Download TAG-72 - Cancer Research

[CANCER
RESEARCH 46, 850-857,
February 1986]
Analysis of a Human Tumor-associated
Glycoprotein (TAG-72) Identified by
Monoclonal Antibody B72.3
Virginia G. Johnson, Jeffrey Schlom, Andrew J. Paterson, Jeffrey Bennett, John L. Magnani, and David Colcher1
Laboratory of Tumor Immunology and Biology, National Cancer Institute [V. G. J., J. S., A. J. P., J. B., D. C.], and National Institute of Arthritis, Diabetes, and Digestive
and Kidney Diseases [J. L. M.], NIH, Bethesda, Maryland 20892
ABSTRACT
Monoclonal antibody B72.3 binds a high-molecular-weight tu
mor-associated glycoprotein identified as TAG-72. This study
reports the partial purification and characterization of TAG-72
from a xenograft of a human carcinoma cell line, LS-174T, which
expresses high levels of this antigen. The tumor homogenate
was initially fractionated by Sepharose CL-4B chromatography.
The high-molecular-weight TAG-72, found in the exclusion vol
ume, was then subjected to two sequential passages through
B72.3 antibody affinity columns. At each step of the procedure,
TAG-72 content was quantitated using a competition radioimmunoassay, and the degree of purification was expressed as the
ratio of antigen in units to total protein. The three-step procedure
produced a purification of TAG-72 with minimal contamination
by other proteins as shown by polyacrylamide gel electrophoresis, followed by staining with Coomassie Blue or periodic acid/
Schiff reagent. The density of affinity-purified TAG-72, as deter
mined by cesium chloride gradient ultracentrifugation, was found
to be 1.45 g/ml. This density determination, together with the
high molecular weight of TAG-72, its resistance to Chondroitinase digestion, the presence of blood group-related oligosaccharides, and sensitivity to shearing into lower-molecular-weight
forms suggest that TAG-72 is a mucin-like molecule.
INTRODUCTION
Monoclonal antibody B72.3 was generated using a membraneenriched fraction from a mammary carcinoma metastasis (1).
Immunoperoxidase studies have demonstrated that B72.3 reacts
with approximately 50% of breast carcinomas (2) and to greater
than 80% of colon carcinomas (3) but shows no significant
reactivity with normal adult liver, spleen, heart, breast, uterus,
bone marrow, colon, stomach, salivary gland, lymph node, or
kidney (2, 3). Monoclonal antibody B72.3 is useful in RIAs2 to
detect the presence of the antigen in the serum of patients with
colon carcinomas3 and to monitor antigenic expression in human
Received 7/31/85; revised 11/1/85; accepted 11/4/85.
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.
1To whom requests for reprints should be addressed.
2 The abbreviations used are: RIA, radioimmunoassay; TAG, tumor-associated
glycoprotein; BSA, bovine serum albumin; PBS, phosphate-buffered saline (80 mw
Na2HPO4:1.5 HIM KH2PO4:2.5 mw KCI:140 mw NaCI:0.5 mM MgCI2:1.0 mw CaCI2,
pH 7.2); TBS, Tris-buffered saline (20 RIM Tris-HCI:150 mM NaCI, pH 7.2); SDS,
sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; PAS, periodic
acid-Schiff reagent; Le", Lewis' antigen; Le*, Lewis6 antigen; Le*, Lewis" antigen;
SPRIA, solid-phase radioimmunoassay.
3 A. J. Paterson, J. Schlom, H. F. Sears, J. Bennett, and D. Colcher. A radioim
munoassay for the detection of a tumor-associated glycoprotein (TAG-72) in human
colon carcinomas using monoclonal antibody B72.3, Int. J. Cancer, in press.
4J. Lundy, F Mornex, A. Keenan, J. Greiner, and D. Colcher. Radioimmunodetection of human colon carcinoma xenografts
athymic mice, Cancer, in press.
RESEARCH
Mucins are ambiguously classified as glycoproteins of high
molecular weight which contain large amounts of carbohydrate
joined in 0-glycosidic linkages to serine and threonine residues
(9,10). Biologically, mucins are usually viscous secretions whose
main function is lubrication and protection (11). Important
changes have been found to take place in mucins and heavily
glycosylated glycoproteins during transformation (12-14). Many
of the tumor-associated antigens identified by monoclonal anti
bodies have been characterized as mucins. For example, mono
clonal antibody 19-9 has been used to detect a mucin in the
serum of gastrointestinal and pancreatic cancer patients (15).
Similarly, Du-Pan-2 (16) detects a mucin-like antigen isolated
from a human pancreatic adenocarcinoma, and OC125 (17, 18),
MOv1, and M0v2 (19) monoclonal antibodies react with mucinlike molecules in ovarian carcinomas.
The purpose of the present study was to purify and charac
terize the TAG-72 molecule(s). The breast tumor metastasis
used to generate monoclonal antibody B72.3 provided a rich
source of TAG-72, but limited availability necessitated the iden
tification of an alternate and reproducible source for TAG-72
purification studies. We have recently shown that, although
B72.3 reacts with a high percentage of breast and colon carci
nomas (2, 3), it reacts with only 1 of 18 colon carcinoma cell
lines and 1 of 25 breast carcinoma cell lines (8). LS-174T, a colon
carcinoma cell line (20), expresses low levels of TAG-72 when
grown in culture. However, the three-dimensional spatial config
uration of the carcinoma cell line has been shown to be important
for the expression of TAG-72 (8). Consequently, the LS-174T
cells, when grown as a xenograft in athymic mice, exhibit a 100fold increase in antigen expression to levels comparable to that
seen in metastatic tumor masses from patients (8). The immunohistological evaluation of LS-174T xenografts identifies it as a
well-differentiated adenocarcinoma demonstrating monoclonal
antibody B72.3 reactivity with both extracellular mucin and intracytoplasmic antigen. This antigenic distribution is similar to that
found in mucinous adenocarcinomas of the human colon. Be
cause of the ease of production and reproducibility of this tumor,
LS-174T xenografts were chosen as the source of TAG-72 for
this study.
MATERIALS AND METHODS
Antibody Preparation. Monoclonal antibody B72.3 was generated
using a membrane-enriched fraction of cells from a mammary carcinoma
metastasis as described elsewhere (1). The antibody was purified from
ascitic fluid by ammonium sulfate precipitation followed by ion-exchange
in visceral organs of congenitally
CANCER
carcinoma xenografts (4,5). Monoclonal antibody B72.3 has also
been used in immunocytochemical studies to detect occult tumor
cells in cytological preparations of effusions (6, 7). Clinical trials
are now ongoing using B72.3 to detect colon carcinoma lesions
in situ. The antigen detected by B72.3 is a high-molecular-weight,
tumor-associated glycoprotein designated as TAG-72 (8).
VOL
46 FEBRUARY
1986
850
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
chromatography (2, 4).
B72.3 (lgG1 ) and goat anti-mouse IgG heavy chain were labeled with
Na125lusing lodogen (Pierce, Rockford, IL) as described previously (5).
MOPC-21, a myeloma lgG1 (Litton Bionetics, Inc., Rockville, MD) was
used as a control immunoglobulin to demonstrate the specificity of B72.3
antibody binding.
Tumor Preparation. LS-174T, a human colon adenocarcinoma cell
line (20), was obtained from the American Type Culture Collection and
grown in Eagle's minimal essential medium with nonessential amino
acids, supplemented with 10% fetal calf serum and gentamicin (50 /ig/
ml). Cells were removed from culture flasks with 0.1% trypsin containing
0.5 m«EDTA and washed twice with culture medium without serum.
Four-wk-old female, athymic mice were obtained from Chartes River
(Kingston, NY). Mice were inoculated s.c. with 1 x 106 cells in 0.1 ml of
culture medium. Tumors were harvested when they reached approxi
mately 1 cm in diameter (15-20 days after implantation), quick frozen in
liquid nitrogen, and stored at -70°C. Larger tumors were not used due
to necrosis.
A breast tumor metastasis to the liver, which was used as the
immunogen for the production of B72.3, was also used in this study.
Tumor Extraction. Tumor extracts were prepared by three different
methods, (a) The first method used nitrogen decompression and sonication. Tumors were finely minced in 3 volumes (w/v) of TBS containing
0.1 triM phenylmethylsulfonyl fluoride, 1.0 mw e-aminocaproic acid, and
1% aprotinin (0.2 trypsin inhibitor units/ml). The tumor was homogenized
for 2 min (Silverson homogenizer, top speed) and then subjected to
nitrogen pressure at 1000 psi (Parr Instrument Co., Moline, IL). The
homogenate was centrifuged for 15 min at 1000 x g, ana the resultant
supernatant was sonicated (Heat Systems Ultrasonics, Plainview, NY)
for 1 min at the maximum setting. After sonication, the homogenate was
further clarified by centrifugation at 10,000 x g for 15 min, and the
supernatant was used for further studies, (b) The second method was
Omni-mix homogenization. Tumors, ranging in size from 3-5 g, were
GLYCOPROTEIN
(TAG-72)
absorbance at 280 nm was determined. A small aliquot from each fraction
was removed and diluted 1:100 in H2O, and the presence of antigen was
determined by SPRIA as described below. The fractions containing the
antigen were pooled, dialyzed extensively in TBS to remove Nal, con
centrated using Aquacide II, and redialyzed against TBS. The concen
trated, bound peak from the first affinity column was then loaded onto a
second B72.3 affinity column consisting of 20 ml of HW 65F affinity
matrix coupled to 40 mg of B72.3. Chromatography was carried out as
described above.
Aliquots were saved at each step of the purification procedure, and
the protein concentration was determined by the method of Lowry ef al.
(21). Quantitation of the antigen at each step was done using a compe
tition radioimmunoassay.
SPRIA. Samples to be tested in a SPRIA (50 ¿il)
were dried in 96-well
polyvinyl chloride microtiter plates. When appropriate, antigen-coated
wells were incubated with modifying agents such as 0.1 M Nal04 (50 n\)
for 30 min at room temperature in the dark. To minimize nonspecific
protein absorption, microtiter wells were treated with 100 p\ of 5% BSA
(w/v) in PBS and incubated for 1 h at 37°C.The BSA was removed, and
1Z5l-labeled B72.3 was added (75,000 cpm in 25 /il). After an overnight
incubation at 4°C,unbound antibody was removed by washing with 1%
BSA (w/v) in PBS. The bound 1Z5I-B72.3 was detected by cutting individ
ual wells from the plate and measuring the radioactivity in a gamma
counter.
Solid-phase radioimmunoassay and immunostaining of thin-layer chro-
minced in TBS containing protease inhibitors as described above and
homogenized using the Son/all Omni-mixer (Sorvall Instruments, Wil
mington, DE) at maximum speed for 1 min at 4°C.This was followed by
centrifugation at 1000 x g for 15 min. The supernatant was further
clarified by centrifugation at 10,000 x g for 15 min, and the supernatant
was used for further studies, (c) Disaggregation with frosted glass slides
was the third method used to prepare tumor extracts. LS-174T tumors
matograms of glycolipids were performed as previously described (23).
Competition Radioimmunoassay. The amount of TAG-72 was quan
tified using a competition RIA. Dilutions of samples to be tested, as well
as those for the standard curve, were incubated with 100 v\ of 125I-B72.3
(100,000 cpm) in 250-jil polyethylene microtest tubes (Brinkman Instru
ments, Inc., Westbury, NY) at 4°Covernight. The tubes were centrifuged
at 8000 x g for 5 min, and 50 n\ of the supernatant were transferred in
triplicate to 96-well polyvinyl chloride microtiter plates that were precoated with 5 ng of LS-174T tumor extract per well. Following an
overnight incubation at 4°C,the plates were washed with 1% BSA (w/
v) in PBS, and the radioactivity was counted. Competition curves were
plotted for each sample tested as the percentage of bound radioactivity
as a function of antigen competitor protein concentration. The linear
portion of the competition curve was compared to a standard extract
containing TAG-72. This standard is based on an extract of the breast
are characterized by large pools of extracellular antigen as shown by
immunoperoxidase analysis. In an attempt to remove this extracellular
antigen by gentle extraction procedures, tumors were disaggregated
between frosted glass microscope slides, and the resulting cellular
suspension was washed in TBS. The majority of the cells were left intact
by this method, and these intact cells, as well as cellular debris, were
removed by centrifugation as described above, leaving the extracellular
antigen in the supernatant.
Protein concentration was determined by the method of Lowry ef al.
(21 ) using BSA as a standard. Tumor extracts were aliquoted and stored
frozen at -20°C.
tumor metastasis that was used to generate the B72.3 monoclonal
antibody. One unit of TAG-72 is defined as the amount of TAG-72 found
Antigen Purification. LS-174T tumors were homogenized using the
Omni-mixer. All steps of the purification were carried out at 4°C. The
material that specifically bound to the wells coated with B72.3 was
assayed for binding by monoclonal antibodies specific for carbohydrate
sequences by the solid-phase radioimmunoassay described above. The
tumor homogenate (75 mg) was resuspended in 10 ml of TBS and loaded
onto a Sepharose CL-4B column (5.5 x 25 cm), previously equilibrated
in TBS. The column was eluted with TBS, and 7.5-ml fractions were
cide II (Calbiochem, San Diego, CA), and dialyzed against TBS. The
pooled peak was then loaded onto a B72.3 affinity column prepared
using the 1,1'-carbonyldiimidazole-activated
affinity matrix Reacti-Gel
IL). One hundred ml of packed gel were
coupled with 200 mg of B72.2 according to the method of Heam et al.
(22). The column was washed with TBS, and the bound antigen was
eluted with 3 M Nal in TBS. Fractions (5.0 ml) were collected, and
CANCER RESEARCH
nant assay similar to that previously described (15). Monoclonal antibody
B72.3 (50 fig/ml) in PBS containing 0.1% NaN3 was added to the wells
of a microtiter plate and incubated overnight at 4°C. The plates were
washed and incubated with TBS containing 1% BSA and 0.1% NaN3 at
pH 8.0 for 2 h at 22°C.This was followed by an incubation under similar
conditions with affinity-purified
TAG-72 diluted in Tris:BSA buffer. The
second monoclonal antibodies used in this assay are specific for blood
group oligosaccharides and are all of the IgM isotype. Antibody AH6252 binds Le" oligosaccharides (Chembiomed, Edmonton, Alberta, Can
ada); antibody 10c17 binds Le" oligosaccharide (24); F-8 binds Le"
collected. Absorbance at 280 nm was determined, and an aliquot from
each fraction was diluted 1:100 in H2O for use in a SPRIA.
Fractions containing TAG-72 were pooled, concentrated using Aqua-
HW-65F (Pierce, Rockford,
in 1 nQ of the breast tumor metastasis extract.
Double Determinant Assay. Specific carbohydrate sequences were
detected on TAG-72 by monoclonal antibodies using a double determi
oligosaccharide (25); anti-A antibody binds blood type A oligosaccharides
(Monocarb, Inc., Lund, Sweden); anti-B antibody binds blood type B
oligosaccharides
oligosaccharides
(Monocarb, Inc.); antibody 102 binds blood type H 2
(26); and antibody CSLEX1 is specific for sialylated Le*
oligosaccharides (27). Binding of the second antibody was detected
using 125l-labeled goat anti-mouse IgM antibody. Monoclonal antibody
19-9 (15) was also used to detect the presence of sialylated Le' oligo
saccharide. Since antibody 19-9 is an lgG1, a direct binding assay, as
VOL. 46 FEBRUARY
1986
851
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
described above, was used rather than a double determinant assay.
SDS:PAGE. Electrophoresis was carried out in 3-10% (w/v) polyacrylamide gels (14 x 12.5 x 1.5 or 3.0 mm) according to the method of
Laemmli (28) using a stacking gel of 3% (w/v) acrylamide.
Proteins in gels were visualized by staining with Coomassie Blue R250
(29). Gels were stained for carbohydrate according to the procedure of
Fairbanks et al. (30) using the PAS reagent.
Western Blotting Procedures. Proteins separated by SDSPAGE
were transferred to nitrocellulose paper (Schleicher and Schuell, Keene,
NH) at 0.35 amp for 2 h according to the method of Towbin ef al. (31).
Nitrocellulose paper was incubated for 2 h in 50 ITIM Tris-HCI:150 mw
NaCI:5 m« EDTA:5% BSA (w/v):0.05% Nonidet P-40 (v/v), pH 7.2, to
minimize nonspecific adsorption. The nitrocellulose paper was then in
cubated with B72.3 (10 Mg/ml) followed by 125l-conjugated goat antimouse IgG heavy chain (5 x 105 cpm/ml). These antibodies were diluted
in the buffer described above. Incubation of each antibody was for 1.5 h
at 37°C with constant rocking. After incubation with each antibody, the
nitrocellulose paper was extensively washed with 50 mw Tris-HCI (pH
7.2):150 mM NaCI:5 mM EDTA:1% BSA (w/v):0.5% Triton X-100 (v/
v):0.1% SDS (w/v). The Western blots were air dried and exposed to
Kodak XAR X-ray film at -70°C for 4-16 h using intensifying screens.
Density Gradient Ultracentrifugation. Ultracentrifugation of affinitypurified TAG-72 and LS-174T tumor extract was carried out in 5 ml of
cesium chloride isopyknic density gradients. The antigen was dissolved
in a CsCI solution with a starting density of 1.42 g/ml, and the gradients
were formed by centrifugation in a Beckman SW 50.1 rotor at 150,000
x g for 72 h at 5°C.Fractions (0.2 ml) were collected, the density was
GLYCOPROTEIN
(TAG-72)
glycoprotein complex (34) found in a breast tumor metastasis to
the liver which was used as the immunogen for the production
of B72.3. This antigen complex has been termed tumor-associ
ated glycoprotein 72 or TAG-72 (8). Because of the limited
availability of this breast tumor metastasis, an alternate source
of TAG-72 was needed for the purification and characterization
of the antigen. LS-174T xenografts in athymic mice were chosen
for this purpose. It was found that these xenografts exhibit
antigen levels comparable to those seen in metastatic tumor
masses from patients, and they demonstrate an identical stand
ard curve in competition RIA to the breast metastasis used as
the original immunogen, thus indicating a similar amount of TAG72 per gram of tumor (8). Immunoperoxidase analysis of LS174T xenografts using monoclonal antibody B72.3 on formalinfixed sections reveals cell-associated antigen as well as large
extracellular mucinous pools of antigen. This is similar to the
antigenic distribution observed in mucinous adenocarcinoma of
the human colon and provides a sufficient source of antigen.
Effect of Different Methods of Extraction on TAG-72. TAG72 identified by Western blotting techniques from LS-174T xen
ografts shows a range of apparent molecular weights ranging
from 200,000 to greater than 1 million (Fig. 1, Lane B). We
recognize that the molecular weight determined by SDSPAGE
determined by refractive index, and an aliquot was diluted 1:10 for SPRIA
to localize the antigen.
Chondroitinase Digestion. Affinity-purified TAG-72 and LS-174T tu
mor extract were dialyzed overnight in enriched Tris buffer (250 mM Tris-
I
HCI: 176 mM NaC2H3O2:250 mM NaCI, pH 8.0) (32). Chondroitinase ABC
(Sigma, St. Louis, MO) was dissolved in the buffer described above at a
concentration of 5 units/ml. The substrate (50 ng) was incubated with
0.5 unit of Chondroitinase ABC at 37°Cfor 1.5 h. Activity of the enzyme
was confirmed using chondroitin sulfate type A and type C (Sigma) as
substrate according to the method of Yamagata ef al. (33).
Neuraminidase Digestion. Neuraminidase type X (Sigma) was dis
solved in 50 mM sodium acetate buffer, pH 5.1. Digestion was carried
out at 37°C for 1.5 h at an enzyme concentration of 0.04 unit/mg of
-200
200-
protein. Digestion was stopped by adding an equal volume of SDS:PAGE
sample buffer and heating at 100°C for 2 min. Samples were analyzed
by Western blotting procedures as described above.
Protease Digestions. Affinity-purified TAG-72 was digested with trypsin, chymotrypsin type VII, papain, and Pronase (all from Sigma) or
Staphylococcus aureus V-8 protease (Miles Laboratories, Naperville, IL).
Digestions were carried out at 37°C for 1 h, using 50 mM CaCI2 in 40
mM Tris-HCI, pH 8.1, for trypsin digestion; 10 mM CaCI2 in 40 mM TrisHCI, pH 8.1, for chymotrypsin digestion; 2 mM EDTA in 5 mM cysteineHCI, pH 6.2, for papain digestion; 40 HIM Tris-HCI, pH 7.8, for V-8
protease digestion; and 0.39 M NaCI:11 mM KCI:5 mM CaCI2:3.6 mM
MgSO4 in 60 rriM Tris-HCI, pH 7.6, for Pronase digestion.
After digestion, an aliquot was removed, mixed with an equal volume
of SDS:PAGE sample buffer, and separated by electrophoresis, followed
by Western blotting techniques. A sample was also heat inactivated at
100°C for 2 min, and TAG-72 was quantitated by competition RIA.
Heating TAG-72 at 100°C for 2 min did not significantly reduce the
binding of B72.3.
To confirm that the antibody used for the competition RIA was not
affected by any residual protease activity present after heat inactivation,
125I-B72.3 was incubated with heat-inactivated proteases and separated
by SDS:PAGE, followed by autoradiography. The 125I-B72.3 appeared
intact.
RESULTS
Identification of a Source for TAG-72. Monoclonal antibody
B72.3 has been shown to react with an M, 220,000-400,000
CANCER
RESEARCH
8768-
-87
-68
433628-
-43
-36
-28
B
D
Fig. 1. Effect of different extraction methods on TAG-72. The breast tumor
metastasis used as the immunogen for the production of B72.3 and the LS-174T
xenograft were extracted by various methods, and the proteins (10 i±g/\ane)
were
separated by SDS:PAGEon 3-10% polyacrylamidegels. TAG-72 was visualized
using Western blotting techniques. The breast tumor metastasis (LaneA) and LS174T xenograft (Lane ß)were extracted using nitrogen decompression followed
by extensive sonication. LS-174T tumors were also extracted using the Omnimixer (Lane C) or dissociated with frosted glass microscope slides (Lane 0). The
lower-molecular-weightband (arrow) in Lanes C and D depicts the murine immunoglobulinheavy chainfrom the LS-174T xenograft in mice detected by the second
antibody used for the Western blotting procedures.
VOL. 46 FEBRUARY
1986
852
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
GLYCOPROTEIN
(TAG-72)
¡sonly an approximation since the migration of heavily glycosy100-
lated proteins is very different from that of the protein molecular
weight markers. The range of molecular weight of TAG-72 from
LS-174T tumors is higher than that found in the breast tumor
metastasis used as immunogen to generate monoclonal antibody
B72.3 (Fig. 1, Lane A). However, this higher molecular weight
range observed for TAG-72 from LS-174T xenografts compares
90
80
70
60
very well with that observed from other breast and colon tumors
extracted by a similar method from recently obtained, snap
frozen material (data not shown).
Large molecules, especially mucins, are subject to shearing
(9), resulting in breakage of an original high-molecular-weight
form into a variety of lower-molecular-weight fragments. Since
5°
40
3020-
the extraction protocol for the antigen shown in Fig. 1, Lane 6,
consisted of nitrogen decompression followed by sonication, the
different molecular weights observed may be due to fragmenta
tion of an original high-molecular-weight molecule. To test this
hypothesis, more gentle extraction methods were used, such as
homogenization with the Omni-mixer or gentle disruption of the
tumor with frosted glass microscope slides. After extraction, the
proteins were separated by SDS:PAGE on 3-10% polyacrylamide gels, and the antigens were analyzed by Western blotting
procedures.
Dissociation of the tumor with frosted glass microscope slides
(Fig. 1, Lane D) appeared to be the least disruptive technique
utilized. This technique produced a high-molecular-weight band
and an absence of lower-molecular-weight forms. If the highmolecular-weight antigen produced by this method is then soni
cated, a range of lower-molecular-weight forms is generated
similar to that seen with nitrogen decompression and sonication
(data not shown). This demonstrates that TAG-72 fragments
10-
0.1
1.0
Antigen competitor
LS-174T tumor homogenate (•),the antigen peak from the Sepharose CL-4B
column (O), the bound peak eluted from the first B72.3 affinity column (A), and the
bound peak from the second B72.3 affinity column (•)
were used.
06
30
<N 25
m
Å’
0.5 |
>
0.4 S
0
03
02
0.1
102030405060708090
FRACTION
Lane C), but this method of homogenization appeared to be less
destructive to the antigen than nitrogen decompression followed
by sonication. Despite some shearing of TAG-72, homogeniza
tion with the Omni-mixer was chosen as the initial step for
1.6
"
1.4
>
12
S
y>
>
z
S
OD
purification, because it allowed easy and reproducible handling
of large quantities of tumors.
Purification of TAG-72. Purification of TAG-72 from LS-174T
xenografts was performed using a three-step procedure con
sisting of passing the tumor homogenate through a Sepharose
CL-4B column followed by two sequential passages through
B72.3 affinity columns. At each step of the purification, the total
protein was determined, and the antigen was quantitated by a
competition RIA (Fig. 2). Purification was expressed as units of
TAG-72 per ng of total protein.
LS-174T tumors were homogenized using the Omni-mixer and
centrifuged to remove nuclei and mitochondria. The supernatant
was chromatographed on a Sepharose CL-4B column. Essen
tially all of the TAG-72 was found in the void volume of the
column (Fig. 3A), resulting in a 14-fold purification (Table 1, Fig.
2). The exclusion of TAG-72 from the Sepharose CL-4B column
is indicative of its high molecular weight (>1 x 106). The antigen
was eluted with 3 M Nal, which was found to be the best reagent
to desorb the antigen with minimal loss in immunoreactivity. A
small amount of TAG-72 (7%) was present in the flow-through
of this column (Fig. 3B). The bound peak was further purified by
S
25
1.0
S
20
0.8
1
15
06
I
10
04
I
3
02
10 20
40 50 60
FRACTION
Fig. 3. Purification of TAG-72 from LS-174T xenografts. LS-174T tumor ho
mogenate was chromatographed on a Sepharose CL-4B column (A). Fractions
collected were analyzed for the presence of TAG-72 in a SPRIA (A) and for the
amount of total protein by measuring absorbance at 280 nm (O). The fractions
containing TAG-72 were pooled and chromatographed on a B72.3 affinity column
(B). The column was washed with TBS (Fractions 1-19), and bound TAG-72 eluted
with 3 M Nal (Fractions 20-60). Fractions were analyzed for the presence of TAG72 in a SPRIA (A) and for the amount of total protein by measuring absorbance at
280 nm (O). Nal (3 M) absorbs light at 280 nm; therefore the absorbance peak
shown between Fractions 27 and 47 is more reflective of Nal absorbance rather
than protein absorbance.
peak from the Sepharose column was pooled and chromato
graphed on a B72.3 affinity column (Fig. 38). Bound TAG-72
RESEARCH
100
Fig. 2. Quantitäten of TAG-72 by competition radioimmunoassay. Aliquots from
the various steps in the purification protocol were assayed for TAG-72 content by
a competition RIA as described in "Materials and Methods." Serial dilutions of the
with harsh treatments, such as sonication.
Some shearing was evident in the Omni-mix extract (Fig. 1,
CANCER
10.0
(/¿g/ml)
30
passage through a second B72.3 affinity column.
Sequential passage through a second affinity column gave an
increased level of purification. A 21 -fold purification was achieved
with a single affinity column and was increased to a 47-fold
VOL. 46 FEBRUARY
1986
853
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
Table 1
Purification of TAG-72 from LS-174T xenografts
TAG-72was purified from LS-174T tumors by a three-step procedureconsisting
of Chromatographieseparation by Sepharose CL-4B followed by two sequential
passages through B72.3 affinity columns. At each step, the protein was analyzed
by the method of Lowry e( a/., and the TAG-72 was quantitated (as units) in a
competition RIA.
Tumor extract
14.6
Sizing peak
51,314
3,515
1,201
21.1
Affinity No. 1 peak
25,342
4215
Affinity No. 2 peakTotal 8,784Totaleg60,588
188units/ng1.0
46.7Recovery(%)85
1
200-
2
r
(TAG-72)
very different staining profiles. Coomassie-stained components
in the purified TAG-72 preparation are essentially absent.
Periodic acid/Schiff reagent does stain a band in the same
molecular weight range as TAG-72 identified by Western blots
(Fig. 4, Lanes 7 and 8). If equal amounts of protein of the starting
LS-174T tumor homogenate and purified TAG-72 are separated
by SDS:PAGE and stained with PAS reagent, there appears to
be insufficient antigen in the total tumor homogenate to stain
(Fig. 4, Lane 7). The 47-fold purification of TAG-72 was sufficient
to increase the amount of antigen relative to the protein to
produce staining with PAS (Fig. 4, Lane 8).
Analysis of Purified TAG-72 by Density Gradient Ultracentrifugation. The density of affinity-purified TAG-72 was deter
mined by equilibrium ultracentrifugation in CsCI as shown in Fig.
5A. The gradient extended from 1.25 g/ml-1.63 g/ml and was
analyzed for TAG-72 content in a SPRIA. The highest antigenic
activity was detected at a density of 1.45 g/ml, typical of that
found for heavily glycosylated glycoproteins such as mucins (11,
35).
The total LS-174T tumor extract was similarily analyzed by
density gradient ultracentrifugation. The TAG-72 in the total
extract produced essentially an identical density profile, indicat
ing that the TAG-72 molecule isolated by our purification meth
ods did not represent a subset of the original antigen (Fig. 53).
units60,480
-
GLYCOPROTEIN
4
I
68-
se
ft
43-
*
3628-
Fig. 4. Purification of TAG-72 from LS-174T xenografts. LS-174T tumor hcmogenate (Lanes 7, 3, 5, and 7) and TAG-72, purified by the three-step procedure
described here (Lanes 2, 4, 6, and 8), were separated by SDSrPAGEon 3-10%
polyacrylamidegels followed by Western blotting procedures (Lanes 1,2,3, and
4) or by staining with Coomassie Blue (Lanes 5 and 6) or periodic acid/Schiff
reagent (Lanes 7 and 8). Equal antigen (10 units) was added to Lanes 7 and 2.
Equal protein (10 ^g) was added to Lanes 3 to 8.
purification after the second affinity column (Table 1, Fig. 2).
Although the use of two affinity columns reduced the total
recovery of TAG-72 from 42 to 15%, the second affinity column
was necessary to achieve the increased degree of purification.
Purification procedures did not appear to change the molecular
weight range of TAG-72. When equal units of TAG-72 before
and after purification (Fig. 4, Lanes 7 and 2) were compared
using SDS:PAGE and Western blotting procedures, identical
patterns were observed. The purification procedures did not
appear to produce significant shearing of the antigen into lowermolecular-weight components. When the starting material was
compared to the purified TAG-72 on the basis of equal protein
(Fig. 4, Lanes 3 and 4), the purified TAG-72 appeared to have a
more disperse molecular weight-range. Since the purification
produced a 47-fold increase in the units of TAG-72 relative to
the total protein, the range in molecular weight found in Lane 4
is believed to result from the antigen overload of the polyacryl
amide gel.
Coomassie Blue does not stain components in the same
molecular weight range as TAG-72 (Fig. 4, Lanes 5 and 6). This
is not surprising, since heavily glycosylated glycoproteins often
have Coomassie-binding sites masked by carbohydrate. How
ever, Coomassie Blue staining can be used to evaluate the purity
of the TAG-72 produced by the three-step procedure. Equal
protein of starting tumor homogenate and purified TAG-72 sep
arated by SDS:PAGE and stained by Coomassie Blue shows
CANCER
RESEARCH
<•?
OJ
CD
15
-
3
O
m
5 -
CL
U
10
15
FRACTION
20
20
1.6
O
m
15
15
U>
<o
a
1.4 =
10
o
CD
a.
ü
1.3
10
15
20
FRACTION
Fig. 5. Density gradient ultracentrifugation of TAG-72. Affinity-purified TAG-72
(A) and LS-174T tumor homogenate (B) were subjected to equilibrium ultracentri
fugation in cesium chloride. Fractions obtained were monitored for the presenceof
TAG-72 by SPRIA (•),
and density was determined by refractive index measure
ments (O).
VOL. 46 FEBRUARY
1986
854
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
Identification of Oligosaccharide Sequences on TAG-72.
The presence of specific Oligosaccharide sequences on the
affinity-purified TAG-72 molecule(s) was demonstrated using
monoclonal antibodies to specific blood group oligosaccharides
in a double determinant assay. Using this assay, it was found
that the molecule containing the B72.3 epitope also contained
Le", Le", H type 2, and sialylated Le* Oligosaccharide structures
(Table 2). The affinity-purified TAG-72 molecule also bound
monoclonal antibody 19-9, and it can therefore be postulated
that the molecule containing the TAG-72 determinant also con
tains sialylated Le8 Oligosaccharide. However, since the binding
of monoclonal antibody 19-9 was determined by direct binding
SPRIA using purified TAG-72 rather than by a double determinant
assay used for the detection of the other blood group-related
oligosaccharides, this binding may be to a contaminant present
in the purified TAG-72 preparation. Further work addressing this
question is in progress.
Antibody B72.3 does not bind purified glycolipid containing
any of the blood group oligosaccharides tested (Le8, Le6, Le*, A
and B blood groups, H type 2, sialylated Le", and sialylated Le*).
Similarity, antibody B72.3 does not bind glycolipids extracted
from a wide variety of human tumors and tumor cell lines as
determined by immunostaining thin-layer chromatograms (data
not shown). These include the LS-174T xenograft and the breast
tumor metastasis used as immunogen in the production of B72.3.
These properties of monoclonal antibody B72.3 clearly distin
guish it from that of monoclonal antibody 19-9.
Analysis of TAG-72 by Enzymatic Digestions. Digestion of
affinity-purified TAG-72 by Chondroitinase ABC was used to
determine whether the TAG-72 epitope is carried by a proteoglycan or a mucin; chondroitin sulfate proteoglycans are sensitive
to Chondroitinase ABC digestion, but mucins are not. After 1.5
h of digestion, the TAG-72 epitope remained unaffected when
observed by Western blotting procedures (Fig. 6a). This finding,
along with the density determinations and double determinant
assay, indicates that TAG-72 has characteristics consistent with
that of a mucin.
Neuraminidase digestion greatly decreased the ability of B72.3
to bind the antigen (Fig. 60). The sensitivity of TAG-72 to
neuraminidase suggests that either sialic acid forms part of the
antigenic determinant or is necessary for its conformational
integrity.
Table 2
Identification of Oligosaccharide sequences on TAG-72
Specific Oligosaccharide sequences on TAG-72 were detected using a double
determinant assay as described in "Materials and Methods." Specific 12SIbound
GLYCOPROTEIN
(TAG-72)
In addition to an overall reduction in antibody binding, removal
of sialic acid with neuraminidase appeared to decrease the
electrophoretic mobility of TAG-72. Although it is impossible to
distinguish between a loss of antigen binding in the lowermolecular-weight forms from a decrease in electrophoretic mo
bility, the latter is believed to be the case since similar anomalous
behavior in which removal of sialic acid leads to an increase in
apparent molecular weight has been observed following neura
minidase treatment of other glycoproteins such as glycophorin
(36). The presence of large quantities of bound SDS effectively
abolishes the native charge on most proteins and glycoproteins
resulting in a migration that correlates with the molecular weight.
However, it is believed that high levels of sialic acid contribute a
significant additional negative charge to the SDS:protein complex
(36). As a result, migration of these heavily sialylated glycopro
teins is greater than would be expected. The mobilities of desialylated proteins therefore are reduced, reflecting the loss of the
additional negative charge contributed by sialic acid.
To investigate the sensitivity of TAG-72 to proteolytic en
zymes, affinity-purified TAG-72 was incubated for 1 h with trypsin, chymotrypsin, papain, Pronase, or S. aureus V-8 protease.
The sensitivity of TAG-72 to digestion was analyzed by Western
blotting procedures. A competition RIA was used to quantitate
the effect of proteolytic digestion on the antibody-binding sites
and to determine whether the complete TAG-72 molecule was
cleaved into small units that retained immunoreactivity but were
undetectable by Western blotting techniques.
As shown in the Western blots (Fig. 6c), all the proteases
tested exhibited some alterations in the molecular weight of the
molecule carrying the TAG-72 epitope. Digestion by trypsin,
chymotrypsin, and staphylococcal V-8 protease altered the pat
tern of bands observed in Western blots, indicating that the
overall TAG-72 molecule was sensitive to these proteases (Fig.
6c, Lanes 2,3, and 6). However, no reduction in antibody-antigen
binding was detected by competition RIA (Fig. Qd), indicating
that the actual epitope on TAG-72 recognized by B72.3 was
unaffected by these three proteases.
Digestion by papain and Pronase appeared to be the most
destructive (Fig. 6c, Lanes 4 and 5). Competition RIA confirmed
that approximately 75% of the antigenic determinants recognized
by B72.3 were destroyed by either papain or Pronase digestion
(Fig. 6cO-The sensitivity of TAG-72 to these proteases suggests
that either protein is directly involved in the antibody binding site
or is necessary for the conformational integrity of the epitope
recognized by B72.3.
represents the difference in the number of counts between antibody-coated wells
incubated with and without TAG-72 diluted in 1% BSA in PBS. Control wells lacked
either B72.3, TAG-72, or the test antibody. All showed background levels of
radioactivity (data not shown). TAG-72 was partially digested with neuraminidase
as described in "Materials and Methods" to expose cryptic or sialylated Oligosac
charide structures. Each anti-oiigosaccharide antibody used was also tested with
an appropriate positive control.
Specific 1Z5Ibound (cpm)
DISCUSSION
Monoclonal antibody B72.3 is directed against a high-molec
ular-weight determinant called TAG-72. TAG-72 precipitated
from the breast tumor metastasis used as immunogen was found
to have an apparent molecular weight of approximately 220,000400,000 by SDS:PAGE (Ref. 34; Fig. 1/ty However, the work
reported here provides evidence that this molecular weight range
AntibodyAH6-25210c17534F-8CS-LEXAnti-AAnti-B102NoneAntigenKLe"Le*Sialyl
Neuraminidase14,1502,64914,011395005,7940
may reflect a shearing of the native high-molecular-weight anti
gen caused by the sonication step used in the initial extraction.
When LS-174T xenografts or snap-frozen colon and breast
Le*ABHtype2None-Neuraminidase7,5201,8702,97412,10202023,6340+
carcinoma biopsy material was prepared without sonication, the
TAG-72 identified in Western blots exhibited a higher molecular
weight without such an apparent heterogeneity in size. This
antigen is found both in the stacking gel and at the top of a 3CANCER RESEARCH
VOL. 46 FEBRUARY
1986
855
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
Fig.6. Enzymatic digestion of TAG-72. Af
finity-purifiedTAG-72 (a,c, d) or LS-174Ttumor
extract (b) was digested with Chondroitinase
ABC (a), neuraminidase(£>).
or a variety of pro
teases (c, d) as described in "Materials and
Methods." After digestion, proteins (10 ^g/lane)
were separated by SDSrPAGEon 3-10% polyacrylamide gels and analyzed using Western
blotting procedures. The effect of the various
protease digestions on TAG-72 was quantifated by competition RIA (d) and expressed as "
units/ml, a, Lane 7, affinity-purifiedTAG-72con
trol; Lane 2, Chondroitinase ABC-digested
TAG-72. 6, Lane 7, LS-174T tumor extract
control; Lane 2, neuraminidase-digested LS- (O
„_
174T tumor extract, c and d, Lane 1, affinity- DC
iu
purified TAG-72 control; Lane 2, trypsin diges- =
tion (1.0 unit/2.5 ng TAG-72); Lane 3, chymo- ¡^
trypsin digestion (1.0 unit/2.5 »¿g
TAG-72);Lane ^
4, papain digestion (0.1 unit/2.5 n9 TAG-72); ,_
Lane5, Pronasedigestion (0.1 unit/2.5 ^g TAG- ^
72); Lane 6, staphylococcalV-8 protease diges
tion (5.0 units/2.5 ^g TAG-72).
GLYCOPROTEIN
(TAG-72)
2.5II—
—
£O12.0
-Jro—
200—
z0iC1.5
g-coi;-0.5
LO
87—
68—
43—
12
1
123456
123456
10% polyacrylamide separating gel after SDS:PAGE. It is difficult
to assign a molecular weight to heavily glycosylated glycopro-
However, neuraminidase digestion of TAG-72 greatly reduces
antibody binding, thus providing evidence that sialic acid is either
teins on the basis of SDS:PAGE. However, the high molecular
weight (>1 x 106) is apparent, since it is found in the void after
Sepharose CL-4B chromatography.
TAG-72 was chromatographed on a Sepharose CL-4B column
in the presence of 6 M guanidine-HCI. The antigen profile obtained
in SPRIA was similar to that obtained in the absence of guanidine-
involved in the antigenic determinant or responsible for the
conformational integrity of the antibody binding site. Periodate
oxidation also reduces antibody binding, providing further evi
dence that carbohydrate is involved in the epitope. Whether
protein, in addition to carbohydrate, forms part of the TAG-72
antigenic determinant remains unresolved. Papain and Pronase
digestion reduced antibody-antigen binding by approximately
HCI, thus indicating that the molecular weight observed was not
due to the aggregation of low-molecular-weight forms (data not
shown). These results were confirmed by Western blotting pro
cedures.
TAG-72 was purified by a three-step procedure involving Seph
arose CL-4B chromatography, followed by two sequential pas
sages of the extract through B72.3 affinity columns. The Seph
arose column was useful as an initial step to remove many of
the contaminating lower-molecular-weight murine proteins, such
as albumin and murine ¡mmunoglobulin. This column alone pro
duced a 14-fold purification of the antigen. Passage through a
series of two B72.3 affinity columns was able to increase the
purification of TAG-72 to 47-fold.
The 47-fold purification obtained by this three-step procedure
may be deceptively low. The LS-174T xenograft has large extra
cellular pools of antigen in addition to cell-associated antigen.
These extracellular pools result in an unusually high concentra
tion of the antigen. The competition RIA is based on an arbitrary
quantitation of 1 unit of antigen as the amount of antigen
corresponding to 1 #/g of protein in the LS-174T tumor. If this
starting assumption is based on a tumor with extremely high
levels of antigen per gram of tumor, the degree of purification
obtained may be misleading relative to other nonmucinous tu
mors or tissue culture cells which lack the high antigen concen
tration present in extracellular pools. However, since we feel that
the LS-174T xenograft is representative of typical mucinous
adenocarcinomas seen in colon carcinoma patients, we feel that
our choice of both antigen source and the definition of 1 unit of
TAG-72 is justified.
The exact nature of the epitope on the TAG-72 molecule
detected by monoclonal antibody B72.3 is unknown at this time.
CANCER
RESEARCH
75% as shown by competition RIA. Although this provides
evidence that the peptide portion of the molecule is involved in
the epitope or affects its conformational integrity, we cannot rule
out contaminating glycosidase activity in the protease sample.
Further work to elucidate the nature of the TAG-72 epitope is in
progress.
Monoclonal antibodies often recognize carbohydrate determi
nants found on both glycoproteins and glycolipids. Monoclonal
antibody 19-9 provides an example of this (15, 23). It appears
that the TAG-72 epitope is not found on glycolipids. This clearly
distinguishes the TAG-72 epitope from the determinant recog
nized by monoclonal antibody 19-9.
Data are presented here to suggest that the TAG-72 molecule
is a mucin. This conclusion is based on the following observa
tions: (a) TAG-72 has a high molecular weight (>1 x 106) as
shown by its exclusion from a Sepharose CL-4B column; (b) the
density of TAG-72 determined by equilibrium centrifugaron in
CsCI was 1.45 g/ml, indicating a heavily glycosylated glycoprotein (11, 35); (c) TAG-72 demonstrates a change in migration
after neuraminidase digestion, indicating that it is a heavily sialylated molecule with an abundance of O-glycosidically linked
oligosaccharides (36) characteristic of mucins; (d) blood group
antigens commonly found on mucins are found on affinity-purified
TAG-72; and (e) Chondroitinase ABC digestion had no effect on
TAG-72, thus demonstrating that the TAG-72 epitope is not
expressed on a chondroitin sulfate proteoglycan.
The purification of TAG-72 from LS-174T as described in this
paper should enable further characterization of both the mucinlike molecule and epitope defined by monoclonal antibody B72.3.
This purification procedure will also allow one to compare mole-
VOL. 46 FEBRUARY
1986
856
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
HUMAN TUMOR-ASSOCIATED
cules from differing sources that exhibit the B72.3 epitope. The
TAG-72 molecule purified in these studies can now be used
toward the production of "second generation" monoclonal anti
bodies reactive with different determinants on the TAG-72 mol
ecule. These antibodies should subsequently aid in the further
characterization of this tumor-associated antigen.
REFERENCES
1. Colcher, D., Horan Hand, P., Nuti, M., and Schlom, J . A spectrum of
monoclonal antibodies reactive with human mammary tumor cells. Proc. Nati.
Acad. Sci. USA, 78: 3199-3203, 1981.
2. Nuti, M., Teramoto, Y. A., Mariani-Costantini, R., Horan Hand, P., Colcher, D.,
and Schlom, J. A monoclonal antibody (B72.3) defines patterns of distribution
of a novel tumor-associated antigen in human mammary carcinoma cell pop
ulations. Int. J. Cancer, 29: 539-545,1982.
3. Stramignoni, D., Bowen, R., Atkinson, 8., and Schlom, J. Differential reactivity
of monoclonal antibodies with human colon adenocarcinomas and adenomas.
Int. J. Cancer, 31: 543-552, 1983.
4. Keenan, A., Colcher, D., Larson, S., and Schlom, J. Radioimmunoscintigraphy
of human colon cancer xenografts in mice with radioiodinated monoclonal
antibody B72.3. J. NucÃ-.Med., 25. 1197-1203,1984.
5. Colcher, D., Keenan, A., Larson, S., and Schlom, J. Prolonged binding of a
radiolabeled monoclonal antibody (B72.3) used for the in situ radioimmunodetection of human colon carcinoma xenografts. Cancer Res., 44: 5744-5751,
1984.
6. Szpak, C., Johnston, W., Lottich, C., Kufe, D., Thor, A., and Schlom, J.
Patterns of reactivity of four novel monoclonal antibodies (B72.3, DF3, B1.1,
and B6.2) with cells in human malignant and benign effusions. Acta Cytol., 28:
356-367,1984.
7. Johnson, W. W., Szpak, C. A., Lottich, C., Thor, A., and Schlom, J. Use of
monoclonal antibody (B72.3) as an ¡mmunocytochemical adjunct to diagnosis
of adenocarcinoma in human effusions. Cancer Res., 45: 1894-1900, 1985.
8. Horan Hand, P., Colcher, D., Salomon, D., Ridge, J., Noguchi, P., and Schlom,
J. Influence of spatial configuration of carcinoma cell populations on the
expression of a tumor-associated glycoprotein. Cancer Res., 45: 833-840,
1985.
9. Rose, M., Voter, W., Brown, C., and Kaufman, B. Structural features of human
tracheobronchial mucus glycoprotein. Biochem. J., 222: 371-377,1984.
10. Gottschalk, A. Definition of glycoproteins and their delineation from other
carbohydrate-protein complexes. In: A. Gottschalk (ed.), Glycoproteins: Their
Composition, Structure, and Function, pp. 24-30. New York: Elsevier Publish
ing Co., 1972.
11. Fleming, N., Brent, M., Arellano, R., and Forstner, J. Purification and immunofluorescent localization of rat submandibular mucins. Biochem. J., 205:225233, 1982.
12. Feizi, T. Demonstration by monoclonal antibodies that carbohydrate structures
of glycoproteins and glycolipids are onco-developmental antigens. Nature
(Lond.), 314: 53-57, 1985.
13. Gold, D., and Miller, F. Comparison of human colonie mucoprotein antigen
from normal and neoplastic mucosa. Cancer Res., 38: 3204-3211, 1978.
14. Boland, C., Montgomery, C.. and Kim, Y. Alterations in human colonie mucin
occurring with cellular differentiation and malignant transformation. Proc. Nati.
Acad. Sci. USA, 79: 2051 -2055,1982.
15. Magnani, J., Steplewski, Z., Koprowski, H., and Ginsburg, V. Identification of
the gastrointestinal and pancreatic cancer-associated antigen detected by
monoclonal antibody 19-9 in the sera of patients as a mucin. Cancer Res., 43:
5489-5492,1983.
16. Lan, M., Finn. O., Fernsten, P., and Metzgar, R. Isolation and properties of a
human pancreatic adenocarcinoma-associated
antigen DU-PAN-2. Cancer
Res., 45: 305-310,1985.
17. Bast, R., Feeney, M., Zazarus, H., Nadler. L., Colvin, R., and Knapp, R.
GLYCOPROTEIN
(TAG-72)
Reactivity of a monoclonal antibody with human ovarian carcinoma. J. Clin.
Invest., 68: 1331-1337, 1981.
18. Bast, R., and Knapp, R. Monoclonal antibodies reactive with human ovarian
carcinoma. In: G. L. Wright (ed.). Monoclonal Antibodies and Cancer, pp. 241252. New York: Marcel Dekker, Inc., 1984.
19. Miotti, S., Aguanno, S., Canevari, S., Diotti, A., Oriandi, R., Sonnino, S., and
Colnaghi. M. Biochemical analysis of human ovarian cancer-associated anti
gens defined by murine monoclonal antibodies. Cancer Res., 45: 826-832,
1985.
20. Tom, B. H., Rutzky, L. H., Jakstys, M. M., Oyasu, R., Kaye, C. I., and Kahan.
B. D. Human colonie adenocarcinoma cells. 1. Establishment and description
of a new cell line. In Vitro (Rockville), 12:180-191,1976.
21. Lowry, 0., Rosebrough, N., Farr, A., and Randall, R. Protein measurement
with the Folin phenol reagent. J. Biol. Chem., 193: 265-275,1951.
22. Hearn, M., Bethell, G., Ayers, J., and Hancock, W. Application of 1,1'carbonyldiimidazole-activated agarose for the purification of proteins. II. The
use of an activated matrix devoid of additional charged groups for the purifi
cation of thyroid proteins. J. Chromatogr., 785: 463-470,1979.
23. Magnani, J.. Nilsson, B., Brockhaus. M., Zopf, D., Steplewski, Z., Koprowski,
H., and Ginsburg. V. A monoclonal antibody-defined antigen associated with
gastrointestinal cancer is a ganglioside containing sialylated lacto-W-fucopentaose II. J. Biol. Chem., 257:14365-14369,
1982.
24. Brockhaus, M., Magnani, J. L., Blaszczyk, M., Steplewski, Z., Koprowski, H.,
Karìsson,K. A., Larson, G., and Ginsburg, V. Monoclonal antibodies directed
against the human Le" Wood group antigen. J. Biol. Chem.. 256: 1322313225,1981.
25. Huang, L. C., Brockhaus, M., Magnani, J. L., Cuttitta, F., Rosen, S., Minna, J.
D., and Ginsburg, V. Many monoclonal antibodies with an apparent specificity
for certain lung cancers are directed against a sugar sequence found in lactoAMucopentaose III. Arch. Biochem. Biophys., 220: 318-320,1983.
26. Fredman, P., Richert, N. D., Magnani, J. L., Willingham, M. C., Pastan, I., and
Ginsburg, V. A monoclonal antibody that precipitates the glycoprotein receptor
for epidermal growth factor is directed against the human blood group H type
1 antigen. J. Biol. Chem., 258:11206-11210,1983.
27. Iguro, T., Wakisaka, A., Terasaki, P. I., Hirota, M., Suyama, N., Fukushima,
K., ChÃ-a,D., and Kawahara, M. Sialylated Lewis' antigen detected in the sera
of cancer patients. Lancet, 2: 817-818, 1984.
28. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head
of bacteriophage T4. Nature (Lond.), 227: 680-685, 1970.
29. Weber, K., and Osbom, M. Proteins and sodium dodecyl sulfate: molecular
weight determination on polyacrylamide gels and related procedures. In: H.
Neurath, R. Hill, and C.-L. Boeder (eds.), The Proteins, Vol. 1, pp. 180-223.
New York: Academic Press, 1975.
30. Fairbanks, G., Steck, T., and Wallach, D. Electrophoretic analysis of the major
polypeptides of the human erythrocyte membrane. Biochemistry, 10: 26062617,1971.
31. Towbin, H., Staehelin, T., and Gordon, J. Electrophoretic transfer of proteins
from polyacrylamide gels to nitrocellulose sheets: procedure and some appli
cations. Proc. Nati. Acad. Sci. USA, 76: 4350-4354, 1979.
32. Saito, H., Yamagata, T., and Suzuki, S. Enzymatic methods for the determi
nation of small quantities of isomerie chondroitin sulfatases. J. Biol. Chem.,
243:1536-1542,1968.
33. Yamagata, T., Saito, H., Habuchi. 0., and Suzuki, S. Purification and properties
of bacterial chondroitinases and chondrosulfatases. J. Biol. Chem., 243. 15231535, 1968.
34. Horan Hand, P., Colcher, D., Wunderlich, D., Nuti, M., Teramoto, Y., Kufe, D.,
and Schlom, J. Rational basis for the diagnostic, prognostic, and therapeutic
utility of monoclonal antibodies in the management of human breast cancer.
In: B. A. Chabner (ed.). Rational Basis for Chemotherapy, UCLA Symposia on
Molecular and Cellular Biology, Vol. 1, pp. 315-318. New York: Alan R. Uss,
Inc., 1982.
35. Creeth, J., Bhaskar, K., and Horton, J. The separation and characterization of
bronchial glycoproteins by density-gradient methods. Biochem. J., 767: 557569,1977.
36. Gahmberg, C., and Andersson. L. Role of sialic acid in the mobility of membrane
proteins containing 0-linked oligosaccharides on polyacrylamide gel electrophoresis in sodium dodecyl sulfate. Eur. J. Biochem.. 722: 581-586, 1982.
CANCER RESEARCH VOL. 46 FEBRUARY 1986
857
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
Analysis of a Human Tumor-associated Glycoprotein (TAG-72)
Identified by Monoclonal Antibody B72.3
Virginia G. Johnson, Jeffrey Schlom, Andrew J. Paterson, et al.
Cancer Res 1986;46:850-857.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/46/2/850
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 14, 2017. © 1986 American Association for Cancer Research.