Multimerization of an apoptogenic TRAIL-mimicking peptide by using adamantane-based dendrons

Article abstract of DOI:10.1002/chem.201202415

We have developed a straightforward strategy to multimerize an apoptogenic peptide that mimics the natural tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) by using adamantane-based dendrons as multivalent scaffolds. The selective binding a

Full text of DOI:10.1002/chem.201202415

FULL PAPER
DOI: 10.1002/chem.201202415
Multimerization of an Apoptogenic TRAIL-Mimicking Peptide by Using
Adamantane-Based Dendrons**
[
a]
[a, c]
[a, d]
Giuseppe Lamanna,* Cristian R. Smulski,
Neila Chekkat,
[a, d]
[
b]
[b]
[a]
Karine Estieu-Gionnet, Gilles Guichard, Sylvie Fournel,
and Alberto Bianco*
Abstract: We have developed
a
demonstrating that the trimeric and
hexameric forms of the peptide exert
an increased affinity of about 1500-
and 20000-fold, respectively, relative to
the monomer. Moreover, only the tri-
meric and hexameric ligands were able
to induce cell death in TR2 expressing
cells (BJAB), thus confirming that a
multivalent form of the peptide is nec-
essary to trigger a substantial TR2-de-
pendent apoptotic response in vitro.
These results provide interesting in-
sight into the multivalency effect on bi-
ological ligand/receptor interactions for
future therapeutic applications.
straightforward strategy to multimerize
an apoptogenic peptide that mimics the
natural tumor necrosis factor-related
apoptosis-inducing ligand (TRAIL) by
using adamantane-based dendrons as
multivalent scaffolds. The selective
binding affinity of the ligands to
TRAIL receptor 2 (TR2) was studied
by surface plasmon resonance, thus
Keywords: adamantane
·
click
chemistry · dendrimers · multivalen-
cy · peptides · TRAIL
Introduction
bohydrate, peptide, small molecule, or any moiety that binds
[4]
to a receptor. The size, shape, flexibility, and valency of
the central scaffold play important roles in the biological ac-
tivity and the action mechanism of the recognition ele-
Polyvalent interactions that occur in biological systems are
characterized by the simultaneous binding of multiple li-
[1]
[5]
gands to multiple receptors. These interactions can induce
agonist mechanism) or inhibit (antagonist mechanism) a bi-
ment. Different scaffolds from various classes of structure,
(
such as proteins, liposomes, dendrimers, polymers, and solid
[2]
[6]
ological response differently to monovalent complexes. To
understand receptor signaling by multivalent interactions,
new strategies have emerged from synthetic organic chemis-
try and chemical biology by the design of multivalent li-
gands. The latter are composed of a central scaffold bear-
ing multiple copies of a recognition element, such as a car-
supports, have been designed. Because of their nanoscale
size and monodispersity, dendrimers and dendrons (wedge-
shaped dendrimer sections) have appeared as ideal scaffolds
[4,7]
for the design of multivalent ligands.
[
3]
We have recently reported the synthesis of the first- and
the second-generation (G1 and G2, respectively) adaman-
tane based-dendrons displaying the anti-inflammatory ibu-
profen moiety at their periphery in a trimeric or nonameric
+
+
[
a] Dr. G. Lamanna, Dr. C. R. Smulski, N. Chekkat, Prof. S. Fournel,
Dr. A. Bianco
[8]
arrangement, respectively. The multivalency effect on the
drug activity was studied, demonstrating that the multivalent
ibuprofen/dendron conjugates exert an enhanced anti-in-
flammatory activity in vitro relative to free ibuprofen. More-
over, in the proposed dendritic structures, the adamantane
building blocks with their rigid tetrahedral geometry, provid-
ed a three-dimensional tripodal arrangement of the attached
entities. This tripodal arrangement is an ideal recognition
motif for the cell surface, as several important receptors
CNRS, Institut de Biologie Molꢀculaire et Cellulaire
Laboratoire dꢁImmunologie et Chimie Thꢀrapeutiques
5 Rue Renꢀ Descartes, 67084 Strasbourg (France)
1
Fax : (+33)388610680
E-mail: lamanna@unistra.fr
a.bianco@ibmc-cnrs.unistra.fr
[
b] Dr. K. Estieu-Gionnet, Dr. G. Guichard
Institut Europꢀen de Chimie et de Biologie CBMN
Universitꢀ de Bordeaux I - CNRS UMR 5248
2
Rue Robert Escarpit, 33607 PESSAC (France)
[9]
+
form complexes with threefold geometry. On the basis of
these elements, we decided to explore the adamantane-
based dendrons as scaffolds to study multivalent ligand/re-
[
c] Dr. C. R. Smulski
Current address: University of Lausanne
Department of Biochemistry
Boveresses 155, 1066 Epalinges (Switzerland)
[10]
ceptor interactions. We chose an apoptogenic cyclic pep-
[
d] N. Chekkat, Prof. S. Fournel
tide, termed M1, as a ligand, which was developed at Affy-
Current address: UMR 7199 CNRS-Universitꢀ de Strasbourg
Laboratoire de Conception et Application de Molꢀcules Bioactives
4 Route du Rhin, BP 60024- 67401 Illkirch Cꢀdex (France)
[11]
max, which specifically recognizes human tumor necrosis
factor-related apoptosis-inducing ligand (TRAIL) receptor 2
(TR2) and once oligomerized mimics the function of its nat-
7
+
[
] These authors contributed equally to this work.
[12]
ural ligand TRAIL.
[
**] TRAIL=tumor necrosis factor-related apoptosis-inducing ligand.
TRAIL is a type II transmembrane protein that, like
other tumor necrosis factor (TNF) superfamily members,
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201202415.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
1
&
ÞÞ
These are not the final page numbers!

forms homotrimers. Each ligand subunit binds a receptor
molecule that induces an oligomerization process, which is
generally accepted to generate the minimal active unit that
[13]
triggers an effective signal.
the TNF receptor subgroup that contains an intracellular
death domain” and triggers apoptosis. There are five differ-
TRAIL receptors belong to
“
ent TRAIL receptors: two fully functional receptors (i.e.,
TR1 and TR2), two membrane-bound receptors that cannot
transmit an apoptotic signal (i.e., TR3 and TR4), and a solu-
[14]
ble receptor called osteoprotegerin (OPG).
Trimerization or multimerization of TR1 and TR2 results
in the formation of a death-inducing signaling complex
(
DISC), thus leading to caspase activation and cell death by
apoptosis. TRAIL is a ligand with enormous potential in
cancer therapy because of its unique characteristic to induce
apoptosis in tumor cells through the endogenous molecule
[15]
while sparing normal cells. Moreover, it is well established
that death-receptor oligomerization is central for efficient
formation of DISC, and structural studies provide novel in-
[16]
sight into this process.
Indeed, it has been shown that
Scheme 1. The multimerization of M1 peptide by using an amidation ap-
proach: synthesis of trimer 4. Boc=tert-butoxycarbonyl, DIEA=N-diiso-
propylethylamine, EDCꢃHCl=N-(3-dimethylaminopropyl)-N’-ethylcar-
bodiimide hydrochloride, NHS=N-hydroxysuccinimidyl, TFA=trifluoro-
acetic acid, TIPS=triisopropylsilane.
strategies that increase receptor oligomerization amplify
[17]
TRAIL-induced apoptosis.
Therefore, in the present
study, we have analyzed the multimerization effect of the
M1 peptide ligand on TR2 binding by using different gener-
ations of adamantane-based dendrons as multivalent scaf-
folds.
Results and Discussion
We have previously described the synthesis of the G1 ada-
mantane-based dendron 1 with three carboxylic acid groups
[18]
at its periphery. We decided to use this molecule as a scaf-
fold for the construction of the trivalent form of the M1
peptide ligand. We initially activated the acid functions of 1
by forming the corresponding NHS esters in the presence of
EDCꢃHCl, and then we coupled the peptide M1 through
amidation with its lysine residue. HPLC purification on a
C18 column afforded the trimeric ligand 3 in an overall yield
of 25% (Scheme 1). The Boc group at the focal point of 3
was subsequently cleaved with TFA in the presence of a
TIPS scavenger to obtain the necessary building block 4 for
the synthesis of the G2 dendron bearing M1 in a nonameric
arrangement. For this purpose, we decided to explore a syn-
thetic approach based on the click chemistry between azides
and alkynes.
Scheme 2. Synthesis of the first-generation azide dendron 6. Boc O=di-
tert-butyl dicarbonate, HOBt=1-hydroxybenzotriazole, TsCl=para-tol-
uenesulfonyl chloride.
2
In our first experiments, we prepared the G1 dendron 6
functionalized at its periphery with azide groups
7-carboxylate positions of adamantane 5 by means of amida-
tion with EDCꢃHCl/HOBt to afford the G1 azide dendron
6 in good yield after isolation by column chromatography
on silica gel (Scheme 2). Tetrasubstituted adamantane 5 was
synthesized, in turn, according to our protocols in six steps
from 1-bromoadamantane as an inexpensive starting materi-
(
Scheme 2). By starting from the commercially available 5-
amino-1-pentanol, we generated the desired azide-terminat-
ed chain following a four-step sequence: 1) the introduction
of a Boc-protecting group on the amine using Boc O and
2
[19]
triethylamine;
2) the activation of the hydroxy function
by tosylation with TsCl; 3) the conversion of the tosylate
[20]
[18]
al.
into an azide group by treatment with NaN ; and 4) Boc-de-
protection with a solution of HCl in dioxane. The resulting
azide-derivatized chain was finally introduced in the 3-, 5-,
At this point, for the development of the nonameric
ligand, we needed a second adamantane-based dendron to
bring an alkyne moiety at the focal point and bearing the
3
&
2
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Ligand Multimerization with Adamantane-Based Dendrons
FULL PAPER
All the compounds displayed
specific binding to human TR2,
and no detectable interaction
was observed with mouse TR2
(
see Figure 1 and Figure S1 in
the Supporting Information).
We also carried out SPR studies
with the peptide-free dendron 1
to demonstrate that the ligand/
receptor interaction is not influ-
enced by the adamantane-based
building block. Unspecific bind-
ing of 1 to the TR2 protein was
not recorded (see Figure S2 in
the Supporting Information).
SPR experiments with the
mono ACHUNTGRNENUGv alent peptide M1 provid-
ed an equilibrium dissociation
constant (or “binding constant”;
À7
K ) of 1.29ꢃ10 m, calculated
D
from the dissociation kinetic
À3 À1
rate (koff =3.07ꢃ10 s ) divid-
ed by the association kinetic
4
À1 À1
rate
(k =2.39ꢃ10 m s ;
on
Table 1). The ligand in its triva-
lent form 4 substantially im-
proved the binding affinity to
human TR2 by approximately
À11
1
500-fold (K =9.10ꢃ10 m).
D
Interestingly, this effect was
mainly due to a change on the
dissociation rate (koff =1.04ꢃ
Scheme 3. Synthesis of hexamer 10 by using the click-chemistry approach.
M1 peptide at its periphery. For this aim, we linked the
NHS ester 7, which terminates with an alkyne function, to
the free amine of 4, thus obtaining intermediate 8 under
basic conditions (Scheme 3). This latter derivative was in-
volved in a Huisgen cycloaddition with the azide dendron 6
by using a copper(I) catalyst in the presence of sodium as-
corbate. Purification by HPLC on a C18 column afforded
only hexameric ligand 9. We did not observe the formation
of the nonamer molecule probably because of steric hin-
drance. Thus, the third azide group of hexamer 9 is likely in-
accessible to alkyne 8. In the last synthetic step, the depro-
tection of the Boc group of 9 with TFA took place to afford
the corresponding hexamer 10 with the free amine at the
focal point.
Once we obtained multivalent ligands 4 and 10, we stud-
ied their binding affinity to TR2 receptor by surface plas-
mon resonance (SPR) relative to monomer M1. Recombi-
nant human TR2 and the mouse homologue of TR2 were
immobilized on the sensor chip and the peptide-based li-
gands were flushed onto this chip at different concentra-
tions. The recombinant human RANK protein, a related
protein from the TNF receptor family, was used as a con-
trol.
Table 1. Kinetic parameters of the M1 ligand in its monovalent, trivalent,
or hexavalent forms.
À1 À1
À1
Analyte
kon [m
s
]
k
off [s
]
D
K [m]
4
À3
À6
À7
À7
monomer M1 (2.39Æ1.91)ꢃ10 (3.07Æ0.02)ꢃ10
(1.29Æ0.04)ꢃ10
(9.10Æ1.92)ꢃ10
(7.16Æ1.35)ꢃ10
4
À11
À12
trimer 4
(1.15Æ0.22)ꢃ10 (1.04Æ0.33)ꢃ10
4
hexamer 10
(8.01Æ0.23)ꢃ10 (5.74Æ0.54)ꢃ10
À6 À1
10 s ) as the k value for trivalent ligand 4 (k =1.15ꢃ
on
on
4
À1 À1
10 m s ) is almost the same than the k value for mono-
on
mer M1.
Moreover, hexamer 10 exerted a 14-fold increased affinity
À12
to human TR2 (K =7.16ꢃ10 m) relative to trimer 4. Re-
D
constitution of the sensor chip surface after injection at high
concentrations of 10 was only effective by using 50 mm HCl
solution probably due to high stability of the formed ligand/
receptor complexes (Table 1). Here again, the higher bind-
ing affinity of 10 was determined by a strong decrease in the
À7 À1
koff value (koff =5.74ꢃ10 s ), while keeping an association
4
À1 À1
rate (k =8.01ꢃ10 m s ) similar to 4 and M1. The latter
on
results can be interpreted as an improved stability of the
complex ligand/receptor due to the multimerization of pep-
tide M1 that provides a multivalent interaction with TR2.
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
3
&
ÞÞ
These are not the final page numbers!

G. Lamanna, A. Bianco et al.
+
Figure 2. Cell-viability assay (MTS) in which BJAB TR2 cells were
treated with increasing concentrations of trivalent 4 or hexavalent 10
peptides for 16 h. The IC50 values were obtained by a sigmoidal, logistic,
four-parameter fitting and correspond to (31.23Æ1.59) and (255.64Æ
1
3.90) nm for the trivalent and hexavalent peptides, respectively.
[12a]
stantial TR2-dependent apoptotic response in vitro.
Un-
expectedly, hexavalent ligand 10 displayed a decreased ac-
tivity relative to trimer 4. Indeed, the IC50 value of 10 was
approximately 255 nm relative to the trivalent ligand 4,
which displayed an IC50 value of approximately 31 nm
(
Figure 2). This seemingly discordant result, when compared
with SPR data, can be due to unfavorable interactions be-
tween the six M1 peptides of 10 and TR2 on the cell surface
for reasons of steric hindrance, thus avoiding the formation
of a functional signaling complex. The TNF superfamily li-
gands are type-II transmembrane proteins that can be active
in a membrane-bound form or as soluble cytokines after
Figure 1. Sensorgrams of monomer M1, trimer 4, and hexamer 10 on
human TR2. The fittings correspond to the Langmuir binding model.
[9a]
proteolytic shedding.
The C-terminal, extracellular
In addition, we can suggest that the applied strategy for
ligand multimerization by using adamantane-based den-
drons does not alter complex formation but contributes to
increased stability during the dissociation phase.
Encouraged by these positive results, we decided to ana-
lyze the efficiency and specificity of trimer 4 and hexamer
domain of TRAIL forms a homotrimer and mediates recep-
[22]
tor binding and oliogomerization. This ligand-induced oli-
gomerization is generally accepted to trigger the signal cas-
cade, but increasing the oligomerization level improves the
[23]
effectiveness of the signal. This effect can be related to
the differential activity of membrane-bound ligands relative
to their soluble products. Our goal was to set up the basis
for mimicking a highly oligomeric ligand by using a G2 den-
dron that presented the TRAIL receptor 2 agonistic peptide
M1 on its periphery. However, contrary to what it was ex-
pected from SPR data, the hexameric peptide 10 displayed a
decreased cellular activity relative to the trimeric peptide 4.
This behavior can be a consequence of a biological incom-
patible three-dimensional arrangement of both trimers pres-
ent on the hexameric molecule, thus avoiding the formation
of a functional signaling complex. Unfortunately informa-
tion is not available about the three-dimensional distribution
of highly multimeric ligands within the TNF superfamily
members. Further synthetic studies are needed to find the
optimal length of the spacer arms between the peptide and
adamantane core to obtain effective apoptotic signaling,
even with multivalent ligands higher than trivalent forms.
10 to induce cell death in a TR2-dependent manner by
using an in vitro model of human Burkitt lymphoma
+
(
BJAB) that either does or does not express (BJAB TR2
À
[21]
or BJAB TR2 cells, respectively) TRAIL receptor 2. The
BJAB TR2 and BJAB TR2 cells were incubated with in-
+
À
creasing concentrations of either trivalent 4 or hexavalent
1
0 ligand and cell viability was analyzed after 16 h of incu-
+
bation. Whereas a significant death of BJAB TR2 cells was
observed upon treatment with both compounds, no cell
death was seen in BJAB TR2 cells, thus indicating high
À
TR2 selectivity of the multivalent peptides (Figure 2) and
no toxic effect at the assessed concentrations (see Figure S3
in the Supporting Information).
As it has been previously shown, the monomer M1 has no
+
effect on BJAB TR2 cell death, thus confirming that a
multivalent form of the peptide is necessary to trigger a sub-
&
4
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Ligand Multimerization with Adamantane-Based Dendrons
FULL PAPER
Conclusion
51.08, 29.44, 28.30, 28.18, 23.71 ppm; FTIR (neat) n˜ =3362, 2972, 2931,
À1
+
2
861, 2091, 1689, 1515, 1244, 1163 cm ; MS (ESI): m/z: 229.2 [M+H] ;
elemental analysis calcd for C10
C 52.92, H 8.38, N 24.59.
20 4 2
H N O : C 52.61, H 8.83, N 24.54; found:
Our studies have demonstrated that adamantane-based den-
drons are promising scaffolds because they provide a three-
dimensional tripodal arrangement for the multimerization of
peptides such as the TRAIL-mimicking ligand M1. Indeed,
we carried out a straightforward synthesis of trimer 4 and
hexamer 10. These multivalent ligands proved to specifically
bind the human TR2, showing an increased affinity (higher
avidity) relative to the monomer M1 in SPR assays. More-
over, both the trimeric and hexameric ligands could induce
selective cell death in TR2 expressing cells (BJAB). These
results together with the fact that the monomeric M1 pep-
5
-Azidopentylamine:
A solution of HCl in dioxane (4m, 6.5 mL,
2
6.28 mmol) was added to a solution of 5-azido-1-tert-butoxycarbonylami-
nopentane (1.00 g, 4.38 mmol) in CH Cl (8 mL). The resulting mixture
was stirred at room temperature for 1 h. Evaporation of the solvent in
2
2
vacuo afforded the product as a beige solid (0.56 g, 4.38 mmol, quant.).
1
H NMR (CDCl
3
, 400 MHz): d=8.27 (bs, 2H), 3.30 (t, J=6.4 Hz, 2H),
2
.93–3.08 (m, 2H), 1.77–1.87 (m, 2H), 1.60–1.67 (m, 2H), 1.47–1.55 ppm
1
3
(m, 2H); C NMR (CDCl , 125 MHz): d=50.94, 39.74, 28.14, 26.92,
3
À1
23.62 ppm; FTIR (neat) n˜ =2931, 2866, 2091, 1605, 1509, 1282, 1254 cm
MS (ESI): m/z: 129.3 [M+H] ; elemental analysis calcd for C H N
5 12 4
C 46.85, H 9.44, N 43.71; found: C 47.00, H 9.46, N 43.57.
;
:
+
[12a]
NHS triester 2: EDCꢃHCl (0.120 g, 0.63 mmol) and N-hydroxysuccini-
mide (0.066 g, 0.57 mmol) were added to a solution of 1 (0.130 g,
tide could not induce cell death
confirm that a multiva-
lent form of the peptide is necessary to trigger a substantial
TR2-dependent apoptotic response in vitro. Unexpectedly,
hexamer 10 displayed a decreased biological activity relative
to the trimeric ligand 4. This outcome is probably due to un-
favorable interactions between the six M1 peptides of 10
and TR2 on the cell surface for reasons of steric hindrance.
To overcome this limitation, in silico models are needed to
predict the most favorable distance between the ligands and
the adamantane core for the formation of a functional sig-
naling ligand/receptor complex. The insertion of a longer
spacer between the adamantane units might also render the
functional groups more accessible, thus generating higher
multimeric structures.
0
.18 mmol) in dry DMF (3 mL). The resulting mixture was stirred over-
night at room temperature. After the addition of AcOEt (60 mL), the or-
ganic phase was washed with saturated aqueous NH Cl solution (1ꢃ
0 mL) and water (1ꢃ60 mL), dried over Na SO , and filtered. Evapora-
tion of the solvent in vacuo yielded the desired product 2 as a beige solid
4
6
2
4
1
(
0.101 g, 0.10 mmol, 55%). H NMR (CD
3
OD, 300 MHz): d=3.17–3.24
(
m, 6H), 2.84 (s, 12H), 2.64 (t, J=7.2 Hz, 6H), 1.98 (s, 6H), 1.82–1.91
(m, 6H), 1.69–1.79 (m, 6H), 1.49–1.58 (m, 6H), 1.42 (s, 9H), 1.39–
1
3
1
1
2
1
3
.47 ppm (m, 6H); C NMR (CD OD, 75 MHz): d=178.03, 177.95,
70.22, 156.33, 79.78, 52.98, 44.57, 42.57, 42.52, 40.32, 31.54, 29.81, 28.88,
6.89, 26.57, 25.40 ppm; FTIR (neat) n˜ =3352, 2937, 2866, 1810, 1780,
729, 1633, 1522, 1201, 1161, 1065 cm ; MS (ESI): m/z: 1037.2 [M+Na]
À1
+
+
, 1015.0 [M+H] ; elemental analysis calcd for C H N O : C 56.85, H
6.66, N 9.67; found: C 57.01, H 6.69, N 9.23.
4
8
67
7
17
Trimer 4: Peptide M1 (0.065 g, 0.0323 mmol) and DIEA (50 mL) were
added to a solution of NHS triester 2 (0.008 g, 0.0079 mmol) in dry DMF
(
3 mL). The resulting mixture was stirred at room temperature for 48 h.
After removal of the solvent, the crude product was purified by prepara-
tive HPLC (t =13.8 min) to obtain trimer 3 as a white solid (0.024 g,
.0036 mmol, 45%). MS (ESI): m/z: 6710.7 [M+H] .
The obtained trimer 3 (0.019 g, 0.0028 mmol) was dissolved in TFA/H
Experimental Section
r
+
0
2
O/
General: All the starting materials, chemicals, and anhydrous solvents
TIPS (1.5 mL, 44:5:1), and the resulting mixture was stirred at room tem-
were obtained from commercial suppliers and used without purification.
1
13
perature for 3 h. Purification by preparative HPLC (t =12.9 min) afford-
r
Peptide M1 was purchased from PolyPeptide. H and C NMR spectra
were recorded using a Brucker 300, 400, or 500 MHz spectrometer; the
protons of the residual solvent were used to reference the chemical shift
in ppm. Coupling constants J are reported in Hertz (Hz), and the splitting
patterns are designated as s (singlet), d (doublet), t (triplet), q (quartet),
m (multiplet), and br (broad). IR spectra were measured on a Perkin–
Elmer Spectrum One ATR-FTIR spectrometer. ESI mass spectra were
recorded on ThermoFisher Finnigan LCQ Advantage Max instrument.
Elemental analyses were performed at the Micro-Analytical Facility Core
of the University of Strasbourg. All the peptide-based compounds were
purified by preparative HPLC on a Beckman instrument equipped with a
System Gold 166 NMP detector with a Macherey–Nagel Nucleodur 100–
ed the desired compound 4 as a white solid (0.018 g, 0.0027 mmol, 83%).
MS (ESI): m/z: 6609.9 [M+H] .
+
1,3,5-Tri(5-azido-pentylcarbamoyl)-7-tert-butoxycarbonylaminoadaman-
tane (6): EDCꢃHCl (0.348 g, 1.82 mmol) and HOBt (0.246 g, 1.82 mmol)
were added to a solution of 5 (0.100 g, 0.26 mmol) in dry DMF (3 mL).
The resulting mixture was stirred at room temperature for 30 min. A sol-
ution of 5-azidopentylamine (0.200 g, 1.56 mmol) and DIEA (0.70 mL,
3.90 mmol) in dry DMF (2 mL) was added to the reaction mixture, which
was stirred at 658C for 72 h. After the addition of AcOEt (80 mL), the
organic phase was washed with saturated aqueous NH
80 mL), saturated aqueous NaHCO solution (2ꢃ80 mL), and brine (1ꢃ
80 mL), dried over Na SO , and filtered. The solvent was removed in
4
Cl solution (2ꢃ
3
1
6 C18 column (gradient: 5–65% B in 20 min; eluent: A=H
2
O+0.1%
2
4
TFA, eluent B=MeCN+0.08% TFA; l=220 nm). The following com-
pounds were synthesized according to reported procedures: 5-O-tosyl-1-
vacuo and the crude product was purified by column chromatography on
silica gel with AcOEt as the eluent to yield the G1 azide dendron 6 as an
[
19,20]
1
tert-butoxycarbonylaminopentane,
1,3,5-tri(5-carboxy-pentylcarbamo-
orange oil (0.128 g, 0.18 mmol, 70%). H NMR (CDCl
3
, 300 MHz): d=
yl)-7-tert-butoxycarbonylaminoadamantane (1), and 1,3,5-tricarboxy-7-
tert-butoxycarbonylaminoadamantane (5).
5.92 (t, J=5.7 Hz, 3H), 4.77 (s, 1H), 3.15–3.27 (m, 12H), 1.98 (s, 6H),
[
18]
1.78–1.90 (m, 6H), 1.44–1.61 (m, 12H), 1.37 (s, 9H), 1.29–1.37 ppm (m,
1
3
6
4
2
3
H); C NMR (CDCl , 75 MHz): d=174.9, 154.1, 79.3, 51.8, 51.0, 43.1,
5
-Azido-1-tert-butoxycarbonylaminopentane: Sodium azide (0.54 g,
1.6, 39.21, 39.2, 28.9, 28.3, 28.28, 23.8 ppm; FTIR (neat) n˜ =3326, 2926,
855, 2086, 1701, 1630, 1519, 1244, 1158 cm ; MS (ESI): m/z: 714.8
M+H] ; elemental analysis calcd for C33
8
.39 mmol) was added to a solution of 5-O-tosyl-1-tert-butoxycarbonyl-
À1
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
aminopentane (1.00 g, 2.79 mmol) in dry DMF (20 mL). The resulting
mixture was stirred at 708C for 24 h in an atmosphere of argon. After the
addition of CH Cl (150 mL), the organic phase was washed with ice-cold
deionized water (3ꢃ150 mL), dried over Na SO , and filtered. After re-
moval of the solvent, the crude product was purified by column chroma-
tography on silica gel with CH Cl as the eluent to give the product as a
colorless oil (0.56 g, 2.45 mmol, 88%). H NMR (CDCl
+
[
H
55
N
13
O
5
: C 55.52, H 7.77, N
2
5.51; found: C 55.11, H 7.90, N 25.03.
2
2
5-Hexynoic NHS ester 7: EDCꢃHCl (0.256 g, 1.33 mmol) and N-hydrox-
ysuccinimide (0.123 g, 1.07 mmol) were added to a solution of commer-
cial 5-hexynoic acid (0.100 g, 0.89 mmol) in dry DMF (2 mL). The result-
ing mixture was stirred at room temperature for 24 h. After the addition
of AcOEt (60 mL), the organic phase was washed with saturated aqueous
2
4
2
2
1
3
, 300 MHz): d=
.78 (bs, 1H), 3.17 (t, J=6.7 Hz, 2H), 2.96–3.06 (m, 2H), 1.33 (s, 9H),
4
1
1
3
.25–1.55 ppm (m, 6H); C NMR (CDCl
3
, 125 MHz): d=155.82, 78.77,
4 2 4
NH Cl solution (1ꢃ60 mL) and water (1ꢃ60 mL), dried over Na SO ,
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
5
&
ÞÞ
These are not the final page numbers!

G. Lamanna, A. Bianco et al.
and filtered. Evaporation of the solvent in vacuo afforded the desired
product 7 as a yellow oil (0.164 g, 0.78 mmol, 88%). H NMR (CDCl ,
3
ty. G.L. is grateful to the Agence Nationale de la Recherche (NANO-
THER, project no. 2010 NANO 008 01). N.C. was given a grant by the
French “Ministꢄre de la Recherche”. We are also grateful to Andrew
Thorburn for providing the BJAB cell lines.
1
4
6
00 MHz): d=2.78 (s, 4H), 2.72 (t, J=7.4 Hz, 2H), 2.30 (dt, J=2.8 and
1
3
.8 Hz, 2H), 1.99 (t, J=2.8 Hz, 1H), 1.87–1.95 ppm (m, 2H); C NMR
, 125 MHz): d=169.07, 167.99, 82.29, 69.67, 29.43, 25.36, 23.13,
7.31 ppm; FTIR (neat) n˜ =3281, 2942, 2105, 1810, 1780, 1727, 1201,
(
CDCl
3
1
1
À1
+
064 cm ; MS (ESI): m/z: 210.1 [M+H] ; elemental analysis calcd for
: C 57.41, H 5.30, N 6.70; found: C 57.54, H 5.97, N 6.44.
[
1] M. Mammen, S. K. Choi, G. M. Whitesides, Angew. Chem. 1998,
10, 2908–2953; Angew. Chem. Int. Ed. 1998, 37, 2754–2794.
10 4
C H11NO
1
Alkyne trimer 8: Amine 4 (0.016 g, 0.0024 mmol) and DIEA (30 mL)
were added to a solution of 7 (0.001 g, 0.0048 mmol) in dry DMF (2 mL).
The resulting mixture was stirred at 458C for 48 h. After removal of the
[
2] a) M. D. Carrithers, M. R. Lerner, Chem. Biol. 1996, 3, 537–542;
b) I. A. Graef, L. J. Holsinger, S. Diver, S. L. Schreiber, G. R. Crab-
tree, EMBO J. 1997, 16, 5618–5628.
r
solvent, the crude product was purified by preparative HPLC (t =
[
3] a) L. L. Kiessling, J. E. Gestwicki, L. E. Strong, Curr. Opin. Chem.
Biol. 2000, 4, 696–703; b) L. L. Kiessling, J. E. Gestwicki, L. E.
Strong, Angew. Chem. 2006, 118, 2408–2429; Angew. Chem. Int. Ed.
2006, 45, 2348–2368.
1
5
3.6 min) to obtain the alkyne 8 as a white solid (0.009 g, 0.0013 mmol,
6%). MS (ESI): m/z: 6703.3 [M+H] .
+
Hexamer 10: Compound 6 (357 mg, 0.5 mmol), sodium ascorbate (99 mg,
.5 mmol), and copper(II) sulfate pentahydrate (75 mg, 0.3 mmol) were
added to a solution of alkyne trimer 8 (12 mg, 1.8 mmol) in THF/H
5 mL, 1:1) in an argon atmosphere,. The reaction mixture was stirred at
room temperature for 48 h in an argon atmosphere. After lyophilization,
the crude product was purified by preparative HPLC (t =15.5 min) to
0
[4] R. Roy, Curr. Opin. Struct. Biol. 1996, 6, 692–702.
[5] R. Roy, D. Pagꢀ, S. F. Perez, V. V. Bencomo, Glycoconjugate J. 1998,
15, 251–263.
[6] L. L. Kiessling, L. E. Strong, J. E. Gestwicki, Annu. Rep. Med.
Chem. 2000, 35, 321–330.
2
O
(
r
obtain hexamer 9 as a white solid (5 mg, 0.35 mmol, 71%). MS (ESI):
m/z: 14123.7 [M+H] .
[7] a) E. K. Woller, M. J. Cloninger, Org. Lett. 2002, 4, 7–10; b) M. J.
Cloninger, Curr. Opin. Chem. Biol. 2002, 6, 742–748; c) W. B. Turn-
bull, S. A. Kalovidouris, J. F. Stoddart, Chem. Eur. J. 2002, 8, 2988–
3000; d) O. Rolland, C. O. Turrin, A. M. Caminade, J. P. Majoral,
New J. Chem. 2009, 33, 1809–1824.
+
The obtained hexamer 9 (5 mg, 0.3 mmol) was dissolved in TFA/H
TIPS (0.5 mL, 44:5:1), and the resulting mixture was stirred at room tem-
perature for 2 h. Purification by preparative HPLC (t =14.1 min) afford-
ed the desired compound 10 as a white solid (4 mg, 0.3 mmol, 79%). MS
2
O/
r
[8] G. Lamanna, J. Russier, H. Dumortier, A. Bianco, Biomaterials
2012, 33, 5610–5617.
+
(
ESI): m/z: 14023.0 [M+H] .
[
9] a) J. L. Bodmer, P. Schneider, J. Tschopp, Trends Biochem. Sci. 2002,
SPR analysis: The Biacore 3000 system, sensor chip CM5, surfactant P20,
amine-coupling kit containing NHS, EDCꢃHCl, and ethanolamine were
from Biacore (Uppsala, Sweden). Biosensor assays were performed with
HBS-EP buffer as a running buffer (10 mm 4-(2-hydroxyethyl)-1-piperaz-
2
4
7, 19–26; b) S. E. Gibson, M. P. Castaldi, Angew. Chem. 2006, 118,
834–4837; Angew. Chem. Int. Ed. 2006, 45, 4718–4720; c) M. C.
Bewley, K. Springer, Y. B. Zhang, P. Freimuth, J. M. Flanagan, Sci-
ence 1999, 286, 1579–1583.
A
C
H
T
U
N
G
T
R
E
N
N
U
N
G
ineethanesulfonic acid (HEPES), 150 mm sodium chloride, 3 mm ethyl-
[
10] V. Humblet, P. Misra, K. R. Bhushan, K. Nasr, Y. S. Ko, T. Tsukamo-
to, N. Pannier, J. V. Frangioni, W. Maison, J. Med. Chem. 2009, 52,
544–550.
11] Y. M. Angell, A. Bhandari, M. N. De Francisco, B. T. Frederick,
J. M. Green, K. Leu, K. Leuther, R. Sana, P. J. Schatz, E. A. White-
horn, K. Wright, C. P. Holmes, Adv. Exp. Med. Biol. 2009, 611, 101–
ACHTUNGTRENNUNGe nediaminetetraacetic acid (EDTA), and 0.005% v/v surfactant P20;
pH 7.4). The low carboxylated dextran matrix (B1) was activated with a
À1
mixture of 0.2m EDCꢃHCl and 0.05m NHS (35 mL) at 5 mLmin
.
[
Human RANK, human TR2, and mouse TR2-Fc fusion proteins were
immobilized with the standard Biacore protocol at
a density of
À2
0
.05 pmolmm . Soluble peptides were diluted in the running buffer and
1
03.
injected (250 mL), followed by a dissociation phase (500 mL) at a flow
À1
[12] a) V. Pavet, J. Beyrath, C. Pardin, A. Morizot, M. C. Lechner, J. P.
Briand, M. Wendland, W. Maison, S. Fournel, O. Micheau, G. Gui-
chard, H. Gronemeyer, Cancer Res. 2010, 70, 1101–1110; b) K.
Gorska, J. Beyrath, S. Fournel, G. Guichard, N. Winssinger, Chem.
Commun. 2010, 46, 7742–7744.
rate of 30 mLmin . The sensor chip surface was regenerated after each
experiment by injecting HCl (20 mL, 25 mm (or 50 mm for the hexamer
1
0)). The kinetic parameters were calculated using Biaeval 4.1 software.
Analysis was performed by using the simple Langmuir binding model.
Specific binding profiles were obtained after subtracting the response
signal from the control RANK channel and from a blank-buffer injec-
tion. The fitting to each model was judged by the chi square c2 value and
the randomness of residue distribution compared to the theoretical
model. The use of others binding models did not give better fits.
[
13] a) S. G. Hymowitz, H. W. Christinger, G. Fuh, M. Ultsch, M. OꢁCon-
nell, R. F. Kelley, A. Ashkenazi, A. M. de Vos, Mol. Cell 1999, 4,
5
63–571; b) A. Ashkenazi, Nat. Rev. Cancer 2002, 2, 420–430.
[
[
14] D. Mꢀrino, N. Lalaoui, A. Morizot, P. Schneider, E. Solary, O. Mi-
cheau, Mol. Cell. Biol. 2006, 26, 7046–7055.
15] a) A. Ashkenazi, P. Holland, S. G. Eckhardt, J. Clin. Oncol. 2008, 26,
Cell-viability assay: BJAB cells were grown in RPMI1640 medium
(
Lonza, Basel Switzerland) supplemented with 10% fetal calf serum. The
3
621–3630; b) E. Cretney, K. Takeda, H. Yagita, M. Glaccum, J. J.
5
À1
cells were seeded in 96-well plates at 5ꢃ10 cellswell and treated with
the indicated concentrations of multivalent peptides for 16 h. Stock pepti-
des were prepared at a concentration of 1 mm in water, and all the dilu-
tions were performed in the cell-culture medium. The CellTiter 96 aque-
ous non-radioactive cell proliferation assay (MTS; Promega, WI, USA)
was used to estimate the percentage of viable cells by following the man-
ufacturerꢁs instructions. Nontreated cells were used to normalize the
Peschon, M. J. Smyth, J. Immunol. 2002, 168, 1356–1361; c) K.
Takeda, M. J. Smyth, E. Cretney, Y. Hayakawa, N. Kayagaki, H.
Yagita, K. Okumura, J. Exp. Med. 2002, 195, 161–169.
[
16] F. L. Scott, B. Stec, C. Pop, M. K. Dobaczewska, J. J. Lee, E. Mono-
sov, H. Robinson, G. S. Salvesen, R. Schwarzenbacher, S. J. Riedl,
Nature 2009, 457, 1019–1022.
17] a) D. Berg, M. Lehne, N. Mꢅller, D. Siegmund, S. Mꢅnkel, W.
Sebald, K. Pfizenmaier, H. Wajant, Cell Death Differ. 2007, 14,
[
1
00% value.
2
021–2034; b) T. W. Poh, S. Huang, J. L. Hirpara, S. Pervaiz, Cell
Death Differ. 2007, 14, 1813–1825.
[
18] G. Lamanna, J. Russier, C. Mꢀnard-Moyon, A. Bianco, Chem.
Commun. 2011, 47, 8955–8957.
Acknowledgements
[
19] B. E. Kane, M. K. Grant, E. E. El-Fakahany, D. M. Ferguson,
Bioorg. Med. Chem. 2008, 16, 1376–1392.
The authors thank the Centre National de la Recherche Scientifique
CNRS), the Agence Nationale de la Recherche (ANR-07-PCVI-0031–
1 (S.F.) and ANR-08-PCVI-0034–01 (G.G.)), and the Ligue Nationale
(
0
[20] T. Wang, H. An, B. D. Haly, P. Dan Cook, J. Heterocycl. Chem.
2000, 37, 687–696.
Contre le Cancer, Comitꢀ de la Dordogne (G.G.). C.R.S. was awarded a
grant by the Agence Nationale de la Recherche and Strasbourg Universi-
[21] L. R. Thomas, R. L. Johnson, J. C. Reed, A. Thorburn, J. Biol.
Chem. 2004, 279, 52479–52486.
&
6
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!

Ligand Multimerization with Adamantane-Based Dendrons
FULL PAPER
[
22] D. W. Banner, A. DꢁArcy, W. Janes, R. Gentz, H. J. Schoenfeld, C.
Broger, H. Loetscher, W. Lesslauer, Cell. 1993, 73, 431–445.
23] a) N. Holler, A. Tardivel, M. Kovacsovics-Bankowski, S. Hertig, O.
Gaide, F. Martinon, A. Tinel, D. Deperthes, S. Calderara, T. Schulth-
ess, J. Engel, P. Schneider, J. Tschopp, Mol. Cell. Biol. 2003, 23,
1428–1440; b) L. E. Haswell, M. J. Glennie, A. Al-Shamkhani, Eur.
J. Immunol. 2001, 31, 3094–3100.
Received: July 6, 2012
[
Revised: October 3, 2012
Published online: && &&, 0000
Chem. Eur. J. 2012, 00, 0 – 0
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
&
7
&
ÞÞ
These are not the final page numbers!

G. Lamanna, A. Bianco et al.
Peptide Design
G. Lamanna,* C. R. Smulski,
N. Chekkat, K. Estieu-Gionnet,
G. Guichard, S. Fournel,
A. Bianco* .................... &&&& — &&&&
Multimerization of an Apoptogenic
TRAIL-Mimicking Peptide by Using
Adamantane-Based Dendrons
TRAIL-blazers: Multimerization of an
apoptogenic peptide that mimics the
natural tumor necrosis factor-related
apoptosis-inducing ligand (TRAIL) by
using adamantane-based dendrons
remarkably increased the selective
binding affinity to TRAIL receptor 2
(TR2; see scheme). Only multimeric
forms of the ligands could induce cell
death in TR2-expressing cells, thus
triggering a substantial TR2-dependent
apoptotic response in vitro. This multi-
valency effect is important in biologi-
cal ligand/receptor interactions.
&
8
&
www.chemeurj.org
ꢂ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 0000, 00, 0 – 0
ÝÝ
These are not the final page numbers!