MUC1 has generated considerable interest as a tumor marker and potential target for tumor killing. To date, most antibodies against MUC1 recognize epitopes within the highly immunogenic α chain tandem repeat array. A major shortcoming of such antibodies is that the MUC1 α chain is shed into the peripheral circulation, sequesters circulating antitandem repeat array antibodies, and limits their ability to even reach targeted MUC1-expressing cells. Antibodies recognizing MUC1 epitopes tethered to the cell surface would likely be more effective. MUC1 α subunit binding the membrane-tethered β subunit provides such an epitope. By use of a novel protocol entailing immunization with cDNA encoding full-length MUC1 (MUC1/TM) followed by boosting with the alternatively spliced MUC1/X isoform from which the tandem repeat array has been deleted, we generated monoclonal antibodies, designated DMC209, which specifically bind the MUC1 α/β junction. DMC209 is exquisitely unique for this site; amino acid mutations, which abrogate MUC1 cleavage, also abrogate DMC209 binding. Additionally, DMC209 specifically binds the MUC1 α/β junction on full-length MUC1/TM expressed by breast and ovarian cancer cell lines and on freshly obtained, unmanipulated MUC1-positive malignant plasma cells of multiple myeloma. DMC209 is likely to have clinical application by targeting MUC1-expressing cells directly and as an immunotoxin conjugate. Moreover, the novel immunization procedure used in generating DMC209 can be used to generate additional anti-MUC1 α/β junction antibodies, which may, analogously to Herceptin, have cytotoxic activity. Lastly, sequential immunization with MUC1/TM cDNA acting as a nonspecific adjuvant followed by protein of interest may prove to be a generalizable method to yield high-titer specific antibodies. (Cancer Res 2006; 66(23): 11247-53)

MUC1 is a glycoprotein highly expressed in several human epithelial malignancies, including breast, prostate, ovarian, and pancreatic carcinomas, as well as on the malignant plasma cells of multiple myeloma (16). Although alternative splicing can generate a variety of MUC1 isoforms (713), the most intensively studied MUC1 protein is a type I transmembrane protein (MUC1/TM) composed of a heavily glycosylated extracellular domain containing a tandem repeat array, a transmembrane domain, and a cytoplasmic domain (9, 14, 15). MUC1/TM is proteolytically cleaved soon after its synthesis, generating two subunits, α and β, which specifically recognize and bind each other in a strong noncovalent interaction (Fig. 1, MUC1/TM; refs. 9, 1619). Cleavage of MUC1 into the two subunits occurs in the SEA module (9, 18, 19), a highly conserved domain found in several cell-tethered mucin-like proteins (20). Shedding of α subunit from the cell membrane results in soluble tandem repeat–containing MUC1 in the peripheral circulation, and it is this molecule that is used to determine serum MUC1 levels in patients (21, 22).

Figure 1.

MUC1 protein isoforms and MUC1 fusion proteins. Left, cleaved transmembrane MUC1 protein containing the tandem repeat array (MUC1/TM). Alternative splicing uses a splice donor indicated by S.D.→ resulting in one of two alternative splice acceptors (S.A. Y→ and S.A. X→) resulting in the MUC1/X and MUC1/Y (MUC1/X Δ1-18) isoforms. The numbers (1, 18, and 63) refer to the location of amino acids in the extracellular domain of MUC1/X. The cleavage site in MUC1/X occurs between amino acids 62 and 63. Isoform MUC1/Y does not contain amino acids 1 to 18 (see also schematics in Fig. 3). As described in the text, antibody DMC209 recognizes the cleaved α/β junction present in both the MUC1/X and MUC1/TM proteins but does not react with the uncleaved MUC1/Y protein.

Figure 1.

MUC1 protein isoforms and MUC1 fusion proteins. Left, cleaved transmembrane MUC1 protein containing the tandem repeat array (MUC1/TM). Alternative splicing uses a splice donor indicated by S.D.→ resulting in one of two alternative splice acceptors (S.A. Y→ and S.A. X→) resulting in the MUC1/X and MUC1/Y (MUC1/X Δ1-18) isoforms. The numbers (1, 18, and 63) refer to the location of amino acids in the extracellular domain of MUC1/X. The cleavage site in MUC1/X occurs between amino acids 62 and 63. Isoform MUC1/Y does not contain amino acids 1 to 18 (see also schematics in Fig. 3). As described in the text, antibody DMC209 recognizes the cleaved α/β junction present in both the MUC1/X and MUC1/TM proteins but does not react with the uncleaved MUC1/Y protein.

Close modal

The presence of the soluble α subunit MUC1 protein in the circulation presents a singular difficulty in delivering adequate amounts of anti-MUC1 antibodies to directly target MUC1-expressing malignant cells. Because the most immunogenic part of MUC1 is the tandem repeat array, all anti-MUC1 antibodies generated to date almost exclusively recognize epitopes in that immunogenic region. Sequestration of antitandem repeat antibodies by the soluble, circulating MUC1 α subunit severely limits the amount of antibody that can successfully bind MUC1 on the cell surface. Furthermore, deposition of immune complexes of antitandem repeat antibodies and its soluble, circulating MUC1 target can lead to significant end-organ damage.

In recent years, numerous efforts have been made to generate effective anti-MUC1 antibodies using the full-length MUC1/TM molecule as immunogen (2325). The major obstacle hampering those attempts is that immunization with the whole MUC1/TM molecule invariably results in an antibody response composed almost in its entirety of antibodies recognizing epitopes on the highly immunogenic tandem repeat array. For ultimate application in in vivo targeting of MUC1-expressing tumor cells, such antibodies pose all the shortcomings inherent in antirepeat antibodies as detailed above.

Antibodies recognizing MUC1 epitopes tethered to the cell surface potentially obviate these difficulties. Although conceptually simple, generation of monoclonal antibodies (mAb) to tethered MUC1 first requires characterization of cell-bound, nonshedding epitopes. The junction formed by the MUC1 α subunit binding the membrane-tethered β subunit provides such an epitope.

We recently investigated the mechanism whereby the cleaved junction composed of the MUC1 α and β subunits is formed (9). In the course of those studies, we analyzed the ‘cleavageability’ of the MUC1/TM, MUC1/Y, and MUC1/X proteins (Fig. 1; ref. 9). The MUC1/Y and MUC1/X isoforms are generated from mRNAs spliced at two distinct sites that use donor and acceptor sites located upstream and downstream to the tandem repeat array (7, 12), and consequently, in both isoforms the tandem repeat array and flanking sequences are spliced out (Fig. 1). The extracellular domains both of MUC1/X and MUC1/Y are thus considerably less complex than the large tandem repeat array–containing MUC1/TM protein. In fact, the extracellular domain of the MUC1/X protein comprises only the 120-amino acid SEA module fused to the MUC1 30 NH2-terminal amino acids (9). MUC1/Y is identical to MUC1/X, except for an 18-amino acid deletion at the SEA module N terminus (9). Significantly, the MUC1/X isoform is cleaved at an identical site as the full-length MUC1/TM protein and thereby results in the same noncovalent interaction of the α and β subunits (9). In contrast, the SEA module NH2-terminal truncation present in the MUC1/Y isoform results in a noncleaved protein (9).

To generate antibodies directly targeting cancer cells, we surmised that antibodies specific for the MUC1 α/β junction would target only membrane-bound MUC1. To circumvent generation of antitandem repeat array antibodies, we used the cleaved MUC1/X protein (7, 9) both as immunogen and as screening reagent. We report here a novel procedure in which mice were initially primed with MUC1/TM DNA to elicit an anti-MUC1/TM response. The resultant immune response contains antibodies reactive with the MUC1/X isoform. To further increase anti-MUC1/X titers, mice were boosted with MUC1/X protein. This resulted in exceptionally high anti-MUC1/X titers. Using this protocol, we were successful in generating antibodies that recognize cleaved MUC1 α/β junction on the cell surface and bind malignant cells expressing the full-length MUC1/TM.

Materials and antibodies. Reagents and chemicals were obtained from Sigma (St. Louis, MO), unless otherwise specified. The anti-MUC1 tandem repeat antibodies (anti-epithelial membrane antigen, Mc5) were obtained from Chemicon International (Temecula, CA).

Cell culture. Cells were grown at 37°C and 5% CO2 in culture medium supplemented with 10% heat-inactivated FCS, 2 mmol/L l-glutamine, 100 IU/mL penicillin, and 25 μg/mL streptomycin. DA3 mouse mammary tumor cells and HK293, human embryonic kidney cells, were grown in DMEM.

Generation of stable DA3 mouse mammary tumor cell transfectants expressing MUC1/TM. DA3 cells were cotransfected with the eukaryotic expression plasmids pCL-MUC1/TM or pCL-MUC/TM truncated at the cytoplasmic PvuII site and pSV2neo (coding for neomycin resistance). Expression constructs were transfected into cells using the calcium phosphate procedure. Stable transfectants were selected with neomycin. Transfectants expressing the MUC1 proteins were identified by immunoblot analysis of cell lysates using affinity-purified polyclonal antibodies directed against the MUC1 cytoplasmic domain.

Generation of MUC1/X and MUC1/Y eukaryotic expression constructs and fusion proteins. The MUC1 fusion proteins used in this study, designated Flag-Yex-hFc and Flag-Xex-hFc, are depicted in Fig. 1. Standard molecular biology methods were used to generate all constructs.

Generation of HK293 transfectants expressing Flag-MUC1/Xex-hFc and Flag-MUC1/Yex-hFc fusion proteins and purification of hFc-tagged fusion proteins. HK293 (human kidney) cells were transiently transfected using calcium phosphate with the eukaryotic pCMV3 expression vectors coding for the Flag-Xex-hFc, Flag-Yex-hFc, or mutant MUC1/X proteins (6 μg DNA/25 cm2 flask). To obtain stable transfectants, neomycin-resistant clones were isolated following addition of neomycin to the culture medium. Conditioned media containing the secreted MUC1 fusion proteins were spun at 15,000 rpm for 20 minutes, and the supernatant was filtered through 0.45-μm filter and stored at −75°C. Protein A-Sepharose 4 Fast Flow (Amersham Pharmacia Biotech, Chalfont St. Giles, United Kingdom) was used to purify the COOH-terminally hFc-tagged Flag-Xex(Yex)-hFc proteins.

Immunization of mice and hybridoma production. Mice were immunized with consecutive i.d. DNA immunizations spaced at 21-day intervals. The injected DNAs consisted of either the pCL-MUC1/TM or pCL-MUC1/X expression vector plasmids. Subsequently, mice were boosted with the extracellular domain of the MUC1/X protein injected with incomplete Freund's adjuvant. Hybridomas were prepared by fusion of the myeloma cell line with immune splenocytes and screened by ELISA assay (see above).

ELISA assay for determining binding of anti-MUC1 antibodies to the MUC1/X and MUC1/Y proteins and to the MUC1 α/β junction. ElisaImmunoAssay plates (Costar, Corning, NY) were coated with polyclonal goat anti-human Fc (Gα-hFc, 4 μg/mL) followed by washing with PBS-Tween 20 (0.05%) and blocking with PBS-Tween 20 plus 5% skimmed milk (Blotto). Spent culture medium containing Flag-Xex(or Yex)-hFc proteins (mutant and wild-type, see ref. 9 for detailed descriptions of these proteins and their mutants) was then applied to the wells to allow binding of the Flag-Xex(Yex)-hFc proteins. Following incubation, samples were removed and the wells were washed with PBS-Tween 20. Wells were then incubated with the mouse mAbs (or hybridoma supernatants) followed by horseradish peroxidase (HRP)-conjugated anti-mouse antibody.

Immunization with MUC1/TM cDNA but not with MUC1/X cDNA induces anti-MUC1 α/β junction antibodies. Mice were immunized either with cDNA expression vectors coding for the MUC1/TM protein (containing the α chain tandem repeat array) or with cDNA coding for the MUC1/X isoform from which the tandem repeat is deleted (Fig. 1). Following four consecutive DNA immunizations, sera were assayed for antibody directed against MUC1/X. Because MUC1/X is cleaved at an identical site as the full-length MUC1/TM molecule, it served as the screening protein to identify anti-MUC1 α/β junction antibodies. This provided several advantages. Because the MUC1/X extracellular domain lacks the central tandem repeat array, antibodies directed against the highly immunogenic tandem repeat array epitopes are not detected. More importantly, as MUC1/X is composed solely of two small interacting MUC1/X α and β subunits that form the native α/β junction, only antibodies directed against epitopes common to both MUC1/X and MUC1/TM are detected. The MUC1 α/β junction is one such major epitope, making this screening procedure a simple detection system for antibodies binding this critical region of interest.

Unexpectedly, all mice (five of five) immunized with the MUC1/X cDNA failed to raise significant anti-MUC1/X antibody titers (Fig. 2A and B). In contrast, five of five mice immunized with MUC1/TM cDNA generated highly reproducible, albeit modest, anti-MUC1/X antibody titers (Fig. 2A and B). These results indicated that, whereas immunization with MUC1/TM cDNA induces antibodies recognizing epitopes common to both the MUC1/X and MUC1/TM proteins, immunization with MUC1/X cDNA failed to elicit anti-MUC1/X antibodies.

Figure 2.

A and B, anti-MUC1/X antibodies generated in mice immunized with DNA coding for either MUC1/TM or MUC1/X. Mice were immunized either with expression vectors containing cDNA coding for the MUC1/TM protein (MUC1/TM, group II) or with cDNA encoding for MUC1/X from which the tandem repeat sequence is deleted (MUC1/X, group I). Sera were assayed for antibody titers directed against the cleaved MUC1/X protein as described in Materials and Methods. All mice immunized with MUC1/TM cDNA showed significant anti-MUC1/X antibody titers, whereas no anti-MUC1/X antibodies were detected in mice immunized with MUC1/X cDNA. A, antibody titers from mice immunized with MUC1/TM cDNA (black lines) and from mice immunized with MUC1/X cDNA (red lines; k = 1,000). B, anti-MUC1/X immunoreactivity of sera from groups I and II mice at a 1:100 dilution. C, anti-MUC1/X titers in mice primed with MUC1/TM DNA followed by a single MUC1/X protein boost. Mice previously primed with MUC1/TM cDNA (A and B) were boosted with a single MUC1/X protein immunization. Sera were assayed for anti-MUC1/X reactivity as described in Materials and Methods 18 or 32 days following the protein boost (violet and green lines, respectively). Bars, SD for titers from individual mice.

Figure 2.

A and B, anti-MUC1/X antibodies generated in mice immunized with DNA coding for either MUC1/TM or MUC1/X. Mice were immunized either with expression vectors containing cDNA coding for the MUC1/TM protein (MUC1/TM, group II) or with cDNA encoding for MUC1/X from which the tandem repeat sequence is deleted (MUC1/X, group I). Sera were assayed for antibody titers directed against the cleaved MUC1/X protein as described in Materials and Methods. All mice immunized with MUC1/TM cDNA showed significant anti-MUC1/X antibody titers, whereas no anti-MUC1/X antibodies were detected in mice immunized with MUC1/X cDNA. A, antibody titers from mice immunized with MUC1/TM cDNA (black lines) and from mice immunized with MUC1/X cDNA (red lines; k = 1,000). B, anti-MUC1/X immunoreactivity of sera from groups I and II mice at a 1:100 dilution. C, anti-MUC1/X titers in mice primed with MUC1/TM DNA followed by a single MUC1/X protein boost. Mice previously primed with MUC1/TM cDNA (A and B) were boosted with a single MUC1/X protein immunization. Sera were assayed for anti-MUC1/X reactivity as described in Materials and Methods 18 or 32 days following the protein boost (violet and green lines, respectively). Bars, SD for titers from individual mice.

Close modal

As a major objective of this study was to generate mAbs recognizing the α/β junction, mice with much higher anti-MUC1/X titers were required. To generate high-titer anti-MUC1/X, we boosted the MUC1/TM cDNA-primed mice with MUC1/X protein. Following a single MUC1/X protein boost, all immunized mice showed a 100-fold increase in anti-MUC1/X titers, from 1:300 (after the MUC1/TM cDNA immunization alone) to a minimum of 1:30,000 following the MUC1/X protein boost (Fig. 2C). High titers were observed 18 days following the MUC1/X protein boost, which increased even higher at day 32 (Fig. 2C). Mice sustained high anti-MUC1/X protein titers up to 6 months following the initial single MUC1/X protein boost (data not shown).

Generation of DMC209 mAbs that recognize the cleaved α/β junction. To best generate anti-α/β junction-specific mAbs, we proceeded to hybridoma formation using spleen cells from mice showing the highest anti-MUC1/X antibody titers. Spent hybridoma media from 10 96-well plates containing fused spleen-myeloma cells were assayed for antibodies recognizing the cleaved MUC1 α/β junction. To identify such antibodies, hybridoma supernatants were screened in parallel against MUC1/X protein and the MUC1/Y isoform (Fig. 1). Whereas MUC1/X is cleaved at an identical site as in the full-length MUC1/TM protein resulting in interacting α and β subunits, MUC1/Y, which does not cleave, does not yield the MUC1 α/β junction epitope. We selected for hybridomas that were MUC1/X positive and MUC1/Y negative.

Screening 950 hybridoma supernatants yielded two hybridomas highly reactive with MUC1/X protein but nonreactive with MUC1/Y. Limiting cell dilution resulted in pure clones, which were designated DMC209, an IgM antibody, and DMC111, an Igγ2a antibody.

Detailed epitope analysis of DMC209 and DMC111 reactivity. Analysis of the specificity of the mAbs generated by clones DMC209 and DMC111 confirmed that both were MUC1/X positive and MUC1/Y negative (Fig. 3A and B). Furthermore, DMC209 and DMC111 were also nonreactive with the noncleaved MUC1/Xex Δ1-11 protein (Fig. 3F) but reacted with cleaved Δ1-7 MUC1/Xex protein (Fig. 3E). MUC1/Xex refers to the extracellular domain of the MUC1 protein, whereas mutants of the MUC1/Xex proteins containing deletions of amino acids 1 to 7 or 1 to 11 are designated by Δ1-7 and Δ1-11 (amino acid 1 is indicated in Fig. 1 next to the MUC1/X protein; also see Fig. 3). To further assess the importance of MUC1 cleavage, the immunoreactivity of DMC111 and DMC209 was tested against MUC1/Xex mutants that harbor a mutation at the critical Ser63 residue (S63; see Fig. 3). MUC1 cleavage occurs directly upstream to this serine residue appearing within the sequence GSVVV, thereby generating an NH2-terminal α subunit terminating at the G (glycine) residue and a β subunit initiating with the S63 residue. We previously found that only S63→C and S63→T mutant proteins, in addition to MUC1/X, generated cleaved MUC1 proteins, whereas all other mutations at this residue resulted in noncleaved proteins (9). These analyses showed that, whereas DMC209 and DMC111 bound equally well to the wild-type MUC1/X protein and to the cleaved S63→C and S63→T mutant proteins (Fig. 3C), all other S63 mutant uncleaved MUC1/X proteins (Fig. 3D) were completely nonreactive with DMC209. In contrast, DMC111 reacted equally well with both the cleaved MUC1/X proteins (wild-type and cleaved S63→C and S63→T mutant proteins; Fig. 3C) and the uncleaved S63 mutant proteins (Fig. 3D). These results confirm that the epitopes recognized by the two mAbs are different. More importantly, only DMC209 is completely dependent on MUC1 cleavage for its reactivity. Significantly, DMC209 was nonreactive with both the MUC1 α and β subunits when each was tested in isolation (Fig. 3G and H); DMC209 reacts with the two subunits only when presented as interacting entities. These analyses provided strong evidence that DMC209 is cleavage dependent and that it recognizes the junction of the interacting α and β subunits. Further analyses (described below and data not shown) revealed robust binding of DMC209 to cells expressing MUC1/TM, the major MUC1 isoform, and significantly less binding with the noncleavage-dependent DMC111 antibody. Henceforth, our reported analyses are restricted to the cleavage-dependent antibody DMC209.

Figure 3.

Detailed analysis of the epitopes recognized by DMC209 and DMC111 mAbs. Wild-type MUC1/Xex and MUC1/Yex proteins (A and B; “ex” designates the extracellular domains of the MUC1 proteins, see Fig. 1) as well as point-mutated MUC1/Xex (C and D) and internally deleted MUC1/Xex (E and F) were assayed for their reactivity with the DMC209 and DMC111 antibodies. The extracellular domains of the MUC1/X and MUC1/Y proteins (MUC1/Xex and MUC1/Yex) were purified as secreted proteins comprising Flag and hFc epitopes at their NH2 terminus and COOH terminus, respectively. ElisaImmunoAssay plates were coated with polyclonal goat anti-human Fc (Gα-hFc). Flag-Xex(or Yex)-hFc proteins (mutant and wild-type; see ref. 9 for detailed descriptions of these proteins and their mutants) were then applied to the wells to allow binding. The location of the numbered amino acids is indicated in Fig. 1, MUC1/X protein. Wells were incubated with the mouse mAbs DMC111 or DMC209 followed by HRP-conjugated anti-mouse antibody to detect bound antibody. +, binding of antibodies DMC111 and DMC209; −, lack of binding; nr, not relevant.

Figure 3.

Detailed analysis of the epitopes recognized by DMC209 and DMC111 mAbs. Wild-type MUC1/Xex and MUC1/Yex proteins (A and B; “ex” designates the extracellular domains of the MUC1 proteins, see Fig. 1) as well as point-mutated MUC1/Xex (C and D) and internally deleted MUC1/Xex (E and F) were assayed for their reactivity with the DMC209 and DMC111 antibodies. The extracellular domains of the MUC1/X and MUC1/Y proteins (MUC1/Xex and MUC1/Yex) were purified as secreted proteins comprising Flag and hFc epitopes at their NH2 terminus and COOH terminus, respectively. ElisaImmunoAssay plates were coated with polyclonal goat anti-human Fc (Gα-hFc). Flag-Xex(or Yex)-hFc proteins (mutant and wild-type; see ref. 9 for detailed descriptions of these proteins and their mutants) were then applied to the wells to allow binding. The location of the numbered amino acids is indicated in Fig. 1, MUC1/X protein. Wells were incubated with the mouse mAbs DMC111 or DMC209 followed by HRP-conjugated anti-mouse antibody to detect bound antibody. +, binding of antibodies DMC111 and DMC209; −, lack of binding; nr, not relevant.

Close modal

Binding of DMC209 to MUC1/TM on intact cells. DMC209 recognizes the cleaved α/β junction located in the MUC1/X protein. As this cleavage site is identical to that found in the full-length MUC1/TM protein (9), it was of critical importance to determine whether DMC209 binds to the MUC1/TM protein as displayed on intact cells. We approached this issue by first examining whether mouse cells not expressing human MUC1/TM become reactive with DMC209 following transfection with human MUC1/TM. Nontransfected parental mouse mammary tumor cells were found to be completely nonreactive with DMC209 (Fig. 4I,, A), although DMC209 clearly bound to the same cells expressing either the full-length MUC1/TM protein (Fig. 4B) or an isoform lacking the cytoplasmic domain of MUC1/TM (Fig. 4I,, C). To confirm that MUC1/TM is expressed on these cells, antibody to the α chain tandem repeat array showed, as expected, MUC1 expression on the transfected cells (Fig. 4I I, B).

Figure 4.

DMC209 flow cytometric analyses of MUC1/TM-expressing cells. Parental nontransfected mouse mammary tumor cells and cells transfected with cDNA coding for full-length MUC1/TM or MUC1/TM with its cytoplasmic tail truncated (DA3/PAR, DA3/TM, and DA3/TM minus CYT, respectively) were reacted with DMC209 and analyzed by flow cytometry as described in Materials and Methods. Similarly, the ovarian cancer cell line HEY and breast cancer cell lines T47D and MCF-7 were analyzed following incubation with DMC209. Red and purple tracings, reaction of cells with the fluorescently labeled secondary antibody alone or with primary DMC209 antibody followed by secondary antibody, respectively. II, green line, reactivity with an anti-MUC1 tandem repeat mouse mAb.

Figure 4.

DMC209 flow cytometric analyses of MUC1/TM-expressing cells. Parental nontransfected mouse mammary tumor cells and cells transfected with cDNA coding for full-length MUC1/TM or MUC1/TM with its cytoplasmic tail truncated (DA3/PAR, DA3/TM, and DA3/TM minus CYT, respectively) were reacted with DMC209 and analyzed by flow cytometry as described in Materials and Methods. Similarly, the ovarian cancer cell line HEY and breast cancer cell lines T47D and MCF-7 were analyzed following incubation with DMC209. Red and purple tracings, reaction of cells with the fluorescently labeled secondary antibody alone or with primary DMC209 antibody followed by secondary antibody, respectively. II, green line, reactivity with an anti-MUC1 tandem repeat mouse mAb.

Close modal

Having established that the DMC209 mAbs bound MUC1/TM in transfected cells, we proceeded to confirm that the antibodies also bind tumor cell lines known to express MUC1/TM. HEY (an ovarian cancer cell line) and the breast cancer cell lines T47D and MCF-7 all showed significant DMC209 binding (Fig. 4I , D-F).

DMC209 binds MUC1-expressing multiple myeloma cells in fresh bone marrow aspirates. For ultimate clinical application, antibodies to the MUC1 α/β junction must be able to target MUC1-expressing malignant cells. To investigate whether DMC209 antibodies can selectively bind tumor cells in an in vivo–like setting, malignant plasma cells in freshly obtained bone marrow aspirates from patients with multiple myeloma were used. The heterogeneous cell population in the aspirates allows direct assessment of DMC209-binding specificity: DMC209 should bind multiple myeloma cells, which express MUC1, whereas not bind, or minimally bind, non-MUC1-expressing cells present in the same aspirate sample. To do these analyses, antibodies DMC209, CD38, CD45, and CD138 were labeled with different fluorophores, and the cell aspirate was analyzed by flow cytometry. DMC209 cell binding in the aspirate material can thereby be readily compared with simultaneous staining with anti-syndecan (CD138) and CD38 both well-recognized myeloma markers (26). As expected (27), DMC209-positive myeloma cells were CD45 negative (data not shown; ref. 27).

Plotted side scatter (reflecting cell granularity) versus red fluorescently labeled DMC209-positive cells assessed the composition of the mixed cell populations present in the bone marrow aspirate (Fig. 5, left). The discrete population of cells with significant DMC209 reactivity was gated and designated R1 (red dots within the R1 gate, Fig. 5, left). The location of the DMC209-positive cells was identified in the CD38/CD138 scatter plot (Fig. 5, right, red dots). Results with freshly obtained bone marrow aspirates from three multiple myeloma patients revealed DMC209 immunoreactivity in 26% of CD138+ cells in one aspirate and in 37% in another, whereas expression was low (5%) in the third aspirate (patients I, II, and III, respectively; Fig. 5). DMC209 seems to identify a subpopulation of malignant plasma cell population in patients I and III. In these samples, a mixture of DMC209-positive (Fig. 5, right, red dots) and DMC209-negative (Fig. 5, right, black dots) cells is seen in the CD38/CD138-positive population. In addition, two samples that were negative for anti-MUC1 tandem repeat array mAb H23 were also DMC209 negative (data not shown). DMC209 binding to MUC1-expressing malignant plasma cells and lack of binding to non-MUC1-expressing cell lineages in the same freshly obtained, unmanipulated aspirate sample underscore the specificity of MUC1 α/β junction binding in distinguishing MUC1-expressing tumor cells.

Figure 5.

Flow cytometry analysis of DMC209 reactivity with freshly obtained bone marrow cells from patients with multiple myeloma. Bone marrow aspirates from three separate patients (patients I, II, and III) were simultaneously analyzed with DMC209 and with antibodies to CD138 and CD38, each labeled with fluorochromes fluorescing at different emission wavelengths. Left, side scatter analyses were done against DMC209. DMC209-positive cells were gated and designated R1 (left, red dots), and their location in the CD38/CD138 scatter plot was identified (right, red dots).

Figure 5.

Flow cytometry analysis of DMC209 reactivity with freshly obtained bone marrow cells from patients with multiple myeloma. Bone marrow aspirates from three separate patients (patients I, II, and III) were simultaneously analyzed with DMC209 and with antibodies to CD138 and CD38, each labeled with fluorochromes fluorescing at different emission wavelengths. Left, side scatter analyses were done against DMC209. DMC209-positive cells were gated and designated R1 (left, red dots), and their location in the CD38/CD138 scatter plot was identified (right, red dots).

Close modal

Tumor-associated MUC1 has been extensively studied as a tumor marker and as a target in cancer therapeutics (2833). In an attempt to target MUC1, a variety of anti-MUC1 antibodies have been generated, almost all of which react with epitopes contained within the highly immunogenic tandem 20-amino acid repeat sequence of the MUC1 α chain. Antibodies binding the tandem repeat array do have potential advantages for therapeutic use because the presence of multiple repeat MUC1 epitopic immunogens may result in a large number of anti-MUC1 antibodies binding MUC1 at distinct sites.

This approach, however, has serious drawbacks, limiting the clinical use of antitandem repeat antibodies. The level of soluble circulating α subunit MUC1 containing the tandem repeat array is elevated in patients with MUC1-expressing malignancies such that circulating α chain can bind and sequester much or all of the antibodies directed against the repeat sequence. Consequently, the amount of anti-MUC1 even reaching the targeted MUC1-expressing cancer cells may be very limited. Furthermore, formation and deposition of MUC1-anti-MUC1 circulating immune complexes can result in severe end-organ damage. These shortcomings likely contributed to the limited success of antitandem repeat MUC1 antibodies in clinical trials to date (28, 3336).

To circumvent the inherent shortcomings of antitandem repeat antibodies, we elected to generate anti-MUC1 antibodies uniquely targeting a cell-bound MUC1 epitope. Because the MUC1 β subunit is stably tethered to the cell membrane and strongly binds the α subunit (Fig. 1; refs. 9, 1619), the region comprising the junction of the two cleaved subunits represents a potentially valuable cell-bound MUC1 epitope. Antibodies specific to the α/β junction should directly target MUC1-expressing cells and escape sequestration by circulating tandem repeat array–containing MUC1 α subunit.

To generate anti-α/β junction antibodies, we immunized mice with cDNA encoding the MUC1 α/β junction. The MUC1/X protein from which the tandem repeat array is deleted (7, 9) represents an isoform far less complex than the complete tandem repeat array–containing MUC1/TM molecule. Because both MUC1/TM and MUC1/X are cleaved at an identical site and result in the same noncovalent interaction of the α and β subunits (9), we reasoned that immunization with cDNA coding for MUC1/X would elicit anti-α/β junction antibodies, avoiding formation of antibodies to the tandem repeat array.

Contrary to expectations, immunization with cDNA encoding the MUC1/X protein failed to elicit strong anti-MUC1/X antibody immune responses, whereas mice immunized with MUC1/TM cDNA generated significant and highly reproducible anti-MUC1/X antibody titers. Furthermore, immunizing mice primed with MUC1/TM cDNA with a single boost of MUC1/X protein led to exceptionally high anti-MUC1/X antibody titers. We concluded that immunization with MUC1/TM cDNA induces antibodies, which recognize epitopes common to both the MUC1/X and MUC1/TM proteins that can be significantly enhanced by subsequent MUC1/X protein immunization. This cDNA/protein two-step ‘heteroimmunization’ protocol successfully yielded junction-specific DMC209 mAbs, which solely target cell-tethered MUC1, and represents a novel paradigm to generate antibodies against MUC1-expressing malignant cells.

Beyond specifically generating anti-MUC1 α/β junction antibodies, our findings may have far-reaching implications for producing high-titer antibodies against intractable protein immunogens, which heretofore have elicited poor immune responses. We surmise that the cis-localization of the highly immunogenic upstream tandem repeat array in the MUC1 sequence potentiates the development of antibodies against the poorly immunogenic downstream α/β junction region. Once antibody synthesis against the α/β junction region is initiated by primary immunization with MUC1/TM cDNA, markedly elevated antibody titers were obtained by a subsequent protein boost. The MUC1 tandem repeat array cDNA may serve as a general molecular adjuvant potentiating production of antibodies directed to protein domain(s) encoded by downstream sequences. In our specific case, the downstream sequences encoded the MUC1 α/β junction region, but one could envisage replacement of this region with any cDNA encoding a protein of interest followed by boost immunization with that same protein. High titers of antibody may thereby be obtained against protein domains that by themselves elicit poor immune responses.

Characterization of monoclonal anti-MUC1 α/β junction antibody DMC209 showed that its immunoreactivity was clearly dependent on the cleavage status of the MUC1 protein: all amino acid point mutations at S63 of the GSVVV cleavage site that abrogated cleavage (9, 19) also abolished DMC209 reactivity. As DMC209 immunoreactivity is preserved with cleaved G[S63C]VVV and G[S63T]VVV point mutants, we conclude that DMC209 is likely a conformational antibody that does not recognize a specific amino acid sequence at the cleavage site but rather recognizes the MUC1 α/β junction structure resulting from cleavage. This is the precise specificity required of an antibody for use in targeting MUC1-expressing cells: the antibody should react only with membrane-tethered MUC1 composed of cleaved and interacting α and β subunits. That DMC209 answers to these specifications is illustrated by its binding to cell transfectants expressing MUC1, MUC1-expressing cancer cell lines, and MUC1-expressing multiple myeloma cells. DMC209 activity in bone marrow aspirates is perhaps the most directly informative. It shows that, in an unmanipulated heterogeneous cell population, DMC209 specifically binds the MUC1-expressing malignant plasma cells of myeloma while remaining nonreactive with the other cell lineages present in the sample. In addition, DMC209 reactivity in the aspirates is concordant with immunoreactivity of anti-CD138 (anti-syndecan), a well-recognized plasma cell marker (37). Our finding of CD138-positive cells differentially expressing the DMC209 epitope is concordant with previous findings showing variability of MUC1 expression on the plasma cells of multiple myeloma (3, 38, 39). Significantly, DMC209 recognizes a distinct subpopulation of multiple myeloma cells, suggesting heterogeneity within the malignant plasma cell population. Cellular heterogeneity may have implications in the variable clinical course of the disease.

These results suggest that the novel cDNA/protein immunization protocol described can yield high titers of antibodies to the MUC1 α/β junction to directly target MUC1-expressing malignant cells. It further opens the possibility of generating therapeutically useful antibody-immunotoxin or antibody-radioisotope conjugates.

Finally, MUC1 acts as a signaling receptor protein that undergoes extensive cytoplasmic tyrosine phosphorylation analogous to the epidermal growth factor receptor family of proteins (9, 13, 27, 4043). The intriguing possibility exists therefore that, just as Herceptin mAb targets erbB2-expressing adenocarcinomas (44, 45), anti-MUC1 α/β junction-specific antibodies may have direct or indirect cytotoxic activity. Such investigations are presently being undertaken.

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.

1
Girling A, Bartkova J, Burchell, et al. A core protein epitope of the polymorphic epithelial mucin detected by the monoclonal antibody SM-3 is selectively exposed in a range of primary carcinomas.
Int J Cancer
1989
;
43
:
1072
–6.
2
Hilkens J, Buijs F, Hilgers, et al. Monoclonal antibodies against human milk-fat globule membranes detecting differentiation antigens of the mammary gland and its tumors.
Int J Cancer
1984
;
34
:
197
–206.
3
Burton J, Mishina D, Cardillo T, et al. Epithelial mucin-1 (MUC1) expression and MA5 anti-MUC1 monoclonal antibody targeting in multiple myeloma.
Clin Cancer Res
1999
;
5
:
3065
–72s.
4
Hadden JW. The immunology and immunotherapy of breast cancer: an update.
Int J Immunopharmacol
1999
;
21
:
79
–101.
5
O'Connor JC, Julian J, Lim SD, Carson DD. MUC1 expression in human prostate cancer cell lines and primary tumors.
Prostate Cancer Prostatic Dis
2005
;
8
:
36
–44.
6
Yonezawa S, Byrd JC, Dahiya R, et al. Differential mucin gene expression in human pancreatic and colon cancer cells.
Biochem J
1991
;
276
:
599
–605.
7
Baruch A, Hartmann M, Zrihan-Licht S, et al. Preferential expression of novel MUC1 tumor antigen isoforms in human epithelial tumors and their tumor-potentiating function.
Int J Cancer
1997
;
71
:
741
–9.
8
Hartman M, Baruch A, Ron I, et al. MUC1 isoform specific monoclonal antibody 6E6/2 detects preferential expression of the novel MUC1/Y protein in breast and ovarian cancer.
Int J Cancer
1999
;
82
:
256
–67.
9
Levitin F, Stern O, Weiss M, et al. The MUC1 SEA module is a self-cleaving domain.
J Biol Chem
2005
;
280
:
33374
–86.
10
Obermair A, Schmid BC, Stimpfl M, et al. Novel MUC1 splice variants are expressed in cervical carcinoma.
Gynecol Oncol
2001
;
83
:
343
–7.
11
Oosterkamp HM, Scheiner L, Stefanova MC, Lloyd KO, Finstad CL. Comparison of MUC-1 mucin expression in epithelial and non-epithelial cancer cell lines and demonstration of a new short variant form (MUC-1/Z).
Int J Cancer
1997
;
72
:
87
–94.
12
Zrihan-Licht S, Vos HL, Baruch A, et al. Characterization and molecular cloning of a novel MUC1 protein, devoid of tandem repeats, expressed in human breast cancer tissue.
Eur J Biochem
1994
;
224
:
787
–95.
13
Baruch A, Hartmann M, Yoeli M, et al. The breast cancer-associated MUC1 gene generates both a receptor and its cognate binding protein.
Cancer Res
1999
;
59
:
1552
–61.
14
Gendler SJ, Lancaster CA, Taylor-Papadimitriou J, et al. Molecular cloning and expression of human tumor-associated polymorphic epithelial mucin.
J Biol Chem
1990
;
265
:
15286
–93.
15
Ligtenberg MJ, Vos HL, Gennissen AM, Hilkens J. Episialin, a carcinoma-associated mucin, is generated by a polymorphic gene encoding splice variants with alternative amino termini.
J Biol Chem
1990
;
265
:
5573
–8.
16
Ligtenberg MJ, Kruijshaar L, Buijs F, van Meijer M, Litvinov SV, Hilkens J. Cell-associated episialin is a complex containing two proteins derived from a common precursor.
J Biol Chem
1992
;
267
:
6171
–7.
17
Lillehoj EP, Han F, Kim KC. Mutagenesis of a Gly-Ser cleavage site in MUC1 inhibits ectodomain shedding.
Biochem Biophys Res Commun
2003
;
307
:
743
–9.
18
Parry S, Silverman HS, McDermott K, Willis A, Hollingsworth MA, Harris A. Identification of MUC1 proteolytic cleavage sites in vivo.
Biochem Biophys Res Commun
2001
;
283
:
715
–20.
19
Macao B, Johansson DG, Hansson GC, Hard T. Autoproteolysis coupled to protein folding in the SEA domain of the membrane-bound MUC1 mucin.
Nat Struct Mol Biol
2006
;
13
:
71
–6.
20
Bork P, Patthy L. The SEA module: a new extracellular domain associated with O-glycosylation.
Protein Sci
1995
;
4
:
1421
–5.
21
Fujita K, Denda K, Yamamoto M, Matsumoto T, Fujime M, Irimura T. Expression of MUC1 mucins inversely correlated with post-surgical survival of renal cell carcinoma patients.
Br J Cancer
1999
;
80
:
301
–8.
22
McGuckin MA, Walsh MD, Hohn BG, Ward BG, Wright RG. Prognostic significance of MUC1 epithelial mucin expression in breast cancer.
Hum Pathol
1995
;
26
:
432
–9.
23
Hamann PR, Hinman LM, Beyer CF, et al. A calicheamicin conjugate with a fully humanized anti-MUC1 antibody shows potent antitumor effects in breast and ovarian tumor xenografts.
Bioconjug Chem
2005
;
16
:
354
–60.
24
Koumarianou AA, Hudson M, Williams R, Epenetos AA, Stamp GW. Development of a novel bi-specific monoclonal antibody approach for tumour targeting.
Br J Cancer
1999
;
81
:
431
–9.
25
Snijdewint FG, von Mensdorff-Pouilly S, Karuntu-Wanamarta AH, et al. Antibody-dependent cell-mediated cytotoxicity can be induced by MUC1 peptide vaccination of breast cancer patients.
Int J Cancer
2001
;
93
:
97
–106.
26
Wijdenes J, Vooijs WC, Clement C, et al. A plasmocyte selective monoclonal antibody (B-B4) recognizes syndecan-1.
Br J Haematol
1996
;
94
:
318
–23.
27
Pulido R, Cebrian M, Acevedo A, de Landazuri MO, Sanchez-Madrid F. Comparative biochemical and tissue distribution study of four distinct CD45 antigen specificities.
J Immunol
1988
;
140
:
3851
–7.
28
Scholl SM, Balloul JM, Le Goc G, et al. Recombinant vaccinia virus encoding human MUC1 and IL2 as immunotherapy in patients with breast cancer.
J Immunother
2000
;
23
:
570
–80.
29
Agrawal B, Krantz MJ, Reddish MA, Longenecker BM. Rapid induction of primary human CD4+ and CD8+ T cell responses against cancer-associated MUC1 peptide epitopes.
Int Immunol
1998
;
10
:
1907
–16.
30
de Bono JS, Rha SY, Stephenson J, et al. Phase I trial of a murine antibody to MUC1 in patients with metastatic cancer: evidence for the activation of humoral and cellular antitumor immunity.
Ann Oncol
2004
;
15
:
1825
–33.
31
Vlad AM, Kettel JC, Alajez NM, Carlos CA, Finn OJ. MUC1 immunobiology: from discovery to clinical applications.
Adv Immunol
2004
;
82
:
249
–93.
32
Apostolopoulos V, Pietersz GA, McKenzie IF. MUC1 and breast cancer.
Curr Opin Mol Ther
1999
;
1
:
98
–103.
33
von Mensdorff-Pouilly S, Snijdewint FG, Verstraeten AA, Verheijen RH, Kenemans P. Human MUC1 mucin: a multifaceted glycoprotein.
Int J Biol Markers
2000
;
15
:
343
–56.
34
Mitchell MS. Cancer vaccines, a critical review—Part II.
Curr Opin Investig Drugs
2002
;
3
:
150
–8.
35
Karanikas V, Thynne G, Mitchell P, et al. Mannan mucin-1 peptide immunization: influence of cyclophosphamide and the route of injection.
J Immunother
2001
;
24
:
172
–83.
36
Taylor-Papadimitriou J, Burchell J, Miles DW, Dalziel M. MUC1 and cancer.
Biochim Biophys Acta
1999
;
1455
:
301
–13.
37
Costes V, Magen V, Legouffe E, et al. The Mi15 monoclonal antibody (anti-syndecan-1) is a reliable marker for quantifying plasma cells in paraffin-embedded bone marrow biopsy specimens.
Hum Pathol
1999
;
30
:
1405
–11.
38
Paydas S, Sahin B, Gonlusen G, Hazar B, Zorludemir S. MUC1 expression in plasmacytoma.
Leuk Res
2001
;
25
:
221
–5.
39
Treon SP, Maimonis P, Bua D, et al. Elevated soluble MUC1 levels and decreased anti-MUC1 antibody levels in patients with multiple myeloma.
Blood
2000
;
96
:
3147
–53.
40
Mukherjee P, Tinder TL, Basu GD, Gendler SJ. MUC1 (CD227) interacts with lck tyrosine kinase in Jurkat lymphoma cells and normal T cells.
J Leukoc Biol
2005
;
77
:
90
–9.
41
Quin RJ, McGuckin MA. Phosphorylation of the cytoplasmic domain of the MUC1 mucin correlates with changes in cell-cell adhesion.
Int J Cancer
2000
;
87
:
499
–506.
42
Wang H, Lillehoj EP, Kim KC. Identification of four sites of stimulated tyrosine phosphorylation in the MUC1 cytoplasmic tail.
Biochem Biophys Res Commun
2003
;
310
:
341
–6.
43
Zrihan-Licht S, Baruch A, Elroy-Stein O, Keydar I, Wreschner DH. Tyrosine phosphorylation of the MUC1 breast cancer membrane proteins. Cytokine receptor-like molecules.
FEBS Lett
1994
;
356
:
130
–6.
44
Mendelsohn J, Baselga J. The EGF receptor family as targets for cancer therapy.
Oncogene
2000
;
19
:
6550
–65.
45
Baselga J, Norton L, Albanell J, Kim YM, Mendelsohn J. Recombinant humanized anti-HER2 antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts.
Cancer Res
1998
;
58
:
2825
–31.