Oxazole-modified glycopeptides that target arthritis-associated class II MHC Aq and DR4 proteins

Article abstract of DOI:10.1039/c003640d

The glycopeptide CII259-273, a fragment from type II collagen (CII), can induce tolerance in mice susceptible to collagen-induced arthritis (CIA), which is a validated disease model for rheumatoid arthritis (RA). Here, we describe the design and synthesis of a small series of modified CII259-273 glycopeptides with oxazole heterocycles replacing three potentially labile peptide bonds. These glycopeptidomimetics were evaluated for binding to murine CIA-associated A^q and human RA-associated DR4 class II major histocompatibility complex (MHC) proteins. The oxazole modifications drastically reduced or completely abolished binding to A^q. Two of the glycopeptidomimetics were, however, well tolerated in binding to DR4 and they also induced strong responses by one or two DR4-restricted T-cell hybridomas. This work contributes to the development of an altered glycopeptide for inducing immunological tolerance in CIA, with the long-term goal of developing a therapeutic vaccine for treatment of RA.

Full text of DOI:10.1039/c003640d

View Article Online / Journal Homepage / Table of Contents for this issue
Organic &
Biomolecular
Chemistry
www.rsc.org/obc
Volume 8 | Number 13 | 28 June 2010 | Pages 2873–3084
ISSN 1477-0520
FULL PAPER
Ida E. Andersson et al.
Oxazole-modified glycopeptides
PERSPECTIVE
Muhammad Moniruzzaman et al.
that target arthritis-associated class II Activation and stabilization of
MHC Aq and DR4 proteins
enzymes in ionic liquids
1477-0520(2010)8:13;1-0

View Article Online
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Oxazole-modified glycopeptides that target arthritis-associated class II MHC
Aq and DR4 proteins†
Ida E. Andersson,a Tsvetelina Batsalova,b Balik Dzhambazov,b Lotta Edvinsson,a Rikard Holmdahl,b
Jan Kihlberg*a,c and Anna Linusson*a
Received 26th February 2010, Accepted 14th April 2010
First published as an Advance Article on the web 20th May 2010
DOI: 10.1039/c003640d
The glycopeptide CII259-273, a fragment from type II collagen (CII), can induce tolerance in mice
susceptible to collagen-induced arthritis (CIA), which is a validated disease model for rheumatoid
arthritis (RA). Here, we describe the design and synthesis of a small series of modified CII259-273
glycopeptides with oxazole heterocycles replacing three potentially labile peptide bonds. These
glycopeptidomimetics were evaluated for binding to murine CIA-associated Aq and human
RA-associated DR4 class II major histocompatibility complex (MHC) proteins. The oxazole
modifications drastically reduced or completely abolished binding to Aq. Two of the
glycopeptidomimetics were, however, well tolerated in binding to DR4 and they also induced strong
responses by one or two DR4-restricted T-cell hybridomas. This work contributes to the development
of an altered glycopeptide for inducing immunological tolerance in CIA, with the long-term goal of
developing a therapeutic vaccine for treatment of RA.
In our ongoing studies to probe Aq binding and T-cell
Introduction
recognition of CII259-273, we have previously used structure-
The disease pathology of rheumatoid arthritis (RA), an autoim-
based design to modify parts of the peptide backbone by amide
mune disease that causes chronic inflammation of peripheral
bond replacement with different isosteres. In an initial study, the
Ala261-Gly262 amide bond was replaced with ketomethylene, (E)-
cartilaginous joints, can be studied using the mouse model
collagen-induced arthritis (CIA).1 A common feature of human
RA and murine CIA is the association with certain antigen-
presenting class II major histocompatibility complex (MHC)
proteins, e.g. DR1 and DR4 in RA,2–4 and Aq in CIA.5,6 Class
II MHC proteins bind and present peptide antigens to circulating
CD4+ T cells, which is a key step in the development of immune
and autoimmune responses. T-cell hybridomas obtained from
mice with CIA are activated by the CII259-273 glycopeptide, a
fragment from type II collagen (CII), when presented by Aq.7,8 It
has been shown in a vaccination study that the CII259-273 gly-
copeptide can protect mice from developing CIA.9 More recently,
pulmonary vaccination with solubilized Aq protein in complex
with CII259-273 was shown to give effective protection against
disease development.10 Arthritis severity was also reduced in mice
with an established chronic relapsing disease. If translation of these
results to patients suffering from RA would be possible, it would
provide completely new approaches to treat and ultimately cure
RA. However, a drawback of using glycopeptides as therapeutics
is their limited metabolic stability. Introduction of non-peptidic
motifs in the CII259-273 glycopeptide to stabilize its structure
towards degradation could improve the half-life and thereby its
effectiveness as a therapeutic vaccine.
alkene, and aminomethylene isosteres.11 A second study, currently
in progress, explores the effects of introducing (E)-alkene and
ethylene isosteres at the Ile260-Ala261, Ala261-Gly262, and Gly262
-
Phe263 positions.12 The design of the modified glycopeptides has
been guided by a comparative model of the Aq protein11 (Fig. 1).
Experimentally determined binding features support the proposed
binding pose of the glycopeptide ligand in the Aq binding site. That
is, the side chains of Ile260 and Phe263 anchor the glycopeptide in the
P1 and P4 pockets of Aq while the b-D-galactosyl residue attached
to the hydroxylysine side chain (GalHyl264) extends out of the
binding site to interact with a T-cell receptor approaching from
above.7,8,13–15 T-cell hybridomas specific for CII259-273 with the
GalHyl264 moiety have also been generated using transgenic mice
expressing human DR4 and human CD4 co-receptor,16 suggesting
an important role for this residue in RA as well as CIA. However,
compared to the Aq system, the DR4 binding epitope is shifted
three amino acids towards the C terminus of CII.17 Hence, the side
chain of Phe263 occupies the prominent P1 pocket while the side
chain of Glu266 is positioned in the shallow P4 pocket in the DR4
binding site.
In this study, we introduce oxazole heterocycles in the CII259-
273 backbone as structurally more diverse bioisosteres in com-
parison to the amide bond isosteres studied previously by us.11,12
Oxazoles have been described as conformationally restricted
templates in peptidomimetic design,18–20 and oxazole-containing
inhibitors of antigen presentation by RA-associated HLA-DR
proteins have been reported.21 Herein, we describe the design of
three oxazole-modified glycopeptides and explore their effects
on class II MHC binding and T-cell recognition. The design
was primarily focused on targeting the CIA-associated murine
aDepartment of Chemistry, Umea˚ University, SE-901 87 Umea˚, Sweden.
E-mail: jan.kihlberg@chem.umu.se, anna.linusson@chem.umu.se
bMedical Inflammation Research, Department of Medical Biochemistry and
Biophysics, Karolinska Institute, SE-171 77 Stockholm, Sweden
cAstraZeneca R&D Mo¨lndal, SE-431 83 Mo¨lndal, Sweden
† Electronic supplementary information (ESI) available: 1H NMR and 13
C
NMR spectra of all new isolated compounds and HPLC chromatograms
of the oxazole-modified glycopeptides. See DOI: 10.1039/c003640d
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2931–2940 | 2931
©

View Article Online
Fig. 1 Comparative model of the class II MHC Aq protein11 in complex with the CII259-270 glycopeptide from type II collagen. The glycopeptide is
bound in an extended conformation with the side chains of Ile260 and Phe263 anchored in the P1 and P4 pockets of Aq, respectively. The b-D-galactosyl
moiety (shown in ball and stick) attached to the side chain of Hyl264 in position P5 is extending out of the binding site and is thereby able to interact with
a T-cell receptor approaching from above.
Aq system, but the RA-associated human DR4 system was also
included in the evaluations. The glycopeptidomimetics are relevant
for the DR4 system as well since, as mentioned above, both DR4
and Aq bind and present the CII259-273 epitope.
added. This introduced more flexibility to facilitate orientation of
the aromatic group into the P4 pocket of Aq.
The designed glycopeptidomimetics 2, 3, and 4 were docked
into the comparative model11 of Aq to explore the effects of the
introduced oxazole moieties on the protein–ligand interactions.
Binding poses with anchoring side chains of the oxazole-modified
glycopeptides positioned in their respective pockets in the Aq
protein were obtained for all designed ligands (Fig. 3). In addition,
Results and discussion
Design of oxazole-modified glycopeptides
Oxazole mimetics were designed for the three positions between
the important Aq anchoring residues Ile260 and Phe263 in 1 to
prevent cleavage of these peptide bonds, and to investigate binding
to Aq as well as T-cell recognition (Fig. 2). The conformationally
restricted oxazole scaffold was introduced in the glycopeptide
backbone by heterocyclization of the b carbon of one amino
acid residue onto the preceding carbonyl group. For the Ala-
Gly oxazole 3, an extra b carbon was added in order to allow
the formation of the heterocyclic five-membered ring. In the case
of the Gly-Phe oxazole 4, the oxazole was functionalized with a
benzyl instead of a phenyl group, i.e. an extra methylene group was
Fig. 3 Docked binding poses for oxazole-modified glycopeptides (carbon
atoms colored green) each superposed with the CII259-270 glycopeptide
1 (carbon atoms colored grey), extracted from their respective model
of the Aq/glycopeptide complex. Only residues 259 to 263 are shown.
The modified glycopeptides are shown in the following order: (A)
glycopeptide 2 modified with Ile260y[oxazole]Ala261, (B) glycopeptide 3
modified with Ala261y[oxazole]Gly262, and (C) glycopeptide 4 modified
Fig. 2 The CII259-273 glycopeptide (1) with the amide bonds be-
tween the Ile260-Ala261, Ala261-Gly262, and Gly262-Phe263 dipeptide frag-
ments highlighted. Replacement of either of the three indicated amide
bonds in 1 with oxazole moieties gave modified CII259-273 glycopep-
tides with either Ile260y[oxazole]Ala261 (2), Ala261y[oxazole]Gly262 (3), or
Gly262y[oxazole]Phe263 (4) bioisosteres.
with Gly262y[oxazole]Phe263
.
2932 | Org. Biomol. Chem., 2010, 8, 2931–2940
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Fig. 4 Proposed model of the complex between DR4 and the CII259-270 glycopeptide from type II collagen based on a DR4 crystal structure.22 The
glycopeptide is primarily anchored by the side chain of residue Phe263 in the P1 pocket of DR4. Gly262 is positioned in the vicinity of the P1 pocket while
Ile260 and Ala261 are positioned on the binding site border. The b-D-GalHyl264 residue (Gal in ball and stick) in position P2 is extending out of the binding
site, as in the complex with Aq.
the oxygen and nitrogen atoms in the oxazole rings of the
Ile-Ala oxazole 2 and Ala-Gly oxazole 3 were located in the
vicinity of the corresponding atoms in the original amide bonds
after superposition with the modeled complex of the CII259-
273 glycopeptide 1 and Aq. In the case of the Gly-Phe oxazole
4, docking revealed that there was an oxygen–nitrogen atom
mismatch, although this would still allow for hydrogen-donor–
acceptor interactions (Fig. 3).
prepared by deprotonation of Schiff base 9 with sodium hydride
followed by reaction with phenylacetaldehyde (Scheme 2). After
work-up, the crude was immediately subjected to acidic aqueous
hydrolysis to give 10, followed by coupling with Boc-protected
glycine to afford 11 (50% yield from 9). Oxidation with Dess–
Martin periodinane to give 12 followed by cyclodehydration using
A model of the CII259-270 glycopeptide in complex with a
DR4 crystal structure22 was constructed to explore how the three
modified glycopeptides could bind in the DR4 binding site (Fig. 4).
As discussed above, the DR4 epitope is shifted three amino acids
towards the C terminus of CII259-273 compared to the complex
of CII259-273 and Aq. Assuming a similar binding mode as for
the CII259-273 glycopeptide 1, the Ile260-Ala261 and Ala261-Gly262
oxazole moieties in 2 and 3, respectively, would be positioned
on the border of the DR4 binding site. The Gly262-Phe263 oxazole
moiety in 4 would, however, be positioned in the vicinity of the P1
pocket in the modeled DR4 complex.
Synthesis of oxazole dipeptide mimetics
Oxazole dipeptide mimetics were synthesized suitably protected
for Fmoc-based solid-phase peptide synthesis. The Boc protected
Ile-Ala and Ala-Gly oxazole derivatives 523 and 624,25 (Scheme 1)
were synthesized as described earlier. From these intermediates,
the target molecules 7 and 8 were easily obtained in three
steps by hydrolysis of the methyl ester and Boc deprotection,
followed by Fmoc protection. The Gly-Phe oxazole derivative was
Scheme 2 Synthesis of Gly-Phe oxazole derivative 14. Reagents and
conditions: (a) i) NaH, phenylacetaldehyde, THF, -78 → 0 ◦C; ii) HCl
(1 M aq), THF, 0 ◦C; (b) Boc-Gly-OH, EDC·HCl, HOAt, Et3N, CH2Cl2,
rt; (c) Dess–Martin periodinane, CH2Cl2, rt; (d) PPh3, I2, Et3N, CH2Cl2,
0 ◦C → rt; (e) i) TFA, CH2Cl2, rt; ii) FmocOSu, NaHCO3, H2O, acetone,
rt.
Scheme 1 Synthesis of Ile-Ala and Ala-Gly oxazole derivatives 7 and 8.
Reagents and conditions: (a) i) LiOH, H2O, MeOH, rt; ii) TFA, CH2Cl2,
rt; iii) FmocOSu, NaHCO3, H2O, acetone, rt.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2931–2940 | 2933
©

View Article Online
triphenylphosphine, iodine, and triethylamine afforded the highly
functionalized oxazole intermediate 13.26 The Boc and t-Bu groups
were removed with TFA followed by N-Fmoc protection to furnish
the target Gly-Phe mimetic 14. Oxazole dipeptide mimetics 7, 8,
and 14 were finally incorporated into the CII259-273 glycopeptide
using solid-phase peptide synthesis to afford the corresponding
oxazole-modified glycopeptides 2, 3, and 4 (Fig. 2) in 28–33% yield
and >98% purity. The structures of these glycopeptidomimetics
were confirmed by 1H NMR spectroscopy and MALDI-TOF mass
spectrometry.
Immunological evaluation
The oxazole-modified glycopeptides were evaluated in two dif-
ferent assay systems (Fig. 5 and 6). First, competitive binding
experiments were performed to study glycopeptide binding to
the murine Aq and human DR4 proteins, respectively. Second,
the ability to stimulate Aq and DR4 restricted T-cell hybridomas
specific for CII259-273 carrying the GalHyl264 moiety were tested
using IL-2 secretion as read out.
In the Aq binding assay, only a weak tendency of binding was
observed for Ala-Gly oxazole 3 at the highest tested concentration
(500 mM) while Ile-Ala oxazole 2 and Gly-Phe oxazole 4 displayed
no affinity (Fig. 5). In line with these results, the Aq-restricted
T-cell hybridoma HCQ.3 also produced a weak response for
3 at 150 mM, while 2 and 4 did not induce T-cell responses.
Hence, the introduced oxazole moieties were not well tolerated
in the Aq system. It is noteworthy that both binding to Aq and
recognition by T-cell hybridoma HCQ.3 were lost completely when
the modified dipeptide moiety involved one of the residues that
anchor CII259-273 in Aq, that is Ile260 or Phe263 (Fig. 1 and 2).
The Ala261-Gly262 position in between the two anchoring residues
was, however, somewhat more tolerant towards the introduced
oxazole bioisostere. This position has previously been modified by
three different amide bond isosteres, where both the ketomethylene
and the (E)-alkene isostere were well tolerated in binding to
Aq.11 The ketomethylene isostere also induced strong responses
in a panel of T-cell hybridomas, and this further illustrates
the tolerance towards modifications of this position in CII259-
273.
In the DR4 binding experiment, it was found that all oxazole-
modified glycopeptides bound to DR4 (Fig. 6). However, Ile-
Ala oxazole 2, Ala-Gly oxazole 3, and unmodified glycopeptide
1 displayed significantly higher affinity than Gly-Phe oxazole 4
at the highest test concentration (500 mM). As illustrated in the
model of the complex between CII259-270 and DR4 (Fig. 4) the
oxazole moieties of 2 and 3 are positioned on the DR4 binding site
border. In contrast, the Gly-Phe oxazole moiety in 4 is found in the
vicinity of the P1 pocket, where it influences binding negatively.
Interestingly, both Ile-Ala oxazole 2 and Ala-Gly oxazole 3
elicited a strong response by the DR4-restricted T-cell hybridoma
hDR11.2 while Gly-Phe oxazole 4 was not recognized (Fig. 6).
Hybridoma mDR17.2 also recognized 2 strongly, while both
3 and 4 failed to activate this hybridoma. Thus, mDR17.2 seemed
to be more sensitive towards oxazole modifications closer to the
DR4 binding site than hDR11.2. Overall, recognition of oxazole-
modified glycopeptides 2–4 by the two T-cell hybridomas agreed
well with the DR4 binding data and with the model of the complex
between CII259-270 and DR4.
Fig. 5 Top: Inhibition of binding of a fixed concentration of biotinylated
CII259-273 Lys264 peptide to recombinant Aq protein upon incubation with
increasing concentrations of glycopeptides 1, 2, 3, or 4, respectively (the
modified position is indicated within parentheses). Aq-bound biotinylated
CII259-273 Lys264 peptide was detected in a time-resolved fluoroim-
munoassay using europium-labeled streptavidin. The points represent the
average of triplicates and error bars are set to one standard deviation.
Bottom: Response of the Aq restricted T-cell hybridoma HCQ.3 after
incubation with syngeneic spleen cells and increasing concentrations of
glycopeptides 1, 2, 3, or 4, respectively (the modified position is indicated
within parentheses). Glycopeptide binding to Aq proteins on the spleen
cells allows recognition by HCQ.3, resulting in IL-2 secretion into the
supernatant. Secreted IL-2 was subsequently quantified by a sandwich
ELISA using the DELFIA system. The points represent the average of
triplicates and error bars are set to one standard deviation. Note: The
standard deviation for 3 at 150 mM is too small to be seen visually.
Conclusions
Three oxazole analogues of the CII259-273 glycopeptide have
been designed and synthesized. Their ability to bind to the
CIA-associated murine Aq and RA-associated human DR4
proteins were evaluated, as well as their ability to induce an
immune response in T-cell hybridomas restricted by these class
II MHC proteins. The oxazole modifications were not well
tolerated in the Aq system where binding to Aq and the T-cell
response either was drastically reduced or completely abolished.
In the DR4 system, the Ile-Ala and Ala-Gly oxazoles bound well
to DR4 and also elicited strong immune responses by one or
both of the DR4-restricted T-cell hybridomas. Thus, these two
2934 | Org. Biomol. Chem., 2010, 8, 2931–2940
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
glycopeptidomimetics could be of interest for vaccination
studies in mice expressing human DR4 protein and human CD4
co-receptor, in particular if the oxazoles provide stabilization
towards enzymatic degradation. Moreover, it can be concluded
that neither Aq nor DR4 tolerated the oxazole bioisostere motif
when it included an anchoring residue. The results presented in
this paper and related studies11,12 form a basis for selecting altered
MHC ligands for in vivo evaluation in autoimmune disease models
with the long-term goal of developing a therapeutic vaccine for
RA.
Experimental section
Molecular modeling
The molecular graphics images were produced using the UCSF
Chimera package27 from the Resource for Biocomputing, Visu-
alization, and Informatics at the University of California, San
Francisco (supported by NIH P41 RR-01081).
Modeling of CII259-270 glycopeptide bound to Aq. Backbone
coordinates for CII259-270 were directly sampled from the pdb file
(pdb code 1MUJ28) of a crystal structure of a human CLIP peptide
bound to the Ab protein. The peptide sequences were manually
aligned so that Ile260 of CII259-270 overlapped with Met91 of CLIP
positioned in the P1 pocket of Ab. One hundred independent mod-
els were constructed for the side chains in CII259-270 that differed
from the ones in CLIP using a Boltzmann-weighted randomized
modeling procedure as implemented in MOE29 without energy
minimization. Selection of side chain conformations of CII259-
270 was guided by the indicated rotamer preference of the ligands
found in the template28 and two other related crystal structures
(pdb codes 1LNU30 and 2IAD31). These crystal structure proteins
had more than 90% sequence identity with Aq in the binding site.
The ability of the rotamers to participate in hydrogen bonds,
hydrophobic and p-interactions, as well as steric factors with
amino acids in the comparative model of the Aq protein11 were
also considered. Hydroxylation and subsequent glycosylation of
the Lys264 side chain and capping of the C terminal with N-methyl
amide were done manually. The CII259-270 glycopeptide was
energy minimized with the comparative model of the Aq protein in
two steps using MacroModel, within Maestro.32 First, the complex
was minimized with the protein backbone atoms constrained with
2
a force constant of 100 kJ mol-1 A and the maximum number
˚
of iterations set to 1000. Second, minimization was performed
without any constraints and the maximum number of iterations
set to 5000. All other parameters were at their default settings.
Fig. 6 Top: Inhibition of binding of a fixed concentration of biotinylated
CLIP peptide to recombinant DR4 protein upon incubation with increas-
ing concentrations of glycopeptides 1, 2, 3, or 4, respectively (the modified
position is indicated within parentheses). The points represent the average
of duplicates and error bars are set to one standard deviation. Middle
and bottom: Response of DR4 restricted T-cell hybridomas hDR11.2 and
mDR17.2 after incubation with syngeneic spleen cells and increasing
concentrations of glycopeptides 1, 2, 3, or 4, respectively (the modified
position is indicated within parentheses). The points represent the average
of triplicates and error bars are set to one standard deviation. The assays
are described further in the legend of Fig. 5 and in detail in the experimental
section.
Modeling of CII259-270 glycopeptide bound to DR4. The
CII259-270 glycopeptide modeled above (but not energy mini-
mized with Aq) was superposed with the collagen ligand in the DR4
crystal structure22 (pdb code 2SEB) so that Phe263 was positioned
in the P1 pocket of DR4 using the MOE29 software. Energy
minimization of the CII259-270/DR4 complex was performed
in two steps using MacroModel within Maestro,32 as described
above.
Docking study. The oxazole-modified CII259-267 glycopep-
tides, with N-methyl amide capped C terminals, were docked into
the binding site of the Aq comparative model11 using GOLD.33–35
The initial 3D conformations of the modified glycopeptides were
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2931–2940 | 2935
©

View Article Online
generated from smiles using the CORINA36 software. The binding
(250 ¥ 21.2 mm, 5 mm) using the same eluent as for the analytical
site was defined as a radius of 18 A from the x-, y-, and
HPLC, a flow-rate of 11 mL min-1, and detection at 214 nm.
˚
z-coordinates 49.76, 34.94, and 78.55, respectively. A docking
constraint was set to prevent orientation of Lys264 down in the P4
pocket, as biological results show that the galactose moiety points
out of the binding site to make contact with the T-cell receptors.7,8
More specifically, the Lys264 Ne atom of CII259-267 and the
bGlu14 Ca atom located in the P4 pocket of Aq were restricted
to be at least 10 A apart and at most 20 A apart. The dockings
were terminated after 50 genetic algorithm (GA) runs and
default parameter settings were used, except for the following:37
popsiz 140, n_islands 4, maxops 75000, niche_siz 6, migratewt 14,
initial_virtual_pt_match_max 6.0, and start_vdw_linear_cutoff 3.
The generated docking poses of the modified glycopeptides were
energy minimized in the Aq binding site using the Dock function
in MOE29 with the MMFF94x force field and only including Aq
General procedure for hydrolysis of methyl ester, Boc
deprotection, and Fmoc protection (7 and 8). LiOH (2 equiv.)
dissolved in H2O (1.8 mL/mmol methyl ester) was added
dropwise to methyl ester 5 or 6 (1 equiv.) dissolved in MeOH
(5.7 mL/mmol methyl ester) at 0 ◦C followed by stirring for
40 min. The cooling bath was removed and the solution was
stirred for an additional 80–95 min. The methanol was removed
under reduced pressure, ethyl acetate and H2O were added,
and the aqueous phase was acidified by addition of HCl (1 M,
aq). The phases were separated and the aqueous phase was
extracted with ethyl acetate. The combined organic phases were
dried over Na2SO4, filtered, and concentrated under reduced
pressure. The obtained crude carboxylic acids were used without
further purification. Trifluoroacetic acid (2 mL/mmol methyl
ester) was added to the carboxylic acid (1 equiv.) dissolved in
CH2Cl2 (9.7 mL/mmol methyl ester) followed by stirring for
1 h at rt. The solution was concentrated and the residue was
redissolved in chloroform and concentrated three times to remove
residual trifluoroacetic acid. The residue was dissolved in H2O
(5.7 mL/mmol methyl ester) and NaHCO3 (3.4 equiv.) was
added, followed by addition of acetone (5.7 mL/mmol methyl
˚
˚
˚
residues with atoms within 7 A of the ligand.
The docked and refined binding poses of the oxazole-modified
glycopeptides were compared to the modeled CII259-270 gly-
copeptide based on shape-based similarity of the sequence Gly259
-
Phe263-N. The top ranked pose for each ligand was selected based
on its ShapeTanimoto value calculated with ROCS.38
ester)
and
N-(9-fluorenylmethoxycarbonyloxy)succinimide
Chemistry
(1.05 equiv.). After stirring for 6 h at rt, the acetone was removed
under reduced pressure and CHCl3 was added. The aqueous
phase was acidified by addition of HCl (1 M, aq) and the phases
were separated. The aqueous phase was extracted with CHCl3,
and the combined organic phases were then washed with brine,
dried over Na2SO4, filtered and concentrated under reduced
pressure. The residue was purified by flash chromatography
(n-heptane–EtOAc–AcOH 4 : 1 : 1%→1 : 1 : 1%).
Compounds 523 and 624,25 were synthesized as described in the
cited literature references. All reactions were carried out under an
inert atmosphere with dry solvents under anhydrous conditions,
unless otherwise stated. CH2Cl2 was distilled from calcium hydride
whereas THF was distilled from potassium. DMF was distilled
˚
under reduced pressure and then dried over 3 A molecular sieves.
MeOH was dried over 3 A molecular sieves. TLC analysis was
˚
2-[1-(S)-(9H-Fluoren-9-ylmethoxycarbonylamino)-2-(S)-methyl-
butyl]-oxazole-4-carboxylic acid (7). The general procedure
described above applied to 523 (270 mg, 0.867 mmol) afforded
the Fmoc-protected building block 7 (273 mg, 75%) as a white
amorphous solid. [a]2D0 -49.0 (c 1.0 in MeOH). 1H NMR
(CD3OD): d 8.41 (s, 1H, oxazole CH), 7.76 (d, J = 7.5 Hz,
2H, Fmoc-arom), 7.66–7.61 (m, 2H, Fmoc-arom), 7.39–7.33 (m,
2H, Fmoc-arom), 7.30–7.25 (m, 2H, Fmoc-arom), 4.73–4.67
(m, 1H, NCH), 4.45–4.32 (m, 2H, Fmoc-CH2), 4.18 (t, J =
6.6 Hz, 1H, Fmoc-CH), 2.03–1.91 (m, 1H, CHCH3), 1.59–1.47
(m, 1H, CH2CH3), 1.29–1.16 (m, 1H, CH2CH3), 0.91 (t, J =
7.4 Hz, 3H, CH2CH3), 0.82 (d, J = 6.8 Hz, 3H, CHCH3); 13C
NMR (CD3OD): d 166.3, 164.1, 158.4, 145.6, 145.2 and 145.1
(1C), 142.6, 134.7, 128.7, 128.1 (splitted), 126.2, 120.9, 67.8
(Fmoc-CH2), 55.4 (NCH), 48.5 (Fmoc-CH), 39.4 (CHCH3), 26.3
(CH2CH3), 15.8 (CHCH3), 11.2 (CH2CH3). HRMS (ESI+) calcd
for [M + H]+ 421.1758, found 421.1745.
performed on silica gel 60 F254 (Merck) with detection by UV
light and staining with alkaline aqueous KMnO4 followed by
heating. Flash column chromatography was performed on silica
˚
gel (Matrex, 60 A, 35–70 mm, Grace Amicon). Optical rotations
were measured with a Perkin-Elmer model 343 polarimeter at
20 ◦C. 1H and 13C NMR spectra of the building blocks and their
intermediates were recorded at 298 K on a Bruker DRX-400
spectrometer at 400 and 100 MHz, respectively, and calibrated
using the residual peak of solvent as internal standard [CDCl3
(CHCl3 dH 7.26 ppm, CDCl3 dC 77.0 ppm), CD3OD (CD2HOD
dH 3.31 ppm, CD3OD dC 49.0 ppm) and DMF-d7 (DMF-d6 dH 2.92
and 2.75, DMF-d7 dC 34.89 and 29.76)]. J values are given in Hz.
First-order chemical shifts and coupling constants were obtained
from one-dimensional spectra; carbon and proton resonances were
1
assigned from COSY, NOESY and HETCOR experiments. H
NMR spectra of glycopeptides 2, 3 and 4 were recorded at 298 K
on a Bruker Avance spectrometer at 500 MHz in H2O–D2O (9 : 1)
1
with H2O (dH 4.76) as internal standard. COSY, TOCSY, H-
2-[1-(S)-(9H-Fluoren-9-ylmethoxycarbonylamino)-ethyl]-oxa-
zole-4-carboxylic acid (8). The general procedure described
above applied to 624,25 (267 mg, 0.988 mmol) afforded the Fmoc-
protected building block 8 (325 mg, 87%) as a white amorphous
13C-HSQC and ROESY experiments were used for assignment of
proton resonances and determination of chemical shifts. HRMS
data were recorded with electrospray ionization (ESI+). Analytical
reversed-phase HPLC was performed on a Beckman System Gold
1
solid. [a]2D0 -51.3 (c 1.0 in MeOH). H NMR (CD3OD): d 8.41
R
HPLC equipped with a Supelco Discoveryꢀ Bio Wide Pore C18
(s, 1H, oxazole CH), 7.77 (d, J = 7.5 Hz, 2H, Fmoc-arom),
7.67–7.61 (m, 2H, Fmoc-arom), 7.40–7.34 (m, 2H, Fmoc-arom),
7.33–7.26 (m, 2H, Fmoc-arom), 4.96–4.89 (m, 1H, NCH), 4.37 (d,
J = 6.7 Hz, 2H, Fmoc-CH2), 4.20 (t, J = 6.7 Hz, 1H, Fmoc-CH),
column (250 ¥ 4.6 mm, 5 mm) using a flow-rate of 1.5 mL min-1
and detection at 214 nm. Preparative reversed-phase HPLC was
R
performed using a Supelco Discoveryꢀ Bio Wide Pore C18 column
2936 | Org. Biomol. Chem., 2010, 8, 2931–2940
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
1.54 (d, J = 7.1 Hz, 3H, CH3); 13C NMR (CD3OD): d 167.4,
163.9, 158.1, 145.9, 145.2, 142.6, 134.6, 128.8, 128.1 (splitted),
126.2 (splitted), 120.9, 67.9 (Fmoc-CH2), 48.4 (Fmoc-CH), 46.4
(NCH), 19.1 (CH3). HRMS (ESI+) calcd for [M + H]+ 379.1288,
found 379.1274.
2-(2-tert-Butoxycarbonylamino-acetylamino)-3-oxo-4-phenyl-
butyric acid tert-butyl ester (12). Alcohol 11 (603 mg, 1.48 mmol)
dissolved in CH2Cl2 (15 mL) was treated with Dess–Martin
periodinane (4.6 mL, 2.22 mmol, 15 wt% solution in CH2Cl2) at rt.
After stirring for 75 min, Na2S2O3 (aq, satd, 5 mL) and NaHCO3
(aq, satd, 5 mL) were added followed by stirring for 20 min. The
phases were separated and the aqueous phase was extracted with
CH2Cl2. The combined organic phases were dried over Na2SO4,
filtered, and concentrated. Purification with flash chromatography
(n-heptane–EtOAc 3 : 1) afforded 12 (414 mg, 69%) as a colorless
oil. 1H NMR (CDCl3): d 7.36–7.25 (m, 3H, Ph), 7.22–7.19 (m, 2H,
Ph), 7.10 (d, J = 6.1 Hz, 1H, NHCH), 5.24 (d, J = 6.1 Hz, 1H,
NCH), 5.10 (br s, 1H, NHCH2), 4.07–3.97 (m, 2H, CH2Ph), 3.88–
3.83 (m, 2H, NCH2), 1.48 (s, 9H, C(CH3)3), 1.45 (s, 9H, C(CH3)3);
13C NMR (CDCl3): d 198.8, 169.2 (splitted), 164.5, 155.9 (broad),
132.6 (Ph), 129.6 (Ph), 128.7 (Ph), 127.4 (Ph), 84.4 (C(CH3)3), 80.4
(broad, C(CH3)3), 62.3 (NCH), 47.6 (CH2Ph), 44.0 (NCH2), 28.2
(C(CH3)3), 27.9 (C(CH3)3).
2-(R/S)-(2-tert-Butoxycarbonylamino-acetylamino)-3-(R/S)-
hydroxy-4-phenyl-butyric acid tert-butyl ester (11). Sodium hy-
dride (60% dispersion in mineral oil, 240 mg, 6.00 mmol) was
added in one portion at 0 ◦C to N-(diphenylmethylene)glycine
tert-butyl ester (1.53 g, 5.17 mmol) dissolved in THF (32 mL).
After stirring for 15 min, the solution was cooled to -78 ◦C, and
after an additional 15 min, freshly distilled phenylacetaldehyde
(0.90 mL, 8.05 mmol) was added dropwise over 10 min. Stirring
was continued at -78 ◦C for 45 min, and the mixture was
then allowed to reach 0 ◦C and was stirred an additional 2 h.
The reaction was quenched by addition of ice-cold NaHCO3
(aq, satd), the phases were separated and the aqueous phase
was extracted with EtOAc. The combined organic phases were
washed with brine, dried over Na2SO4, filtered and concentrated
under reduced pressure. The crude product was dissolved in THF
5-Benzyl-2-(tert-butoxycarbonylamino-methyl)-oxazole-4-car-
boxylic acid tert-butyl ester (13). Ketone 12 (396 mg, 0.98 mmol)
dissolved in CH2Cl2 (15 mL) was treated with PPh3 (516 mg,
1.97 mmol), I2 (503 mg, 1.98 mmol), and Et3N (0.32 mL,
3.95 mmol) at 0 ◦C. After stirring for 30 min, the cooling bath
was removed and the reaction mixture was stirred an additional
2 h. H2O (10 mL) was then added followed by stirring for 1 h. The
phases were separated and the aqueous phase was extracted with
CH2Cl2. The combined organic phases were dried over Na2SO4,
filtered, and concentrated. Purification with flash chromatography
(n-heptane–EtOAc 4 : 1) afforded 13 (315 mg, 83%) as a colorless
oil. 1H NMR (CDCl3): d 7.33–7.21 (m, 5H, Ph), 5.15 (br s,
1H, NH), 4.41–4.34 (m, 2H, NCH2), 4.32 (s, 2H, CH2Ph), 1.59
(s, 9H, C(CH3)3), 1.43 (s, 9H, C(CH3)3); 13C NMR (CDCl3): d
161.0, 159.6, 156.7, 155.4, 136.1 128.7, 128.6, 128.5, 126.9, 82.1
(C(CH3)3), 80.0 (broad, C(CH3)3), 37.8 (NCH2), 32.0 (CH2Ph),
28.2 (2C, C(CH3)3).
◦
(140 mL), and aqueous HCl (1 M, 17.5 mL) was added at 0 C
followed by stirring for 70 min. The THF was removed under
reduced pressure, CH2Cl2 was added and the aqueous phase was
neutralized by addition of NaHCO3 (aq, satd). The phases were
separated and the aqueous phase was extracted with CH2Cl2.
The combined organic phases were dried over Na2SO4, filtered,
and concentrated. The residue was filtered through a plug of
silica (CH2Cl2–MeOH 1 : 0→50 : 1) to afford amino alcohol 10
(1.01 g), which was used without further purification
in the next step. N-(tert-Butoxycarbonyl)glycine (703 mg,
4.01 mmol), N-ethyl-N¢-(3-dimethylaminopropyl)carbodiimide
hydrochloride (EDC·HCl, 807 mg, 4.21 mmol) and 1-hydroxy-
7-azabenzotriazole (HOAt, 573 mg, 4.21 mmol) were added to 10
(1.01 g, 4.0 mmol) dissolved in CH2Cl2 (8 mL) at rt. The mixture
was stirred for 5 min and then triethylamine (0.36 mL, 4.44 mmol)
was added dropwise. After stirring for 12 h, H2O (5 mL) was added
and the phases were separated. The aqueous phase was extracted
with CH2Cl2 and the combined organic phases were dried over
Na2SO4, filtered, and concentrated. The residue was purified by
flash column chromatography (n-heptane–EtOAc 4 : 1→1 : 1) to
afford 11 (1 : 1 mixture of diastereomers, 1.07 g, 50% from 9) as a
white foam. 1H NMR (CDCl3) (1 : 1 mixture of diastereomers): d
7.34–7.19 (m, 10H, Ph), 7.02 (d, J = 6.8 Hz, 1H, NHCH), 6.82 (d,
J = 8.4 Hz, 1H, NHCH), 5.24–5.17 (m, 1H, NHCH2), 5.17–5.09
(m, 1H, NHCH2), 4.64–4.59 (m, 2H, NHCH), 4.31 (ddd, J =
9.1, 4.6, and 2.3 Hz, 1H, CHOH), 4.22 (ddd, J = 7.2, 6.5, and
2.9 Hz, 1H, CHOH), 3.93–3.85 (m, 2H, NCH2), 3.85–3.79 (m,
2H, NCH2), 2.85 (dd, J = 13.8 and 4.6 Hz, 1H, CH2Ph), 2.80
(d, J = 6.5 Hz, 2H, CH2Ph), 2.71 (dd, J = 13.8 and 9.1 Hz, 1H,
CH2Ph), 1.48 (s, 9H, C(CH3)3), 1.47 (s, 9H, C(CH3)3), 1.46 (s, 9H,
C(CH3)3), 1.45 (s, 9H, C(CH3)3); 13C NMR (CDCl3) (1 : 1 mixture
of diastereomers): d 170.3, 169.8, 169.7, 168.7, 156.0 (2C, broad),
137.8 (Ph), 137.3 (Ph), 129.43 (Ph), 129.38 (Ph), 128.7 (Ph), 128.5
(Ph), 126.8 (Ph), 126.6 (Ph), 83.3 (C(CH3)3), 82.7 (C(CH3)3), 80.5
(C(CH3)3), 80.3 (C(CH3)3), 74.1 (CH2OH), 73.1 (CH2OH), 58.2
(NHCH), 56.2 (NHCH), 44.3 (2C, NHCH2), 40.4 (CH2Ph), 39.6
(CH2Ph), 28.3 (C(CH3)3), 28.3 (C(CH3)3), 28.0 (C(CH3)3), 28.0
(C(CH3)3).
5-Benzyl-2-[(9H-fluoren-9-ylmethoxycarbonylamino)-methyl]-
oxazole-4-carboxylic acid (14). Trifluoroacetic acid (0.8 mL) was
added to 13 (153 mg, 0.396 mmol) dissolved in CH2Cl2 (3.8 mL)
followed by stirring for 6 h at rt. The solution was concentrated,
and the residue was redissolved in CHCl3 and concentrated
three times to remove residual trifluoroacetic acid. The residue
was dissolved in H2O (3.5 mL) and NaHCO3 (166 mg, 1.98 mmol)
was added, followed by addition of acetone (3.5 mL) and N-(9-
fluorenylmethoxycarbonyloxy)succinimide (142 mg, 0.421 mmol).
After stirring for 16 h at rt, the acetone was removed under
reduced pressure and CHCl3 was added. The aqueous phase was
acidified to pH 2 by addition of HCl (1 M, aq) and the phases were
separated. The aqueous phase was extracted with CHCl3 and the
combined organic phases were then washed with brine, filtered,
and concentrated under reduced pressure. Purification by flash
chromatography (n-heptane–EtOAc–AcOH 4 : 1 : 1%→1:2 : 1%)
afforded 14 (104 mg, 57%) as a white amorphous solid. 1H NMR
(DMF-d7): d 8.11 (t, J = 5.6 Hz, 1H, NH), 7.94 (d, J = 7.5 Hz,
2H, Fmoc-arom), 7.76 (d, J = 7.4 Hz, 2H, Fmoc-arom), 7.48–7.41
(m, 2H, Fmoc-arom), 7.38–7.28 (m, 6H, Fmoc-arom and Ph),
7.28–7.21 (m, 1H, Ph), 4.49–4.43 (m, 4H, NCH2 and CH2Ph),
4.36–4.24 (m, 3H, Fmoc-CH and Fmoc-CH2); 13C-NMR (DMF-
d7): d 163.5, 162.6, 160.5, 157.6, 156.9, 144.5, 141.4, 137.3, 129.0,
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2931–2940 | 2937
©

View Article Online
Table 1 1H NMR chemical shifts for CII259-273 with Ile260y[oxazole]Ala261 (2)a
Residue
NH
Ha
Hb
Hg
Others
Gly259
3.87b
5.02
4.02b
4.63
4.26
Ile260y[oxazole]Ala261
8.96
8.56
8.12
8.46
2.06
1.45, 1.22, 0.87 (CH3)
1.59b
0.85 (Hd), 8.35 (oxazole)
Gly262
Phe263
Hyl264
3.09, 3.01
1.99, 1.75
7.21 (Hd), 7.26 (He and Hz)
4.01 (Hd), 3.15 and 2.96 (He), 7.61
(eNH2), c
Gly265
Glu266
Gln267
Gly268
Pro269
Lys270
Gly271
Glu272
Thr273
8.02
8.16
8.45
8.25
3.88b
4.34
4.36
4.10, 3.97
4.39
2.07, 1.91
2.10, 1.96
2.40b
2.34b
7.48 and 6.81 (dNH2)
2.24, 1.89
1.82, 1.75
1.97b
1.43b
3.56b (Hd)
8.44
8.34
8.20
8.07
4.28
1.66b (Hd), 2.97b (He), 7.49 (eNH2)
3.93b
4.45
2.14, 1.96
4.31
2.45b
1.16
4.31
a Measured at 500 MHz and 298 K in water containing 10% D2O with H2O (dH 4.76 ppm) as internal standard. b Degeneracy has been assumed. c Chemical
shifts for the galactose moiety: d 4.42 (H1), 3.90 (H4), 3.74 (H6), 3.67 (H5), 3.57 (H3), and 3.51 (H2).
128.8, 128.0, 127.4, 127.1, 125.7, 120.4, 66.8 (Fmoc-CH2), 47.3
(Fmoc-CH), 38.3 (NCH2), 31.7 (CH2Ph). HRMS (ESI+) calcd for
[M + H]+ 455.1601, found 455.1588.
Glycyl-L-isoleucyl-L-alanyly[oxazole]glycyl-L-phenylalanyl-5-
O-(b-D-galactopyranosyl)-5-hydroxy-L-lysylglycyl-L-glutam-1-yl-
L-glutaminylglycyl-L-prolyl-L-lysylglycyl-L-glutam-1-yl-L-threonine
(3). Synthesis was performed with building block 8 on solid
phase (51 mmol) according to the general procedure described
above. This afforded the trifluoroacetate salt of 3 (34.1 mg, 33%
yield based on the amount of resin used) as a white amorphous
solid. MS (MALDI-TOF) calcd 1663.81 [M + H]+, found 1663.78.
1H NMR data are given in Table 2.
General procedure for solid-phase glycopeptide synthesis.
Glycopeptides 2, 3, and 4 were synthesized in mechanically
agitated reactors on Tentagel-S-PHB-Thr(tBu)-Fmoc resins
using standard solid-phase methodology essentially as described
elsewhere.39 Couplings were performed in DMF and monitored
using bromophenol blue as indicator. Na-Fmoc amino acids
with standard side-chain protecting groups (4 equiv.) were
activated with 1-hydroxybenzotriazole (HOBt, 6 equiv.) and 1,3-
diisopropylcarbodiimide (DIC, 3.9 equiv.). (5R)-Na-(Fluoren-9-
ylmethoxycarbonyl)-Ne-benzyloxycarbonyl-5-O-(2,3,4,6-tetra-O-
acetyl-b-D-galactopyranosyl)-5-hydroxy-L-lysine11,40 (1.5 equiv.)
and building blocks 7, 8 and 14 (1.5 equiv) were activated
Glycyl-L-isoleucyl-L-alanyl-glycyly[oxazole]-L-phenylalanyl-5-
O-(b-D-galactopyranosyl)-5-hydroxy-L-lysylglycyl-L-glutam-1-yl-
L-glutaminylglycyl-L-prolyl-L-lysylglycyl-L-glutam-1-yl-L-threonine
(4). Synthesis was performed with building block 14 on solid
phase (50 mmol) according to the general procedure described
above. This afforded the trifluoroacetate salt of 4 (28.4 mg, 28%
yield based on the amount of resin used) as a white amorphous
solid after lyophilization. MS (MALDI-TOF) calcd 1663.78 for
[M + H]+, found 1663.80. 1H NMR data are given in Table 3.
with
O-(7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium
hexafluorophosphate (HATU, 1.5 equiv) and 2,4,6-collidine
(3.0 equiv.) and coupled for 24 h. Fmoc deprotection after each
coupling cycle was accomplished by treatment with 20% piperidine
in DMF for 10 min. The glycopeptides were cleaved from the
resins with trifluoroacetic acid–H2O–thioanisole–ethanedithiol
MHC-peptide binding assay
The inhibition assay was performed in 96-well microtiter assay
plates essentially as described elsewhere.15,41 Briefly, purified sol-
uble recombinant class II MHC proteins (Aq assay: 0.5 mM Aq;
DR4 assay: 1 mM DR4) were incubated with a fixed concentration
of a biotinylated tracer peptide (3 mM, Aq assay: CII259-273-bio
Lys264; DR4 assay: CLIP-bio with sequence KPVSKMRMAT-
PLLMQALPM) and increasing concentrations of test peptides 1,
2, 3, or 4 (Aq assay: 0, 4, 20, 100, 500 and 2500 mM; DR4 assay: 0,
0.8, 4, 20, 100 and 500 mM) in PBS for 48 h at room temperature.
This mixture also contained a cocktail of protease inhibitors
(CompleteTM, Boehringer, Mannheim). During this incubation,
new 96-well microtiter assay plates were precoated with mAb
(10 mg mL-1, Aq assay: Y3P 10 mAb; DR4 assay: L243 mAb)
◦
(35 : 2 : 2 : 1) for 3 h at 40 C with workup performed essentially
as described elsewhere.39 Purification by reversed-phase HPLC
was followed by deacetylation with NaOMe in MeOH (20 mM,
1 mL/mg peptide) for 2–3 h at rt (monitored by analytical
reversed-phase HPLC). Neutralization by addition of AcOH and
concentration under reduced pressure gave a residue that was
purified using reversed-phase HPLC followed by lyophilization.
Glycyl-L -isoleucyly[oxazole]alanyl-glycyl-L -phenylalanyl-5-
O-(b-D-galactopyranosyl)-5-hydroxy-L-lysylglycyl-L-glutam-1-yl-
L-glutaminylglycyl-L-prolyl-L-lysylglycyl-L-glutam-1-yl-L-threonine
(2). Synthesis was performed with building block 7 on solid
phase (50 mmol) according to the general procedure described
above. This afforded the trifluoroacetate salt of 2 (29.5 mg, 29%
yield based on the amount of resin used) as a white amorphous
solid after lyophilization. MS (MALDI-TOF) calcd 1649.77
[M + H]+, found 1650.03. 1H NMR data are given in Table 1.
◦
by incubating for 2 h at room temperature or overnight at 4 C
followed by blocking with PBS containing 2% low fat milk and
washing with PBS containing 0.1% Tween 20. 100 mL of the
incubated mixtures containing class II MHC proteins, biotinylated
2938 | Org. Biomol. Chem., 2010, 8, 2931–2940
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Table 2 1H NMR chemical shifts for CII259-273 with Ala261y[oxazole]Gly262 (3)a
Residue
NH
Ha
Hb
Hg
Others
Gly259
3.83b
4.18
5.13
4.73
4.31
Ile260
8.46
8.89
8.30
8.51
1.79
1.57
3.16
2.00, 1.74
1.40, 1.13, 0.83 (CH3)
1.62b
0.81 (Hd)
Ala261y[oxazole]Gly262
8.28 (oxazole),
Phe263
Hyl264
7.27 (Hd), 7.31 (He and Hz)
4.02 (Hd), 3.16 and 2.96 (He), 7.60
(eNH2), c
Gly265
Glu266
Gln267
Gly268
Pro269
Lys270
Gly271
Glu272
Thr273
7.98
8.25
8.47
8.26
3.89b
4.36
4.36
4.10, 3.97
4.39
2.09, 1.94
2.10, 1.96
2.42b
2.34b
7.48 and 6.81 (dNH2)
2.25, 1.90
1.82, 1.75
1.98b
1.43b
3.57b (Hd)
8.44
8.34
8.20
8.06
4.28
1.66b (Hd), 2.97b (He), 7.49 (eNH2)
3.94b
4.45
2.14, 1.96
4.30
2.45b
1.15
4.30
a Measured at 500 MHz and 298 K in water containing 10% D2O with H2O (dH 4.76 ppm) as internal standard. b Degeneracy has been assumed. c Chemical
shifts for the galactose moiety: d 4.41 (H1), 3.89 (H4), 3.74 (H6), 3.66 (H5), 3.60 (H3), and 3.51 (H2).
Table 3 1H NMR chemical shifts for CII259-273 with Gly262y[oxazole]Phe263 (4)a
Residue
NH
Ha
Hb
Hg
Others
Gly259
3.83b
4.16
4.30
4.45
Ile260
8.41
8.42
8.56
1.77
1.28
1.39, 1.11, 0.82 (CH3)
0.80 (Hd)
Ala261
Gly262y[oxazole]Phe263
4.76b (oxazole-CH2), 7.24 (Hd),
7.34 (He), 7.28 (Hz)
Hyl264
8.38
4.45
2.12, 1.92
1.72b
4.05 (Hd), 3.19 and 2.99 (He), 7.62
(eNH2), c
Gly265
Glu266
Gln267
Gly268
Pro269
Lys270
Gly271
Glu272
Thr273
8.73
8.15
8.41
8.20
3.95b
4.36
4.34
4.03, 3.91
4.38
2.05, 1.87
2.09, 1.94
2.31b
2.32b
7.46 and 6.79 (dNH2)
2.24, 1.89
1.82, 1.74
1.96b
1.42b
3.53b (Hd)
8.43
8.34
8.19
8.07
4.27
1.65b (Hd), 2.96b (He), 7.49 (eNH2)
3.93b
4.45
2.14, 1.96
4.31
2.45b
1.15
4.31
a Measured at 500 MHz and 298 K in water containing 10% D2O with H2O (dH 4.76 ppm) as internal standard. b Degeneracy has been assumed. c Chemical
shifts for the galactose moiety: d 4.43 (H1), 3.88 (H4), 3.74 (H6), 3.65 (H5), 3.60 (H3), and 3.52 (H2).
tracer peptide, and test peptides in PBS were then transferred
to the plates precoated with mAb followed by incubation for
2 h at room temperature or overnight at 4 ◦C. After washing
with PBS, containing 0.1% Tween 20, the amount of biotinylated
tracer peptide bound to the recombinant Aq or DR4 proteins
captured in the wells was quantified using the dissociation-
Michae¨lsson et al.42 with slight modifications. In brief, 5 ¥ 104
T-cell hybridomas were co-cultured with 5 ¥ 105 syngeneic spleen
cells and increasing concentrations of test peptides 1, 2, 3, or 4 (0,
0.01, 0.048, 0.24, 1.2, 6.0, 30, and 150 mM) in a volume of 200 mL in
96-well flat-bottom microtiter plates. After 24 h, 100 mL aliquots
of the supernatants were removed and frozen in order to kill any
transferred T-cell hybridomas. The contents of IL-2 in the culture
supernatant were measured by sandwich ELISA (purified rat anti-
mouse IL-2 as capturing mAb and the biotin rat anti-mouse IL-2
as detecting mAb, both from PharMingen, Los Angeles, CA) using
R
enhanced lanthanide fluoroimmunoassay (DELFIAꢀ) system
based on the time-resolved fluoroimmunoassay technique with
europium-labeled streptavidin (Wallac, Turku), according to the
manufacturers instructions. The Aq experiment was performed in
triplicates while the DR4 experiment was performed in duplicates.
R
the DELFIAꢀ system (Wallac, Turku, Finland), according to the
manufacturers instructions. Recombinant mouse IL-2 served as a
positive control. The experiment was performed in triplicates.
T-cell activation assay
The response of the glycopeptide-specific and Aq or DR4 restricted
T-cell hybridoma lines, that is, the amount of IL-2 secreted
following incubation of the hybridoma with antigen-presenting
spleen cells expressing Aq or DR4 and increasing concentrations
of the glycopeptides, was determined essentially as described by
Acknowledgements
This work was funded by grants from the Swedish Research
Council, the Swedish strategic science foundation (SSF), the
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 2931–2940 | 2939
©

View Article Online
Biotechnology Fund at Umea˚ University, the JC Kempe Foun-
dation (SJCKMS), and the EU project Masterswitch HEALTH-
F2-2008-223404.
18 M. Falorni, G. Giacomelli, A. Porcheddu and G. Dettori, Eur. J. Org.
Chem., 2000, 3217–3222.
19 A. Plant, F. Stieber, J. Scherkenbeck, P. Lo¨sel and H. Dyker, Org. Lett.,
2001, 3, 3427–3430.
20 E. Mann and H. Kessler, Org. Lett., 2003, 5, 4567–4570.
21 E. F. Rosloniec, T. Brandstetter, S. Leyer, F. W. Schwaiger and Z. A.
Nagy, J. Autoimmun., 2006, 27, 182–195.
22 A. Dessen, C. M. Lawrence, S. Cupo, D. M. Zaller and D. C. Wiley,
Immunity, 1997, 7, 473–481.
23 B. Wagner, D. Schumann, U. Linne, U. Koert and M. A. Marahiel,
J. Am. Chem. Soc., 2006, 128, 10513–10520.
24 A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno and D. R. Williams, Org.
Lett., 2000, 2, 1165–1168.
References
1 J. S. Courtenay, M. J. Dallman, A. D. Dayan, A. Martin and B.
Mosedale, Nature, 1980, 283, 666–668.
2 G. P. Astorga and R. C. Williams Jr, Arthritis Rheum., 1969, 12, 547–
554.
3 P. Stastny, J. Clin. Invest., 1976, 57, 1148–1157.
4 P. K. Gregersen, J. Silver and R. J. Winchester, Arthritis Rheum., 1987,
30, 1205–1213.
25 A. D. Wist, L. Gu, S. J. Riedl, Y. Shi and G. L. McLendon, Bioorg.
Med. Chem., 2007, 15, 2935–2943.
26 P. Wipf and C. P. Miller, J. Org. Chem., 1993, 58, 3604–3606.
27 E. F. Pettersen, T. D. Goddard, C. C. Huang, G. S. Couch, D. M.
Greenblatt, E. C. Meng and T. E. Ferrin, J. Comput. Chem., 2004, 25,
1605–1612.
5 P. H. Wooley, H. S. Luthra, J. M. Stuart and C. S. David, J. Exp. Med.,
1981, 154, 688–700.
¨
6 U. Brunsberg, K. Gustafsson, L. Jansson, E. Michae¨lsson, L. Ahrlund-
Richter, S. Pettersson, R. Mattsson and R. Holmdahl, Eur. J. Immunol.,
1994, 24, 1698–1702.
28 Y. Zhu, A. Y. Rudensky, A. L. Corper, L. Teyton and I. A. Wilson,
J. Mol. Biol., 2003, 326, 1157–1174.
7 J. Broddefalk, J. Ba¨cklund, F. Almqvist, M. Johansson, R. Holmdahl
and J. Kihlberg, J. Am. Chem. Soc., 1998, 120, 7676–7683.
8 B. Holm, J. Ba¨cklund, M. A. F. Recio, R. Holmdahl and J. Kihlberg,
ChemBioChem, 2002, 3, 1209–1222.
9 J. Ba¨cklund, A. Treschow, R. Bockermann, B. Holm, L. Holm, S.
Issazadeh-Navikas, J. Kihlberg and R. Holmdahl, Eur. J. Immunol.,
2002, 32, 3776–3784.
29 MOE (Molecular Operating Environment), version 2007.09, Chemical
Computing Group Inc., Montreal, Canada.
30 X. Liu, S. Dai, F. Crawford, R. Fruge´, P. Marrack and J. Kappler, Proc.
Natl. Acad. Sci. U. S. A., 2002, 99, 8820–8825.
31 C. A. Scott, P. A. Peterson, L. Teyton and I. A. Wilson, Immunity,
1998, 8, 319–329.
32 Maestro, version 9.1, Schro¨dinger, LLC, New York, NY.
33 G. Jones, P. Willett and R. C. Glen, J. Mol. Biol., 1995, 245, 43–53.
34 G. Jones, P. Willett, R. C. Glen, A. R. Leach and R. Taylor, J. Mol.
Biol., 1997, 267, 727–748.
10 B. Dzhambazov, K. S. Nandakumar, J. Kihlberg, L. Fugger,
R. Holmdahl and M. Vestberg, J. Immunol., 2006, 176, 1525–
1533.
11 I. E. Andersson, B. Dzhambazov, R. Holmdahl, A. Linusson and J.
Kihlberg, J. Med. Chem., 2007, 50, 5627–5643.
35 GOLD (Genetic Optimisation for Ligand Docking), version 3.1.1, The
Cambridge Crystallographic Datacenter, Cambridge, U. K.
12 I. E. Andersson, T. Batsalova, B. Dzhambazov, R. Holmdahl, A.
Linusson and J. Kihlberg, unpublished work.
36 CORINA
(COoRdINAtes),
Molecular
Networks
GmbH,
Na¨gelsbachstraße 25, 91052 Erlangen, Germany, Accessed Nov
13, 2009.
37 C. D. Andersson, I. E. Andersson, J. Kihlberg and A. Linusson,
unpublished work.
13 E. F. Rosloniec, K. B. Whittington, D. D. Brand, L. K. Myers and J. M.
Stuart, Cell. Immunol., 1996, 172, 21–28.
14 A. Corthay, J. Ba¨cklund, J. Broddefalk, E. Michae¨lsson, T. J.
Goldschmidt, J. Kihlberg and R. Holmdahl, Eur. J. Immunol., 1998,
28, 2580–2590.
38 ROCS (Rapid Overlay of Chemical Structures), version 2.4.2, OpenEye
Scientific Software, Inc.: Santa Fe, NM.
15 P. Kjelle´n, U. Brunsberg, J. Broddefalk, B. Hansen, M. Vestberg, I.
˚
39 J. Broddefalk, M. Forsgren, I. Sethson and J. Kihlberg, J. Org. Chem.,
1999, 64, 8948–8953.
Ivarsson, A. Engstro¨m, A. Svejgaard, J. Kihlberg, L. Fugger and R.
Holmdahl, Eur. J. Immunol., 1998, 28, 755–767.
40 B. M. Syed, T. Gustafsson and J. Kihlberg, Tetrahedron, 2004, 60,
16 J. Ba¨cklund, S. Carlsen, T. Ho¨ger, B. Holm, L. Fugger, J. Kihlberg, H.
Burkhardt and R. Holmdahl, Proc. Natl. Acad. Sci. U. S. A., 2002, 99,
9960–9965.
5571–5575.
41 C. M. Hill, A. Liu, K. W. Marshall, J. Mayer, B. Jorgensen, B. Yuan,
R. M. Cubbon, E. A. Nichols, L. S. Wicker and J. B. Rothbard,
J. Immunol., 1994, 152, 2890–2898.
17 E. C. Andersson, B. E. Hansen, H. Jacobsen, L. S. Madsen, C. B.
Andersen, J. Engberg, J. B. Rothbard, G. Sønderstrup McDevitt, V.
Malmstro¨m, R. Holmdahl, A. Svejgaard and L. Fugger, Proc. Natl.
Acad. Sci. U. S. A., 1998, 95, 7574–7579.
˚
42 E. Michae¨lsson, M. Andersson, R. Holmdahl and A. Engstro¨m,
Eur. J. Immunol., 1992, 22, 1819–1825.
2940 | Org. Biomol. Chem., 2010, 8, 2931–2940
This journal is
The Royal Society of Chemistry 2010
©