threshold concentration may bind
`high-affinity but not low-affinity antibod(cid:173)
`ies. Glycoproteins containing the same
`carbohydrate sequence may also be in(cid:173)
`volved in antibody binding (26, 27).
`
`REFERENCES
`
`1. KOBATA, A., AND GINSBURG, V. (1969) J. Bi.ol. Clw.m.
`244, 5496.
`2. BROCKHAUS, M., MAGNANI, J. L., HERLYN, M.,
`BLAszczYK, M., Sl'EPLEWSKI, Z., KOPROWSKI, H.,
`AND GINSBURG, V. (1982) Arch. BWcJunn.. Bio(cid:173)
`phys. 217, S47-G51.
`3. HAKOMORI, 8., NUDELMAN, E., KAN'NAGI, R., AND
`LEVERY, S. B. (1982) BWcJunn.. Biaphys. Res.
`Commun. 109, 36-44.
`4. H UANG, L. C., BROCKHAUS, M., MAGNANI, J. L.,
`CUTTrITA, F., ROSEN, S., MINNA, J. D., AND
`GINSBURG, V. (1983) Arch. Biochem Biqphys.
`220, 318-320.
`5. H UANG, L. C., CJVIN, C. I., MAGNANI, J. L., SHAPER,
`J. H., AND GINSBURG, V. (1983) B/,ood 61, 1020-
`1023.
`6. GIRARDET, C., LADISCH, S., liEuMANN, D., MACH,
`J.P., AND CARREL, S. (1983) Int. J. Cancer 32,
`177-183.
`7. Goor, H. C., THORPE, s. J., HoUNSELL, E. F., RuM(cid:173)
`IPOLD, H., KRAFT, D., F~RSTER, 0., AND FEIZI,
`T. (1983) Eur. J. Immunol. 13, 306-312.
`8. ANDREWS, R. G., TOROK-SToRB, B., AND BERNSTEIN,
`I. D. (1983) Bl.ood 62, 124-132.
`9. HAKOMORI, S., NUDLEMAN, E., LEVERY, S., SoLTER,
`D., AND KNOWLES, B. B. (1981) Biochem. Bio(cid:173)
`phy& Re& Commun. 100, 1578-1586.
`10. Gom, H. C., FEiz1, T., KAPADIA, A., KNOWLES,
`B. B., SoLTER, D., AND E VANS, M. J . (1981) Na(cid:173)
`ture (Lcnu:Um) 292, 156-158.
`11. BALL, E. D., GRAZIANO, R. F., SHEN, L., AND FAN(cid:173)
`GER, M. W. (1982) Proc. Nail Acad. &i. USA
`79, 5374-5378.
`12. BALL, E. D., GRAZIANO, R. F., AND F ANGER, M. W.
`(1983) J. Immunol. 130, 2937-2941.
`13. KANNAGI, R., NUDELMAN, E., LEVERY, S. B., AND
`
`HAKOYORI, S. (1982) J. Biol. Chem 257, 14865-
`14874.
`14. HAKOMORI, S., NUDELMAN, E., LEVERY, S. B., AND
`KANNAGI, R. (1984) J. Biol. Chem 259, 4672-
`4680.
`15. KOBATA, A. (1972) in Methods in Enzymology
`(Ginsburg, V., ed.), Vol. 28, p. 262, Academic
`Press, New York.
`16. SALACINSKI, P.R. P ., MCLEAN, C., SYKES, J.E. C.,
`CLEMENT-JONE$, V. V., AND LoWRY, P. J. (1981)
`Anal. BWcJunn.. 117, 136-146.
`17. SVENNERHOLM, L., AND FREDMAN, P. (1980)
`Biochim. Binphys. Acta 617, 97-109.
`18. YOUNG, w. w .. JR., MACDONALD, E. M. s .. Now(cid:173)
`INSKY, R. C., AND HAKOMORI, S. (1979) J. Exp.
`Med. 150, 1008-1019.
`19. MAGNANI, J. L., NILSSON, B., BROCKHAUS, M., ZOPF,
`D., STEPLEWSKI, Z., KOPROWSKI, H., AND GINS(cid:173)
`BURG, v. (1982)J. Biol Chem 257, 14365-14369.
`20. FuKUSHI, Y., HAKOMORI, S., NUDELMAN, E., AND
`CocHRAN, N. (1984) J. Biol. Chem 259, 4681-
`4685.
`21. GINSBURG, V., F'REDMAN, P ., AND MAGNANI,
`J. L. (1984) in Genes and Antigens in Cancer
`Cells: The Monoclonal Antibody ;\pproa<'h
`(Reithmiiller, G., Koprowski, H., von Kleist, S.,
`and Munk, K., eds.), pp. 44-50, Karger Verlag,
`Basel.
`22. SCHWARTING, G. S., AND MARCUS, D. M. (1977) J.
`Immunol. 118, 1415-1419.
`23. TSAJ, C.-M., ZOPF, D . A., WISTAR, R., AND GINS(cid:173)
`BURG, V. (1976) J. lmmunol. 117, 717-721.
`24. ZoPF, D. A., TSAI, C.-M., AND GINSBURG, V. (1979)
`in Carbohydrate-Protein Interaction (Gold(cid:173)
`stein, I. J., ed.), pp. 90-101, Amer. Chem. Soc.,
`Washington, D. C.
`25. YOUNG, W. W., JR., JOHNSON, H . 8., TAMURA, Y.,
`KARLSSON, K.-A., LARSON, G., PARKER,
`J. M. R., KHARE, D. P ., SPOHR, U., BAKER,
`D. A., HINDSGAUL, 0., AND LEMIEUX, R. U. (1983)
`J. Biel Chem 258, 4890-4894.
`26. URDAL, D. L., BRENTNALL, T. A., BERNSTEIN, I. D.,
`AND fiAKOMORI, S.-1. (lgg,q) Blood 62, 1022-1026.
`27. SKUBITZ, K. M., PESsANO, S., BOTI'ERO, L., FERRER,
`D., ROVERA, G., AND AUGUST, J. T. (1983) J.
`Immunol. 131, 1882-1888.
`
`6 of 6
`
`Celltrion, Inc., Exhibit 1007
`
`