MAY/JUNE 2001 Volume 12, Number 3 © Copyright 2001 by the American Chemical Society
COMMUNICATIONS Synthesis and Biological Evaluation of an Anticancer Vaccine Containing the C-Glycoside Analogue of the Tn Epitope Francesco Peri, Laura Cipolla, Maria Rescigno, Barbara La Ferla, and Francesco Nicotra* Department of Biotechnology and Biosciences, University of Milano-Bicocca, Piazza della Scienza 2, I-20126 Milano, Italy. Received November 28, 2000
The C-saccharide analogue of the GalNAc (Tn epitope) has been covalently linked to the T cell epitope peptide 328-340OVA using a chemoselective convergent synthetic approach. In this way, a nonhydrolyzable synthetic vaccine was obtained composed by a B epitope conjugated to a T cell epitope. This compound was tested in a proliferation assay with spleen cells from DO11.10 mice. The molecule was recognized by transgenic T cells although at a slightly lower efficiency if compared with the reference peptide OVA. An additional experiment with dendritic cells fixed with glutaraldehyde shows that the glycopeptide can bind to extracellular MHC molecules without need of internalization and processing and that the C-glycoside part does not interfere with TCR recognition. These observations constitute an important starting point for the use of this molecule as vaccine against the Tn-expressing TA3-Ha mouse mammary carcinoma.
Tumor immunotherapy is based on the theory that tumor-associated antigens (TAA) become immunogenic if presented to a properly trained immunosystem (1). Many of the tumor antigens are constituted by sugars, because the malignant cells are commonly characterized by incomplete glycosylation or neoglycosylation, and a large number of tumor-associated carbohydrate antigens (TACA) expressed on glycolipids and glycoproteins have been identified (2, 3). In particular, the monosaccharide R-GalNAc, called Tn antigen, has been extensively studied, since it is expressed on mucin-type glycoproteins by the majority of human adenocarcinomas as a consequence of aberrant glycosylation, whereas it is hidden in normal cells (4). * To whom correspondence should be addressed. Phone: +39.02.64483457. Fax: +39.02.64483565. E-mail:
[email protected].
Figure 1. Chemical structure of the synthetic vaccine 1.
An immune response directed against carbohydrate antigens results in the induction of antibodies that could eradicate the micrometastases and the circulating tumor cells in the blood stream, thus providing protection against tumor. However, carbohydrate-based vaccines have so far been unsuccessful in inducing detectable T cell immunity (5). To overcome this limitation, and in
10.1021/bc000143a CCC: $20.00 © 2001 American Chemical Society Published on Web 04/12/2001
326 Bioconjugate Chem., Vol. 12, No. 3, 2001
Peri et al.
Figure 2. Overlay of RP-HPLC chromatograms of the conjugation reaction. Conditions: A: 0.1% TFA in water, B: 0.1% TFA in CH3CN; gradient: from 0% to 100% B in 30 min; detection at 214 nm. After 90 min at rt, the HPLC profile of reaction crude shows the complete disappearance of aminooxy-peptide 3 and the appearance of the conjugation product 1 constituted by two peaks (syn and anti oxime isomers).
order to induce an immunological memory toward a carbohydrate epitope, a suitable vaccine has to provide an helper T cell response for B cell induction and IgG production. According to this concept, semisynthetic vaccines have been prepared by conjugation of a carbohydrate B-epitope with a protein and it has been shown that these molecules are capable to induce IgM and IgG anti-Tn antibody responses (6). Totally synthetic vaccines composed by a carbohydrate B-epitope covalently linked to the lipopeptide tripalmitoyl-S-glycerylcysteinylserine as a combined carrier and adjuvant system have been prepared (7, 8, 9). Vaccines with a more complex molecular structure have been developed by assembling covalently a carbohydrate (B-epitope), a peptide T-epitope and the lipopeptide dipalmitoyloxypropyl-N-palmitoylcysteine (Pam3-Cys) (10). However, the outcome of the immune response for this molecule has not been reported. Following the same concept we designed a totally synthetic vaccine (Figure 1) by linking covalently a sugar B and a peptide T cell epitope through a spacer. In our design, the spacer containing an oxime bond has the function to keep the saccharide and the peptide apart, thus minimizing a possible negative interference of the sugar in the formation of the complex between peptide and MHCII1 and its recognition by the TCR. We verified our hypothesis in vitro by testing the capacity of the glycopeptide to form a complex with MHCII on dendritic cells (DC). However, the main innovation in the design of our vaccine consists of the use of the R-C-glycosidic analogue of the GalNAc instead of the glyco-amino acid GalNAcSer/Thr as Tn epitope. The C-glycosidic bond is stable toward acids, bases, and enzymatic hydrolysis; this makes 1 a promising candidate for drug development. Finally, we tested the ability of dendritic cells to present glycopeptide 1 to TCR. DC are the only antigen presenting cells capable of priming a T cell response; thus, vaccines targeted to DC are potentially more powerful (11). 1 Abbreviations: Boc, tert-butyloxycarbonyl; DC, dendritic cell(s); DIPEA, diisopropylethylamine; GalNAc, N-acetyl-galactosamine; MHCII, major histocompatibility complex II; Fmoc, 9-fluorenylmetoxycarbonyl; TCR, T-cell receptor(s); SPPS, solidphase peptide synthesis; TFA, trifluoroacetic acid; TIS, triisopropylsilane.
Scheme 1a
a Reagents and conditions: i, TFA, TIS, H O (95: 2.5: 2.5); 2 ii, 2 (1.2 equiv), 3 (1 equiv), acetate buffer 0.1 M, pH 4.5.
The 328-340OVA peptide in 1 corresponds to a part of the epitope for the TCR-OVA transgenic T cells derived from DO11.10 mice (12). The 327-339OVA peptide was chosen as a model antigen to provide T cell help either in DO11.10 mice or in mice that have been previously immunized with soluble OVA in an immunogenic fashion. The latter will mimic a situation of immunized animals and it will be a “proof of principle” for the use of a recall T-helper antigen, such as the tetanus toxoid in humans. In particular, it will be possible to test the effectiveness of the synthetic vaccine 1 in providing protection against challenge with a highly invasive mouse mammary carcinoma, TA3-Ha, which expresses the Tn antigen (13). The synthesis of the vaccine was accomplished according to a convergent strategy based on the chemoselective coupling (14) of the R-C-glycosyl ketone 2 with peptide 3 bearing an aminooxy group at the N-terminal end (Scheme 1). The preparation of the C-glycosyl analogue
Communications
Figure 3. OVA peptide in compound 1 is recognized by TCROVA transgenic T cells. Spleen cells from DO11.10 mice were incubated with the indicated concentrations of 1 and of reference 327-339OVA peptide. The proliferative response was measured after 2 days as 3H-thymidine incorporation by TCR-OVA T cells. As shown, compound 1 was recognized although with slightly lower efficiency as compared to the reference peptide.
of GalNAc has been reported by our group (15); the peptide sequence was assembled on Wang resin using the Fmoc/tert-butyl solid-phase strategy (16) (SPPS, Scheme
Bioconjugate Chem., Vol. 12, No. 3, 2001 327
1). The aminooxy functionality was introduced by solidphase condensation of N-hydroxysuccinimide activated Boc-aminooxyacetic acid in the presence of the base DIPEA. Aminooxy antigen 3 was obtained with 40% overall yield after cleavage from resin, complete side chain deprotection, and RP-HPLC purification. The chemoselective oxime bond formation between the Tnketone and the aminooxy antigen was carried out in aqueous solution (acetate buffer, pH 4.5) by monitoring the reaction by RP-HPLC (Figure 2); the coupling turned out to be complete after 90 min at rt (35% yield after HPLC product purification). Compound 1 presents two isomeric forms in equilibrium at rt, deriving from the cis/ trans isomerism of the oxime bond, detectable by HPLC analysis and characteristic of oxime-linked glycoconjugates) (14). The glycopeptide was tested in a proliferation assay with spleen cells from DO11.10 mice in order to assess if the carbohydrate group interferes with the capacity of TCR-OVA transgenic T cells to recognize the peptide in association with MHC II. As shown in Figure 3, compound 1 was recognized by transgenic T cells although at a slightly lower efficiency if compared with the reference peptide OVA. Next, we analyzed whether the glycopeptide 1 has to be internalized and processed by DC in order to be presented to T cells. We used a wellcharacterized nonimmortalized DC cell line (D1 cells)(17) which is the prototype of immature DC able to
Figure 4. Internalization and processing of compound 1 is not necessary for OVA peptide presentation. D1 cells were either fixed (A) or nonfixed (B) with 0.001% glutaraldehyde for 1 min on ice, loaded with the different peptides and incubated with the costimulation-insensitive hybridoma (BO97.10) which is specific for the 327-339OVA peptide. Activation of the hybridoma was tested by measuring the IL-2 produced in culture supernatants as a function of 3H-thymidine incorporation by a CTL line which is dependent on IL-2 for its growth (CTLL-2). Peptides 1 and 3 were similarly presented to T cells by fixed and nonfixed cells, indicating that processing is not required for efficient antigen presentation. (C) As a control, fixed DC were unable to present whole ovalbumin which requires processing for the generation of the OVA peptide.
328 Bioconjugate Chem., Vol. 12, No. 3, 2001
present antigens to T cells in vitro. DC were thus fixed with 0.001% glutaraldehyde in phosphate-buffered saline (PBS) before exposure to the peptides to avoid the internalization and processing of the compounds. To overcome the need of costimulation which is essential for T cell priming and cannot be provided by fixed DC, a costimulation-insensitive hybridoma (BO97.10) recognizing the same OVA peptide in association with MHC II molecules was used. As a control we analyzed the capacity of fixed D1 cells to present ovalbumin which has to be internalized and degraded to produce the OVA327-339 peptide recognized by the hybridoma. Both the glycopeptide 1 and the aminooxypeptide 3 could be presented by fixed D1 cells (Figure 4A) with similar efficiencies as non fixed cells (Figure 4B) indicating that the peptide can bind to extracellular MHC molecules and that the carbohydrate does not interfere with TCR recognition. The antigen presentation is very similar in both 1 and 3 but less efficient than in the OVA epitope, suggesting that the aminooxy linker interferes with MHC binding or with TCR recognition. As expected, the whole ovalbumin could not be presented by fixed cells (Figure 4C). Thus, we have validated our vaccine 1 in vitro by showing that it can be presented by DC and that the C-glycosidic analogue of GalNAc, stable toward enzymatic and chemical hydrolysis, does not influence the antigen specificity of the compound. Future perspectives of this work will aim to study the capacity of compound 1 to induce an antibody response to GalNAc in immunized animals in vivo. The requirement of T cell help delivered by the OVA peptide for antibody production will also be evaluated and this will be correlated with the capacity of vaccine 1 to protect mice from lethal challenge with the TA3-Ha mouse mammary carcinoma. ACKNOWLEDGMENT
We gratefully acknowledge Antonella Leone and Felice Daverio for their contribution to the experimental work. LITERATURE CITED (1) Danishefsky, S. J., and Allen, J. R. (2000) From the Laboratory to the Clinic: A Retrospective on Fully Synthetic Carbohydrate-Based Anticancer Vaccines. Angew. Chem., Int. Ed. 39, 836-863. (2) Hakomori, S. (1989) Aberrant glycosylation in tumors and tumor-associated carbohydrate antigens. Adv. Cancer Res. 52, 257-331.
Peri et al. (3) Toyokuni, T., and Singhal, A. K. (1995) Synthetic carbohydrate Vaccines Based on Tumour-Associated Antigens. Chem. Soc. Rev. 231-242. (4) Springer, G. F. (1984) T and Tn, general carcinoma autoantigens. Science 224, 1198-1206. (5) Francis, M. J. (1991) in Vaccines: Recent Trends and Progress (Gregoriadis, G., Allison, A. C., Poste, G., Ed.) Plenum Press, New York. (6) Toyokuni, T., Hakomori, S., and Singhal, A. K. (1994) Synthetic carbohydrate vaccines: synthesis and immunogenicity of Tn antigen conjugates. Bioorg. Med. Chem. 2(11), 1119-1132. (7) Wiesmuller, K. H., Jung, G., and Hess, G. (1989) Novel lowmolecular-weight synthetic vaccineagainst foot-and-mouth disease containing a potent B-cell and macrophage activator. Vaccine 7(1), 29-33. (8) Bessler, W. G., and Jung, G. (1992) Synthetic lipopeptides as novel adjuvants. Res. Immunol. 143(5), 548-553. (9) Wiesmuller, K. H., Bessler, W. G., and Jung, G. (1992) Solidphase peptide synthesis of lipopeptidevaccines eliciting epitopespecific B-, T-helper and T-killer cell response. Intl. J. Pept. Protein Res. 40(3-4), 255-260. (10) Reichel, F., Ashton, P. R., and Boons, G.-J. (1997) Synthetic carbohydrate-based vaccines: synthesis of an L-glycero-Dmanno-heptose antigen-T-epitope-lipopeptide conjugate. Chem. Commun. 2087-2088. (11) Banchereau, J., and Steinman, R. M. Dendritic cells and the control of immunity. Nature 392(6673), 245-252. (12) Murphy, K. M., Heimberger, A. B., and Loh, D. Y. (1990) Induction by antigen of intrathymic apoptosis of CD4+CD8+TCRlo thymocytes in vivo. Science 250(4988), 1720-1723. (13) Shingal, A., Fohn, M., and Hakomori, S.-I. (1991) Induction of R-N-Acetylgalactosamine-O-serine/Threonine (Tn) Antigenmediated Cellular Immune Response for Active Immunoterapy in Mice. Cancer Res. 51, 1406-1411. (14) Peri, F., Cipolla, L., La Ferla, B., Dumy, P., and Nicotra, F. (1999) Chemoselective synthesis of C-glycopeptides: an highly convergent approach to O- and N-linked glycopeptide analogues. Glycoconjugate J. 16(8), 399-404. (15) Cipolla, L., La Ferla, B., Lay, L., Peri, F., and Nicotra, F. (2000) Synthesis of C-analogue of N-acetylgalactosamine. Tetrahedron Asymmetry 11, 295-303. (16) Fields, G. B., and Noble, R. L. (1990) Solid-phase peptide synthesis utilizing 9-fluorenylmethoxycarbonyl amino acids. Intl. J. Pept. Protein Res. 35(3), 161-221. (17) Winzler, C., Rovere, P., Rescigno, M., Granucci, F., Penna, G., Adorini, L., Zimmermann, V. S., Davoust, J., and Ricciardi-Castagnoli, P. (1997) Maturation stages of mouse dendritic cells in growth factor-dependent long-term cultures. J. Expl. Med. 185(2), 317-328.
BC000143A
