Content uploaded by Mepur H. Ravindranath
Author content
All content in this area was uploaded by Mepur H. Ravindranath on Mar 20, 2017
Content may be subject to copyright.
(CANCERRESEARCH56. 5666-5671. December 15, 19961
ABSTRACT
A human B-lymphoblastoid cell clone, L55-81, that produces human
monoclonal antibody (MAb) to ganglioside G@ was established from
peripheral blood B lymphocytes of a melanoma patient. L55-81 secretes
IgMK light chain antibody in a serum-free medium. GM2 specificity of the
antibody was tested by immune adherence assay, TLC immunostaining,
and ELISA. Anti-GM2antibody was shown to have the ability to kill the
GM2-rich human melanoma cell line M14 in the presence of human or
rabbit complement. A purified L55-81 MAb (>99.5% purity in protein
concentration) was biotinylated and tested for its reactivity to various
histological-type biopsied tumor and normal tissues in an avidin-biotin
detection system. L55-81 MAb (20 tag/mI) reacted with several types of
tumor tissues such as melanoma (7 of 10), colon carcinoma (4 of 5), ovary
carcinoma (4 of 5), breast carcinoma (1 of 5), kidney carcinoma (1 of 5),
and prostate carcinoma (1 of 5). None of the normal tissues derived from
24 different organs and adjacent normal tissues surrounding the cancer
ous tissues were stained. Production of the antibody in a serum-free
medium, the cytotoxic potential with human complement, the inability to
react to normal tissues, and the ability to target antigen-specific target
cells make L55-81 a potential therapeutic agent for the treatment of
cancers expressing ganglioside GM2.
INTRODUCTION
Ganglioside GM2 [GalNAcj3l-4(NeuNAccs2-3)Galf3l-4Glcj3l-lce
ramide] is a major tumor-differentiation antigen in human melanoma
(1—3)and hepatoma (4). Ganglioside antigens have been one of the
prominent antigens used as targets for the study of immunotherapy of
cancer using MAbs3 (5, 6). Murine MAbs reactive to ganglioside
antigens have shown partial or complete responses in the treatment of
neuroblastoma and melanoma patients (7—15).However, the effec
tiveness is usually only transient, and higher doses or repeated anti
body injections were not attempted in these trials due to the toxicity
or human anti-mouse antibody responses in patients. Because human
MAbs do not usually recognize human normal tissues, they offer
potential advantages over murine MAbs and chimeric human-mouse
MAbs for cancer therapy. Human MAbs do not elicit human anti
mouse antibody responses and interact more efficiently with the
human effector system to activate potentially therapeutically useful
functions. In spite of these obvious advantages, the number of human
MAt, cell lines that stably secrete a high density of human MAb has
been few; thus, their therapeutic potential has not yet been deter
mined. In 1982 and 1983, our laboratory developed two human MAbs,
L72 and L55, that had specificity for ganglioside GD2 and GM2,
respectively (1, 16, 17), and, more recently, developed human MAb
L612 to ganglioside GM3 (18). The immunoglobulin class of these
antibodies is 1gM with a biologically functional Fc portion.
These cell lines were initially grown in a culture medium contain
Received 7/2/96; accepted 10/16/96.
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.
I Supported by NIH/National Cancer Institute Grants CA30647 and CA 12582.
2 To whom requests for reprints should be addressed, at Department of Biotechnology
Sciences, John Wayne Cancer Institute, 2200 Santa Monica Boulevard, Santa Monica, CA
90404.Phone:(310)449-5218;Fax:(310)998-3915.
3 The abbreviations used are: MAb, monoclonal antibody; IA, immune adherence;
AJCC, American Joint Committee on Cancer.
ing FCS. After repeated recloning and testing of the cell lines in
several different commercially available serum-free media, some but
not all cell lines were successfully adapted to a serum-free medium. In
an effort to establish cell lines that could produce human MAb to GM2
in a serum-free medium, we cloned two new human B cell lines
(L55-8 1 and L55-637) and successfully adapted one (L55-81) of these
cell lines to AIM-V. The adaptation of human 1gM MAb-producing
cell lines to a serum-free medium is essential for clinical therapeutic
trials because it is nearly impossible to purify 1gM antibody to a
clinical grade from spent medium containing animal serum within a
reasonable cost.
In this investigation, we have assessed three major preclinical
criteria of the suitability of the antibody for passive immunotherapy of
cancer: (a) the epitope specificity; (b) antitumor cytotoxic potential;
and (c) the level of reactivity to cancer and normal tissues. The
L55-81 cell line not only produces human MAb in serum-free medium
but also secretes a greater amount of antibody (10—30 @g/ml)than our
original L55 cell line (1—4@tg/ml)in standard plastic flasks. When the
cell line is grown in a hollow fiber system (AcuSyst-Maximizer 1000;
Cellex, Minneapolis, MN), the antibody concentration increased to
200—400 @g/ml.Using a standard procedure of 1gM purification, we
were able to achieve over 99.5% purity of L55-8l antibody. The
present study was designed to investigate at a preclinical level the
therapeutic usefulness of human MAb to GM2. The antibody was
demonstrated to be highly specific to cancer tissues and killed anti
gen-positive cancer cells in the presence of complement. As suggested
in a previous report citing the original L55 antibody (17), the L55-81
antibody reacted not only to one particular histological type of cancer
but to a variety of cancer tissues. No normal tissues obtained from 24
different organs or normal parts of tissues surrounding cancer tissues
were positive for the antibody. This indicates that the antibody may be
applicable to the treatment of many different cancers without causing
toxicity.
MATERIALS AND METHODS
Establishment of Human MAb to Ganglioside G@. Human B cell lines
that produce human MAb specific for ganglioside GM2were established from
peripheral blood B lymphocytes of a melanoma patient using the EBV trans
formation technique as described previously (17). The transformed B-lympho
blastoid cells that produced human MAb to GM2 were viably frozen. The cells
were thawed, and additional clonings were performed to obtain two relatively
higher-producer L55-8l and L55-637 cell lines. These antibodies were of 1gM
class with Klight chains. After recloning three times in serum-free medium
(AIM-V; Life Technologies, Inc., Gaithersburg, MD), L55-8l was successfully
adapted to the medium, whereas L55-637 failed to grow. Human 1gMsecreted
in the spent tissue culture medium was measured by a double sandwich ELISA.
Functional antibodies were monitored by IA assay using GM2-rich human
melanoma cell line M14 (17) and ELISA using purified GM2(19, 20). The
epitope specificity of the antibodies was determined by ELISA (20) and TLC
immunostaining (21) using a variety of gangliosides and neutral glycolipids as
antigen sources.
Glycolipids. Glycosphingolipids from melanoma cell lines used in TLC
immunostaining assay to determine the specificity of human MAb were
extracted and purified according to methods reported previously (2). Commer
cially available human and bovine gangliosides and neutral glycolipids used in
ELISA were purchased from Sigma (St. Louis, MO), KG. Pullman (Munich,
5666
Development of a Human Monoclonal Antibody to Ganglioside G@ with Potential
for Cancer Treatment1
Yumiko Nishinaka, Mepur H. Ravindranath, and Reiko F. Irie2
Departments of Biotechnology Sciences fY. N., R. F. I.] and Glycolipid immunotherapy (M. H. RI, John Wayne Cancer Institute, Santa Monica, California 90404
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
Table 1 Reactivity of 155-81 human MAb in IA assay to five human melano
lines with different ganglioside profilesma
cellMelanoma
cell linesIA
score― Gangliosides―(nmol/g
dry weight of cell)
L55-81 L6l2 GM2 GD2GM3GD3Ml4
M24
M12
Ml5
M25+++
+ 4.0 2.4 7.3
+ — 8.2 0.7 0.3
— +++ <0.1 <0.1 18.1
— +++ 0.5 <0.1 20.6
— — <0. 1 0. 1 1 1 .74.2
0.4
2.8
2.0
10.0
HUMAN MAb TO GANGLIOSIDE GM2
Germany), or Advanced Immuno Chemical, Inc. (Long Beach, CA). A quan
titative determination of the individual glycosphingolipids was performed by
TLC densitometry, as described in a previous report (21).
Complement-dependent Cytotoxicity Assay. Cytotoxicitywas measured
in a 4-h 51Cr release assay. Target cells (1 X 106 cells) were labeled with 160
@tCi51Cr for 1 h at 37°C.After washing, the labeled cells (1.2 X l0@cells)
were mixed with 120 p1 of antibodies diluted with veronal buffer-containing
complement at 1:5 ratio. After a 4-h incubation at 37°C,the 5tCr release
activity in the supernatants was counted. The percentage of specific@ Cr
release was determined according to the following formula.
Percentage of specific release
— (release with antibody and complement— spontaneous release) < 100
— (maximum release — spontaneous release)
All test systems were done in triplicate for each experiment.
Biotinylatlon of the Antibody. Biotinylation of L55-8l human MAb was
performed with biotin hydrazide (Pierce, Rockford, IL) according to the
manufacturer's protocol. Briefly, antibody (0.1—0.5mg/ml) purified by ion
exchange chromatography and gel filtration was dialyzed against 0.1 Macetate
buffer (pH 5.5) containing 0.1 MNaCI. A 0.1 volume of 0.1 MNaIO4 was
added to the antibody solution to oxidize the antibody for 20 mm on ice in the
dark. The oxidized antibody was separated from excess reagent by a Sephadex
G-25 (PD-b) column. A 0.2 volume of 10 mr@ibiotin hydrazide was added to
the oxidized antibody solution. After incubation for 1 h at room temperature,
the biotinylated antibody was separated from excess biotin hydrazide by a
PD-b column.
Immunohistochemistry.Humanmalignantand normal specimenswere
collected shortly after surgery, immediately snap-frozen in liquid nitrogen, and
stored at —70°C.Four-p@mcryostat sections of Tissue-Tek-embedded tissues
were prepared and frozen at —20°Cuntil staining. Cytospin slides for mela
noma cell lines were prepared by spinning cells for S mm at 1000 rpm. Slides
were air-dried overnight at room temperature and stored at 4°Cuntil staining.
Cells and tissues on slides were fixed with cold acetone before immuno
staining. Immunostaining was carried out in an avidin-biotin complex immu
noperoxidase system. Sections were sequentially blocked with hydrogen per
oxide (0.3%) for 5 mm, avidin D blocking solution for 15mm, biotin blocking
solution (Vector SP200l ; Vector Laboratories, Inc., Burlingame, CA) for 15
mm,and4%BSAfor20mm.Biotinylatedantibodywasaddedtothe sections
and incubatedfor 1 h in a humiditychamberat room temperature.Sections
were stained using the Vectastain ABC Elite Reaction Kit solution (Vector
Laboratories, Inc.) and counterstained with Harris-hematoxylin. Staining in
tensity of the tissues was graded as + + + (very strong), + + (strong), + (mod
crate), ±(weak), and —(negative).
RESULTS
Binding Specificity of L55-81 Human MAb to GM2. The binding
specificity of L55-8l human MAb to the cell surface of melanoma cell
lines was assessed by IA assay using five different melanoma cell
lines expressing different GM2 antigen density. L55-81 demonstrated
positive reactivity to melanoma cell lines with a relatively high GM2
3.0
E 2.5
@2.0
@ 1.5
11.0
,@ 03
0.0 AsGM1 GM! GD1a GT1b GD1b GM3 GM2 GD3 GD2
Glycolipids Antigens
Fig. 1. Various glycolipids were tested for the binding activity of L55-81 human MAb
in ELISA. The human MAb (800 @sg/ml)was diluted with PBS-4% human serum
albumin: U, 1:50; @,1:100: D, 1:200. Glycolipids (3 nmol/well) were coated on each
well, and all tests were done in duplicate.
content, such as M14 and M24. Other cell lines (M12, MIS, and M25)
that expressed low levels of GM2 showed no binding of this human
MAb at a dose of 10 @tg/mi(Table 1). As a control, the binding of
anti-GM3 human MAb L-612 (18) to these cell lines was assessed
using the same assay. Cell lines with a high GM3 content (Ml2 and
MiS) had strong antibody binding, whereas those with relatively less
GM3 (such as M14, M24, and M25) had no or moderate binding.
These results demonstrated that these human MAbs had a differential
reactivity among cells with different ganglioside expression and sug
gested that L55-81 human MAb had specificity for GM2.
To confirm the specificity of L55-8l human MAb for GM2, ELISA
and TLC immunostaining were performed using a variety of ganglio
sides and neutral glycolipids. In ELISA, a significant absorbance
value was obtained only with GM2, and no positive reactivity was
obtained with asialo GMI, GM!, GD1,,,GT1b, GDIb, GMI, GD3, and GD2
(Fig. 1). For TLC immunostaining, the total gangliosides extracted
from M14 and M24, as well as a purified GM2 from M14, were used
as antigen sources. Fig. 2 is a representative staining result with the
L55-8l MAb. GM2 was the only ganglioside to react with the human
MAb, indicating that the human MAb has no cross-reactivity with
other gangliosides.
Antitumor Activity of L55-81 Human MAb in the Presence of
Complement. The abilityof the L55-8l humanMAbto mediatelysis
of the human melanoma M14 cell line, which has a high expression of
GM2on its membrane, was evaluated using@@ Cr release assay. When
M14 cells were exposed to antibody in the presence of human or
rabbit complement, significant@ ‘Cr release activity was observed at a
range of 0.2—120 j.tg/ml antibody (Fig. 3). In the absence of comple
ment, the human MAb did not show any cytotoxic activity, even at
120 @tg/ml. An irrelevant human 1gM MAb derived from the same
patient as L55-8l MAb had no 5tCr releasing activity against M14
cells at 120 tag/mi (data not shown).
Immunohistochemical Analysis of Human MAb L55-81. Pun
fled L55-81 human MAb was biotinylated as described in “Materials
and Methods.―This method involved the labeling of the carbohydrate
moiety of immunoglobulins (22, 23). The reactivity of the biotinylated
L55-81 was examined on 9 different histological-type biopsied tumor
tissues and normal tissues derived from 24 different organ sites (2—4
tissues per organ site). The optimal concentration of the biotinylated
L55-8l was predetermined with cytospin slides of the positive control
cell line M14. The reactivity ofMl4 was + + + at 20 @ag/ml(Fig. 4A),
+ + at 10 @gIml, + at 1 @ag/ml, ± at 0.1 p.g/ml, and — with no
antibody. Subsequent experiments on biopsied tissues were performed
a Ten pg/mi of human MAb were used in the IA assay. IA titer was graded using the
following scale: >75% of cells adhered, + + + ; 50% of cells adhered, + + ; 25%of cells
adhered, + ; and —,no adherence was observed.
b Gangliosides of these cell lines were purified and analyzed as described in “Materials
and Methods.―
5667
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
HUMAN MAb TO GANGLIOSIDE GM2
A
I
GM3
120 24 4.8 0.96 0.192 0.038
L55-81 human MAb (j.tg/ml)
Fig. 2. A, purified gangliosides from melanoma
cell lines Ml4 and M24 run on a TLC plate and
visualized with resorcinol-HCI reagent. STD,
standard gangliosides; M14 total, total gangliosides
from Ml4 cell line; M24 total, total gangliosides
from M24 cell line. B, TLC immunostaining of the
above-mentioned gangliosides with L55-81 human
MAb. Purified °M2from Ml4 and total ganglio
sides from Ml4 and M24 were run on a TLC plate.
Twenty @sg/m1of L55-81 human MAb were used.
GM2@--
@@bf'*@GD3GDIa
GD2@:—:T_GDIbSTD.M14
totalM24 totalM14 GM2M14 totalM24 total
with 20 ;.tg/mi of the antibody. The staining results on biopsied tumor
tissues are summarized in Table 2, and representative staining is
shown in Fig. 4, B and C. Positive reactivity of L55-81 was demon
strated on various types of tumor tissues: 70% of melanoma (7 of 10),
80% of colon carcinoma (4 of 5), 80% of ovary carcinoma (4 of 5),
20% of breast carcinoma (1 of 5), 20% of kidney carcinoma (1 of 5),
and 20% of prostate carcinoma (1 of 5). Of the five specimens each
tested, positive reactivity was not seen on lung carcinoma, pancreas
carcinoma, and lymphoma. Background staining was not observed,
and noncancerous tissues were clearly negative. Of the 24 normal
tissues tested (1—4cases/organ), none showed any detectable reaction
to the antibody. These normal tissues included colon (two), lung
(two), ovary (two), cerebellum (three), breast (two), heart (four),
thyroid (four), lymph node (two), eye (three), kidney (two), liver
(four), muscle (three), skin (two), spleen (two), salivary gland (two),
spinal cord (two), pons (two), cerebrum (one), esophagus (four), testis
(two), larynx (four), stomach (two), duodenum (four), and bladder
(two). A high statistical significance (P < 0.0001), based on Fisher's
exact test, was obtained with tumor tissues of melanoma, colon
carcinoma, and ovary carcinoma. However, due to the small number
of tissues tested, prostate, breast, and kidney carcinomas did not reach
a statistically significant level (P = 0.0746; Table 2). Two of three
cultured lung carcinoma cell lines were positive in our previous
studies with the original L55 human MAb (24), but none of the five
lung carcinomas tested showed positive staining. A definitive conclu
sion of the negativity or positivity of various histological-type malig
nancies must wait until more specimens are tested.
DISCUSSION
In 1982, we reported for the first time the establishment of a human
anti-GM2 MAb using B lymphocytes from a melanoma patient (17).
The availability of human MAb enabled us to define GM2 as a human
tumor-associated molecule that could induce immune responses in
man (11, 17). Since then, three murine anti-GM2 MAbs (25—27)and
one human MAb (28) have been reported, all of which have a lesser
degree of specificity.
The human MAb L55-8l reacted specifically to GM2 [Ga1NAc(3l-
4(NeuNAca2-3)Ga131-4Glcj3l-lceramide] but not with GM1 [Galf3l-
3GalNAc@l-4(NeuNAca2-3)Gal@l-4Glc@l-lceramide], °D2 [Gal
NAc@31-4(NeuNAca2-8)(NeuNAca2-3)Galf3l-4Glc@1-lceramide],
or GM3 (NeuNAca2-3Gal@l4Glc@1-1ceramide), indicating that the
epitope recognized by L55-81 contains both GaINAc and sialic acid
bound to lactose. This study confirms the monospecificity of the
MAb. In contrast, the human MAb MAb3-207 recently developed by
Yamaguchi et a!. (28) showed affinity for both GM2 and GD2, sug
gesting a lack of specificity and difference in the epitope specificity.
One of the major goals of this study was to establish an anti-GM2
antibody-producing cell line that can grow and produce a high level of
antibody in a serum-free medium. After several recloning procedures,
the L55-81 cell line was successfully adapted to AIM-V serum-free
medium. The amount of antibody produced in the spent tissue culture
medium is unusually high (10—30 .ag/ml; 5-day culture at an initial
4
Fig. 3. Complement-dependent cytotoxic activity of L55-8l human MAb to a GM2-
positive human melanoma Ml4 cell line. All tests were done in triplicate. The percentage
of cytolysis was shown by the percentage of 51Crrelease calculated according to the
formula described in “Materialsand Methods.―U, in the presence of human complement;
D, in the presence of rabbit complement. 5668
B
@ @..:.@ _
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
HUMAN MAb TO GANGLIOSIDE GM2
Fig. 4. Immunohistochemical staining of M14 cell line
(A), melanoma (B), and colon carcinoma (C) with bioti
nylated L55-81 in avidin-biotin complex peroxidase
assay. Twenty pg/mi of biotinylated L55-81 were used for
staining.
, @:‘@ @—-..i_..i@ @1..
@ .,,@ @@44,•@@ @.
•,@:.. . @. .“ #@__ p'@ a
- •—@@a • ‘ .@
@@ ..@ ,@ s_-.' @M@@_F @q'
5669
A
.@‘
‘!.
- ,; ..@.—o',@ @. .@ -@ .,@@
@ ,@4;@!4;.@,!@ 0 7―4 3.@?
-.:#@ ,_ .@ @/. & @,. _;
- ‘p 1 ,@@ @a-' • _ 2'
-@@ @,
I- 1'@@@ . -@e-@'
@. .
.)
‘:@.-@
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
HUMAN MAb TO GANGLIOSIDE 0M2
Table 2 Reactivity of biotinylated L55-81 antibody on biopsied tumor and normal
tissues in avidin-biotin complex peroxidase assay―
Tissues tested No. positive/no. tested (%)b
Tumor tissues
Melanoma 7/10 (70%) <0.0001
Colon ca. 4/5 (80%) <0.0001
Ovary ca. 4/5 (80%) <0.0001
Prostate ca. 1/5 (20%) 0.0746
Breast ca. 1/5 (20%) 0.0746
Kidney ca. 1/5 (20%) 0.0746
Pancreas ca. 0/5 (0%) 1.0000
Lung ca. 0/5 (0%) 1.0000
Lymphoma 0/5 (0%) 1.0000
Normal tissues
Cerebellum 0/3 (0%)
Cerebrum 0/1 (0%)
Pons 0/2 (0%)
Spinal cord 0/2 (0%)
Eye 0/3 (0%)
Skin 0/2 (0%)
Muscle 0/3 (0%)
Breast 0/2 (0%)
Salivary gland 0/2 (0%)
Larynx 0/4 (0%)
Esophagus 0/4 (0%)
Stomach 0/2 (0%)
Duodenum 0/4 (0%)
Colon 0/2 (0%)
Liver 0/4 (0%)
Lung 0/2 (0%)
Heart 0/4 (0%)
Thyroid 0/4 (0%)
Spleen 0/2 (0%)
Lymph node 0/2 (0%)
Bladder 0/2 (0%)
Kidney 0/2 (0%)
Ovary 0/2 (0%)
Testis 0/2 (0%)
‘IFresh snap-frozen surgical tissues specimens were used for each immunostaining.
Twenty @sg/mlof biotinylated L55-8l were used for staining.
b Degree of reactivity for each cancer type is as follows: of the seven positive
melanomas, one + + + , four + + , one + , and one ±; of the four positive colon ca., one
+++,one ++,one +,andone ±;ofthefourpositiveovaryca.,three ++ andone +;
one positive prostate ca., + ; one positive breast ca., + + + ; one positive kidney ca., + +.
C Significance between tumor tissues versus normal tissues. P based on Fisher's exact
test. P < 0.05 is considered a significant value.
concentration of l0@cells/ml) and is equivalent to that produced by
the same cell line grown in RPM! 1640 with 5% FCS. Of the many
subclones producing anti-GM2 antibodies established in our labora
tory, this is the only cell line that adapted to a serum-free medium and,
as far as we know from literature, the only serum-free cell line
available today that produces human MAb to GM2. The antibody has
been purified to almost 100% purity with relative ease and at a
reasonably low cost.
In the present study, we investigated the potential usefulness of
L55-81 human MAb for the treatment of cancer. The reactivity of
L55-8 I human MAb to various human cancer tissues and normal
tissues was tested by direct immunohistochemical staining using
biotinylated antibody. As suggested in previous reports citing the
original L55 human MAb (17), L55-8l human MAb reacts not only to
one particular histological type of cancer but to a variety of cancer
tissues. The antibody, however, shows dramatic negativity to normal
tissues. None of the tissues obtained from 24 normal organ sites were
positive. GM2 @5a biosynthetic intermediate of the abundant brain
ganglioside@ . When gangliosides are biochemically extracted
from brain tissues, GM2 is found as a minor ganglioside (<5% of total
gangliosides; Refs. 29 and 30). Our immunological assay, however,
did not detect GM2 in either compartment of brain tissues. The
differences in the ceramide portion of the ganglioside between ma
lignant cells and normal brain tissues might be responsible for the
binding of the anti-GM2 antibody (31). Nudelman et a!. (32) reported
that ganglioside GD3, obtained from malignant melanoma, contained
a different ceramide composition from that in the brain and suggested
that the ceramide portion might influence the antibody binding to the
carbohydrate structure of the ganglioside. Dohi et a!. (33) tested
the antibody-binding of a murine-human chimeric anti-GM2 MAb
(KM966) to human normal tissues. KM966 showed positive reactivity
to glycolipids extracted from normal brain tissues, whereas an immu
nohistological staining for KM966 was negative on normal brain
tissues. A similar phenomenon has been found in an anti-GM3 anti
body binding on tissues. Hakomori et a!. (34) hypothesized that
differential reactivity between normal cells and melanoma with anti
GM3 antibody might be due to the differences in conformation and
density of GMS on these respective membranes.
When comparing the frequency of positivity among the different
cancer tissues tested, we observed some unexpected results, namely,
nonneural-origin cancer tissues such as colon and ovarian carcinomas
that synthesized gangliosides at the lower level showed positive
reactivity to L55-81 as frequently and strongly as the neuroectoder
mal-origin cancer melanoma. Even among melanoma cell lines, the
cellular content of GM2 measured by biochemical assays did not
always reflect the binding of L55-81 to the cell surface. Experimental
results showed that M24, which has high GM2 content (8.2 nmol/g dry
weight), had lower cell surface reactivity with L55-81 than Ml4 (4.0
nmol/g dry weight; Table 1). There are three possible explanations:
(a) antigenic molecules detected by L55-8l on carcinoma cells are not
GM2but simplycross-reactwith GM2.In this regard,Patel et al. (35)
reported that a murine anti-GD2 MAb, MAb3F8, cross-reacted with
the glycoprotein neural cell adhesion molecule and that both GD2 and
the neural cell adhesion molecule have NeuAccs2-8NeuAc terminal
residue; (b) the carbohydrate moiety of GM2 is relatively short as
compared to that of many other gangliosides or glycoproteins. The
reactivity of L55-81 may have been interfered with partially by these
longer cell surface molecules in certain melanoma tissues; and (c)
gangliosides are normally located not only on cell membranes but also
in the Golgi apparatus, where they are biosynthesized (36). It is
possible that a larger amount of GM2 in the Golgi apparatus reflected
the total GM2 content rather than the amount of GM2 on the cell
surface membranes.
Since we reported GM2 as a target for cancer immunotherapy, it has
been tested in clinical Phase I/I! trials in the form of a pure molecule
or GM2-rich melanoma cell vaccine (37, 38). Patients who received
these vaccines in active specific immunotherapy produce 1gM anti
bodies to GM2, the levels of which correlated to good prognosis.
Recently, the Eastern Cooperative Oncological Group has initiated a
double-blind randomized trial specifically targeting GM2 in melanoma
patients with AJCC stage II or III melanoma with purified GM2
conjugated to adjuvants. The therapeutic potential of anti-GM2 anti
body has also been suggested in our preliminary experiments, which
used the original L55 human MAb. A patient with G@2-positive
cutaneous melanoma was treated with the human MAb via intrale
sional injection (39). A complete regression of the melanoma was
observed with 3.3 mg of the antibody. In the same experiment,
treatment with the antibody was ineffective in another patient with
G@2-negative melanoma, confirming immunological specificity of the
treatment efficacy by the antibody. The study has since been discon
tinued due to the high cost of clinical grade L55 human MAb. Due to
the high secretion rate of antibody by the L55-8l cell line and the
adaptation of the cells in a serum-free medium, which has reduced the
purification process, we are able to reduce the cost of production for
purified L55-8l significantly. To conclude, the new human MAb to
GM2 secreted from the highly stable cell line L55-8l has strong
antitumor activity in the presence of complement and may be one of
the most potent MAbs for the treatment of cancer.
5670
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
HUMAN MAb TO GANGLIOSIDE G@
ganglioside GM3 antigen on human cancers. Cancer Rca., 53: 5244—5250, 1993.
19. Tai, T., Cahan, L. D., Paulson, J. C., Saxton, R. E., and Irie R. F. Human monoclonal
antibody against ganglioside 002: use in development of enzyme-linked immunosor
bent assay for the monitoring of anti-G@2in cancer patients. J. NatI. Cancer Inst., 73:
627—633,1984.
20. Ravindranath, M. H., Ravindranath, R. M. H., Morton, D. L., and Graves, M. C.
Factors affecting the fine specificity and sensitivity of serum antiganglioside anti
bodies in ELISA. J. Immunol. Methods, 169: 257—272,1994.
21. Iwamori, M., Sunada, S., lshihara, E., Moki, M., Fujimoto, S., and Nagai, Y.
Differential expression of fucosyl GM, and a disialoganglioside with a Neu-Aca2-
6GaINAc linkage (GDI,) in various rat ascites hepatoma cells. FEBS Lett., 198:
66—70,1986.
22. O'Shannessy, D. J., and Quarles, R. H. Specific conjugation reactions of the oligo
saccharide moieties of immunoglobulins. J. AppI. Biochem.. 7: 347—355,1985.
23. O'Shannessy, D. J., and Quarles, R. H. Labeling of the oligosaccharide moieties of
immunoglobulins. J. Immunol. Methods, 99: 153—161,1987.
24. Inc. R. F., Tai, T., and Morton, D. L. Antibodies to tumor-associated gangliosides
(GM2 and GD2): potential for suppression of melanoma occurrence. in: M. Torisu and
T. Yoshida (eds.), Basic Mechanisms and Clinical Treatment of Tumor Metastases,
pp. 371—384.San Diego, CA: Academic Press, 1984.
25. Natoli, E. J., Livingston, P. 0., Pukel, C. S., Lloyd, K. 0., Wiegandt, H., Szalay, 3.,
Oettgen, H. F., and Old, L. 3. A murine monoclonal antibody detecting N-acetyl- and
N-glycolyl-G@2: characterization of cell surface reactivity. Cancer Res., 46: 41 16—
4120, 1986.
26. Bjerkvig, R., Engebraaten, 0., Laerum, 0. D., Fredman, P., Svennerholm, L.. Vrionis.
F. D., Wikstrand, C. J., and Bigner, D. D. Anti-GM2monoclonal antibodies induce
necrosis in GM2-rich cultures of a human glioma cell line. Cancer Res., 51: 4643—
4648,1991.
27. Miyake, M., Ito, M., Hitomi, S., Ikeda, S., Taxi, T., Kurata, M., Hino, A., Miyake, N.,
and Kannagi, R. Generation of two murine monoclonal antibodies that can discrim
mate N-glycolyl and N-acetyl neuraminic acid residues of °M2gangliosides. Cancer
Res., 48: 6154—6160,1988.
28. Yamaguchi, H., Furukawa, K., Fortunato, S. R., Livingston. P. 0., Lloyd. K. 0.,
Oettgen, H. F., and Old, L. 3. Human monoclonal antibody with dual GM2/GD2
specificity derived from an immunized melanoma patient. Proc. NaIl. Acad. Sci.
USA, 87: 3333—3337,1990.
29. Ando, S., Chang, N-C., and Yu, R. K. High-performance thin-layer chromatography
and densitometric determination of brain ganglioside compositions of several species.
Anal. Biochem., 89: 437—450,1978.
30. Nagai, Y., and Iwamori, M. Ganglioside distribution at different levels of organiza
tion and its biological implications. In: R. W. Ledeen, R. K. Yu, M. M. Rapport, and
K. Suzuki (eds.), Ganglioside Structure, Function, and Biomedical Potential, Vol.
174, pp. 135—146.New York: Plenum Press, 1984.
31. Itonori, S., Hidari, K., Sanai, Y., Taniguchi, M., and Nagai, Y. Involvement of the
acyl chain of ceramide in carbohydrate recognition by an anti-glycolipid monoclonal
antibody: the case of an anti-melanoma antibody, M259O, to G@3-ganglioside.
Glycoconj. 3., 6: 551—560,1989.
32. Nudelman, E., Hakomori, S., Kannagi, R., Levery, S., Yeh, M. Y., Hellstrom, K. E.,
and Hellstrom, I. Characterization of a human melanoma-associated ganglioside
antigen defined by a monoclonal antibody, 4.2. 3. Biol. Chem., 257: 12752—12756,
1982.
33. Dohi, T., Nakamura, K., Hanai, N., Taomoto, K., and Oshima. M. Reactivity of a
mouse/human chimeric anti-GM2antibody KM966 with brain tumors. Anticancer
Rca., 14: 2577—2582,1994.
34. Hakomori, S. General concept of tumor-associated carbohydrate antigens: their chem
ical, physical, and enzymatic basis. In: H. F. Oettgen (ed). Gangliosides and Cancer,
pp.57—67.NewYork:VCHPublishers,1989.
35. Patel, K., Rossell, R. 3., Pemberton, L. F., Cheung, N. K., Walsh, F. S., Moore, S. E.,
Sugimoto, T., and Kemshead, 3. T. Monoclonal antibody 3F8 recognizes the neural
cell adhesion molecule (NCAM) in addition to the ganglioside G02. Br. 3. Cancer, 60:
861—866,1989.
36. Teuamanti, G. An ouiline of ganglioside metabolism. in: R. W. Ledeen, R. K. Yu, M.
M. Rapport. and K. Suzuki (eds.), Ganglioside Structure, Function, and Biomedical
potential, Vol. 174, pp. 197—211. New York: Plenum Press, 1984.
37. Morton, D. L., Foshag, J., Hoon, D. S. B., Nizze, J. A., Wanek, L. A., Chang, C.,
Davtyan, D. G., Gupta, R. K., Elashoff, R.,and Irie, R. F. Prolongation of survival in
metastatic melanoma after active specific immunotherapy with a new polyvalent
melanoma vaccine. Ann. Surg., 216: 463—482,1992.
38. Livingston, P. 0. Approaches to augmenting the immunogenicity of melanoma
gangliosides: from whole melanoma cells to ganglioside-KLH conjugate vaccines.
Immunol. Rev., 145: 147—166,1995.
39. Inc. R. F., Matsuki, T., and Morton, D. L. Human monoclonal antibody to ganglioside
GM2 for melanoma treatment. Lancet, 8641: 786—788, 1989.
5671
ACKNOWLEDGMENTS
We thank the staff of Immunosciences,Inc. (New York, NY) for their
technical and pathological expertise in immunohistochemicalstudies. We also
thank Lan Sze for her technical assistance and Christina Riley for her editorial
assistance.
REFERENCES
1. Tai, T., Paulson, J. C., Cahan, L. D., and Inc. R. F. Ganglioside GM2as a human
tumor antigen (OFA-I-l). Proc. NatI. Aced. Sci. USA, 80: 5392—5396,1983.
2. Tsuchida, T., Saxton, R. E., Morton, D. L., and inc. R. F. Gangliosides of human
melanoma. J. Natl. Cancer Inst., 78: 45—54,1987.
3. Hamilton,W. B., Helling,F., Lloyd, K. 0., and Livingston,P. 0. Ganglioside
expression on human malignant melanoma assessed by quantitative immune thin
layer chromatography. Int. J. Cancer, 53: 566—573,1993.
4. Higashi, H., Hirabayashi, Y., Hirota, M., Matsumoto, M., and Kato, S. Detection of
ganglioside GM2in sera and tumor tissues of hepatoma patients. Jpn. J. Cancer Rca.,
78:1309—1313,1987.
5. Inc. R. F., and Ravindranath, M. H. Gangliosides as targets for monoclonal antibody
therapy of cancer. In: C. Borrebaech and J. Larrick (eds.), Therapeutic Monoclonal
Antibody, pp. 75—94.New York: Stockton Press, 1990.
6. Inc. R. F., and Morton, D. L. Regression of cutaneous metastatic melanoma by
intralesional injection with human monoclonal antibody to ganglioside GD2. Proc.
Natl. Acad. Sci. USA, 83: 8694—8698,1986.
7. Houghton, A. N., Mintzer, D., Cordon-Cardo, C., Welt, S., Fliegel, B., Vadhan, S.,
Carswell, E., Melamed, M. R., Oeugen, H. F., and Old, L. J. Mouse monoclonal IgG3
antibody detecting GD3 ganglioside: a Phase I trial in patients with malignant
melanoma. Proc. Nail. Acad. Sci. USA, 82: 1242—1246,1985.
8. Vadhan-Raj, S., Cordon-Cardo, C., Carswell, E., Mintzer, D., Dantis, L., Duteau, C.,
Templeton, M. A., Oettgen, H. F., Old, L. J., and Houghton, A. N. Phase I trial of a
mousemonoclonalantibodyagainstGD3gangliosidein patientswith melanoma:
inductionof inflammatoryresponsesat tumorsites.J. Clin.Oncol.,6: 1636—1648,
1988.
9. Bajorin, D. F., Chapman, P. B., Wong, G., Coit, D. G., Kunicka, J., Dimaggio, J.,
Cordon-Cardo, C., Urmacher, C., Dantes, L., Templeton, M. A., Liu, J., Oettgen,
H. F., and Houghton, A. N. Phase I evaluation of a combination of monoclonal
antibody R24 and interleukin 2 in patients with metastatic melanoma. Cancer Res.,
50: 7490—7495,1990.
10. Cheung, N. K., Burch, L., Kushner, B. H., and Munn, D. H. Monoclonal antibody 3F8
can effect durable remissions in neuroblastoma patients refractory to chemotherapy:
a Phase II trial. Prog. Clin. Biol. Res., 366: 395—400,1991.
11. Saleh, M. N., Khazaeli, M. B., Wheeler, R. H., Dropcho, E., Liu, T., Urist, M., Miller,
D. M., Lawson, S., Dixon, P., Russel, C. H.,and LoBuglio, A. F. Phase I trial of the
murine monoclonal anti-GD2 antibody l4G2a in metastatic melanoma. Cancer Res.,
52: 4342—4347, 1992.
12. Saleh, M. N., Khazaeli, M. B., Wheeler, R. H., Allen, L., Tilden, A. B., Grizzle, W.,
Reisfeld, R. A., Yu, A. L., Gillies, S. D., and LoBuglio, A. F. Phase I trial of the
chimeric anti-G02 monoclonal antibody chl4.l8 in patients with malignant mela
noma. Hum. Antib. Hybrid., 3: 19—24,1992.
13. Handgretinger, R., Bander, P., Dopfer, R., Klingebiel, T., Reuland, P., Treuner, J.,
Reisfeld, R. A., and Niethammer, D. A Phase I study of neuroblastoma with the
anti-ganglioside GD2 antibody l4G2a. Cancer Immunol. Immunother., 35: 199—204,
1992.
14. Murray, J. L., Cunningham, J. E., Brewer, H., Mujoo, K., Zukiwski, A. A., Podoloff,
D. A., Kasi, L. P., Bhadkamkar, V., Fritsche, H. A., Benjamin, R. S., Legha, S. S.,
Ater, J. L., Jaffe, N., Itoh, K., Ross, M. I., Bucana, C. D., Thompson, L., Cheung, L.,
and Rosenblum, M. G. Phase I trial of murmnemonoclonal antibody l4G2a admin
istered by prolonged intravenous infusion in patients with neuroectodermal tumors.
J. Clin. Oncol., 12: 184—193,1994.
15. Handgretinger, R., Anderson, K., Lang, P., Dopfer, R., Klingebiel, T., Schrappe, M.,
Reuland, P., Gillies, S. D., Reisfeld, R. A., and Niethammer, D. A Phase I study of
human/mouse chimeric anti-ganglioside GD2 antibody chl4. 18 in patients with neu
roblastoma. Eur. J. Cancer, 31: 261—267,1995.
16. Cahan, L. D., Inc. R. F., Singh, R., Cassidenti, A., and Paulson, J. C. Identification
ofahumanneuroectodennaltumorantigen(OFA-I-2)asgangliosideG02.Proc.Nail.
Acad. Sci. USA, 79: 7629—7633,1982.
17. inc. R. F., Sze, L. L., and Saxton, R. E. Human antibody to OFA-I, a tumor antigen
produced in vitro by Epstein-Barr virus-transformed human B-lymphoid cell lines.
Proc. NatI. Acad. Sci. USA, 79: 5666—5670,1982.
18. Hoon,D. S.B.,Wang,Y.,Sze,L.,Kanda,H.,Watanabe,T.,Morrison,S. L.,Morton,
D. L., and inc. R. F. Molecular cloning of a human monoclonal antibody reactive to
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
1996;56:5666-5671. Cancer Res
Yumiko Nishinaka, Mepur H. Ravindranath and Reiko F. Irie
with Potential for Cancer Treatment
M2
G
Development of a Human Monoclonal Antibody to Ganglioside
Updated version
http://cancerres.aacrjournals.org/content/56/24/5666
Access the most recent version of this article at:
E-mail alerts related to this article or journal.Sign up to receive free email-alerts
Subscriptions
Reprints and
.pubs@aacr.orgDepartment at
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Permissions
.permissions@aacr.orgDepartment at
To request permission to re-use all or part of this article, contact the AACR Publications
on March 20, 2017. © 1996 American Association for Cancer Research.cancerres.aacrjournals.org Downloaded from
