In normal epithelia MUC1 expression is found in small amounts and restricted to the apical cell surface. However, in >80% of most premalignant lesions that are precursors to cancer and other adenocarcinomas [28, 37], MUC1 expression is highly up-regulated; for instance in breast cancer it is expressed up to 10-fold higher than in normal cells . MUC1 expressed on tumor cells displays striking alterations with respect to shortened glycan chains (Tn-antigen (GalNAc1-O-S/T) and Thomsen-Friedenreich (TF) or T antigen (Gal01-3GalNAca1-O-S/T) ), increased sialylation (mono- and di-sialylated TF-antigen, mono-sialyl Tn-antigen, sialyl-core1), and a shift in the carbohydrate core-type [39-41]. As an example, normal lactating mammary gland expresses MUC1 with primarily long polylactosamine-type chains based on the elongation of the core2 structure (GlcNAc01-6[Gal01-3] GalNAca1-O), and is dominated by neutral glycans with linear and branched backbone structures that can comprise up to 16 monosaccharide units [42, 43]. During malignant transformation the core2 forming p6-glucosaminyltransferase level is low or not expressed and leads to the accumulation of corel structures  that now serve as substrates of Gal-specific a3-sialyltransferase and GalNAc-specific p6-sialyltransferases. In addition, the increased presence of CMP-NeuAc:Galp1-3GalNAca3-sialyltrans-ferase may compete with the core2 enzyme for substrate . Thus, the length of O-linked glycans on tumor MUC1 is restricted by a high degree of sialic acid that dominates over neutral carbohydrates [19, 45]. This underglycosylation results in unmasking of previously cryptic protein core epitopes and short carbohydrate epitopes. This is important for the humoral and cellular immune response, because the peptide core is more accessible for peptide specific anti-MUCl antibodies and for enzymatic processing into peptides for presentation to T cells, compared to MUC1 on normal epithelia.
The new epitopes on MUC1 are tumor-specific targets for both T and B-cells. The DTR sequence within the tandemly repeated icosapeptide sequence is characterized as the immunodominant epitope  recognized by tumor-specific anti-MUC1 antibodies. Because the VNTR domain of MUC1 contains 25-125 tandem repeats, the immunodominant DTR epitope is presented many times by every MUC1 molecule. Although chemical analyses have shown that the threonines in the DTR-motif can be glyco-sylated [47, 48], it appears to be a low affinity site for glycosyltranferases and thus it is usually presented in its non-glycosylated form as a knob-like structure . This self-stabilizing and tandemly repeated structure  is assumed to favor immune recognition, especially for non-MHC restricted T cells . Other studies revealed that the presence of a GalNAc moiety on the threonine in the DTR motif increases anti-genicity [52, 53]. Vaccination studies in mice also showed that a 60mer MUC1 glyco-peptide composed of three tandem repeats with GalNAc (Tn) or NeuAc-GalNAc (sialyl-Tn) moieties in each of the five glycosylation sites per repeat, elicited the strongest MUC1-specific antibody response compared to less glycosylated or unglyco-sylated MUC1 peptides . Interestingly, in up to 50% of the tandem repeats the DTR motif is replaced by an ESR motif, often in concert with a Pro to Ala replacement in position +10 relative to the DT or ES in the repeat (PDT/ESRPAPGSTAPP/ AAHGVTSA). These variant repeats often occur in clusters . The amino acid changes revealed a higher conformational flexibility of ES/P peptides and reduced glycosylation density of ES/A sequence-variant repeats compared to DT/ P peptides. Moreover, sera of healthy controls, pregnant woman and patients with benign tumors contained IgGs preferentially directed to variant repeat clusters, whereas sera of cancer patients showed predominant specificity to the invariant DTR peptide .
The most unique aspect of the DTR motif is that due to its stable structure identically repeated multiple times, it is recognized by human T-lymphocytes as an unprocessed epitope on the native MUC1 protein expressed on the surface of tumor cells [57, 58]. This MHC-unrestricted recognition of native MUC1 epitopes induces calcium mobilization, phosphorylation of ZAP-70 and proliferation of T cells (CTL) . MUC1 processing by DCs is influenced by O-glycosylation such that a high degree of complex O-glycosylation on MUC1 is unable to prime strong helper T cell responses when presented by dendritic cells in vitro . This is due to the high density of O-glycans that protect the molecule from proteolytic degradation. In contrast, less complex O-linked glycans, like core-type structures, are not removed from MUC1 when processed by DCs and presented on MHC class II molecules . Efficient processing of MUC1 glycopeptides is also a matter of site-specific glycosylation . Elongated glycans proximate to the minor cleavage sites of the tandem repeats, AH I GV and PG I ST ( I represents the cleavage site), can sterically impede the access of the protease (Cathepsin L) and glycosylation adjacent to the major cleavage site VT I SA prevents cleavage completely .
Considering the undesirable Mr. Hyde tumor promoting characteristics of MUC1, it was gratifying to discover that in its tumor form this molecule is under immune surveillance. Because immune responses against MUC1 can be found in cancer patients, an expectation can be made that if these responses were boosted, they may eliminate or control the growth of cancer cells that express MUC1. Work to date has shown that MUC1 is a promising target for immunotherapeutic strategies to treat cancer in humans [63-65]. Several phase I studies have been completed so far and demonstrated that MUC1 vaccines are safe and in a small number of cases also effective in controlling disease. In one study, a vaccine consisting of a non-glycosylated synthetic MUC1 peptide corresponding to five 20-amino acid long repeats and SB-AS2 adjuvant, was tested for safety, toxicity, and ability to elicit or boost MUC1-specific immune responses in patients with resected or locally advanced pancreatic cancer without prior chemotherapy or radiotherapy . Another clinical phase I study used the same MUC1 peptide and incomplete Freund's adjuvant as a vaccine for patients with pancreatic or bile duct cancer. This study showed again safety of the vaccine . In a yet another phase I trial, autologous mature dendritic cells pulsed with HLA-A2-binding MUC1 peptides were used in patients with metastatic renal cell carcinoma (RCC). This study showed that MUC1 peptide-pulsed dendritic cells could induce clinical and immunologic responses in some patients . Moreover, the study revealed epitope spreading, whereby some patients developed T-cell responses to antigens not used for vaccinations, such as adipophilin, telomerase, or oncofetal antigen.
The MUC1 peptide from the tandem repeat region was the chosen antigen in all these vaccine formulations because this region differs the most between normal epithelial cells and tumor cells and thus the immune response induced against this form would not be expected to target normal tissues. There is another form of MUC1 that is equally promising in its immunogenicity and safety and that is the MUC1 glycopep-tide that bears tumor specific carbohydrates. This form induces strong helper T cell responses that are able to help generation of MUC1-specific antibodies, preferentially of IgGj isotype and ADCC, and/or effective T cell mediated cytotoxicity .
Was this article helpful?