Mucin-like Antigens in a Human Pancreatic Cancer Cell Line Identified by Murine Monoclonal Antibodies SPan-1 and YPan-1 Jenny J. L. Ho, Yong-suk Chung, Yasuhisa Fujimoto, et al. Cancer Res 1988;48:3924-3931. Published online July 1, 1988.
Updated Version
E-mail alerts Reprints and Subscriptions Permissions
Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/48/14/3924
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 July 11, 2011 Copyright © 1988 American Association for Cancer Research
(CANCER RESEARCH 48, 3924-3931, July 15, 1988]
Mucin-like Antigens in a Human Pancreatic Cancer Cell Line Identified by Murine Monoclonal Antibodies SPan-1 and YPan-11 Jenny J. L. Ho,2 Yong-suk Chung, Yasuhisa Fujimoto, Ning Bi, Whitney Ryan, Shi-zhen Yuan, James C. Byrd, and Young S. Kim Gastrointestinal Research Laboratory (151M2), Veterans Administration Medical Center, and Department of Medicine, University of California, San Francisco, California 94121
ABSTRACT The antigenic determinant recognized by monoclonal antibody SPan-1 is greatly elevated in sera of patients with pancreatic cancer but not in sera of normal individuals. Here we describe the mucin-like characteris tics of the SPan-1 antigen isolated from culture medium and xenografts of the human pancreatic cancer cell line SW-1990. YPan-1, another pancreatic cancer associated monoclonal antibody, also reacts with the SPan-1 antigen. The SPan-l/YPan-1 antigens have densities of 1.4-1.5 u, ml and elute in the void volume of Sepharose CL-2B columns. They are resistant to degradation by chondroitinase ABC, nitrous acid, and hyaluronidase but susceptible to protease digestion and reductive ^-elimination. All these characteristics suggest that the SPan-1 and YPan-1 determinants are carried on mucinous antigens. Both SPan-1 and YPan-1 immunoreactivities are unaffected by boiling or by alkylation and reduction of the mucins while they are abolished by mild periodate oxidation or neuraminidase and are markedly decreased by wheat germ agglutinin. Thus, their antigenic determinants are composed principally of carbohydrates with sialic acid, an absolute requirement for reactivity. However, the epitope specificities of SPan-1 and YPan-1 are different since YPan-1 does not compete with SPan-1 for binding to antigen. Moreover, YPan-1 and SPan-1 can be distinguished from several other sialic acid requiring, cancer associated antibodies such as B72.3, (MIA 1, DU-PAN-2, OC-125, and 19-9 by either their epitope characteristics or their tissue reactivity patterns.
INTRODUCTION Immunological detection of mucin-like high molecular weight glycoproteins has become increasingly important in the diagnosis of various cancers, including cancer of the pancreas (1-5). We have previously reported (6) the production of a monoclonal antibody, YPan-1, against the human pancreatic cancer cell line Capan-2, which reacts with glycoproteins re leased by these cells into their culture medium (7). YPan-1 had a 89%, 90%, and 46% positive immunoreactivity with formalin fixed cancerous pancreas, stomach, and colon, respectively (6). Although YPan-1 reacts with normal pancreas it has little or no reaction with normal colon or stomach. In well differentiated pancreatic carcinoma, YPan-1 reacted with both the luminal border of cells and luminal contents. Another monoclonal antibody, SPan-1 (8), was recently produced in our laboratory using a high mucin containing pancreatic cancer cell line, SW1990 (9), as the immunogen. Immunohistochemical studies using SPan-1 showed 89, 67, and 62% positive reaction with cancers of the pancreas, stomach, and colon, respectively. SPan1 also reacts with normal pancreas but minimally with normal colon or stomach (10). Furthermore, 93% of sera from pan creatic cancer patients had elevated levels of the SPan-1 deterReceived 3/17/87; revised 2/19/88; accepted 4/14/88. 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. 1This study was supported in part by USPHS Grant CA24321 from the National Cancer Institute and by the Veterans Administration Medical Research Service. Y. S. K. is a Medical Investigator of the Veterans Administration. 2To whom requests for reprints should be addressed, at Gastrointestinal Research Laboratory (151M2), Veterans Administration Medical Center, 4150 Clement St., San Francisco, CA 94121.
minant. No normal individuals and only 23 and 13% of gastric and colonie cancer patients, respectively, had elevated serum levels (8). In this study, we describe the biochemical and imnmiiodici mail properties of the SPan-1 and YPan-1 reactive antigens isolated from both spent culture medium and xenograft tumors of SW-1990 cells. MATERIALS AND METHODS SPan-1 and YPan-1. The murine monoclonal antibodies SPan-1 (8) and YPan-1 (6) were produced against the human pancreatic cancer cell lines SW-1990 and Capan-2, respectively. Both SPan-1 and YPan1 are IgMs and were purified from spent hybridoma culture media by chromatography on Sephacryl S-300 as previously described (8). Purification of Kadiolabeled Medium Antigen. SW-1990 cells grown to confluence in Dulbecco's modified Eagle's medium 21 supplemented with 5% fetal bovine serum, 100 units/ml penicillin, and 100 ng/m\ streptomycin, were washed with sterile PBS3 and then incubated in Dulbecco's modified Eagle's medium 16 containing 5 ng/m\ of both human transferrin and bovine insulin (Sigma), 1% nonessential amino acids, 2 m\i glutamine, penicillin/streptomycin, and 10 ¿iCi/mlof tritiated glucosamine (specific radioactivity, 40 Ci/mmol; ICN Kadi ochemicals). After 24 h the medium was collected, 0.1 mM PMSF was added, and the medium was centrifuged at 100,000 x g for l h and concentrated under vacuum with a collodion membrane (Schleicher & vhnciD. Concentrated medium was chromatographed on 1.5- x 94-cm columns of Sepharose CL-2B in PBS containing 0.02% NaNj. Two-mi fractions were collected. Smaller columns (1.2 x 18 cm) and O.S-ml fractions were used for analytical purposes. SPan-1 and YPan-1 antigen levels of the fractions were determined either by direct binding or by "competition" ELISAs. To determine radioactivity, aliquots of column fractions were mixed with AquaMix (West Chem) and counted in a Beckman LS-8100 liquid scintillation counter (Beckman Instruments, Inc.). The void volume fractions were further purified by CsCl density gradient centrifugation according to the method of Carlstedt et al. (11). Eight 1.5-ml fractions were collected. Densities were determined gravimetrically. Purification of Xenograft Antigens. Xenograft tumors (1-1.S cm in diameter) were produced by the s.c. injection of 5 x 10*SW-1990 cells/ site into athymic BALB/c-nu/nu mice (National Cancer InstituteFrederick Cancer Research Facility, Frederick, MD). Tumors were homogenized in 0.1 M NH4HCO3, 0.5 M NaCl, and 0.1 mM PMSF (10% w/v) with a Polytron Omni-Mixer (Brinkmann Instruments). The homogenate was centrifuged at 45,000 x g for 45 min and the super natant fluid was dialyzed overnight against 10 mM Tris-Cl (pH 8.0). After centrifugation at 20,000 x g for 45 min the supernatant fluid was fractionated by CsCl density gradient centrifugation as described above for medium antigen. Fractions having the most SPan-1 and YPan-1 immunoreactivities were further purified by HPLC with a Beckman Model 332 HPLC and an Altex Spherogel TSK-G (4000) SWG column d.d.. 30 x 2.15 cm). Samples were eluted with 0.2 M ammonium acetate at a flow rate of 6.0 ml/min. ELISA. /3-Galactosidase enzyme-linked immunoassays were per formed as previously described (8). The concentration of purified xen3 The abbreviations used are: PBS, phosphate buffered saline; BSA, bovine serum albumin; ELISA, enzyme linked immunosorbent assay; HPLC, high per formance liquid chromatography; PMSF, phenylmethylsulfonyl fluoride; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; WGA, wheat germ agglutinin.
3924
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-l AND YPan-:
ograft mucins used was 2 Mg hexose/ml unless otherwise specified. Lectins were preincubated with antigen for 2 h at room temperature. WGA and succinylated WGA (Triticum vulgaris) were obtained from Vector Laboratories, and fucose binding protein from Lotus tetragonolohus was purchased from ICN ImmunoBiologicals. All other lectins were obtained from Sigma. In the competition ELISA, antibody was preincubated with inhibitor overnight at 4°C.A range of antibody concentrations was tested to determine the appropriate subsaturating levels. Double Determinant Radioimmunoassay. The procedure was similar to that described elsewhere (8). In competition studies the inhibiting antibodies were incubated with antigen for 2 h at 37°Cprior to addition of radiolabeled SPan-l or YPan-1. Partially purified monoclonal anti body 19-9 (4.8 mg/ml) was a gift from Dr. V. Zurawski, Jr. (Centocor, Inc.), DU-PAN-2 ascitic fluid was a gift from Dr. Richard S. Metzgar, and CSLEX-1 ascitic fluid was a gift from Dr. Paul I. Terasaki. Purified myeloma IgM was a gift from Caltag Laboratories. Chemical and Enzymatic Treatments. After purified SPan-l /YPan-1 antigens from culture media or xenograft were treated with 0.05 M NaOH and l M NaBH4 at 37°Cfor 16 h, they were neutralized and
intensifying screen [Cronex Lighting Plus, DuPont (20)], in the case of 125I,at -70°C. For fluorography dried gels were exposed to untreated film without a screen. Proteins and glycoproteins were visualized by staining the gels with 0.25% Coomassie Brilliant Blue R-250 or periodic acid-Schiff (21). The following proteins were used for molecular weight standards: myosin, /i-galactosidase, BSA, and ovalbumin (Sigma). Affinity Chromatography on WGA-Agarose. Purified xenograft mu ein was applied to 1-ml columns of WGA-agarose (Sigma), equilibrated with PBS, and eluted with increasing concentrations of jV-acetylglucosamine in PBS. One-mi fractions were collected. Chemical Analyses. Protein concentration was determined by the method of Lowry et al. (22) using BSA as standard. Hexose concentra tion was determined by the phenolsulfuric method (23) with galactose as standard. 0-Linked oligosaccharide was determined by the fluores cent method of Crowther and Wetmore (24), using a Perkin-Elmer LS2 filter fluorimeter at an excitation wavelength of 340 nm and an emission wavelength of 383 nm.
RESULTS
desalted, and the borate was removed according to the method of Edge Enzymatic and Chemical Treatment of Medium Antigens. [3H] et al. ( 12). Mucins were also reduced and alkylated ( 13) or treated with Glucosamine labeled material released by SW-1990 cells into nitrous acid ( 14) or periodate ( 15). The following enzymatic conditions culture medium eluted principally at the excluded volume of were used: (a) bovine testicular hyaluronidase (0.5 mg/ml in citrate buffer, pH 5.5); (b) Clostridium perfringens neuraminidase (type X, Sepharose CL-2B columns (Fig. 1). The void volume fraction 0.25 unit/ml in PBS, 2 h, 37°C);(c) Proteus vulgaris chondroitinase ABC (0.5 unit/mi in 0.05 M Tris-acétatebuffer, pH 7.3); and (d) Streptomyces griseus Pronase E (protease type XIV, in 0.10 M Tris-CI containing 0.05 M CaCI2, pH 8.0) at a 1:100 protease:mucin protein ratio. All enzymes were obtained from Sigma. Enzyme digests (37°C overnight unless otherwise stated) were boiled for 10 min prior to testing with antibody. As positive controls hyaluronidase and chon droitinase ABC were shown to be active against their respective sub strates under the conditions used. Immunoprecipitation. Protein A-positive Staphylococcus aureus cells (Boehringer Mannheim Biochemicals) were prewashed three times with Buffer A [0.5 M NaCl-1 mM EDTA-0.1 mM PMSF-0.1% BSA-1% Nonidet P-40 in 10 mM Tris, pH 8.0 (16)] and then resuspended in Buffer A (10% original dry w/v). Radiolabeled medium antigens pre pared as described above in PBS containing 0.1 mM PMSF were diluted 1:1 with Buffer A to a final volume of 400 n\. After nonspecific precipitation (incubation with 100 ^1 of washed Protein A for 1 h on ice with occasional shaking followed by centrifugation for 2 min at 12,000 x g in a microcentrifuge) the supernatant fluid was incubated successively with: (a) YPan-1, SPan-l, or myeloma (P3-X63-Ag 8.653 or SP-2/O-Ag 14) ascitic fluid (final dilution 1:100) for 3 h at 4°Cwith occasional shaking; (b) rabbit anti-mouse IgM second antibody (Zymed Laboratories; final dilution, 1:60) overnight at 4°C;and finally (c) 100
- 0.1
o X
n'o
12 CHONDROITINASE
10
ABC
p\ of washed Protein A for 1 h on ice. The pellets obtained after centrifugation for 2 min at 12,000 x g were washed once with Buffer A, once with Buffer A containing 0.1% SDS, and twice with 10 mM Tris-CI, pH 7.6, containing 0.1% Nonidet P-40. Radioactivity in the pellets was determined after solubilization with Omni-Sol (ICN Biomédical). SDS-PAGE. SDS-PAGE was performed according to the method of Laemmli (17) with a 7% polyacrylamide running and a 4% stacking gel. Samples were heated at 100'C in 2.5% mercaptoethanol and 1.2% SDS for 5 min prior to electrophoresis. For fluorography the gels were incubated in a solution of 5% glycerol in Autofluor (National Diagnos tics) for 30 min and then dried. Immunoblotting was done by first electrophoretically transferring the proteins to nitrocellulose (18) and then incubating the nitrocellulose with the following solutions at room temperature: (a) 2% BSA in 10 mM Tris (pH 7.4); 0.15 M NaCl, and 0.1% NaN., for 90 min; (¿>) YPan-1, SPan-l, or myeloma ascitic fluid diluted 1:500 in the BSA solution; (c) rabbit anti-mouse IgG plus IgA plus IgM (Zymed Laboratories) diluted 1:500 in the BSA solution for 90 min; and (d) '"I labeled Protein A (106 cpm/ml) in the BSA solution for 60 min. Protein A was obtained from Zymed Laboratories and iodinated as described for the monoclonal antibodies (8). The sheet was washed with PBS-Nonidet P-40 between incubations. Preflashed (19) X-Omat ARS film (Kodak) was exposed to the nitrocellulose with an
-|0.5
- 0.4
- 0.3
• 0.2
- 0.1
40
50
60
70
80
FRACTION NUMBER Fig. I. Effect of chondroitinase ABC on radiolabeled medium antigen. Centrifuged, dialyzed, and concentrated culture medium from SW-1990 cells grown for 24 h in tritiated glucosamine was incubated with 0.5 unit/ml /'. vulgaris chondroitinase ABC in 0.05 M Tris-acétatebuffer (pH 7.3) overnight at 37'C. The samples were eluted from a 1.5- x 94-cm column of Sepharose CL-2B with PBS and fractions of 2 ml each were collected; 0.5 ml of each fraction was used for determination of radioactivity. SPan-l and YPan-1 immunoreactivities were determined with a direct binding ELISA. V¡,end of the included volume as determined by elution of tritiated glucosamine in a separate experiment. The calculated V, corresponds to Fraction 28.
3925
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-1 AND YPan-1
also contained most of the SPan-1 and YPan-1 immunoreactivity as determined by ELISA. Low levels of radioactivity, YPan1 antigen, and SPan-1 antigen were detected in fractions at or just before the included volume. Extracellular high molecular weight glycoconjugates could be either glycoproteins or proteoglycan. However, the elution profiles of [3H]glucosamine la beled material, of SPan-1 antigen, and of YPan-1 antigen were unaffected by pretreatment of the medium with chondroitinase ABC (Fig. 1; YPan-1 data not shown), which hydrolyzes chondroitin, chondroitin sulfates, hyaluronic acids, and dermatan sulfate (25). Lack of digestion by bovine testicular hyaluronidase (Fig. 2) further substantiated the conclusion that hyalu ronic acids were not responsible for the void volume radioactiv ity or immunoreactivity (YPan-1 not shown). Similar results were obtained when medium antigens were treated with nitrous acid, ruling out the possibility of heparan sulfate (14). Rechromatography of the void volume fractions on another Sepharose CL-2B column in the presence of 4 M guanidine-HCl, to dis sociate any high molecular weight proteoglycan aggregates (26), did not alter the elution profile of labeled molecules or immunoreactivities. However, when medium was digested with neuraminidase (Fig. 2), much of the radioactivity in the void volume disappeared with the appearance of a low molecular weight radioactive peak. Immunoreactivity with SPan-1 and YPan-1 n 0.4
(YPan-1 not shown) was completely abolished by sialic acid removal, concomitant with marked reduction in radioactivity. Density of Medium Antigens. When the void volume fractions from Sepharose CL-2B columns containing the most radioac tivity and SPan-1/YPan-1 immunoreactivities were pooled and centrifuged in the presence of CsCl, the radioactivity was found predominantly at a density of 1.4-1.5 g/ml (Fig. 3). The same fractions also contain the main portion of both SPan-1 and YPan-1 immunoreactivities. ^-Elimination and Protease Digestion of SPan-1 and YPan-1 Medium Antigens. When medium antigens purified by gel fil tration and density gradient centrifugation were subjected to alkaline borohydride treatment to release O-linked oligosaccharides by /3-elimination. the radioactivity was completely dis placed from the void volume of Sepharose CL-2B to the in cluded volume (Fig. 4). Immunoreactivity with SPan-1 was not detectable by direct binding ELISA but was detected in a competition ELISA, indicating that released oligosaccharides did not bind to microtiter plates but were recognized by anti body. Those fractions containing the greatest amounts of radio activity proved also to be the most inhibitory to SPan-1 (Fig. 4). Similarly, when purified SPan-1 antigens were digested with Pronase E, immunoreactivity was detected only with the com petition ELISA (Fig. 4). After 24 h of enzymatic hydrolysis, most of the radioactivity and SPan-1 inhibitory activity were shifted toward the included volume. In contrast to the results of alkaline borohydride treatment, both 3H and SPan-1 inhibi tory activity were broadly distributed between the void volume and the included volume, indicating the presence of immunoreactive glycopeptides of heterogeneous sizes. In the case of YPan-1, it was not possible to detect antigenic activity after either ^-elimination or protease hydrolysis. Isolation of SPan-1 and YPan-1 Antigens from SW-1990 Xenografts. We also carried out similar studies using SW-1990 xenografts grown in athymic nude mice. As shown in Table 1, both SPan-1 and YPan-1 antigens were present in CsCl frac tions corresponding to densities of 1.4-1.5 g/ml, a result similar
i o X
n'e
NEURAMINIDASE 1.6 ïS 1.4«
1.2LUO
10
15
20
25
30
¿
40
FRACTION NUMBER
1.0:•.'•'*.
12345678
Fig. 2. Effect of hyaluronidase and neuraminidase on medium antigens. Radiolabelcd culture medium was prepared as described in legend to Fig. 1. Part of the sample was incubated with 0.25 unit/ml C. perfringens neuraminidase in PBS for 2 h at 37'C and part in 0.5 mg/ml of bovine testicular hyaluronidase in citrate buffer (pH 5.5) for 16 h at 37'C. The column was 1.2 x 18 cm, and 0.5-ml
Fig. 3. Density gradient centrifugation of medium antigens. The void volume fractions after gel filtration of Sepharose CL-2B were centrifuged in the presence of 0.54 g/ml CsCl for 48 h at 4'C. and 1.5 ml fractions were collected: 0.2 ml of
fractions were collected; 0.2 ml of each fraction was used for determination of radioactivity. V,, end of the included volume. The calculated K>corresponds to Fraction 11. SPan-1 immunoreactivity was determined by direct binding ELISA.
each fraction was used for determination of radioactivity. Density was determined gravimetrically. YPan-1 and SPan-1 immunoreactivities of the dialyzed fractions were determined by direct binding ELISA.
FRACTION
NUMBER
3926
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-1 AND YPan-1 ALKALINE
chromatography of CsCl Fraction 6 on a TSK-G (4000) SWG column. The peak of immunoreactivity and carbohydrate was at the excluded volume with an apparent molecular weight greater than that of thyroglobulin (M, 660,000). Reductive ^-Elimination and Protease Digestion of Purified Xenograft Antigens. Alkaline borohydride treatment of xeno graft antigen shifted SPan-1 and YPan-1 immunoreactivity to the included volume of Sepharose CL-2B columns (Fig. 6). Treatment of the xenograft antigen with Pronase resulted in low molecular weight glycopeptides which inhibit both SPan-1 and YPan-1 immunoreactivities. Although YPan-1 immuno reactivity was decreased by both treatments, a low level of YPan-1 inhibitory activity was detectable with the competition ELISA using a lower antibody concentration than in the control (Fig. 6). Comparison of Immunoprecipitated Media and Purified Xen ograft Antigens by SDS-PAGE. When the [3H]glucosamine
8OROHYORIDE
-,0.25
JO.OS 10
15
20
25
30
35
40
FRACTION NUMBER
Fig. 4. Effect of alkaline borohydride and protease on purified medium anti gens. Medium antigens purified by gel filtration and density gradient centrifugation were treated with 0.05 M NaOH and 1 M NaBH«at 37'C for 16 h (see "Materials and Methods" for details) or Streptomyces griseus Pronase E at 5 jig/ ml for 16 h at 37'C in 0.1 M Tris-Ci buffer containing 0.05 M CaCl2 (pH 8.0). Column conditions were the same as those described in legend to Fig. 2; 0.2 ml of each fraction was used for determination of radioactivity. SPan-l immunoreactivity was determined by competition ELISA.
to those obtained with spent culture medium (Fig. 3). The highest protein concentration was found in the lowest density fraction (Fraction 1). Hexose was present in two major peaks. Fractions 6 and 8. Fraction 6 also contained the highest SPan1 and YPan-1 immunoreactivities and O-linked oligosaccharides. The high phenolsulfuric positive reaction of Fraction 8 may be due to nucleic acids since this fraction has a maximum absorbance at 260 nm as compared to 280 nm for Fraction 6. In the case of xenograft antigen, we substituted HPLC for conventional column chromatography to isolate the high mo lecular molecules. As shown in Fig. 5, YPan-1 and SPan-1 immunoreactivity were found in the void volume region after
labeled high molecular weight fraction was immunoprecipitated with SPan-1 or YPan-1,76-86% of the total label was recovered in the pellets. Under the same conditions, the myeloma ascites used as control precipitated only 2-4% of the total counts. This result indicates that the bulk of the [3H]glucosamine labeled high molecular weight glycoprotein is recognized by SPan-1 and by YPan-1. In agreement with the results of column chro matography (Fig. 1), [3H]glucosamine labeled glycoproteins immunoprecipitated with YPan-1 or SPan-1 did not penetrate into a 7% resolving gel (Fig. 7). Thus even under denaturing conditions and in the presence of mercaptoethanol, these anti gens have molecular sizes substantially greater than 200,000, the largest molecular weight standard used (myosin). Similarly, after immunoblotting of purified xenograft antigen a substantial portion of SPan-1 and YPan-1 immunoreactivity remained in the 4% gel or, in the case of SPan-1, at the top of the 7% gel. No immunostaining was observed when myeloma ascitic fluids were used in place of the first antibody. Periodic acid-Schiff reactive material was also observed in the 4% gel and at the top of the 7% gel. General Characteristics of SPan-1 and YPan-1 Epitopes. The above results demonstrate that the YPan-1 and SPan-1 epitopes are present on mucin-like molecules both in spent medium and in xenograft tumors. Neither boiling nor reduction and alkylation of xenograft mucins had any effect on the reactivity of either SPan-1 or YPan-1. Incontrasi, pretreatment of xenograft mucin with 0.1 ITIMperiodate reduced YPan-1 and SPan-1 immunoreactivity by 38% and 12%, respectively, while 1 mM periodate abolished reactivity of both antibodies. Neuraminidase pretreatment also destroyed all SPan-1 and YPan-1 reac tivities. These results suggest that the carbohydrates of the
Table 1 Characterization of density gradient fractions Fraction Immunoreactivity0 SPan-1 YPan-1Carbohydrate Hexose* oligosaccharidesfProtein'' O-linked
0.1050.100
11.39.15
0.1200.075
10.96.25
0.0980.057
12.44.10
Density*0.120 " Determined * Determined r Determined d Determined ' Determined
0.1100.062
0.0900.05621.71.10 0.2300.162
14.72.20
1.200.145 1.260.187 1.280.190 1.320.290 by ELISA (absorbance at 415 nm). by phenolsulfuric assay (mg/ml). by fluorescence assay in the presence of NaOH (relative fluorescence units). by assay of Lowry et al. (mg/ml). gravimetrically (g/ml).
61.60.75 1.350.495
1.400.295
3927
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
0.1800.093
32.20.12 1.490.063
0.0700.450
9.10.12 1.52
MUCIN-LIKE ANTIGENS OF SPan-l AND YPan-I CONTROL
o.er OB060.40.2*••-
SPAN-1f *'•'
\••oocüU .».*•"'•
po
p°oOoo°oO c1 —°
f\\ß•«'
1.0 r
ALKALINE
U^^^^vs
BOROHYDRIDE
0.8
0.4
o.e
-D O
0.3
m z
0.2
o en
0.4
S 5
POOOO°O
0.2
V
o 3> a>
0.1
lÃ-
0.8
PROTEASE
..^•v
0.6 FRACTION NUMBER
Fig. S. 11l'I < of partially purified xenograft mucins. CsCI fractions containing the maximum amount of SPan-l and YPan-1 immunoreactivities (usually with densities of 1.4-1.S g/ml) were pooled and passed through a TSK-G (4000) SWG column (30 x 2.15 cm). Three-mi fractions were collected. Vhend of the included volume. YPan-1 and SPan-l immunoreactivities were determined by a direct binding ELISA, and hexose was determined by the phenolsulfuric method. Bar, fractions which were pooled and concentrated for SDS-PAGE (Fig. 7).
0.4
0.2
0oo0OocboooooooO 10
mucin molecule play the major role in YPan-1 and SPan-l binding. Mucins were also preincubated with several lectins such as WGA, succinylated WGA, fucose binding protein (L. tetragonolobus), Bandeiraea simplicifolia (BS-II; W-acetylglucosamine), Arachis hypogaea (galactose; /V-acetylgalactosamine), Ulex europaeus (fucose), and Dolichos biflorus (JV-acetylgalactosamine) at concentrations up to 1 mg/ml in an attempt to block binding of SPan-l and YPan-1 to specific monosaccharides. No signif icant effect was observed except in the case of WGA (Table 2). WGA inhibited binding by both SPan-l and YPan-1. Since WGA is reported to bind both sialic acids and jV-acetylglucosamine we also tried the succinylated form of the lectin which binds only /V-acetylglucosamine (27). Succinylated WGA had much less of an inhibitory effect on the binding of SPan-l and YPan-1 to mucin, suggesting that WGA inhibits SPan-l and YPan-1 through competition for sialic acid and not JV-acetylglucosamine. Mild alkali treatment of xenograft mucin in the presence of borohydride produced a fraction which, after desalting and removal of borate, could partially inhibit SPan-l and YPan-1 (Fig. 8). The recovered oligosaccharide fraction had a markedly reduced ability to inhibit the antibodies (especially YPan-1) when compared to a comparable amount of starting material (based on hexose determination). Relationship of the SPan-l and YPan-1 Antigenic Determi nants. Purified YPan-1 did not specifically inhibit SPan-l bind ing to purified xenograft mucins in a double determinant radioimmunoassay developed with purified SPan-l antibody (Table 3). This indicates that the epitopes for the two are distinct.
20
18
25
FRACTION
30
35
40
NUMBER
Fig. 6. Effect of alkaline borohydride and protease on purified xenograft mucins. Purified xenograft mucins were incubated with 0.05 M NaOH and l M NaBH4 for 16 h at 37'C or with Pronase E in 0.05 M Tris-CI containing 0.05 M ( ;i( l... pH 8.0, for 16 h at 37'C. Chromatography conditions are the same as those described in legend to Fig. 2. YPan-1 and SPan-l immunoreactivities were determined by competition ELISA.
B 4%
.
7%
•4 205
K
•4116 K •« 66 K •« 45
12
K
34
Fig. 7. Comparison of immunoprecipitated radiolabeled media antigens with immunoblots of xenograft antigens. Radiolabeled medium immunoprecipitated as described in "Materials and Methods" and HPLC purified xenograft mucin (indicated by the bar in Fig. 5) were subjected to SDS-polyacrylamide electrophoresis using a discontinuous 4%/7% gel system. Xenograft antigens were then transferred electrophoretically to nitrocellulose paper for immunostaining with mouse monoclonal antibodies (see "Materials and Methods"). Periodic acidSchiff staining was done directly on the gels. Medium antigen II) immunoprecip itated with SPan-l (Lane I: 21,200 cpm, developed 14 days) and YPan-1 (Lane 2; 16,820 cpm, developed 14 days). Xenograft antigen (A) immunoblotted with SPan-l (Lane 3) and YPan-1 (Lane 4) or PAS stained (Lane 5). Molecular weight markers are ovalbumin (M, 45,000), BSA (M, 66,000), 0-galactosidase (M, 116.000), and myosin (M, 205,000). A',thousands.
3928
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-1 AND YPan-l
Table 2 Inhibition by wheat germ agglutinin SPan-1mg/ml2.5 WGA1914
WGA31
17 90 9 8471Succinylated13 2YPan-lWGA91 12
1.25 6453 0.31 0.08WGA71" 47Succinylated
Percentage of inhibition; mean of triplicate determinations.
0.5r
SPAN-1
YPAN-1
°^2--*?3/0
O.4HI0 m
O.2
*>"••.•
OLIGOS. 0-.0-«"*
0.3CO§ 0.2V)
0.1 MUCIN•/.-•-*•^III!
CO"
MUCIN
0.1OLIGOS..7°o/°
2
8
32
128 512 2048
RECIPROCAL
2
OF INHIBITOR
8
32
128
DILUTION
Fig. 8. Immunoreactivity of intact mucin and isolated oligosaccharides. oli gosaccharides obtained by reaction of purified xenograft mucins with 0.05 M NaOH in presence of l M NaBH4 were desalted as described in "Materials and Methods." Immunoreactivity was determined by a competition ELISA. The highest concentration of intact mucin or oligosaccharide tested was 2 jig hexose/ ml. Table 3 Competitive radioimmunoassay Double determinant radioimmunoassay with radiolabeled SPan-1 and SW1990 HPLC purified xenograft mucin as antigen. Proteintaj0.02 control5557 ±209"
0.10 0.5 2.0IgM
±540 ±669 ±372 5796 ±810 6576 ±275 4859 ±264 5646 ±459 4812 + 521 6088 + 71 1132 ±182 2115± 249 4943 ±613YPan-l6252 4852 ±593SPan-15329 724 ±23619-95579 852 ±196
' Mean cpm + SD; n = 3. 6 r-
SPAN-1
- SPAN-1*
ilarly, radiolabeled YPan-l is also bound if unlabeled SPan-1 is used first as the catcher. The amount of radioactivity is a function of antigen level and requires at least 0.1 Mg/well of either unlabeled antibody to be added as catcher. In an attempt to dissociate YPan-l immunoreactivity from that of SPan-1, purified xenograft mucins were separated by affinity chromatography on WGA-lectin columns. No SPan-1 or YPan-l an tigen was eluted until 0.1 M /V-acetylglucosamine was used. Significant amounts of both SPan-1 and YPan-l immunoreactivities eluted from the column in parallel when a step gradient of yV-acetylglucosamine (0.1,0.2,0.3,0.4, and 0.5 M) was used. Thus, the two antigenic determinants were not dissociable by this method. Comparison of SPan-1 and YPan-l with CSLEX-1, DU-PAN2, and 19-9. In addition to SPan-1 and YPan-l, SW-1990 xenograft mucin is recognized by three other sialic acid depend ent monoclonal antibodies, 19-9, CSLEX-1, and DU-PAN-2 (Fig. 10). When CSLEX-1 or DU-PAN-2 was used as compet itor in SPan-1 or YPan-l double determinant radioimmunoassays at concentrations of up to 2 mg/ml protein, no specific inhibition of the binding of radiolabeled SPan-1 or YPan-l to xenograft mucin was seen. 19-9 antibody, however, does inhibit binding by SPan-1 (Table 3). DISCUSSION The SPan-1/YPan-l antigens from SW-1990 spent medium are high density glycoproteins which have molecular weights greater than 200,000 even after dissociating treatments such as 4 M guanidine hydrochloride and heating in the presence of SDS and 0-mercaptoethanol. While high molecular weight carbohydrate containing substances secreted by pancreatic can cer cells may include proteoglycans in addition to mucins, the SPan-1/YPan-l antigens are not affected by proteoglycan hydrolyzing enzymes and chemicals such as chondroitinase ABC, hyaluronidase, and nitrous acid. Furthermore, the density of the SPan-l/YPan-1 antigens, 1.4-1.5 g/ml, while much higher than that of serum-type glycoproteins and of carbohy drate-free proteins, is lower than that typically found for pro teoglycans (28). Most of the [3H]glucosamine labeled high
5 -
E U
':'O\CSLEX-1
\YPAN-1
4 -
19-9:
\o^0^ - SPAN-1*
\SPAN-1 3 . iXAX/A^A-YPAN-1 - YPAN-1*
CD
\I
\0 •\
1 -
':-A * ^A YPAN-1^ v\ 1
4
16
RECIPROCAL
64
UiQcom1.11.00.90.80.7 0.6
\XAX
°\
0.50.40.30.20.110
\^DU-PAN-2*^^^X\.lili
,jbÈ ^ . 256
OF ANTIGEN
1024
\
Mlt
4096
16384
DILUTION
Fig. 9. Heterologous and homologous double determinant radioimmunoassays. The 96-well polyvinyl chloride microtiter plates were first coated with either purified YPan-1 or SPan-1. Different dilutions of purified xenograft mucin were applied (initial concentration, 40 fig hexose/ml). Radioiodinated SPan-1 or YPan-l was then added (10s cpm/well). *, labeled antibody.
li^s— + 2
4
8
16
32
1
64 128 256
RECIPROCAL OF ANTIGEN DILUTION
However, molecule. significant unlabeled
the two determinants are found on the same mucin In double determinant radioimmunoassays (Fig. 9) a amount of radiolabeled SPan-1 becomes bound if YPan-l (heterologous) is used as the "catcher." Sim-
Fig. 10. Reactivity of CSLEX-1, DU-PAN-2, and 19-9 with xenograft mucin. Immunoreactivity was measured by ELISA with purified xenograft mucin as the antigen. IgM specific ff-galactosidase conjugated second antibody was used for detection of CSLEX-1 and DU-PAN-2 while IgG specific second antibody was used for 19-9. Optimal concentration of each test antibody was used for the assay. The initial dilution of antigen used was 10 pg hexose/ml.
3929
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-1 AND YPan-1
molecular weight glycoprotein in medium is immunoprecipitable with either SPan-1 or YPan-1. Moreover, the glucosamine labeled molecules are susceptible to Pronase digestion and reductive ¿¡-elimination. These results indicate that both the SPan-1 and YPan-1 antigenic determinants are carried on mucins. The antigens of B72.3 (29), 19-9 (1), and DU-PAN-2 (30) have previously been shown to be mucins by experiments similar to those presented here. Binding of SPan-1 and YPan-1 to mucin is extremely sensi tive to periodate oxidation and requires sialic acid as indicated by neuraminidase and WGA sensitivity. Denaturation of the protein by boiling or by reduction and alkylation had no effect. Pronase digestion or /^-elimination resulted in YPan-1 immunoreactivity only for the xenograft antigen, probably because of the higher level of antigen in xenograft preparations. The marked reduction of the ability of isolated mucin oligosaccharides to block YPan-1 and SPan-1 immunoreactivities (Fig. 8) may be due to their having lower binding affinities for isolated oligosaccharides than for intact mucins, as has been reported (31) for other antibodies. The above results suggest that the SPan-1 and YPan-1 antibody binding sites on mucin molecules are composed mainly of carbohydrates and require sialic acid. Attempts to further characterize their epitopes with lectins were unsuccessful. This may, in part, be due to steric hindrance of sialic acids or neighboring oligosaccharides which prevent ac cess of the lectins to their appropriate binding sites. Although SPan-1 and YPan-1 determinants are distinct, they can be found on the same mucin molecules as shown by heterologous double determinant radioimmunoassays. The two immunoreactivities could not be dissociated by WGA affinity column chromatography. The possible existence of SPan-1 or YPan-1 enriched populations of mucins is presently being investigated by means of SPan-1 and YPan-1 antibody affinity columns. An increasing number of sialic acid requiring monoclonal antibodies have been reported to be cancer associated. These include B72.3 (32), CSLEX-1 (33), DU-PAN-2 (34), OC 125 (35), and 19-9 (36). The SPan-1 and YPan-1 antigenic deter minants are distinct from those of CSLEX-1 and DU-PAN-2 since the latter antibodies do not inhibit SPan-1 or YPan-1 binding to SW-1990 xenograft mucins. SPan-1 and YPan-1 can also be distinguished from B72.3 in that SPan-1 has a 95% (10) and YPan-1 a 81% reactivity (6) with normal pancreatic tissue while B72.3 has little if any reactivity with most adult normal tissues including pancreas (32). OC 125 is distinct from SPan1 and YPan-1 in that its antigenic determinant is heat sensitive (37). On the other hand, 19-9 does compete with SPan-1 for binding to SW-1990 mucins. Thus, the SPan-1 epitope is similar to that of 19-9. However, there is evidence that 19-9 and SPan-1 epitopes are not identical. While 19-9 does not react with colonie cancer tissues from patients with the Lewis", Lewis* negative phenotype, SPan-1 does (8). Experiments are in progress to isolate specific oligosaccharides from xenograft mucins which bind to YPan-1 or SPan-1 in order to obtain more details of the epitopes of these two antibodies and to determine how the antibody binding site of SPan-1 relates to that of 19-9 and of other antibodies similar to 19-9 such as C50 (38). Several of the mucin reactive antibodies which have a high degree of specificity for malignancy when assayed in serum show less specificity when pancreatic tissues are examined. This is the case with SPan-1 (10), 19-9 (36), and DU-PAN-2 (39). This suggests that although the epitopes themselves may not be unique to malignancy, their appearance in high levels in the bloodstream is (40). At present, little is known about the
mechanisms or factors producing these high levels. It may be due to increased synthesis and/or release, or changes in the mode of release, for example, directly into the bloodstream (basolaterally rather than apically into the duct lumen), result ing from loss of polarity of cancer cells or to changes in the way such molecules are degraded. Because mucins isolated from SW-1990 carry several potentially clinically useful epitopes such as SPan-1, YPan-1, 19-9, and DU-PAN-2, the SW-1990 cell in culture and in vivo would offer an ideal model system for the study of the factors regulating the biosynthesis and release of such pancreatic cancer associated products. REFERENCES 1. Magnani, J. I ., Steplewski, /.. Koprowski, II., and Ginsburg, V. Identifica tion of the gastrointestinal and pancreatic cancer-associated antigen detected by monoclonal antibody 19-9 in the sera of patients as a mucin. Cancer Res., «.•5489-5492,1983. 2. Kannagi, R., Fukushi, Y., Tachikawa, T., Noda, A., Shin, S., Shigeta, K., Hiraiwa, N., Fukuda, Y., Inamoto, T., Hakomori, S. I., and Imura, H. Quantitative and qualitative characterization of human cancer-associated serum glycoprotein antigens expressing fucosyl or sialyl-fucosyl type 2 chain polylactosamine. Cancer Res., 46: 2619-2626, 1986. 3. Paterson, A. J., Schlom, J., Sears, H. F., Bennett, J., and Colcher, D. A radioimmunoassay for the detection of a human tumor-associated glycopro tein (TAG-72) using monoclonal antibody B72.3. Int. J. Cancer, 37: 659666, 1986. 4. Sawabu, N., Toya, D., Takemori, Y., Hattori, N., and Fukui, M. Measure ment of a pancreatic cancer-associated antigen (DU-PAN-2) detected by a monoclonal antibody in sera of patients with digestive cancers. Int. J. Cancer, 3 7:693-696, 1986. 5. Burchell, J., Gendler, S., Taylor-Papadimitriou, J., Girling, A., Lewis, A., Millis, R., and Lamport, D. Development and characterization of breast cancer reactive monoclonal antibodies directed to the core protein of the human milk mucin. Cancer Res., 47: 5476-5482, 1987. 6. Yuan, S., Ho, J. J. L., Yuan, M., and Kim, Y. S. Human pancreatic cancerassociated antigens detected by murine monoclonal antibodies. Cancer Res., «.•6179-6187,1985. 7. Ho, J. J. L., Yuan, S., Yuan, M., and Kim, Y. S. Characterization of the antigen from human pancreatic cancer cells recognized by monoclonal anti body, YPan-1. Gastroemerology, 86: 1113, 1984. 8. Chung, Y. S., Ho, J. J. L., Kim, Y. S., Tanaka, H., Nakata, B., Hiura, A., Motoyoshi, H., Satake, K., and Umeyama, K. The detection of human pancreatic cancer-associated antigen in the serum of cancer patients. Cancer (Phila.), 60: 1636-1643, 1987. 9. Kyriazis, A. P., McCombs, W. B., Ill, Sandberg, A. A., Kyriazis, A. A., Sloane, N. H., and Lépera,R. Establishment and characterization of human pancreatic adenocarcinoma cell line SW1990 in tissue culture and the nude mouse. Cancer Res., 43: 4393-4401, 1983. 10. Chung, Y. S., Ho, J. J. L., Satake, K., Tanaka, H., Nakata, B., Sowa, M., Umeyama, K., Yuan, M., and Kim, Y. S. Distribution of SPan-1 antigen in normal and cancerous human gastrointestinal tissues. Gastroenterology, 94: A71, 1988. 11. Carlstedt, I., Lindgren, H., Sheehan, J. K., Ulmsten, U., and Wingerup, L. Isolation and characterization of human cervical-mucus glycoproteins. Biochem. J., 2il: 13-22, 1983. 12. Edge, A. S. B., Van Langenhov, A., Reinhold, V., and Weber, P. Character ization of O-glycosidically linked oligosaccharides of rat erythrocyte mem brane sialoglycoproteins. Biochemistry, 25: 8017-8024, 1986. 13. Carlstedt, I., Lindgren, H., and Sheehan, J. K. The macromolecular structure of human cervical-mucus glycoproteins. Studies on fragments obtained after reduction of disulphide bridges and after subsequent trypsin digestion. Biochem. J., 213:427-435, 1983. 14. Kosher, R. A., and Searls, R. L. Sulfated mucopolysaccharide synthesis during the development of Rana pipiens. Dev. Biol., 32: 50-68, 1973. 15. Woodward, M. P., Young, W. W., Jr., and Bloodgood, R. A. Detection of monoclonal antibodies specific for carbohydrate epitopes using periodate oxidation. J. Immunol. Methods, 78: 143-153, 1985. 16. Gum, J. R., Kam, W. K., Byrd, J. C, Hicks, J. W., Sleisenger, M. H., and Kim, Y. S. Effects of sodium butyrate on human colonie adenocarcinoma cells. Induction of placental-like alkaline phosphatase. J. Biol. Chem., 262: 1092-1097, 1987. 17. Laemmli, U. K. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (Lond.), 227:680-685, 1970. 18. 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. 19. Laskey, R. A., and Mills, A. D. Quantitative film detection of 3H and MC in polyacrylamide gels by fluorography. Eur. J. Biochem., 56: 335-341, 1975. 20. Swanstrom, R., and Shank, P. R. X-ray intensifying screens greatly enhance the detection by autoradiography of the radioactive isotopes 32P and '"I. Anal. Biochem., 86: 184-192, 1978.
3930
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
MUCIN-LIKE ANTIGENS OF SPan-1 AND YPan-l
21. Zacharius, R. M., Zeli, T. E.. Morrison, J. H., and Woodlock, J. J. Glycoprotein staining following electrophoresis on acrylamide gels. Anal. Biochem., 30: 148-152. 1969. 22. Lowry, O. H., Rosebrough, N. J., Fair, A. L., and Randall, R. J. Protein measurement with the Polin phenol reagent. J. Biol. Chem., 193: 265-275, 1951. 23. Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F. Colorimetrie method for determination of sugars and related substances. Anal. Chem., 28: 350-356, 1956. 24. Crowther, R. S., and Wetmore, R. F. Fluorometric assay of O-linked glycoproteins by reaction with 2-cyanoacetamide. Anal. Biochem., 163: 170-174, 1987. 25. Kim, K. C, Rearick, J. I., Nettesheim, P., and Jetten, A. M. Biochemical characterization of mucous glycoproteins synthesized and secreted by ham ster trachéalepithelial cells in primary culture. J. Biol. Chem., 260: 40214027, 1985. 26. Hoffman, P. Selective aggregation of proteoglycans with hyaluronic acid. J. Biol. Chem., 254: 11854-11860, 1979. 27. Monsigny, M., Sene, C., Obrenovitch, A., Roche, A-C., Delmotte, F., and Boschetti, E. Properties of succinylated wheat-germ agglutinin. Eur. J. Biochem., 98: 39-45, 1979. 28. lozzo, R. V., and Wight, T. N. Isolation and characterization of proteoglycans synthesized by human colon and colon carcinoma. J. Biol. Chem., 257: 11135-11144,1982. 29. Johnson, V. G., Schlom, J., Paterson, A. J., Bennett, J., Magnani, J. L., and Colcher. D. Analysis of a human tumor-associated glycoprotein (TAG-72) identified by monoclonal antibody B72.3. Cancer Res., 46: 850-857, 1986. 30. Lan, M. S., Finn, O. J., Fernsten. P. D., and Metzgar, R. S. Isolation and properties of a human pancreatic adenocarcinoma-associated antigen, 1)1 PAN-2. Cancer Res., 45: 305-310, 1985. 31. Painter, T. J., Watkins, W. M., and Morgan, W. T. J. Serologically active
32. 33. 34. 35.
36. 37. 38.
39.
40.
fucose-containing oligosaccharides isolated from human blood-group A and B substances. Nature (Lond.), 206: 594-597, 1965. Thor, A., Ohuchi, N., Szpak, C. A.. Johnston, W. W., and Schlom, J. Distribution of oncofetal antigen tumor-associated glycoprotein-72 denned by monoclonal antibody B72.3. Cancer Res., 46: 3118-3124. 1986. Kawahara, M., Chia, D., Terasaki, P. I., Roumanas. A., Sugich, L., Hermes, M.. and Iguro, T. Detection of sialylated Lewis" antigen in cancer sera using a sandwich radioimmunoassay. Int. J. Cancer, .«>: 421-425, 1985. Metzgar, R. S., Gaillard. M. T., Levine, S. J., Tuck, F. L., Bossen, E. H., and Borowitz, M. J. Antigens of human pancreatic adenocarcinoma cells defined by murine monoclonal antibodies. Cancer Res., 42: 601-608, 1982. Hanisch, F-G., Uhlenbruck, G., Dienst, C., Stottrop, M., and Hippauf. E. Ça125 and Ça19-9: two cancer-associated sialylsaccharide antigens on a mucus glycoprotein from human milk. Eur. J. Biochem., 149: 323-330, 1985. Atkinson, B. F., Ernst, C. S., Herlyn, M., Steplewski, Z., Sears, H. F., and Koprowski, H. Gastrointestinal cancer-associated antigen in immunoperoxidase assay. Cancer Res., 42:4820-4823, 1982. Davis, H. M., Zurawski. V. R., Jr., Bast, R. C., Jr., and Klug, T. L. Characterization of the CA 125 antigen associated with human epithelial ovarian carcinomas. Cancer Res., 46:6143-6148, 1986. Musson, O., Mansson, J-E., Lindholm, L.. Holmgren, J., and Svennerholm, L. Sialosyllactotetraosylceramide, a novel ganglioside antigen detected in human carcinomas by a monoclonal antibody. FEBS Lett., 182: 398-402, 1985. Borowitz, M. J., Tuck, F. L., SnuU-l.tr. W. F., Fernsten, P. D., and Metzgar, R. S. Monoclonal antibodies against human pancreatic adenocarcinoma: distribution of DU-PAN-2 antigen on glandular epithelia and adenocarcinomas. J. Nati. Cancer Inst., 72: 999-1005. 1984. Kalthoff, H., Kreiker, C., Schmiege!, W-H., Greten, H., and Thiele, H-G. Characterization of CA 19-9 bearing mucins as physiological exocrine pan creatic secretion products. Cancer Res., 46: 3605-3607, 1986.
3931
Downloaded from cancerres.aacrjournals.org on July 11, 2011 Copyright © 1988 American Association for Cancer Research
