European Journal of Pharmaceutical Sciences, 6 (1998) 121–129
Human carcinoma-associated antigen (HCA), isolated from the endometrial carcinoma cell line KLE-1 and ascitic fluid of a patient with ovarian carcinoma; comparison with epiglycanin Torunn Thingstad a , Svein Haavik a , Katrine Hansen a , Knut Sletten c , John F. Codington b , a, Hilde Barsett * a
Department of Pharmacognosy, Institute of Pharmacy, P.O. Box 1068, Blindern, N-0316 Oslo, Norway b Boston Biomedical Research Institute, Boston, MA 02114, USA c The Biotechnology Centre of Oslo, P.O. Box 1125, N-0316 Oslo, Norway Received 13 July 1996; accepted 9 July 1997
Abstract Human carcinoma-associated antigen (HCA), detected by the mouse monoclonal anti-epiglycanin antibody, AE-3, has been isolated from ascitic fluid taken from a patient with metastatic ovarian adenocarcinoma and from spent medium of the human endometrial carcinoma cell line KLE-1 and compared with epiglycanin. The ascitic fluid and the spent medium were concentrated by a Filtron Ultrasette 100 K Omega membrane and fractionated by gel filtration on Sepharose CL-2B. The active fractions which consisted mainly of glycoproteins having relative molecular weights in the range 1000–2000 kDa compared to dextrans, were further purified by affinity chromatography on a column of immobilized AE-3. The active fraction was subjected to SDS–PAGE and blotted onto a PVDF membrane. The amino acid composition of HCA isolated from the two sources, were related but not identical and both showed some differences from the amino acid composition of epiglycanin. They all have, however, compositions typical of mucin-type glycoproteins. The isoelectric point for HCA from both KLE-1 and ascitic fluid were determined to be at pH 1.8 and the buoyant densities were about 1.4 g / ml as determined by cesium trifluoroacetate gradient centrifugation. 1998 Elsevier Science B.V. Keywords: Epiglycanin; Mucin-type glycoprotein; Human carcinoma antigen; KLE-1; Ovarian carcinoma
1. Introduction Most carcinomas generate and secrete mucins expressing carbohydrate structures that are not found in normal tissues (Hilkens, 1988). These changes can include the increased size of N-linked oligosaccharides that occurs with transformation (Warren et al., 1978) and the expression of incomplete or truncated O-linked mucin-type oligosaccharides which characterize Tn and the ThomsenFriedenreich (TF) determinants (Springer, 1984). Both secreted and membrane forms of certain carcinoma-associated mucins can protect the cancer cells from attack by the immune system (Codington et al., 1983; Fung and Longenencker, 1991). The potential importance of tumor cell surface mucins was first recognized in the mouse TA3 mammary adenocarcinoma (Codington et al., 1972). Allotransplantable *Corresponding author. Tel.: 147 22856573; fax: 147 22854402; e-mail:
[email protected] 0928-0987 / 98 / $19.00 1998 Elsevier Science B.V. All rights reserved. PII S0928-0987( 97 )00076-6
ascitic sublines of this tumor were found to possess a high concentration of a cell surface glycoprotein, epiglycanin, which was absent from non-allotransplantable sublines (Codington et al., 1983). Epiglycanin has been shown to protect the tumor cells from the immune system of their host, by blocking recognition of the histocompatibility complex (Fung and Longenencker, 1991). Recently it has been shown to inhibit a6,b4-mediated cell adhesion and E-cadherin cell–cell interaction (Kemperman et al., 1994). The molecular weight of epiglycanin is 500 000, 80% of which is carbohydrate, consisting mainly of Galb(1→3)GalNAc side chains linked to serine or threonine in the polypeptide chain (Van den Eijnden et al., 1979). Epiglycanin has been found to be released into ascitic fluid and serum (Cooper et al., 1974) and it has been shown that ascitic fluid and serum from cancer patients with carcinoma, contain epiglycanin crossreacting glycoproteins (Codington et al., 1984). Both poly- and monoclonal antibodies have been developed against epiglycanin (Codington et al., 1984, 1993). Certain anti-
122
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
epiglycanin antibodies possess a specificity for a glycoprotein present in all types of carcinomas tested but absent in individuals without such diseases (Codington et al., 1993). This carcinoma-related glycoprotein is called the human carcinoma-associated antigen (HCA). Recently an assay has been developed for determination of HCA in the serum of carcinoma patients, the COD Test HCATM EIA (Human Carcinoma Antigen (HCA) Enzyme Immunoassay Test Kit). This assay is capable of identifying the serum of advanced cancer (carcinoma) with a specificity and sensitivity of at least 90% (unpublished results). The purpose of the present study was to isolate, partly characterize and compare the HCA from two different sources, human ascitic fluid collected from a patient with ovarian carcinoma and spent medium from a human endometrial carcinoma cell line, KLE-I, grown in culture (Richardson et al., 1984), and to compare it with epiglycanin.
2. Experimental procedures Bovine serum albumin (BSA), naphthol AS-MX phosphate, O-dianisidine, tetrazotized (Fast Blue B-salt) and p-nitrophenyl phosphate were obtained from Sigma Chemical Company (St. Louis, MO, USA). Alkaline phosphatase-labeled goat anti-mouse IgM (m chain) and anti-mouse Ig were obtained from Boehringer Mannheim (Indianapolis, IN, USA). Dextrans, Sepharose CL-2B and gel filtration standards were obtained from Pharmacia AB (Uppsala, Sweden). Microtiterplates (NUNC Maxisorp) were obtained from Nunc (Copenhagen, Denmark).
2.1. Identification of HCA-producing cell lines Normal and cancerous human endometrial cell lines were grown in medium containing equal volumes of Ham F12 and Dulbecco’s modified Eagle’s media with 15% fetal calf serum (Gibco), 60 mU / ml insulin, 20 U / ml penicillin, 20 U / ml streptomycin and 0.5 mg / ml amphotericin B (Richardson et al., 1984). Samples of spent medium from confluent cells were withdrawn, centrifuged to remove insoluble material and tested for HCA content by the enzyme competitive binding assay described below.
2.2. Ascitic fluid The ascitic fluid from which the HCA was isolated was taken by paracentesis from a 75-year-old woman at Massachusets General Hospital, Boston, USA. The patient’s illness was characterized as metastatic serous adenocarcinoma. Three tumor masses were observed, one of bowel or ovarian origin (oval in shape and of dimensions, 2.031.430.7 cm). The second tumor (dimensions 93937 cm) extended from the anterior surface at the
peritoneal reflection. A third mass (1.531.030.5 cm) was attached to the anterior endometrial surface. The serosal surface contained multiple fibrous adhesions. One omental lymph node was identified and was positive for metastatic carcinoma. Approximately 5.5 l of fluid were obtained. This was immediately cooled to 48C, centrifuged for 20 min at 80003g, and stored at 2208C until used.
2.3. Epiglycanin Epiglycanin was isolated from the ascites fluid of A / WySn mice 7 days following i.p. injection of 10 5 TA3-Ha cells, as described previously (Cooper et al., 1979).
2.4. Monoclonal antibodies Monoclonal antibodies were prepared and purified as described previously (Haavik et al., 1992). Briefly C57BL / J mice were immunized by subcutaneous injections of 10 6 viable TA3-Ha ascites cells and boosted with purified epiglycanin. Spleen cells of the immunized mice were fused with BALB7c (NS1) mouse myeloma cells, as described by others (Galfre and Milstein, 1981). Hybridomas producing antibodies against epiglycanin were screened by ELISA using microtitre plates coated with purified epiglycanin.
2.5. Enzyme competitive binding assay of HCA To the wells of a Nunc Maxisorp 96-well plate were added 100 ml of epiglycanin (50 ng / ml) in PBS, pH 7.50. Plates were incubated for 16 h at 48C. The wells were blocked by incubation for 45 min with 0.5% BSA in PBS, pH 7.50 (PBS–BSA). Two solutions were added: (a) 50 ml of a standard solution of epiglycanin or sample; and (b) 50 ml of purified monoclonal anti-epiglycanin IgM. Both antibody and antigen were dissolved in PBS–BSA. The mixture was incubated on a shaker for 16 h at 208C. The wells were washed three times (Skatron microplate washer) with PBS containing 0.05% Tween 20 and incubated for 2 h at 208C with alkaline phosphatase-labeled goat anti-mouse IgM dissolved in PBS–BSA. The wells were washed three times and then 100 ml substrate solution containing 1 mg / ml p-nitrophenyl phosphate in 0.1 M ethanolamine, pH 10.0, was added. The absorbance was read at 405 nm in a Bio-Rad Microplate reader coupled to a Macintosh SE computer using the Microplate Manager TM program.
2.6. Concentration by a Filtron Omega Ultrasette 100 K membrane Spent medium of the KLE-I cell line and the human ascitic fluid were concentrated by a Filtron Omega Ultrasette equipped with a 100 K membrane (48C).
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
2.7. Gel filtration chromatography on a Sepharose CL2 B column Samples of concentrated (10:1) medium and ascitic fluid were subjected to gel filtration chromatography on a column (53180 cm) of Sepharose CL-2B eluted with PBS buffer, pH 7.50, at 1 ml / min. The collected fractions were assayed for HCA content by anti-epiglycanin AE-3 in a competitive binding assay, as described above, for carbohydrate by the phenol–sulfuric acid method (Dubois et al., 1956) and for protein by reading the absorbance at 280 nm. Fraction A (Fr. A) was pooled and concentrated by the Filtron Ultrasette using an Omega 100 K membrane and stored at 2208C The column was pre-equilibrated with the dextrans: T2000, T500 and T70, with thyroglobulin, bovine serum albumin and K 2 Cr 2 O 7 .
2.8. Affinity chromatography on a column of immobilized anti-epiglycanin antibody A 50-ml sample of Fr. A from the Sepharose CL-2B column was applied at 0.2 ml / min onto a column (0.535 cm) of Carbolink TM (Pierce) with immobilized anti-epiglycanin AE-3 antibody and then recirculated through the column at the same rate for at least 20 h. The column was washed with PBS, pH 7.5, and then eluted with 4 M guanidinium chloride in PBS, pH 7.5. Fractions of 2.0 ml were collected, the absorbance read at 280 nm, and tested for HCA activity with anti-epiglycanin AE-3 and G-1 antibodies.
2.9. SDS–PAGE Homogeneous SDS–PAGE in 5% running and 3% stacking gels was run in a Protean TM Mini Dual Slab gel apparatus (Bio-Rad) using the discontinuous buffer system described by Laemmli (1970). Prior to electrophoresis the samples were heated with 5% SDS and 15 mM 2-mercaptoethanol for 5 min at 1008C. After electrophoresis the proteins were electrophoretically transferred to Immobilon TM -PSQ PVDF Transfer Membrane (Millipore). Gels and blots were stained for proteins with Coomassie Brilliant Blue R-250. To detect antibody-binding components, the blots were incubated with anti-epiglycanin antibodies AE-3 and G-1, and mouse IgM binding components were visualized by incubation with goat antimouse IgM conjugated with alkaline phosphatase. This was followed by the addition of substrate buffer containing 0.5 mg / ml Naphthol AS-MX phosphate and 2 mg / ml O-dianisidine, tetrazotized (Fast Blue B-salt) in 0.1 M Tris–HCl buffer, pH 10.0.
2.10. Determination of amino acid composition The high-molecular weight HCA band excised from the PVDF membrane, was subjected to hydrolysis under
123
vacuum in 6 M HCl at 1108C for 24 h. After removal of HCl under reduced pressure, the amino acid composition was determined using a Biocal JC 5000 automatic amino acid analyser.
2.11. Density gradient centrifugation To a sample of Fr. A from the Sepharose CL-2B column, cesium trifluoroacetate was added to a final concentration of 47%. The mixture was centrifuged in a Sorvall T-1250 rotor at 130 0003g for 72 h at 148C (Devaraj et al., 1992). Fractions of 1.0 ml were collected by pumping from the bottom of the tube. The densities of the fractions were determined by weighing aliquots. HCA activity in the collected fractions was determined by the enzyme competitive binding assay, as described above.
2.12. Isoelectric focusing A sample of Fraction A from the Sepharose CL-2B column was electrofocused in a Rotofor TM cell (Bio-Rad) for 4 h at 12 W at 118C with Biolyte (Bio Rad) pH 3–10 ampholyte. Fractions were collected and their absorbance at 280 nm and pH measured. The fractions were then neutralized by addition of 0.5 M Na-phosphate buffer, pH 7.50, and assayed for HCA-activity.
3. Results and discussion The cultured endometrial carcinoma cell line KLE-I (Richardson et al., 1984) was shown to secrete higher levels of human carcinoma-associated antigen (HCA), defined by activity against anti-epiglycanin antibodies, into its culture medium than any of several other human carcinoma cell lines tested (results not shown). The average content of HCA in spent medium of the KLE-I was about 500 ng / ml, determined as epiglycanin. The non-carcinoma endometrial cells tested, did not secrete HCA, and no activity was detected with the monoclonal antibody AE-3. The content of HCA in human ascitic fluid collected from a patient with ovarian carcinoma was about 700 ng / ml, determined as epiglycanin. More than 95% of the HCA activity of the spent KLE-I medium and the ascitic fluid was retained in the high-molecular weight fraction after concentration on Filtron Ultrasette, using a 100-kDa cutoff membrane. Quantitation of HCA in KLE-I and human ascitic fluid was determined by the enzyme competitive binding assay, described in the Section 2. Preferably, the standard and test material should have been identical in the enzyme competitive binding assay, but purified HCA was not available. The use of epiglycanin as a reference material for this assay was believed to be satisfactory as titration curves of epiglycanin and HCA isolated from KLE-I and human ascitic fluid, were shown to be parallel (Fig. 1). The
124
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
Fig. 1. Titration curves of epiglycanin (j), partially purified HCA (Fr. A from the Sepharose CL-2B column) from the KLE-1 cell line (m) and partially purified HCA (Fr. A from the Sepharose CL-2B column) from ascitic fluid (d) by the enzyme competitive binding assay.
specificities of several monoclonal anti-epiglycanin antibodies with affinity constants in the range of 10 8 to 10 10 l / mol has been described (Haavik et al., 1992). G-1 and AE-3, later raised monoclonal IgM anti-epiglycanin antibodies with high-affinity constants, were used for detection of HCA in this study. These antibodies bind to a complex glycopeptide epitope containing the T-antigen, Galb(1→3)GalNAc. AE-3 is used in the COD-test for determination of HCA in the serum of carcinoma patients and an article describing the characterization of AE-3 recognizing HCA is in preparation. HCA from both KLE-1 and ascitic fluid showed, however, high cross-reactivity against all the anti-epiglycanin antibodies. Both G-1 and AE-3 were available in relatively large amounts and they were successfully used in enzyme competitive binding assays capable of quantitating less than 1 ng (2310 215 mol) of epiglycanin. KLE-1 and ascitic fluid were purified using a Sepharose CL-2B gel filtration column. The fractions obtained after gel filtration on Sepharose CL-2B were tested for protein, carbohydrate and HCA, determined by the enzyme competitive assay, and the profiles shown in Fig. 2a,b. The active HCA from both KLE-1 and ascitic fluid was separated from the main protein peak and found in the fraction of the highest molecular weight (Fr. A). This fraction corresponds roughly to material having relative molecular weights in the range of 1000–2000 kDa. The use of 6 M guanidinium hydrochloride as eluent did not change the elution pattern significantly. Similar molecular weight ranges was reported for the DU-PAN-2 antigen from human pancreatic adenocarcinoma (Lan et al., 1985), TAG-72 from human colon carcinoma xenograft LS-174T (Sheer et al., 1988) and the breast cancer-associated glycoprotein 83D4 (Pancino et al., 1991). These molecular
weights are far higher than the molecular weight of epiglycanin, which is about 500 000 (Codington et al., 1979). Purified epiglycanin was also run on the Sepharose CL-2B column as a control, and gave as expected one active peak with a molecular weight of about 500 000 (Fig. 2c). The collected fractions were pooled as indicated in the figure, and fraction A was concentrated as before using the 100-kDA cutoff membrane. Fraction A contained about 80% of the HCA activity that was applied to the column. Upon density gradient centrifugation of the Sepharose CL-2B-isolated HCA from KLE-1 in a cesium trifluoroacetate gradient, one major activity peak was obtained corresponding to a density of about 1.4 g / ml (Fig. 3a). The presence of one symmetrical peak shows that the partially purified HCA from KLE-I medium is a relatively homogenous mucin-type glycoprotein. The density of the active material isolated from ascitic fluid was estimated to be in the region 1.35–1.45 g / ml (Fig. 3b). The active material is present in two to three partly separated peaks and shows that the partially purified HCA from human ascitic fluid is a more heterogeneous glycoprotein. The density distributions of HCA from both KLE-1 and ascitic fluid are, however, typical of mucin-type glycoproteins (Dahiya et al., 1993; Decaens et al., 1993; Ho et al., 1993; Nishida et al., 1993). Density gradient centrifugation of purified epiglycanin in a cesium trifluoroacetate gradient resulted in one narrow peak corresponding to a density of 1.4 g / ml (Fig. 3c). Isoelectric focusing in a Rotofor TM Cell was carried out in order to determine the isoelectric point of the HCA. The determination was complicated by some precipitation of the HCA below pH 2, but the pI of HCA from both KLE-1 and ascitic fluid was estimated to be around pH 1.8. Preliminary carbohydrate analysis of HCA from KLE and ascitic fluid showed presence of mannose, galactose, N-acetylgalactosamine, N-acetylglucosamine, Nacetylneuraminic acid (NANA) and a small amount of fucose. Test for sulfate has not been performed. The active fractions from the Sepharose CL-2B column were further purified by affinity chromatography on a column of immobilized anti-epiglycanin antibody (AE-3). Slow application (0.2 ml / min) and recirculation of the solution through the column for 20 h was necessary for complete binding between the glycoproteins and the immobilized antibody. About 90% of the active material from KLE-1 and about 80% of the active material from the ascitic fluid bound to the column. The column was thoroughly washed and bound HCA was eluted with 4 M guanidinium chloride. Preliminary experiments for analytical purpose eluting the affinity column with a pH gradient (pH 8–2) or a methyl-galactoside gradient (0–2 M) gave, with both gradients, one active peak for HCA from KLE-1 and ascitic fluid. Use of methyl-galactoside resulted, however, in cross-reactivity in the enzyme competitive binding assay. For preparative purposes the affinity column was therefore eluted with 4 M guanidinium chloride. The HCA activity in the fractions was tested with
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
125
Fig. 2. Gel filtration chromatography of (a) concentrated spent medium from the KLE cell line, (b) concentrated ascitic fluid from patient with ovarian carcinoma and (c) epiglycanin on Sepharose CL-2B (53180 cm). Fractions were tested for HCA activity with anti-epiglycanin AE-3 antibody (m), for protein content by reading the absorbance at 280 nm (d) and carbohydrate content by the phenol sulphuric acid method (absorbance at 487 nm) (s). Elution volumes of molecular weight markers are indicated by arrows: (1) dextran T 2000 (2 000 000); (2) thyroglobulin (670 000); (3) bovine serum albumin (67 000); (4) K 2 Cr 2 O 7 (294). Fractions were pooled as indicated.
both AE-3 and G1 anti-epiglycanin antibodies to confirm that the HCA isolated by immobilized AE-3 also showed activity by other anti-epiglycanin antibodies. The elution with 4 M guanidinium chloride gave one active peak, and the active fractions were pooled, dialyzed against distilled water and lyophilized prior to further studies. A concentrated and dialyzed solution of partially purified HCA gave some precipitation. This precipitate was not soluble in
different detergents and dissociating agents tested (results not shown). SDS–PAGE was performed on samples of the material that was applied to the affinity column (Fr. A), the material that passed through the column, the affinity-purified HCA and epiglycanin. The separated bands were transferred to immobilon PVDF membranes and stained for proteins with Coomassie Brilliant Blue, and for HCA activity with the
126
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
Fig. 3. Density gradient centrifugation of (a) Fr. A from the KLE-1 cell line, (b) Fr. A from ascitic fluid from a patient with ovarian carcinoma, and (c) epiglycanin in 47% cesium trifluoroacetate, in a Sorvall T1250 rotor at 130 0003g for 72 h. Fractions of 1 ml were collected from the bottom of the tube and their density and HCA activity determined. HCA activity (mg / ml, j), density (g / ml, s).
anti-epiglycanin antibodies AE-3 and G-1, which gave the same staining pattern. The Coomassie and AE-3 stained blots are shown in Fig. 4. When the gels were stained with Coomassie, several bands were observed. This can be due to heterogeneity of the applied materials but also to degradation of the molecules upon boiling the samples
with SDS and mercaptoethanol. Application of SDS– PAGE to sialomucin analysis is not as simple as for other proteins (Carraway and Spielman, 1986). Because of their high negative charge, they stain poorly with Coomassie Blue and other protein stains. The anti-epiglycanin antibodies are bound, however, only to one broad high-molecular weight band, which means that SDS–PAGE can be seen as an extra purification step. All the samples were located in the running gel after SDS–PAGE, and the blots shown in Fig. 4 represent all of the sample loaded onto the gels. Epiglycanin gave the same staining pattern as HCA when stained with the anti-epiglycanin antibody AE-3, one high-molecular weight band at the top of the running gel. The portion of the PVDF-membranes containing the high-molecular weight band was subjected to hydrolysis (6 M HCl) and subsequent analysis of amino acid composition (Table 1). The anti-epiglycanin antibodies are bound to one broad high-molecular weight band. To ensure that this band was homogeneous, amino acid analysis were performed separately on the upper and the lower parts of the band. The amino acid composition of these bands were identical, which indicates that the broad high-molecular band of HCA on SDS–PAGE is due to differences in glycosylation. Glycine from the transfer buffer, was removed from the PDVF membranes with repeated washings with distilled water before performing the hydrolysis and amino acid analysis. In spite of this, the glycine content is relatively high. This could still be due to some contamination from the transfer buffer, but the glycine content before washing with distilled water was, however, much higher (300–400 residues / 1000 residues). The amino acid composition of the HCA from the ascitic fluid and the KLE-1 show high structural similarity, but the molecules are not identical. They are both characterized by a high proportion of serine, threonine, alanine, glycine and proline. The threonine content, however, is lower in KLE-1 than in the ascitic fluid. In the ascitic fluid, threonine constitutes 13% of the total amino acid content, while the same amino acid constitutes only 7% of the total content in HCA from KLE-1. These data suggest the presence of different forms of HCA from different sources. The amino acid composition of HCA from both KLE-1 and ascitic fluid differ from the one of epiglycanin which is characterized by a higher proportion of serine and threonine (Codington and Haavik, 1992) (Table 1). The relative content of serin:threonine in epiglycanin is 1.5:1, while in HCA from KLE-1 and HCA from ascitic fluid the ratios are 1.8:1 and 1:1, respectively. The isolation procedure described, gives HCA from KLE-1 and ascitic fluid with little or no degradation. The isolated molecules have much higher molecular weight than epiglycanin, and could be mucins in complex with less mucin typical parts. The absence of cysteine in HCA indicates that this secreted mucin lacks the C-terminal cysteine-rich domain found in many mucins (Wu et al., 1994). All three mucins, epiglycanin, HCA from KLE-1 and HCA from ascitic fluid have about 60%
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
127
Fig. 4. SDS–PAGE in 5% running and 3% stacking gel of affinity purified HCA from (a) KLE-1 and (b) ascitic fluid. The proteins were blotted onto PVDF-membranes and stained for proteins with (A) Coomassie Brilliant Blue R-250 and (B) HCA activity with monoclonal anti-epiglycanin antibody AE-3. (1) Solution of HCA before affinity chromatography; (2) solution of HCA not bound to the affinity column; (3) affinity-purified HCA; and (4) epiglycanin. Samples of the membrane containing the HCA were subjected to hydrolysis and subsequent determinations of amino acid compositions. The top of the running gel is the top of the figure.
of the composition accounted for by threonine, serine, proline, glycine and alanine. The reported amino acid composition generally resemble, but are distinct from, those of other characterized carcinoma-associated glycoproteins such as epitectin from the human laryngeal carcinoma (Bhavanandan, 1988), MUC 1 (episialin) mouse tumor-associated mucin (Spicer et al., 1991), ASGP-1 of the 13 762 rat mammary adenocarcinoma cell line (Carraway and Spielman, 1986), the pancreatic tumor mucin from the human pancreatic tumor cell line HPAF (Lan et
al., 1990) or TAG-72 from the human colon carcinoma xenograft, LS-174T (Sheer et al., 1988). The reason for these differences may be different genes coding the mucins, or the differences may be a consequence of different isolation procedures. Epiglycanin is highly glycosylated, and the disaccharide Galb(1→3)GalNac linked to serine or threonine, has been shown to form an integral part of the epitopes for anti-epiglycanin antibodies (Haavik et al., 1992). The HCA from KLE-1 and HCA from ascitic fluid are isolated using anti-epiglycanin antibodies for
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129
128
Table 1 Amino acid composition of HCA in KLE-1 and ascitic fluid determined by analysis after blotting onto PVDF membrane (number of residues / 1000 residues): epiglycanin is included for comparison Amino acid
EPGN
KLE-1
Ascitic fluid
Serine Threonine Gln1Glu Asn1Asp Leucine Alanine Glycine Valine Proline Lysine Isoleucine Arginine Phenylalanine Tyrosine Methionine Histidine Cysteine
230 153 101 49 32 101 119 39 52 27 16 29 13 8 6 26 —
130 71 140 74 66 89 155 51 80 20 29 35 22 20 5 13 —
127 126 118 81 63 70 119 51 85 39 31 28 26 15 9 12 —
detection. The amino acid composition of the core proteins of these mucins differ from epiglycanin. The reason why the anti-epiglycanin antibodies cross-react with HCA may be identical areas in the core protein, and areas of identical glycosylation. The KLE-1 cell line has been shown to express the tumor-associated antigen, TAG-72 (Horan Hand et al., 1992) which is defined by the monoclonal antibody B72.3 (Sheer et al., 1988). In enzyme competitive binding assay, the isolated HCA did not bind the B72.3 antibody. This further indicates that HCA is different from TAG 72, but positive control was not included in the test to confirm the results, since TAG-72 was not available. In preliminary SDS–PAGE experiments it was shown that the anti-epiglycanin antibodies AE-3 and G-1 did not bind episialin. The anti-episialin antibody 139H 2 (Hilkens and Buijs, 1988), which is directed towards the repeat part of episialin, did not bind epiglycanin or HCA from KLE-1 and ascitic fluid. This also indicates that HCA is different from episialin. These comparison studies are in progress and will be presented. Although the HCA from KLE-I, the HCA from human ascitic fluid, and other characterized carcinoma-associated glycoproteins exhibit distinct differences, their amino acid compositions demonstrate that they all may be classified as mucins. There is still limited knowledge about the detailed glycoprotein structures of the mucins expressed on carcinomas. Further characterization of HCA from KLE-1 and ascitic fluid, and study of the antigen from other carcinoma sources, are in progress.
Acknowledgements We gratefully acknowledge Dr. David T. MacLaughlin and Linda P. Merk, Department of Gynecology, Massachusetts General Hospital, Boston, MA, USA, who provided
human carcinoma cells and medium from these cells and Dr. N. Nikrui, Department of Gynecology, Massachusetts General Hospital, Boston, MA, USA, for providing ascitic fluid collected from patients with ovarian cancer. This research was supported in part by grants from the Norwegian Research Council for Science and the Humanities (311.91 / 022, 100593 / 410, 100670 / 410) and a grant from Epigen Inc.
References Bhavanandan, V.P., 1988. Malignancy-related cell surface mucin-type glycoproteins. Ind. J. Biochem. Biophys. 25, 36–42. Carraway, K.L., Spielman, J., 1986. Structural and functional aspects of tumor cell sialomucins. Mol. Cell. Biochem. 72, 109–120. Codington, J.F., Haavik, S., 1992. Epiglycanin—a carcinoma-specific mucin-type glycoprotein of the mouse TA3 tumour. Glycobiology 2, 173–180. Codington, J.F., Sanford, B.H., Jeanloz, R.W., 1972. Glycoprotein coat of the TA3 cell. Isolation and partial characterization of a sialic acid containing glycoprotein fraction. Biochemistry 11, 2559–2564. Codington, J.F., Cooper, A.G., Miller, D.K., Slayter, H.S., Brown, M.C., Silber, C., Jeanloz, R.W., 1979. Isolation and partial characterization of an epiglycanin-like glycoprotein from a new non-strain-specific subline of TA3 murine mammary adenocarcinoma. J. Natl. Cancer Inst. 63, 153–161. Codington, J.F., Das, H.R., Dalianis, T., Klein, G., Miller, S.C., Silber, C., Lampert, L.A., Darby, D.M., Jeanloz, R.W., 1983. Cell surface characteristics of an allotransplantable TA3 ascites subline resulting from a process of immunoselection. Cancer Res. 43, 4373–4381. Codington, J.F., Bhavanandan, V.P., Block, K., Nikrui, N., Ellard, J.V., Wang, P.S., Jeanloz, R.W., 1984. Antibody to epiglycanin and radioimmunoassay to detect epiglycanin-related glycoproteins in body fluids of cancer patients. J. Natl. Cancer Inst. 73, 1029–1038. Codington, J.F., Haavik, S., Hui, Z., Matson, S., Wu, Z., Chen, X., 1993. Immunoassays for determining the concentration of a carcinomaspecific antigen in the blood of cancer. Glycoconjugate J. 10, 319. Cooper, A.G., Codington, J.F., Brown, M.C., 1974. In vivo release of glycoprotein I from the Ha subline of TA3 murine tumor into ascites fluid and serum. Proc. Natl. Acad. Sci. USA 71, 1224–1228. Cooper, A.G., Codington, J.F., Miller, D.K., Brown, M.C., 1979. Loss of strain specificity of the TA3-St subline: evidence for the role of epiglycanin in mouse allogeneic tumor growth. J. Natl. Cancer Inst. 63, 163–169. Dahiya, R., Kwak, K.S., Byrd, J.C., Ho, S., Yoon, W.H., Kim, Y.S., 1993. Mucin synthesis and secretion in various human epithelial cancer cell lines that express the MUC-1 mucin gene. Cancer Res. 53, 1437– 1443. Decaens, C., Nardelli, J., Bara, J., Burtin, P., 1993. Biochemical characterization of a rat oncofetal colonic antigen defined by a monoclonal antibody raised against gastric surface epithelium. Biochem. J. 293, 531–536. Devaraj, N., Devaraj, H., Bhavanandan, V.P., 1992. Purification of mucin glycoproteins by density gradient centrifugation in cesium trifluoroacetate. Anal. Biochem. 206, 142–146. Dubois, M., Gilles, K.A., Hamilton, J.K., Rebers, P.A., Smith, F., 1956. Colorimetric method for determination of sugars and related substances. Anal. Chem. 28, 350–356. Fung, P.Y.S., Longenencker, M.L., 1991. Specific immunosuppressive activity of epiglycanin, a mucin-like glycoprotein secreted by a murine mammary adenocarcinoma (TA3-HA). Cancer Res. 51, 1170–1176. Galfre, G., Milstein, C., 1981. Preparation of monoclonal antibodies: Strategies and procedures. Methods Enzymol. 73 (Part B), 3–46.
T. Thingstad et al. / European Journal of Pharmaceutical Sciences 6 (1998) 121 – 129 Haavik, S., Codington, J.F., Davison, P.F., 1992. Development and characterization of monoclonal antibodies against a mucin type glycoprotein. Glycobiology 2, 217–224. Hilkens, J., Buijs, F., 1988. Biosynthesis of MAM-6, an epithelial sialomucin. J. Biol. Chem. 263, 4215–4222. Hilkens, J., 1988. Biochemistry and function of mucins in malignant disease. Cancer Rev. 11 / 12, 25–54. Ho, J.J.L., Bi, N., Siddiki, B., Chung, Y.S., Yuan, M., Kim, Y.S., 1993. Multiple forms of intracellular and secreted mucins in a pancreatic cancer cell line. Cancer Res. 53, 884–890. Horan Hand, P., Calvo, B., Milenic, D., Yokota, T., Finch, M., Snoy, P., Garmestani, K., Gansow, O., Schlom, J., Kashmiri, S.V.S., 1992. Comparative biological properties of a recombinant chimeric anticarcinoma mAb and a recombinant aglycosylated variant. Cancer Immunol. Immunother. 35, 165–174. Kemperman, H., Wijnands, Y., Wesseling, J., Niessen, C.M., Sonnenberg, A., Roos, E., 1994. The mucin epiglycanin on TA3 / Ha carcinoma cells prevents a(6)ß(4)-mediated adhesion to laminin and kalinin and E-cadherin-mediated cell-cell interaction. J. Cell Biol. 127, 2071– 2080. Laemmli, U.K., 1970. Clevage of structural proteins during the assembly of the head of bacteriophage T 4. Nature 227, 680–685. Lan, M.S., Finn, O.J., Fernsten, P.D., Metzgar, R.S., 1985. Isolation and properties of a human pancreatic adenocarcinoma-associated antigen, DU-PAN-2. Cancer Res. 45, 305–310. Lan, M.S., Batra, S.K., Qi, W.-N., Metzgar, R.S., Hollingsworth, M.A., 1990. Cloning and sequencing of a human pancreatic tumor mucin cDNA. J. Biol. Chem. 265, 15294–15299. Nishida, K., Ueda, R., Sakakibara, K., Minoura, Y., Ohta, M., Ariyoshi, Y., Takahashi, T., 1993. Characterization of OM-B Monoclonal antibody-defined antigen associated with mucinous type human ovarian tumor. Jpn. J. Cancer Res. 84, 65–74.
129
Pancino, G., Osinaga, E., Charpin, C., Mistro, D., Barque, J.Ph., Roseto, A., 1991. Purification and characterization of a breast-cancer-assosiated glycoprotein not expressed in normal breast and identified by monoclonal antibody 83D4. Br. J. Cancer 63, 390–398. Richardson, G.S., Dickersin, R., Atkins, L., MacLaughlin, D.T., Raam, S., Merk, L.P., Bradley, F.M., 1984. KLE: A cell line with defective estrogen receptor derived from undifferentiated endometrial cancer. Gynecol. Oncol. 17, 213–230. Sheer, D.G., Schlom, J., Cooper, H.L., 1988. Purification of the human tumor-associated glycoprotein (TAG-72) defined by monoclonal antibodies CC49 and B72.3. Cancer Res. 48, 6811–6818. Spicer, A.P., Parry, G., Patton, S., Gendler, S.J., 1991. Molecular cloning and analysis of the mouse homologue of the tumor-associated mucin, MUC1, reveals conservation of potential o-glycosylation sites, transmembrane, and cytoplasmic domains and a loss of minisatellite-like polymorphism. J. Biol. Chem. 266, 15099–15109. Springer, G.F., 1984. T and Tn, general carcinoma autoantigens. Science. 224, 1198–1206. Van den Eijnden, D.H., Evans, N.A., Codington, J.F., Reinhold, V., Silber, C., Jeanloz, R.W., 1979. Chemical structure of epiglycanin, the major glycoprotein of the TA3-Ha ascites cell. J. Biol. Chem. 254, 12153– 12159. Warren, L., Buck, C.A., Tuszynski, G.P., 1978. Glycopeptide changes and malignant transformation a possible role for carbohydrate in malignant behavior. Biochim. Biophys. Acta 516, 97–127. Wu, K., Fregien, N., Carraway, K.L., 1994. Molecular cloning and sequencing of the mucin subunit of a heterodimeric, bifunctional cell surface glycoprotein complex of ascites rat mammary adenocarcinoma cells. J. Biol. Chem. 269, 11950–11955.