[CANCER RESEARCH 49, 6645-6651. December 1, 1989]
Five New Epitope-defined Monoclonal Antibodies Reactive with GM2and Human
Glioma and Medulloblastoma Cell Lines'
Folios D. Vrionis, Carol J. Wikstrand, Pam Fredman, Jan-Eric Mânsson,Lars Svennerholm, and Dareil D. Bigner2
Department of Pathology. Duke L'nirersity Medical Center, Durham. North Carolina 27710 ¡F.D. V., C.J. H'., D. D. B.]. and Department of Psychiatry and
Neurochemistry, Gothenburg L'nirersity, St. JörgenHospital, 422 03 Hisings Backa, Sweden ¡P.F., J-E. M., L. S.J
ABSTRACT
5). Furthermore, there is evidence that unique gangliosides like
the mono- and disialogangliosides 3'-isoLM] and 3'6'-isoLD1,
In order to investigate GM2expression in gliomas, the GM2-positive
human glioma cell line (HGI.) D-54 IMG, which contains 0.6 nmol GM2/
mg protein, representing 77% of the total monosialoganglioside fraction,
was used as an immunogen for the production of anti-Gun monoclonal
antibodies. For gangliosidc designations, see IUPAC-IUB (Eur. J.
Biochem., 79: 11-21, 1977) and Svennerholm (J. Neurochem., 10:613623, 1963). Five IgM monoclonal antibodies (DMAb-1 through DMAb5) specifically recognizing the GalNAc/31-4(NeuAca2-3)Gal-terminal
epitope common to «,>.,;
and GalNAc-GDi. are reported. The antibodies
did not react with GM,, GM3,GD2,GDJ,GDI., GDII»
GTii»
and GOib.Purified
anti-GM2MAbs were used to define the expression of the "GM2"terminal
epitope by cultured human malignant and normal cells by radioimmunoassay and membrane immunofluorescence. Among neuroectodermal tis
sue-derived cell lines, DMAb-3, at an optimal concentration of 5 fig/ml,
showed high reactivity (radioimmunoassay binding ratios > 20) with 9 of
19 HGLs, 3 of 5 medulloblastoma, 4 of 5 neuroblastoma, and 1 of 3
melanoma lines. Moderate reactivity (binding ratio, 10-20) was exhibited
by 3 HGL, 2 medulloblastoma, and 1 neuroblastoma lines and low
reactivity (binding ratio, 3-10) by 5 HGL lines; no reactivity was detected
with 2 HGL and 2 melanoma lines. Densitometric evaluation of monosi
aloganglioside extracts from human glioma and medulloblastoma cell
lines in conjunction with immunostaining on thin-layer chromatograms
showed that GM2represents the major monosialoganglioside in 8 of 10
HGL and in 3 of 4 Med lines. In these lines the amount of GM2ranged
from <0.1 to 0.6 nmol/mg protein. These results indicate that GM2
represents a proportionally increased ganglioside of most glioma, medul
loblastoma, and neuroblastoma cells in vitro.
which are undetectable in normal adult brain, are associated
with human gliomas (5, 6). In the only study of ganglioside
content of human medulloblastomas, GM2and GM.,were iden
tified as the major monosialogangliosides of the TE-671 human
medulloblastoma murine xenograft (7).
The GM2ganglioside is a minor component of human nervous
system (8). It was shown to be expressed on a variety of human
cancer cells, including melanomas, breast carcinomas, and sar
comas (9, 10). A mouse monoclonal antibody (MAb 5-3),
generated after immunization with JB-RH mouse melanoma
cells, was demonstrated to be reactive with both /V-acetyl- and
7V-glycolyl-GM2(11). In the same study it was hypothesized that
GM2 represents a differentiation antigen largely restricted to
cells of neuroectodermal origin on the basis of an analysis of
melanoma cell lines and a limited number of glioma and neuro
blastoma cell lines. Furthermore, a monoclonal antibody rec
ognizing A'-acetyl-GM2 in lung carcinomas has been reported
(12).
This study was undertaken to define the expression of GM2
ganglioside on a large number of well-characterized human
glioma cell lines shown to be individually distinct by cytogenetic
criteria and on all five permanent medulloblastoma lines de
scribed. Five mouse MAbs (DMAb-1 through DMAb-5) reac
tive to GM2 were isolated from a fusion involving splenocytes
hyperimmunized to the GMi-positive D-54 MG human glioma
cell line. Analysis of structurally related gangliosides revealed
reactivity to GM2 and GAlNAc-GDla, which both bear the
GalNAc/31-4 (NeuAca2-3)Gal epitope in a terminal position;
these MAbs were used to define the expression of this "GM2"
INTRODUCTION
Gliomas are highly lethal neoplasms that show marked het
erogeneity in their genotypic and phenotypic characteristics ( 1).
Despite this heterogeneity gliomas show a marked shift in
ganglioside expression, namely from the more complex oligosialylated gangliosides like GDia,3 GDib, and GTib found in
normal brain, toward the simpler, less polar gangliosides (2Received 4/12/89; accepted 9/6/89.
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 solely to indicate this fact.
'This work was supported by NIH Grants R37 CAI 1898, NS 20023, and
CA32672 and by grants from the Swedish Medical Research Council (Project
03X-627), Swedish Cancer Society (Project 2260-B88-01X), and the National
Swedish Board for Technical Development (Project 84-4667).
2To whom requests for reprints should be addressed, at P. O. Box 3156, 207
Jones Bldg.. Duke University Medical Center, Durham, NC 27710.
3The abbreviations used are: Gangliosides have been designated according to
CBN recommendations (IUPAC-IUB. 1977) and to the coding system of Sven
nerholm (1963) (8, 31): CD,., IV3NeuAc,II3NeuAc-GgOse4Cer; GA2,GgOse3Cer;
GM2, II3NeuAc-GgOse3Cer; GM3, II3NeuAc-LacCer; NeuAc-GM2. II3NeuAcGgOsejCer; NeuGc-GM2, Il3NeuGc-GgOse3Cer; GMi, II3NeuAc-GgOse4Cer;
GalNAc-isoGMi. IV3NeuAc-GgOse,Cer; GD3.II3(NeuAc)2-LacCer; GalNAc-GD,.,
IV3NeuAc,II3NeuAc-GgOse3Cer;GD2,II3(NeuAc)2-CgOse3Cer:GD,b,II3(NeuAc)2GgOse4Cer; Gâ„¢. IV3NeuAc.II3(NeuAc)2-GgOse4Cer; GQlb, IV3(NeuAc)2,
II3(NeuAc)2GgOse4Cer; GalNAc-3'-isoLMI, IV4GalNAcIV3NeuAc-LcOse4Cer;
3'-isoLMi. IV3NeuAc-LcOse4Cer; 3'6'-isoLD1, IV3NeuAc,III'NeuAc-LcOse4Cer;
HPTLC. high performance thin layer chromatography; MAb, monoclonal anti
body; BSA. bovine serum albumin: PCS, fetal calf serum; DPBS, Dulbecco's
phosphate-buffered saline; RIA, radioimmunoassay; BR, binding ratio; FI, fluo
rescence index; CS-RIA, cell surface radioimmunoassay; ZO, zinc option; HGL,
human glioma lines; MED. medulloblastoma: HPLC, high performance liquid
chromatography.
epitope by human glioma and medulloblastoma cells in culture.
MATERIALS
AND METHODS
Glycolipids. All gangliosides and synthetic derivatives used in this
study were prepared at the laboratory in Gothenburg, Sweden. NeuAcGM2 was purified from Tay-Sachs brain (13) and NeuGc-GM2 from
normal BALB/c mouse liver. The purity and identity of the latter
ganglioside were confirmed with fast atom bombardment-mass spectrometry. GM3and GD3were purified from metastatic melanoma tissue
removed at surgery. GD2was prepared by /i-galactosidase treatment of
Goib (enzyme kindly provided by Dr. George Jourdian, University of
Michigan). GA2(asialo-GM2) was derived from GM2by acid hydrolysis
in l M formic acid for 30 min at 100°C(14). Similarly, GalNAc-isoGMi
was prepared from GalNAc-GDi, (32). The isolation and characteriza
tion of the novel ganglioside GalNAc-3'-isoLM, from meconium are
thoroughly described by Fredman et al. (32). All other glycolipids were
purified from normal adult human brain.
Ganglioside Extraction. Tissue culture cells obtained from a cytogenetically monitored "seed" stock maintained in Dr. Darell Signer's
laboratory and shown to be Mycoplasma free were grown to confluence
in roller bottles and harvested with 0.125% trypsin-0.02% EDTA. After
the pellet was washed and a cell count was obtained, an aliquot was
taken for karyotypic analysis and confirmation of the individuality of
each cell line. The remaining cells were extracted with chloroform:methanol:water (4:8:3, by volume) (15). The extract was applied
6645
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
EXPRESSION
IN GLIOMA AND MEDULLOBLASTOMA
to a silica gel column (Silica Gel 60. 230-400 mesh; EM Science,
Cherry Hill, NJ) in chloroform:methanol:water (65:25:4, by volume).
Most neutral lipids were eluted with 8 bed volumes of the same solvent,
whereas the gangliosides were eluted with chloroform:methanol:water
(30:60:20, by volume). After dialysis against water, the ganglioside
fraction was dissolved in chloroform:methanol:water
(60:30:4.5, by
volume) and applied on a DEAE-Sepharose Fast Flow column. Neutral
lipids were eluted with chloroform:methanol:water (60:30:4.5, by vol
ume) and mono- and oligosialogangliosides with 0.01 and 0.05 M
potassium acetate in methanol, respectively. Quantitative measurement
of the total ganglioside sialic acid was determined by the resorcinol
assay (16). Densitometric scanning of gangliosides (visualized with
resorcinol) was performed with a CAMAG thin layer chromatography
Scanner II (Muttenz, Switzerland) at 620 nm.
HPTLC-Immunostain. Gangliosides were separated by HPTLC on
alumina-backed HPTLC plates (Silica Gel 60; EM Sciences, Cherry
Hill, NJ) in chloroform:methanol:0.25% aqueous KC1 (50:40:10, by
volume) as indicated. The plate was then plasticized in 0.1% polyisobutylmethacrylate in hexane, air dried, and immersed in incubation
buffer (50 MMTris-HCl, pH 7.8, 15 mM NaCl, 1% BSA) for 30 min at
4°C.Primary' antibody incubation (at 5 pg/m\, in ZO-10% FCS) was
done at 4°Cfor 12-16 h, followed by rinsing with DPBS and exposure
to ami minine IgM biotinylated antibody (Zymed Laboratories, Inc.,
South San Francisco, CA) at a dilution of 1:300 in DPBS-0.05% Tween
for 15 min. After washing and incubation with streptavidin-peroxidase
reagent (1:300 for 5 min), the plates were developed with 3,3'-diaminobenzidine (Sigma; 1 mg/ml) containing 0.003% H2O2. Reference
gangliosides were visualized with orcinol reagent (0.1 % orcinol in water,
3% H2SO4).
For analysis of the epitope recognized by MAbs 1-5, HPTLCImmunostain with structurally characterized gangliosides was per
formed. Glycolipids were applied on 5-mm lanes on silica gel-precoated
plastic sheets (Polygram *Sil G; Marchery and Nagel, Duren, Federal
Republic of Germany). Chromatographie
solvent was chloroform:methanol:0.25% aqueous KC1 (50:40:10, by volume). The plate
was then dipped twice for 1 min each in 0.1% polyisobutylmethacrylate
in hexane before preincubation in Tris-BSA (50 itiviTris-HCl, pH 7.8,
15 mM NaCl, 1% BSA) for 15-30 min at room temperature. Subse
quently the plate was overlaid with glycolipid-specific MAb in TrisBSA for 16 h at 4°Cfollowed by alkaline phosphatase conjugated antimouse (IgM + IgG) antibody (Jackson Immunoresearch Lab., Inc.,
Westgrove, PA), also in Tris-BSA, for 3 h at room temperature. Finally,
bound antibody was detected by incubating the plate in 0.1 M glycine
buffer containing 1 mM ZnCl2, 1 mM MgCI2, and 0.1% 5'-bromo-4'chloro-3'-indolylphosphate
as substrate. The plate was washed four
times with PBS in between each step.
Indirect Membrane Immunofluorescence. The assay was performed
as described previously (17). Briefly, cells were harvested with 0.02%
EDTA and plated in round bottomed well microtiter plates at a density
of 5 x IO5 cells/well. Primary antibody (50 n\, 5 Mg/ml) was applied
for 2 h at 4°C,washed twice (with 10% FCS-1 x ZO, 0.02% EDTA),
and followed by incubation with 50 n\ of fluorescein isothiocyanateconjugated anti-mouse IgM at 4°Cfor 2 h. Cells were then washed
three times and mounted with 50% glycerol in DPBS. Cells were scored
for fluorescence under dark-field UV light with a Zeiss Universal
Microscope. The FI was calculated as
FI= 100 x
A-B
where A is the percentage of nonfluorescing cells with the nonspecific
control antibody and B is the percentage of nonfluorescing cells with
the experimental antibody. Experimental antibody activity giving 10%
or more fluorescing cells above that obtained with the control antibody
of the same isotype was considered significant antibody binding ( 17).
In all immunofluorescence assays performed in this study, reactivity of
the MOPC-104E IgM control antibody with both antigen-positive and
antigen-negative cells resulted in Fis of less than 5.
Cell Surface RIA. Confluent cell monolayers in 96-well fiat bottomed
plastic plates (Flow Laboratories) were incubated with 50 M'of primary
antibody in ZO-10% FCS (at the indicated concentration) for 1 h at
CELL LINES
37°C,followed by incubation under the same conditions with 50 n\ of
affinity-purified I25l-goat anti-mouse IgM (Pel Freeze, Brown Deer,
WI) as described previously (18). Cells were solubilized with 10%
NaOH and individual wells were counted in a Packard automatic
gamma counter. Purified IgM isolated from the MOPC-104E myeloma
cell line served as negative control. Binding ratio (BR) was calculated
by dividing experimental cpm by negative control cpm and was consid
ered to be positive if it exceeded 3; such values exceed the mean
background value by >3 SD (19). Nonadherent cell lines were assayed
in suspension (1 x IO5 cells/well) in round bottomed 96-well plates
(Cooke Engineering). In order to investigate internalization of ganglioside-antibody complexes, cell monolayers were incubated with primary
and secondary' antibody at 4°Cversus 37°Cas described above.
Solid Surface RIA. Purified gangliosides in methanol were plated at
various concentrations in plastic microtiter plates as described previ
ously (20). The wells were blocked for a minimum of 30 min at 22°C
with incubation buffer (50 MMTris-HCl, pH 7.8, 15 mM NaCl, 1%
BSA). Primary and secondary antibody application was performed as
described above with the exception that incubation was extended to 2
h at 37°Cfor each step. Results are expressed as for CS-RIA.
Cell Lines. The established permanent human glioma- and medulloblastoma-derived cell lines used in this study have been described
previously (1, 21-25). The neuroblastoma cell lines SK-N-MC, SK-NSH, LAN-1, and LAN-5 were kindly provided by Dr. R. Seeger, UCLA.
The following cell lines were obtained from the American Type Culture
Collection: Tera-1, Tera-2, SK-MEL 28, RD 36, IMR-32, ATCC 1147,
and P3X63/Ag8.653. The human breast carcinoma cell line DU4475
was the gift of Dr. A. Langlois, Duke University Medical Center; the
HL-60 and SB cell lines were obtained from Dr. R. Metzgar, Duke
University Medical Center. The human melanoma cell lines T-8 and
KENT were provided by Dr. N. Levy. The epidermoid carcinoma cell
line A431 was a gift from Dr. C. Stoscheck; the PA-1 cell line was the
gift of Dr. J. Trosko, Michigan State University. All human cell lines
were propagated in 10% FCS-ZO. Their storage and testing for
Mycoplasma or HeLa cell contamination has been described previously
(1).
Immunization. BALB/c mice were immunized with 0.125% trypsinharvested cells of the GM2expressing the D-54 MG human glioma cell
line. Cells were washed twice in DPBS, resuspended in ZO media and
injected i.p. into BALB/c mice ( 1 x 107cells/mouse) at Day 1, Day 21,
and 3-4-week intervals thereafter. Three days prior to fusion an i.v.
boost of 1 x IO6 cells was given. The mice were then sacrificed, and
their splenocytes were fused with the P3X63/Ag8.653 nonsecretor
murine myeloma cell line (18). Hybridoma supernatants were initially
tested for reactivity versus GM2(solid-phase RIA; 20 pmol GM2/well)
and D-54 MG cells (CS-RIA). Positive hybrids in both assays were
expanded and tested against GM2 in HPTLC-immunostain. Reactive
hybrids were cloned in methylcellulose semisolid medium.
Antibody Purification. Three Chromatographie methods were used,
either alone or in tandem. Anti-IgM affinity chromatography was
performed as follows. Ascites fluid was diluted 1:10 with 0.5 M NaCl,
0.01 M phosphate buffer, pH 7.2, and applied to an anti-IgM affinity
column (Sigma Chemical Co., St. Louis, MO). After a washing with
the same buffer, elution of IgM was carried out with 0.01 M citric acid,
pH 3.0. For gel exclusion chromatography on a Sephacryl S-300
column, a buffer consisting of 0.5 M NaCl, 0.01 M Tris, pH 8.2, was
used. HPLC ion-exchange chromatography was performed with an
ABx column from J. T. Baker Chemical Company. Ascites fluid was
diluted 1:10 with 75 mM KH2PO4, pH 6.0, and injected onto the
column. The column was washed with the same buffer and IgM was
eluted with a linear gradient (0-50% 500 mM KH2PO4, pH 6.8, over
60 min with a flow rate of 1 ml/min). All antibody preparations were
dialyzed for 48 h against phosphate buffer and concentrated with a
PM-30 Amicon filter (Amicon Div., Grace & Co., Danvers, MA). Purity
was analyzed with HPLC gel filtration chromatography (with a Seph
acryl S-200 column) and 7.5% sodium dodecyl sulfate-polyacrylamide
gel electrophoresis under reduced conditions. All antibody preparations
met both criteria for purity: a single peak in the HPLC profile and 2
bands on sodium dodecyl sulfate-polyacrylamide gel electrophoresis
gels with molecular weights of approximately 68,000 and 23,000 cor-
6646
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
EXPRESSION IN GLIOMA AND MEDULLOBLASTOMA
responding to the molecular weight of the IgM heavy and light chain,
respectively. Purified antibody was sterilely filtered and stored until
used at -70°C.
CELL LINES
TARGET: 20pmol
500r
Heavy chain determination was done by immunodiffusion using
immunoglobulin heavy chain specific antisera and antibody-producing
hybridoma culture supernatant with a mouse monoclonal typing kit
from ICN Immunobiologicals, Lisle, IL; light chain determination was
done by an enzyme immunoassay with a monoAb-ID enzyme immunoassay kit from Zymed Laboratories, Inc., San Francisco. CA.
GM2/well
no
10
RESULTS
Production and Characterization of Monoclonal Antibodies
Reacting with GMZ. Mice were immunized with trypsin-harvested D-54 MG human glioma cells. The D-54 MG permanent
glioma cell line was selected as an immunogen on the basis of
the HPTLC profile of its monosialoganglioside fraction which
showed a high proportion of GM2 ganglioside (approximately
80% of the total monosialoganglioside
fraction) (26). Five
splenic fusions with outgrowth rates ranging from 24 to 88%
yielded 67 hybrids reactive with both GM2and D-54 MG cells
by RIA. Eighteen IgM hybrids were cloned in Methocel; five
were selected for high specific anti-GM2 titer and stability and
were designated DMAb-1, DMAb-2, DMAb-3, DMAb-4, and
DMAb-5. Monoclonal antibodies were purified by means of
sequential Chromatographie methods from ascites fluid derived
from hybrid tumor-bearing BALB/c«///««
mice. DMAb-1 as
cites fluid was loaded onto a HPLC Bakerbond ABx ionexchange column, and the resulting IgM fraction was further
subjected to gel exclusion chromatography on a Sephacryl S300 column. DMAb-2 and DMAb-3 were affinity purified by
an anti-IgM affinity column. DMAb-4 and DMAb-5 were iso
lated from ascitic fluid by Sephacryl S-300 gel exclusion chro
matography followed by anti-IgM affinity chromatography.
Purity of monoclonal antibody preparations was greater than
95%, as determined by HPLC gel filtration chromatography
and sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
All characterized anti-GM2 monoclonal antibodies were of the
IgM class with /¿-lightchains.
Binding of Monoclonal Antibodies to <.M•
and D-54 MG Ceils.
Purified MAbs were compared at equal concentrations (ranging
from 20 ng/mi to 100 ng/ml) with regard to reactivity with the
GM2-positive D-54 MG human glioma cell line used as immu
nogen and with purified GM2in solid-surface RIA and HPTLCimmunostain. DMAb-3 appeared to have a higher affinity to
GM2than the other MAbs. On the basis of antibody reactivity
to GM2in RIA (20 pmol GM2/well), the concentrations at which
50% binding to GM2for DMAb-1 through DMAb-5 were found
to be DMAb-1, 1.65; DMAb-2, 0.6; DMAb-3, 0.16; DMAb-4,
2.0; and DMAb-5, 1.1 ¿tg/ml(Fig. \A). However, when D-54
MG cells were used as target (Fig. IB), the pattern of reactivity
observed with purified GM2 was not apparent, and the differ
ences observed were not as apparent as with purified GM2Specificity Analysis with Standard Gangliosides. Structurally
characterized glycolipids, including NeuAc-GM2, NeuGc-GM2,
GA2,GMI, GM3,GD3, GD2, GDia, GalNAc-GDia, GDlb, GTib, and
GQIb were tested in HPTLC-immunostain
with DMAb-1
through DMAb-5 at a concentration of 5 ng/m\ (Table 1). The
amount of ganglioside applied to the plate was 250 pmol of
GMI, Go3, GD2, GDia, G-rib, and GQH,; 200 pmol of GM.i and
GDlb; 500 to 3 pmol of GalNAc-GDU; 25 to 1.5 pmol of NeuAcGM2 in doubling dilutions; 500 pmol of NeuGc-GM2; and 500
pmol of GA2-Reactivity was detected only with NeuAc-GM2 and
GalNAc-Goia, but not with any of the other gangliosides tested
with the exception of DMAb-3, which was weakly positive with
20
5
125
031
008
002
ANTIBODY CONCENTRATION l/xg/ml)
TARGET:D-54 MG
20
5
125
031
008
ANTIBODY CONCENTRATION(^g/ml)
002
Fig. 1. Reactivity of monoclonal antibodies DMAb-1, DMAb-2. DMAb-3,
DMAb-4. and DMAb-5 with D-54 MG human glioma cells and GM2as detected
by cell surface or solid surface RIA. In A. plastic plates were coated with 20 pmol
of purified GM2in 50 ^1 of methanol per well. The wells were dried and treated
with incubation buffer (50 MMTris-HCl, pH 7.8, ISmMNaCI, 1% BSA). Primary
and secondary antibody were applied as for CS-RIA described above. In B, D-54
MG cells were plated at a density of 2 x IO4cells/well and incubated with various
concentrations of specific (DMAb-1, DMAb-2, DMAb-3. DMAb-4, DMAb-5) or
nonspecific (MOPC-104E) antibody. After a washing. IMI-goal anti-mouse IgM
was added and binding ratios were calculated (as described in "Materials and
Methods").
NeuGc-GM2 in excess of 100 pmol. The minimum epitope
recognized by all five MAbs was found to consist of a terminal
A'-acetylgalactosaminyl residue linked by a ß
1-4 linkage to «23 yV-acetylneuraminylgalactose.
Reactivity of Anti-GM2 Monoclonal Antibodies for Human
Malignant and Normal Cells in Vitro. Expression of GM2 by
cultured human malignant and normal cells was assayed with
DMAb-1, DMAb-3, DMAb-4, and DMAb-5 versus nonreactive
IgM control by membrane immunofluorescence and cell surface
RIA (Table 2). Binding of all four monoclonal antibodies at an
optimal concentration of 5 ng/m\ (approximately 90% of max
imum binding on D-54 MG cells) was consistent; however, the
reactivity of the monoclonal antibody DMAb-3 with most cell
lines was higher than that of DMAb-1, DMAb-4, and DMAb5, which is in agreement with the titration results with purified
GM2(see Fig. I/I). DMAb-3 reacted with 17 of 19 permanent
HGL. Of the positive lines, high levels of antibody binding
(RIA binding ratios, >20) were exhibited by 9 HGL, moderate
(BR, 10-20) by 3 HGL and low reactivity (BR <10) by 5 HGL.
Quantitative absorption of DMAb-4 (100 ng) was performed
with the following cell lines (from IO4 to IO7 cells), D-54 MG,
D-32 MG, U-l 18 MG, and SK-MEL 28, which demonstrated
high (the first two), moderate, and no reactivity, respectively.
Absorption was carried out at 0°Cfor l h and the absorbed
antibody was tested in CS-RIA on D-54 MG target cells. D-54
MG was most effective, whereas SK-MEL 28 was least effective
6647
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
GM¡EXPRESSION IN GLIOMA AND MEDULLOBLASTOMA CELL LINES
Table I Specificity analysis with standard gangliosides
Binding of DMAb-1 through DMAb-5 to various structurally characterized gangliosides by enzyme immunostaining on thin layer chromatography plates (•,
glucose: D, galactose; O, galactosamine: V, NeuAc, V, NeuGc). Glycolipids were separated by HPTLC on silica gel-coated aluminum sheets in chloroform:methanol:0.25% aqueous KCI (50:40:10. by volume): immunostaining at a primary antibody concentration of 5 jjg/ml was done as described in "Materials and
Methods." Ganglioside reference standards were visualized with orcinol staining. The minimum amount of GM: (NeuAc and NeuGc) and GalNAc-GDi. (in pmol)
reactive with MAbs 1-5 is indicated.
Reactivity" with
Ganglioside
DMAb-1
DMAb-2
DMAb-3
DMAb-4
DMAb-5
GM2(NeuAc)
1.5
1.5
GM2(NeuGc)
O—P
GalNAc-GD1,
O—D—•
GA2
o—C
GD2
1.5
1.5
100
25
10
10
10
GMI
GD1.
GDI»
* Minimum pmol ganglioside reactive in HPTLC-immunostain.
—,absence of reactivity.
in absorbing DMAb-4 binding activity. The number of cells
absorbing 50% DMAb-4 binding activity were IO4 for D-54
MG, 10' for D-32 MG, and 6 x 10s for U-l 18 MG (Fig. 2).
All medulloblastoma and neuroblastoma cell lines tested were
positive with all four antibodies, with TE-671 and Daoy among
medulloblastomas and SK-N-SH and IMR-32 among neuroblastomas showing the highest binding. All MAbs reacted very
strongly (BR > 20) with 1 of 3 melanoma, 1 of 1 rhabdomyosarcoma, and 1 of 1 osteosarcoma cell lines. In addition,
low reactivity was exhibited by 1 of 3 teratocarcinomas. All
other cell lines tested were negative (BR < 3 in RIA and FI <
10) including 1 fibroblast, 2 teratocarcinoma, 1 B-cell leukemia,
1 human myeloid (HL-60), 1 breast carcinoma, and 1 epidermoid carcinoma cell lines.
In general, the results seen in RIA tests with normal and
malignant human cells corresponded with those seen in immunofluorescence. However, three notable exceptions were ob
served. Human glioma cell lines U-251 MGsp, U-373 MG, and
U-l 18 MG were weakly positive or even negative (depending
on the MAb tested) in RIA at 37°C,whereas in immunofluorescent assays all these were found moderately to strongly
positive (Table 2). Subsequent comparison of RIA reactivity at
37°Cand 0°Cover the same incubation period (1 h) showed
the highest binding ratios occurring at 0°Cand the lowest at
37°C.The largest difference in BRs was observed with DMAb4 with 3.2 and 42.2 for U-251 MGsp; 1.2 and 14.1 for U-373
MG; and 2.8 and 16.6 for U-l 18 MG, between assays per
formed at 37°Cand 0°C,respectively. Furthermore, when
U-251 MGsp cells were tested in CS-RIA at 37°Cversus 0°C
for 0.5, 1, 2, and 4.5 h with DMAb-1, DMAb-3, DMAb-4, and
DMAb-5 (at a concentration of 5 ng/ml), a constant highly
significant difference was present with all antibodies tested
throughout the incubation period (P < 0.01; t test), the highest
BRs occurring at 0°Cand after l h of primary antibody incu
bation.
Densitometric Analysis and Immunostaining of Ganglioside
Fractions. To confirm the molecule reacting with the anti-GM2
MAbs in CS-RIA and immunofluorescence assays (Table 2),
ganglioside extractions were performed on most available
glioma and medulloblastoma cell lines (as in "Materials and
Methods"). Monosialoganglioside fractions were separated on
HPTLC and were immunostained with DMAb-3 and DMAb-
6648
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
GMI EXPRESSION IN GLIOMA AND MEDULLOBLASTOMA CELL LINES
Table 2 Terminal "G
L'pitopeexpression in normal and malignant human
cells
>
5s
Confluent cell monolayersin 96-well plastic plates »ereincubated with DMAb1, DMAb-3, DMAb-4. and DMAb-5 for l h at 37'C. After incubation with '"I-
LU
goat anti-mouse IgM. cell-bound cpm were detected with a gamma counter.
Indirect membrane immunofluorescence was performed at 4°Cwith 2 h incuba
s
g
CD
tion for both primary and secondary antibodies. Binding ratios and fluorescence
indices were calculated as described in "Materials and Methods." Binding ratios
are shown as 0-3 (—,negative); 3-10 (+, weakly positive); 10-20 (++, positive);
>20 (+++, strongly positive). Experimental antibody activity giving 10cr or more
z•:
o
uj
m
Qt
fluorescing cells above that obtained with the control antibody of the same isotype
was considered significant antibody binding [Wikstrand et al. (17)].
(5
i/i
CC
Z
FI+++
GliomasD-54
MGU-343
MGD-245
MGU-251
MGD-247
MGD-37
MGU-l
MGD-65
38
MGU-251
MGspD-270
MGU-373
MGD-263
MGD-32
MGD-344
MGU-118MGU-105
Fl+++
FI+++
5«
FI+++
O
IO4
ABSORBING
91+++
65++
56+++
50++
45+
28++
35++
4531++
3427++
29++
28+
21-116+
16-1342++
MGU-410
MGD-259
MGD-32cl2
MGMedulloblastomasTE-671DaoyD283
95+++
91++
81+++
63+++
62+++
57+++
58+++
49+
45+++
51+
38++
37+++
30++
33+
19+
17+
983+++
94+++
87++
73+++
68+++
71+++
61++
41++
41+
56++
4541+
37+++
32+
2834-118+
1182++
94+++
85++
73+++
64+++
65+++
55++
54++
49-151++
4439+
32++
30-13032+
18+
97_
-GLIOMAS
—-GOlo
- -GDlb
GTlb
Y Y YY1
"¿a?1
"W
YY
U-4W
850+
810-
68+++
56++
37++
24++
26+++
•l-
. . 0-263
YT. YY
YY
U-105 ,,.OY
0-270
MG
MG
MG
DAOY
i—ii—j
i—ii—i
12
12
D-M4
MtD
——
l 2
GM2 Iff
-GM3
-
« »-GM2
—-GMI
•-GDlo
YY
D-37MO
5501++
12
12
LAN-I GM2 Ulf
tT
MG
70+++
51++
35++
31++
30+++
.-MEDUll.» -.-MISCaBiASTOMAS
LANEOUS
r
69+++
49++
71++
65+++
49+++
51+++
44+++
46++
39++
46+++
49++
43++
MedD341
26-127++
26++
28++
MedD384
15+++
21+++
26+++
22+++
MedNeuroblastomasSK-N-SHIMR-32LAN-5SK-N-MCLAN-1MelanomasKENTSK-MEL
72+++
52+++
39+++
43++
31++
IO6
Fig. 2. Absorption of DMAb-4 with human malignant cells. DMAb-4. 100 ng
in 200 ^1 ZO lx-10% FCS. was incubated for l h at 0°Cwith 10". 10!. IO6, or
IO7 cells (D-54 MG. D-32 MG, U-l 18 MG. or SK-MEL 28 cells). After spinning
at 2000 rpm for 10 min, the absorbed reagent was tested in CS-RIA on D-54 MG
target cells. Results are expressed as percentage of reactivity of unabsorbed
antibody.
2+++
69+++
42++
25+
15++
32+++
IO5
CELL NUMBER
YY
TE-671
YY
GM2 REF
7102+++
Fig. 3. HPTLC-immunostaining of monosialoganglioside fractions extracted
from human tumor cells with monoclonal antibody DMAb-3. Gangliosides were
separated by high performance thin layer chromatography on silica gel-coated
aluminum sheets in chloroform:methanol:0.25% aqueous KCI in H;O (50:40:10.
by volume). Lane 1. immunostain pattern of monosialoganglioside fractions
42++
66+++
75+++
54+++
(rhabdomyosarcoma)2T
36
derived from the following cell lines: D-54 MG (50 pmol); D-245 MG (100
56+
(osteosarcoma)PA-1
55700— 63+
61+
(teratocarcinoma)ATCC
1440100—13401330500— pmol); U-251 MG (30 pmol); D-247 MG (200 pmol); D-37 MG (100 pmol); U410 MG (300 pmol); D-270 MG ( 150 pmol); U-373 MG (125 pmol); D-263 MG
fibroblast)TERA-1
1147 (adult
(450 pmol); U-105 MG (300 pmol); TE-671 (100 pmol); Daoy (150 pmol):
(teratocarcinoma)TERA-2
20000DMAb-3RIA 50034DMAb-5RIAD283 Med (500 pmol); D384 Med (500 pmol); SK-MEL 28 (500 pmol); TERA(teratocarcinoma)HL-60
2 (500 pmol); LAN-1 (125 pmol). Lane 2, an orcinol-stained chromatogram of
myeloid)SB
(human
leukemia)DU-4475
(B-cell
the same monosialoganglioside fractions. The amounts of total gangliosides
~>1DMAb-4RIA
loaded per lane are: D-54 MG (6 nmol): D-245 MG (1.5 nmol); U-251 MG (2.5
11
carcinoma)A431
(breast
nmol): D-247 MG (1.8 nmol): D-34 MG (2.7 nmol): U-410 MG (3.3 nmol): D(epidermoid carcinoma)DMAb-1RIA
270 MG (1.5 nmol); U-373 MG (4 nmol): D-263 MG (2.7 nmol); U-105 MG
(10 nmol): TE-671 (2.4 nmol); Daoy (4 nmol); D283 Med (15 nmol); D384 Med
5 (Fig. 3). A double band was stained in all glioma and medul- (2.1 nmol); SK-MEL 28 (2.5 nmol); TERA-2 (2.9 nmol); LAN-1 (15 nmol).
Ganglioside reference standards were visualized with oreinol staining. The
loblastoma extracts in contrast to the purified GM2 standard
amounts of GM2 applied to TLC plates for immunostaining and oreinol staining
from Tay-Sachs brain, which was stained as a single band. The were 50 and 500 pmol. respectively. Immunostaining at a primary antibody
concentration of 5 Mg/ml was done as described in "Materials and Methods."
28T8MiscellaneousRD
3+++
5+++
upper band migrated slightly ahead of the GM2 standard,
whereas the lower band migrated between GM2 and GM1. The
double band was present when extracts were run in chloroform:methanol:0.25% aqueous KCI (50:40:10, by volume) but
resolved into a single band migrating as standard GM2 when
extracts were developed in l-propanol:2.5 M ammonia (75:25,
by volume). Alkaline treatment of cell line extracts did not alter
the chromatograpahic migration of the double band suggesting
the lack of an alkali-labile O-acetylated sialic acid residue.
Furthermore, mixing cell line extracts showing a double band
in HPTLC-immunostain
with purified standard GM2 from
Tay-Sachs brain resulted in three bands (in chloroform:
methanol:0.25% aqueous KCI, 50:40:10, by volume), indicating
that the doublet is not due to GM2-degrading enzymes present
in the cell line extracts. These results suggest that the doublet
is due to differences in the ceramide portion of the molecule.
The amount of total ganglioside added in each lane is given
in detail in the legend to Fig. 3. Staining was relatively weak
with the monosialoganglioside extract prepared from D384
6649
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
EXPRESSION
IN GLIOMA AND MEDt'LLOBLASTOMA
CELL LINES
Med and D283 Med. No reactivity was seen with the TERA-2
and SK-MEL 28 cell extracts.
Densitometric scanning of the monosialoganglioside frac
tions (separated on HPTLC and sprayed with resorcinol) from
17 cancer cell lines was used to determine the proportion of
GM2in the monosialo fraction and the amount of GM2in nmol/
mg of protein (Table 3). Cell lines are listed in decreasing
amount of CM: (in nmol of GM2/mg of protein). Two cell lines
for which a protein determination was not performed were
listed according to nmol of GM2/g of wet cell pellet. Glioma
cell lines showed a range in the amount of GM2 from 0.6 for
D-54 MG to <0.1 nmol/mg for D-263 MG. Among medulloblastoma cell lines, Daoy and TE-671 showed a high amount
of GM2,at least 5 times more than D283 Med and D384 Med.
Furthermore, high GM2 content was found in LAN-1 neuro
blastoma cells.
cells in suspension CS-RIA, higher reactivity was observed with
trypsin-treated D-54 MG cells.4
A thorough specificity analysis by HPTLC-enzyme-linked
immunosorbent assay (Gothenburg, Sweden) of all five MAbs
showed absence of reactivity with closely related gangliosides
including GA2,GM.i,GD2,GM1,GDU. GD.i,Gmb, GTu,. and GQib.
This indicated that the terminal GalNAc/il-4(NeuAc«2-3)Gal
was the minimum required epitope recognized by all five MAbs.
It is noteworthy that all MAbs showed some binding to
GalNAc-Gma, a minor component of normal human adult brain
(14). GalNAc-GDla, like NeuAc-GM2, shares the GalNAc/314(NeuAc«2-3)Gal terminal epitope. Similar reactivity was re
ported with monoclonal IgM sera from two patients with gammopathy and neuropathy which were shown to react strongly
with GM2, GalNAc-isoGM1, and GalNAc-Gr>,a (28). All MAbs
reacted exclusively with NeuAc-GM2 with the exception of
DMAb-3, which also reacted weakly with NeuGc-GM2.
DMAb-1 and DMAb-3 through DMAb-5 were used as probes
to detect the presence of GM: on human cultured cells from
DISCUSSION
tumors of neuroectodermal origin. Most gliomas, all medulloblastomas, and all neuroblastomas were found to be positive.
In this study we describe the generation of five new murine
However, antibody reactivity was not restricted to tumor cells
MAbs that react with the GM2 ganglioside, which is the major
of neuroectodermal origin but was also seen with osteosarcoma,
ganglioside of the D-54 MG human glioma cell line (26).
teratocarcinoma, and rhabdomyosarcoma, which do not share
Several modes of immunization were used in order to generate
a common embryological origin. Interestingly, a recent immuspecific GM2-reactive antibodies, including EDTA-harvested
nohistochemical study of lung carcinomas with the MK1-16
D-54 MG cells with and without cyclophosphamide pretreat
MAb, which also recognizes GM2,showed that GM2was present
ment (27); trypsin-harvested D-54 MG cells; and GM2coupled
not only in small cell lung carcinomas as neuroectodermal
to Salmonella minnesota. Cell trypsinization was determined
tumor-associated antigen but also in the squamous cell carci
to be the most effective immunization procedure with regard to noma and adenocarcinoma of the lung (12).
antibody response against GM2. Trypsin, as a proteolytic en
Densitometric scanning in conjunction with HPTLC-immuzyme, may liberate and degrade extracellular matrix as well as nostain confirmed the results obtained by CS-RIA and immuintegral membrane proteins that may mask a partially cryptic
nofluorescence. There was a good correlation between the
antigenic determinant. When DMAb-4 was used retrospectively
amount of GM2 per mg of protein (Table 3) for different cell
versus equal numbers of EDTA or trypsin-detached D-54 MG
lines and the antibody reactivity of the same cell lines seen in
CS-RIA and immunofluorescence (Table 2). TERA-2 and SKTable .1 Densitomctric evaluation of ganglioside extracts from human glioma
MEL 28, which were negative in both CS-RIA and immunoflu
and medullohlasloma cell lines
orescence, also failed to immunostain with DMAb-3, even at
Gangliosidcs were separated by HPTLC in chloroform:melhanol:0.25% KC1
500 pmol/lane (Fig. 3). HPTLC-immunostain showed the pres
(50:40:10, by volume) and visualized with resorcinol. Densitometric scanning was
performed at 620 nm. The amount of GM2in nmol/mg of protein was calculated
ence of two bands located in the GM2region (Fig. 3). Purely on
from the following formula: total sialic acid in the monosialo fraction (in nmol)
the basis of their migration in neutral and alkaline solvents, it
X % GM2in the monosialo fraction (as determined by densitometry)/total amount
of protein (in mg) as determined by the Lowry assay. Two cell lines. U-343 MG
seems unlikely that these bands are related to other known
gangliosides sharing the "GM2" terminal epitope. Furthermore,
(glioma) and Daoy (medulloblastoma), for which a protein determination was not
performed, were listed according to nmol of GM2/g of wet cell pellet (U-343 MG.
the possibility of alkali-labile O-acetylated sialic acid residues
395 nmol/g; D-54 MG, 138 nmol/g; Daoy, 172 nmol/g; TE-671, 106 nmol/g).
or GM2-degrading enzymes present in the cell extracts was
of GM2of
excluded. However, differences in fatty acid length are known
monosialoganglioside
of GM2/mg
fraction8377745999637079II177464692510000"
proteinND°0.60.30.30.20.20.20.10.1<0.1ND0.4<0.1<0.10.600%
lineGliomaU-343
Cell
to give the doublet appearance of GM.iand GD.Ion HPTLC (29,
30). Such differences in the lipophilic constituents of GM2could
MC;D-54
explain the observed ¡mmunostaining pattern in human glioma
MGU-251
MGD-247
line extracts.
MGU-410MGU-373
The present study shows that GM2 represents a major gan
glioside of human glioma, medulloblastoma,
and neuro
MGD-37
blastoma cells in vitro. DMAb-1 through DMAb-5 can also be
MGD-270
MGU-105
used to define shifts in ganglioside composition of biopsies
MGD-263
versus cultured cells, to study the expression of GM2 during
MGMedulloblastomaDaoyTE-671D283
development, and to find applications in tumor diagnosis and
immunotherapy.
MedD384
MedMiscellaneousLAN-1SK-MEL
28TERA-2nmol
ND, not determined.
Note Added in Proof
Since submission of this manuscript, it has been established that TE671, a human cell line initially reported to be derived from a cercbellar
4 F. D. Vrionis, C. J. Wikstrand, P. Fredman. J-E. Mansson, L. Svennerholm.
and D. D. Bigner, unpublished results.
6650
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
GM¡EXPRESSION IN GLIOMA AND MEDULLOBLASTOMA CELL LINES
12. Miyakc, M., Ilo, M.. Hitomi. S.. Ikeda. S., Taki. T., Kurata, M.. Hiño,A.,
Miyake, N., and Kannagi, R. Generation of two murine monoclonal antibod
ies that can discriminate A'-glycolyl and N-acetyl neuraminic acid residues of
GM2gangliosides. Cancer Res.. 48: 6154-6160. 1988.
13. Rosengren. B., Mansson. J-E., and Svennerholm. L. Composition of ganglio
sides and neutral glycosphingolipids of brain in classical Tay-Sachs and
Sandhoff disease: more lyso-GM2 in Sandhoff disease? J. Neurochem.. 49:
834-840. 1987.
14. Svennerholm. L.. Mansson, J-E., and Li, Y-T. Isolation and structural
determination of a novel ganglioside. a disialosylpentahexosylceramide from
human brain. J. Biol. Chem.. 248: 740-742, 1973.
15. Svennerholm. L.. and Fredman, P. A procedure for the quantitative isolation
of brain gangliosides. Biochim. Biophys. Acta. 617: 97-109. 1980.
16. Svennerholm. L. Quantitative estimation of sialic acids. II. A colorimetrie
resorcinol-hydrochloric acid method. Biochim. Biophys. Acta. 24: 604-611.
1957.
17. Wikslrand. C. J.. Mahaley. M. S.. and Bigner. D. D. Surface antigenic
characteristics of human glial brain tumor cells. Cancer Res., 37:4267-4275,
1977.
18. Wikstrand. C. J., and Bigner. D. D. Expression of human fetal brain antigens
by human tumors of neuroectodermal origin as defined by monoclonal
antibodies. Cancer Res., 42: 267-272, 1982.
19. Bourdon, M. A., Wikstrand, C. J., Furthmayr. H.. Matthews. T. J.. and
Bigner. D. D. Human glioma-mesenchymal extracellular matrix antigen
defined by monoclonal antibody. Cancer Res.. 43: 2796-2805, 1983.
20. Fredman. P.. Magnani. J. L.. Nirenberg, M., and Ginsburg, V. Monoclonal
antibody
reacts
many 1984.
gangliosides in neuronal tissue. Arch.
Biochem. A2B5
Biophys..
233:with
661-666."
medulloblastoma (a), is a subline of the rhabdomyosarcoma cell line,
RD, established several years earlier in Dr. McAllister's laboratory (h,
c, d). Therefore, the number of medulloblastoma cell lines investigated
is reduced to four (Tables 2 and 3. Fig. 3).
a. McAllister, R. M., Isaacs, H., Rongey, R., Peer, M., Au, W., Soulap,
S. W., and Gardner. M. B. Establishment of a human medulloblas
toma cell line. Int. J. Cancer, 20: 206-212, 1977.
b. McAllister, R. M., Melnyk, J., Finlestein, J. Z., Adams, E. C., and
Gardner, M. B. Cultivation in vitro of cells derived from a human
rhabdomyosarcoma. Cancer (Phila.), 24: 520-526, 1969.
c. Stratton, M. R., Reeves, B. R., and Cooper, C. S. Misidentified cell.
Nature (Lond.). 337: 311. 1989.
d. Stratton, M. R., Darling, J., Pilkington, G. J., Lantos, P. L., Reeves,
B. R., and Cooper, C. S. Characterization of the human cell line
TE-671. Carcinogenesis (Lond.), 10: 899-905, 1989.
ACKNOWLEDGMENTS
The authors wish to thank Beatrice Brcwington and Laura Shaughnessy for expert technical assistance. Nancy Hall for excellent secre
tarial assistance, and Ann Tamariz for editorial assistance.
REFERENCES
1. Bigner. D. D.. Signer. S. H.. Punten. J.. Westermark. B.. Mahaley. M. S.,
Ruoshlati. E., Herschman. H., Eng. L. F., and Wikstrand, C'.J. Heterogeneity
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
of gcnolypic and phenotypic characteristics of fifteen permanent cell lines
derived from human gliomas. J. Neuropalhol. Exp. Neurol.. 40: 201-229.
1981.
Kostic. D.. and Buchheit. I". Gangliosidcs in human brain tumors. Life Sci.
[II]. 9:589-596. 1970.
Yates. A. J.. Thompson. D. K.. Boesel, C. P.. Albrightson. C.. and Hart. R.
W. Lipid composition of human neural tumors. J. Lipid Res.. 20: 428-436.
1979.
Traylor. T. D.. and Hogan, E. L. Gangliosides of human cerebral astrocytomas. J. Neurochem.. 34: 126-131. 1980.
Fredman. P.. Hoist, H.. Collins. V. P., Ammar. A.. Dellhcdcn. B., Wahren.
B.. Granholm. L., and Svennerholm, L. Potential ganglioside antigens asso
ciated with human gliomas. Neurol. Res.. 8: 123-126. 1986.
Mansson, J-E., Fredman. P., Bigner. D. D.. Molin. K.. Rosengren. B..
Friedman. H. S.. and Svennerholm. L. Characterization of new gangliosides
of the lactotetraose series in murine xenografts of a human glioma cell line.
FEBS Lett.. 201: 109-113. 1986.
Gottfries. J., Mansson, J-E.. Fredman. P.. Wikstrand, C. J.. Friedman. H.
S., Bigner, D. D.. and Svennerholm, L. Ganglioside mapping of a human
medulloblastoma xenograft. Acta Neuropathol.. 77: 283-288. 1989.
Svennerholm. L. Chromatographie separation of human brain gangliosides.
J. Neurochem.. 10:613-623, 1963.
Irie. R. F.. Irie. K.. and Morton, D. L. A membrane antigen common to
human cancer and fetal brain tissues. Cancer Res.. Ì6:
3510-3517. 1976.
Tai. T.. Paulson, J. C., Cahan. L. D.. and Irie, R. F. Ganglioside GM¡as a
human tumor antigen (OFA-l-1). Proc. Nati. Acad. Sci. USA. 80: 53925396. 1983.
Natoli, E. J.. Livingston. P. O., Pukel. C. S.. Lloyd, K. O., Wiegandl. H.,
Szalay. J.. Oettgen. H. F.. and Old. L. J. A murine monoclonal antibody
detecting A'-acetyl and .V-glycolyl-GM2:characterization of cell surface reac
tivity. Cancer Res.. 4f>:4116-4 ¡20.1986.
21. Bigncr. S. H.. Mark, J.. and Bigner. D. D. Chromosomal composition of
four permanent cultured cell lines derived from human gliomas. Cancer
Genet. Cytogenet.. 10: 335-349, 1983.
22. Bigner, S. H., Mark, J., and Bigner, D. D. Chromosomal progression of
malignant human gliomas from biopsy to establishment as permanent lines
in rilro. Cancer Genet. Cytogenet.. 24: 163-176, 1987.
23. Mark. J.. Ponten. J.. and Westermark. B. Cytogenetical studies with G-band
technique of established cell lines of human malignant gliomas. Hereditas.
78: 304-307. 1974.
24. Mark. J., W'estermark. B.. Ponten. J.. and Hugosson. R. Banding patterns
in human glioma cell lines. Hereditas, 87: 243-260, 1977.
25. He. H-M., Skapek, S., Wikstrand, C., Friedman, H.. Kemshead. J., Coakham. H.. Bigner, S., and Bigncr. D. D. Phenotypic analysis of four human
medulloblastoma cell lines and transplantable xenografts. J. Neuropathol.
Exp. Neurol., 48: 48-68. 1989.
26. Poduslo, S. E., Miller. K.. and Jany. Y. Comparison of lipids and lipid
metabolism in a human glioma cell line, its clone and oligodcndroglia. Cancer
Res., 43: 1014-1018. 1983.
27. Livingston. P. O.. Deleo. A. B.. Jones, M., and Oettgen. F. Comparison of
approaches for augmenting the serologie response to the individually specific
methylcholanthrene-induccd sarcoma-meth A: pretreatment with cyclophosphamide is most effective. J. Immunol.. 131: 2601-2605. 1983.
28. Ilyas. A. A., Li, S-O, Chou, D. K., Li, Y-T., Jungalwala, F. B., Dalakas, M.
C, and Quarles, R. H. Gangliosides GM!. IV GalNAcGMlb, and IV4
GalNAcGou as antigens for monoclonal immunoglobulin M in neuropathy
associated with gammopathy. J. Biol. Chem., 263: 4369-4373. 1988.
29. Ando, S., Chang, N-C., and Yu, R. K. High performance thin layer chromalography and densitometric determination of brain ganglioside composi
tions of several species. Anal. Biochem., 89: 437-450. 1978.
30. Ando. S.. and Yu. R. K. Fatty acid and long-chain base composition of
gangliosides isolated from adult human brain. J. Neurosci. Res., 12: 205211. 1984.
31. RJPAC-IL'B Commission on Biochemical Nomenclature (CBN). The no
menclature of lipids. Eur. J. Biochem., 79: 11-21. 1977.
32. Fredman, P., Mansson, J.-E.. Wikstrand, C. J., Vrionis, F. D., Bigner. D.
D.. and Svennerholm. L. A new ganglioside of the lactotetraose series.
GalNAc 3'-isolLMi detected in human meconium. J. Biol. Chem.. 264:
12122-12125. 1989.
6651
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.
Five New Epitope-defined Monoclonal Antibodies Reactive with
G M2 and Human Glioma and Medulloblastoma Cell Lines
Fotios D. Vrionis, Carol J. Wikstrand, Pam Fredman, et al.
Cancer Res 1989;49:6645-6651.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/49/23/6645
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1989 American Association for Cancer Research.