Synthetic MUC1 antitumor vaccine with incorporated 2,3-sialyl-T carbohydrate antigen inducing strong immune responses with isotype specificity

Article abstract of DOI:10.1002/cbic.201800148

The endothelial glycoprotein MUC1 is known to underlie alterations in cancer by means of aberrant glycosylation accompanied by changes in morphology. The heavily shortened glycans induce a collapse of the peptide backbone and enable accessibility of the latter to immune cells, rendering it a tumor-associated antigen. Synthetic vaccines based on MUC1 tandem repeat motifs, comprising tumor-associated 2,3-sialyl-T antigen, conjugated to the immunostimulating tetanus toxoid, are reported herein. Immunization with these vaccines in a simple water/oil emulsion produced a strong immune response in mice to which stimulation with complete Freund’s adjuvant (CFA) was not superior. In both cases, high levels of IgG1 and IgG2a/b were induced in C57BL/6 mice. Additional glycosylation in the immunodominant PDTRP domain led to improved binding of the induced antisera to MCF-7 breast tumor cells, compared with that of the monoglycosylated peptide vaccine.

Full text of DOI:10.1002/cbic.201800148

Accepted Article
Title: Synthetic MUC1 antitumor vaccines with incorporated 2,3-sialyl-
T carbohydrate antigen inducing strong immune responses with
isotype specificity
Authors: Horst Kunz, David Strassburger, Markus Glaffig, Natascha
Stergiou, Sabrina Bialas, Pol Besenius, and Edgar Schmitt
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201800148
Link to VoR: http://dx.doi.org/10.1002/cbic.201800148
A Journal of
www.chembiochem.org

10.1002/cbic.201800148
ChemBioChem
COMMUNICATION
Synthetic MUC1 antitumor vaccine with incorporated 2,3-sialyl-T
carbohydrate antigen inducing strong immune responses with
isotype specificity
D. Straßburger,[a] M. Glaffig,[a] N. Stergiou,[b] S. Bialas,[a] P. Besenius,[a] E. Schmitt,[b] H. Kunz*[a]
Dedication ((optional))
TA-MUC1 can also bind to selectins on blood vessel endothelial
cells facilitating metastasis.[10,11]
Abstract: The endothelial glycoprotein MUC1 is well known to
underlie alterations in cancer by means of aberrant glycosylation
accompanied by changes in morphology. The heavily shortened
glycans induce a collapse of the peptide backbone and enable
accessibility of the latter to immune cells, rendering it a tumor-
associated antigen. We report synthetic vaccines based on MUC1
tandem repeat motifs comprising tumor-associated 2,3-sialyl-T
antigen, conjugated to the immunostimulating tetanus toxoid.
Immunization with these vaccines in a simple water/oil emulsion
produced a strong immune response in mice to which stimulation
with complete Freund’s adjuvant (CFA) was not superior. In both
cases high levels of IgG1, and IgG2a/b were induced in C57BL/6
mice. An additional glycosylation in the immunodominant PDTRP
domain lead to improved binding of the induced antisera to MCF-7
breast tumor cells compared to the mono-glycosylated peptide
vaccine.
An anti-tumor vaccine is thought to prime the organism’s
immune system to enable the recognition of tumor-associated
structures and activate self-defense mechanisms to eliminate
the malignancy. Such an immune response should address solid
tumors as well as circulating malignant cells, thus lowering the
tumor burden and the risk of metastases. Nevertheless, a
cancer immunotherapy through treatment with a vaccine holds
the inherent risk of an autoimmune response due to the
endogenous origin of the epitope and the associated self-
tolerance. In this case the generated antibodies do not
distinguish between healthy and tumor cells due to a lack of
antibody selectivity.
In comparison to peptide epitopes carbohydrate antigens induce
low immune responses mainly due to the involved T-cell-
independent mechanisms.[12,13] Among the tumor-associated
carbohydrate antigens (TACAs) the trisaccharide (2,3)-sialyl T-
antigen (2,3-ST) was found the most abundant in different tumor
cell lines e.g. T47D[14] and HAT-29[15]. Furthermore, highly
sialylated MUC1 is found in invasive and metastasizing
cancers.[16] An immune response directed against these
structures should have a high therapeutic impact on neoplastic
tissues. A supreme objective of such a vaccine is to avoid the
worst-case scenario of auto-immune reaction with fatal outcome
for the patient by thorough design of the vaccine. Recent
experiments with transgenic mice expressing human MUC1
showed no auto-aggressive side effects although the generated
antibodies exhibit almost equivalent binding properties towards
human breast tumor cells.[17]
There is strong evidence that the glycosylation sites within the
MUC1 tandem repeat and the incorporated TACA determine
specificity and selectivity of the induced immune response.[18–20]
It was shown that the saccharides cause conformational
changes in the peptide epitopes[21] impacting on the generated
antibodies. In order to enhance the mandatory selectivity, the
2,3-sialyl T-antigen most frequently occurring on carcinomas
was incorporated into the B-cell epitope of the vaccine on
threonine (T18) of vaccine V1 whose sequence comprises the
VNTR domain. The 20mer tandem repeat was expanded by
proline and alanine in order to complete the STAPPA motif
known to be a CD8+ T-cell epitope.[22,23] In the second vaccine
V2, a Tn antigen was attached additionally to threonine 11 within
the immunodominant PDTRP region. There is evidence that
glycosylation at this site has a strong impact on morphology[24]
and on receptor binding.[25] Although in patients suffering from
e.g. mammary carcinoma, low antibody levels against TA-MUC1
were detected,[26,27] they lack a strong autonomous humoral
response due to suppression by the tumor. To overcome this
hurdle in the synthetic vaccine, the B-cell epitope was
MUC1 has been identified as promising tumor-associated
antigen. It is found on the surface of epithelial cells throughout
the human body.[1] It is a membrane-bound glycoprotein and
features a highly glycosylated extracellular domain containing a
variable number of tandem repeats (VNTR) of proline-, serine-
and threonine-rich repetitive sequences of 20 amino acids
(PAHGVTSAPDTRPAPGSTAP). The O-linked saccharides
make up 50-90% of the molecular weight[2] and cause the brush-
like morphology of the glycoprotein protruding 200-500 nm[3,4]
from the apical cell membrane. Tumorigenesis involves profound
alterations in the metabolism-affecting enzyme expression levels.
The decrease of β-1,6-N-acetylglucosaminyltransferase[5] activity
and upregulated α-2,3- (ST3Gal-I) and α-2,6-sialyltransferases
(ST6GalNAc-II) result in a dramatic shortening of glycan chains
on tumor-associated (TA)-MUC1 preferentially terminated with
sialic acid.[6–8] The hypoglycosylation provokes the collapse of
the hydrophobic peptide backbone rendering it accessible to the
immune system.
Tumor-associated MUC1 also interacts with receptors and
growth factors pushing the proliferation of the tumor further.[9]
[a]
[b]
D. Straßburger, M. Glaffig, S. Bialas, Prof. Dr. P. Besenius,
Prof. Dr. H. Kunz
Institut für Organische Chemie
Johannes Gutenberg-Universität Mainz
Duesbergweg 10-14, 55128 Mainz
E-mail: hokunz@uni-mainz.de
N. Stergiou, Prof. Dr. E. Schmitt
Institut für Immunologie
Johannes Gutenberg-Universität Mainz
Obere Zahlbacher Straße 67, 55101 Mainz
Supporting information for this article is given via a link at the end of
the document
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800148
ChemBioChem
COMMUNICATION
conjugated to the strongly T-cell-stimulating tetanus toxoid
(TTox) which is an established immunostimulant and applicable
in human immunotherapy.
removed with catalytic sodium methoxide in methanol and
positions 4 and 6 were masked as benzylidene acetal. The
glycosylation was carried out under FISCHER HELFERICH
conditions to afford the intermediate T-antigen 15 in excellent
yields. The sialyl donor 20 was prepared from N-
acetylneuraminic acid (NANA) by exhaustive acetylation,
benzylation of the carboxylic function by reaction of the cesium
salt with benzyl bromide. Treatment with acetyl chloride and
subsequent substitution with potassium xanthate yielded
donor 20. Sialylation of the selectively deacetylated acceptor 16
was performed using methyl sulfenyl bromide, prepared in situ
from the disulfide, and silver triflate.[32] The acetal group of
product 21 was acidolytically removed and replaced by acetyl
groups. For the introduction in solid phase syntheses (SPPS),
tert-butyl esters of both, Thr-Tn 7 as well as Thr-2,3-ST 24, were
removed using TFA in DCM. A Tentagel R TRT resin preloaded
(0.17 mmol/g) with Fmoc-L-alanine was used as the solid
support. It should be mentioned that some modifications and
optimizations of the deprotection procedures (vii) and (viii) were
made (Scheme 1): For removal of the benzyl group, Pd(0) was
prepared in situ from the Pd(II) acetate in a mixture of 2-
propanol, water and acetic acid in order to avoid the
The biomimetic synthesis of the TACAs Thr-Tn 7 and Thr-2,3-
ST 24 starts from L-threonine, D-galactose and N-acetyl
neuraminic acid.[28] Peracetylated D-galactose was treated with
HBr in acetic acid. Subsequent elimination with zinc yielded
galactal 2. Through azidonitration[29] the neighboring effect-
lacking azide was introduced on C-2. The generated nitrate
ester 3 on C-1 was converted to α-bromide 4 as the glycosyl
donor. Glycosylation of acceptor 1 was achieved with silver
perchlorate/carbonate.[30] Simultaneous reduction of the azide 5
and conversion to the acetamide 6 were achieved by treatment
with zinc in acetic acid/acetic anhydride mixture in THF. In the
case of Thr-2,3-ST the acetyl protecting groups were removed
with catalytic sodium methoxide in methanol.[31] The galactosyl
donor 14 was synthesized from galactose. Positions 1’,2’ and
3’,4’ were blocked as acetonides and a benzyl ether was
installed on C-6. Acidic cleavage of the acetal protecting groups
and subsequent peracetylation, followed by treatment with HBr
in acetic acid yielded the second glycosyl donor 14. In order to
address the 3-position of Tn 9, acetyl protecting groups were
Scheme 1 On the left: synthesis of the tumor-associated carbohydrate antigens Thr-Tn 7 and Thr-2,3-ST 24 (i) Fmoc-OSu, NaHCO3, acetone, 97% (ii) DCC,
t
CuCl, BuOH, 65% (iii) Ac2O, HClO4 (iv) HBr, AcOH, 98% (v) Zn, AcOH, 87% (vi) CAN, NaN3, MeCN, 40% (vii) LiBr, MeCN, 70% (viii) AgClO4, Ag2CO3,
toluene/DCM, 44% (iix) Zn, AcOH, Ac2O, THF, 55% (ix) NaOMe/MeOH 68%. (x) DCM, TFA, H2O, 91% (xi) NaOMe, MeOH, 68% (xii). α,α-dimethoxytoluene,
p-TosOH, MeCN 82%. (xiii) Hg(CN)2, MS3 Å, CH3NO2, DCM, 92%. (xix) NaOMe/MeOH, 75% (xx) Ac2O, pyridine, quant. (xxi) 1.Cs2CO3, EtOH/H2O, 2.BnBr,
DMF, 67% (xxii) AcCl, H2O, quant. (xxiii) KEX, EtOH, 82% (xxiv) MeSBr, AgOTf, MS 3 Å, MeCN/DCM, 53% (xxv) AcOH, 70%. (xxvi) Ac2O, pyridine, DMAP,
97%. (xxvii) TFA, anisole, quant. On the right: SPPS Synthesis of MUC1 tandem repeat glycopeptides and subsequent protecting group manipulations. On the
bottom: final TA MUC1 tetanus toxoid conjugate vaccines.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800148
ChemBioChem
COMMUNICATION
terminated compounds 27
and 28. The conjugation of
the amino-terminated B-cell
epitopes was achieved via
coupling of their squarate
monoamides[33] 29 and 30
coupling which allows for a
chemoselective linkage of
the glycopeptides to lysines
of TTox.
This concept enables
a
multivalent presentation of
the glycopeptide epitope
considered crucial for an
effective immunostimulation.
Ultrafiltration
over
a
polyether sulfone membrane
(exclusion limit 30 kDa)
followed by lyophilization of
the supernatant yielded
vaccines
(Supporting
Conjugation
V1
and
Information).
of the
V2
Figure 1. ELISA dilution curve of the sera collected from mice treated with vaccines five days after the third immunization.
#1: V1 first CFA/s.c., boost IFA/i.p. BALB/c (3 mice), #2: V1 first CFA/s.c., boost IFA/i.p. C57BL/6 (3 mice), #3: V1 only
IFA, C57BL/6 (3 mice), #4: V2 first CFA/s.c., boost IFA/i.p., BALB/c (3 mice), OD at 410 nm.
glycopeptides to BSA was
achieved by an analogous procedure and gave the coatings for
ELISA experiments C1 and C2 (Supporting Information).
The first immunizations were performed by subcutaneous
injection of a formulation of the vaccines V1 and V2 in complete
Freund’s adjuvant (CFA), a water-in-oil emulsion containing
inactivated mycobacteria tuberculosis. CFA is an effective
adjuvant used for boosting the immune response, but its usage
limited to animal models due to its severe side effects. Therefore
V1 was additionally tested in mice when CFA was replaced by
formation of catalytically inactive palladium aggregates. The
deacetylation was carried out in dilute aqueous solution of 4 mM
sodium hydroxide. The pH of 11.5 of the reaction was steadily
monitored until a constant pH indicated complete conversion.
Subsequent semipreparative HPLC yielded the amino-
incomplete Freund’s adjuvant (IFA,
a simple water-in-oil
emulsion) in the first immunization. Immunizations were
performed three times in groups of three mice each. The
antisera were collected from the tail veins five days after the last
administration. The yielded anti-sera were tested for antibody
levels by ELISA and cell affinity by flow cytometry.
BALB/c wild-type mice generally generate IL-4-dominated
immune responses via MHC II restricted T helper 2 cell support
and, therefore, the activation of a humoral immune response
with high titers of the IgG1 antibody type.[34] For the purpose of
investigating the differences in immune responses, V1 was also
applied in mice of the C57BL/6 strain. The sera collected from
BALB/c and C57BL/6 mice showed high titers for both vaccines
indicating a strong overall immune response (see Supporting
Information). Titer levels of BALB/c mice treated with vaccine V2
(T18 2,3-ST and T11 Tn, #4) did not significantly differ from those
of mice immunized with the mono-glycosylated vaccine V1 (T18
2,3-ST), both administered with CFA. Additionally, in groups of
C57BL/6, V1 was applied with (#2) and without CFA (#3).
Unexpectedly, only insignificant differences were observed in
direct comparison. We hypothesize, that CFA is not necessary
to enhance the efficiency when working with TTox conjugated
vaccines. Furthermore, glycosylation of threonine 11 within the
immunodominant domain (V2) enhanced the immunogenicity
most likely due to a more tumor-selective conformation of the
Figure 2. Isotype analysis of selected sera from BALB/c mice treated with V1
(#1) and V2 (#4) five days after third immunization, OD at 410 nm.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800148
ChemBioChem
COMMUNICATION
glycopeptide epitope.[24,25] The presence of IgG1 antibodies
indicates the specificity of the induced immune response.
Increased levels of both IgG2a and IgG2b after immunization of
BALB/c mice with V2 compared to V1 (Figure 2) should activate
antibody-dependent cell-mediated cytotoxicity (ADCC) and
complement-dependent cytotoxicity (CDC) as a result of the
response to TA-MUC1. The development of T-cell dependent
immune responses is necessary for the switch from IgM to IgG
mandatory for the generation of a long term immunological
memory against tumor-associated MUC1 glycopeptides found
on epithelial cancers. The induced antibodies are not directed
against the squarate linker as was shown by complete
The mice used were 8 weeks old. All mice used for this study
were bred and housed in a specific pathogen-free colony at the
animal facility of Johannes Gutenberg University following
institutionally approved protocols (permission was obtained from
the Landesuntersuchungsamt Koblenz, reference number: 23
177-07/G 08-1-019).
Acknowledgements
This
work
was
supported
by
the
Deutsche
Forschungsgemeinschaft (SFB 1066) and by the Fonds der
Chemischen Industrie.
neutralization with the glycopeptide epitopes.[20]
.
Keywords: MUC1 • synthetic antitumor vaccine • 2,3-sialyl-T
antigen • glycopeptide • regioselective sialylation
Conclusion
The results of the immunological evaluation provide strong
evidence that robust immune responses are induced through
vaccination with both synthetic (T18 2,3-ST)-vaccines V1 and V2.
Surprisingly co-administration with CFA did not show a benefit
for V1. Human MCF-7 breast tumor cells incubated with the
antisera induced by V2 (containing the second glycan in the
PDTRP domain) showed increased binding of the antibodies in
the flow cytometry (Figure 3). Although the incorporation of Tn
antigen at threonine 11 has less impact on the total antibody
titers, their affinity significantly increases probably due to an
improved turn-type conformation in the PDTRP domain [24,25] This
supports the hypothesis that the glycosylation pattern plays a
key role in tumor specificity. The observed class switch to IgG
subtypes, indicates an acquisitioned immunological memory and
the generated antibodies show high affinity to tumor cells.
[1]
[2]
[3]
R. Singh, D. Bandyopadhyay, Cancer Biol. Ther. 2007, 6, 481–486.
S. Nath, P. Mukherjee, Trends Mol. Med. 2014, 20, 332–342.
M. Brayman, A. Thathiah, D. D. Carson, Reprod. Biol. Endocrinol.
2004, 2, 4.
M. M. DeSouza, G. A. Surveyor, R. E. Price, J. Julian, R. Kardon, X.
Zhou, S. Gendler, J. Hilkens, D. D. Carson, J. Reprod. Immunol.
2000, 45, 127–158.
I. Brockhausen, J. M. M. Yang, J. Burchell, C. Whitehouse, J.
Taylor-Papadimitriou, Eur. J. Biochem. 1995, 233, 607–617.
G. F. Springer, Science 1984, 224, 1198–206.
[4]
[5]
[6]
[7]
[8]
A. Harduin-Lepers, Front. Biosci. 2012, E4, 396.
R. Sewell, M. Bäckström, M. Dalziel, S. Gschmeissner, H. Karlsson,
T. Noll, J. Gätgens, H. Clausen, G. C. Hansson, J. Burchell, et al., J.
Biol. Chem. 2006, 281, 3586–3594.
T. M. Horm, J. A. Schroeder, Cell Adh. Migr. 2013, 7, 187–198.
T. Hayashi, T. Takahashi, S. Motoya, T. Ishida, F. Itoh, M. Adachi, Y.
Hinoda, K. Imai, Digestion 2001, 63 Suppl 1, 87–92.
P. Ciborowski, O. J. Finn, Clin. Exp. Metastasis 2002, 19, 339–345.
J. J. Mond, Q. Vos, A. Lees, C. M. Snapper, Curr. Opin. Immunol.
1995, 7, 349–354.
C. M. Snapper, J. J. Mond, J. Immunol. 1996, 157, 2229–2233.
S. Muller, F.-G. Hanisch, J. Biol. Chem. 2002, 277, 26103–26112.
N. T. Marcos, A. Cruz, F. Silva, R. Almeida, L. David, U. Mandel, H.
Clausen, S. von Mensdorff-Pouilly, C. A. Reis, J. Histochem.
Cytochem. 2003, 51, 761–771.
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
S. Yonezawa, E. Sato, Pathol. Int. 1997, 47, 813–830.
N. Stergiou, M. Glaffig, H. Jonuleit, E. Schmitt, H. Kunz,
ChemMedChem 2017, 12, 1424–1428.
[18]
[19]
U. Westerlind, A. Hobel, N. Gaidzik, E. Schmitt, H. Kunz, Angew.
Chem. Int. Ed. 2008, 47, 7551–7556.
B. Palitzsch, N. Gaidzik, N. Stergiou, S. Stahn, S. Hartmann, B.
Gerlitzki, N. Teusch, P. Flemming, E. Schmitt, H. Kunz, Angew.
Chem. Int. Ed. 2016, 55, 2894–2898.
[20]
A. Kaiser, N. Gaidzik, U. Westerlind, D. Kowalczyk, A. Hobel, E.
Schmitt, H. Kunz, Angew. Chem. Int. Ed. 2009, 48, 7551–7555.
A. Kuhn, H. Kunz, Angew. Chem.Int. Ed. 2007, 46, 454–458.
B. Agrawal, M. J. Krantz, M. A. Reddish, B. M. Longenecker, Int.
Immunol. 1998, 10, 1907–1916.
[21]
[22]
[23]
[24]
[25]
[26]
[27]
P. Mukherjee, A. R. Ginardi, T. L. Tinder, C. J. Sterner, S. J.
Gendler, Cancer Res. 2001, 7.
S. Dziadek, C. Griesinger, H. Kunz, U. M. Reinscheid, Chem. - A
Eur. J. 2006, 12, 4981–4993.
J. Burchell, J. Taylor-Papadimitriou, M. Boshell, S. Gendler, T.
Duhig, Int. J. Cancer 1989, 44, 691–696.
Y. Hamanaka, Y. Suehiro, M. Fukui, K. Shikichi, K. Imai, Y. Hinoda,
Int. J. Cancer 2003, 103, 97–100.
S. von Mensdorff-Pouilly, A. A. Verstraeten, P. Kenemans, F. G. M.
Snijdewint, A. Kok, G. J. Van Kamp, M. A. Paul, P. J. Van Diest, S.
Meijer, J. Hilgers, J. Clin. Oncol. 2000, 18, 574–574.
S. Dziadek, C. Brocke, H. Kunz, Chem. Eur. J. 2004, 10, 4150–
4162.
R. U. Lemieux, R. M. Ratcliffe, Can. J. Chem. 1979, 57, 1244–1251.
H. Paulsen, J.-P. Hölck, Carbohydr. Res. 1978, 64, 89–107.
B. Liebe, H. Kunz, Tetrahedron Lett. 1994, 35, 8777–8778.
A. Marra, P. Sinay, Carbohydr. Res. 1989, 190, 317–322.
L. F. Tietze, M. Arlt, M. Beller, K. Glüsenkamp, E. Jähde, M. F.
Rajewsky, Chem. Ber. 1991, 124, 1215–1221.
Figure 3. Flow cytometry analysis of the binding of the antisera (after the 3rd
immunization) to MCF-7 breast tumor cells.
Thus, the synthetic glycopeptide vaccines are promising, potent
for the induction of antitumor-directed immune responses in vivo,
in particular regarding the comparability of the murine and the
human immune system. These results are highly encouraging
for the translation into human cancer therapy, not least because
the glycopeptide vaccines no longer need the co-stimulant CFA
which is not applicable in humans.
[28]
[29]
[30]
[31]
[32]
[33]
[34]
F. P. Heinzel, M. D. Sadick, B. J. Holaday, R. L. Coffman, R. M.
Locksley, J. Exp. Med. 1989, 169.
This article is protected by copyright. All rights reserved.

10.1002/cbic.201800148
ChemBioChem
COMMUNICATION
COMMUNICATION
D. Straßburger, M. Glaffig, N. Stergiou,
S. Bialas, P. Besenius, E. Schmitt*, H.
Kunz*
Page No. – Page No.
Synthetic MUC1 antitumor vaccine
with incorporated 2,3-sialyl-T
carbohydrate antigen inducing a
significant immune response with
isotype specificity
Synthetic antitumor vaccines based on human tumor-associated MUC1
glycopeptides incorporating 2,3-sialyl T antigen and Tn antigen induce a strong
immune response in BALB/c and C57BL/6 mice with antibody subtype selectivity.
Control experiments indicate that CFA is not superior to IFA and the generated
antisera effectively bind to MCF-7 tumor cells. These results suggest that
glycosylation in the PDTRP domain positively effects the immunological response.
This article is protected by copyright. All rights reserved.