Dimeric self-assembly of pyridyl guanidinium carboxylates in polar solvents

Article abstract of DOI:10.1002/chem.201001861

A series of pyridyl guanidinium-carboxylates has been prepared and the dimeric self-assembly of these studied in H₂O/DMSO mixtures, principally using dilution isothermal calorimetry. Compounds 5 and 6, incorporating an aromatic ring in the "tet

Full text of DOI:10.1002/chem.201001861

FULL PAPER
DOI: 10.1002/chem.201001861
Dimeric Self-Assembly of Pyridyl Guanidinium Carboxylates in Polar
Solvents
Muhammad Irfan Ashiq,[a] Biniam F. Tesfatsion,[a] Francesca Gaggini,[a]
Sally Dixon,[a] and Jeremy D. Kilburn*[a, b]
Abstract: A series of pyridyl guanidini-
um–carboxylates has been prepared
and the dimeric self-assembly of these
studied in H2O/DMSO mixtures, prin-
cipally using dilution isothermal calo-
rimetry. Compounds 5 and 6, incorpo-
rating an aromatic ring in the “tether-
ing” region between the guanidinium
and carboxylate groups, demonstrate
the strongest dimerisation in neat
DMSO. X-ray crystal structures of 5
and 6 reveal two different dimerisation
architectures in the solid-state, but
both involve carboxylate–guanidinium
salt bridges as anticipated, and p–p in-
teractions. Compounds 10–16 incorpo-
rating peptidic fragments between the
guanidinium and carboxylate groups,
showed reduced dimerisation strength
with increased amino acid content, but
also sustained dimerisation under in-
creasingly aqueous conditions, up to
50% H2O/DMSO in the case of 14 and
15. The extent of our study in H2O/
DMSO mixtures was determined by
substrate solubility of 10–16, and not
the limit of self-assembly.
Keywords: calorimetry · dimeriza-
tion · molecular recognition · self-
assembly
chemistry
·
supramolecular
Introduction
tional amide hydrogen bond donors, and in which the bind-
ing conformation can be preorganised as a result of two in-
tramolecular hydrogen bonds between the pyridine lone
pair and adjacent guanidinium/amide NHs.[4] Receptor pre-
organisation resulted in entropy-driven binding of acetate,
even in polar media (Figure 1a, Ka =3900mꢀ1 in 10% H2O/
DMSO and 480mꢀ1 in 30% H2O/DMSO) and compares fa-
vourably with other carboxylate receptors.[2h,3e]
Guanidinium salts have been extensively employed as
motifs for carboxylate binding[1,2] and, in particular, are
often incorporated within synthetic receptors for amino acid
and peptide recognition.[3] Since ion-pair stability relies
upon polar interactions, competing solvation serves to
weaken binding in water and synthetic receptors for binding
carboxylate in aqueous media typically employ further, co-
operative binding sites for the anion, and often complex ar-
chitecture (e.g., macrocyclic) for pre-organisation of binding
site directionality. We recently described[2e] a structurally
simple pyridylguanidinium receptor, incorporating two addi-
[a] M. Irfan Ashiq, B. F. Tesfatsion, Dr. F. Gaggini, Dr. S. Dixon,
Prof. Dr. J. D. Kilburn
School of Chemistry, University of Southampton
Southampton, SO17 1BJ (UK)
[b] Prof. Dr. J. D. Kilburn
Present address:
School of Biological and Chemical Sciences, Queen Mary, University
of London
Mile End Road, London, E1 4NS (UK)
Fax : (+44)020-78822848
Figure 1. a) Binding of acetate (nBu4N+ salt) by a preorganised pyridyl
ꢀ
guanidinium receptor (PF6 salt) Ka1:1 =22000mꢀ1 (DG=ꢀ24.8 kJmolꢀ1
,
DH=ꢀ8.0 kJmolꢀ1
, )
TDS=16.8 kJmolꢀ1 in DMSO; 3900mꢀ1 (DG=
ꢀ20.5 kJmolꢀ1, DH=ꢀ7.4 kJmolꢀ1, TDS=13.1 kJmolꢀ1) in 10% H2O/
DMSO; and 480mꢀ1 (DG=ꢀ15.3 kJmolꢀ1, DH=ꢀ2.0 kJmolꢀ1, TDS=
13.3 kJmolꢀ1) in 30% H2O/DMSO;[2e] b)proposed (schematised) dimeri-
sation of a pyridyl guanidinium–carboxylate for quantification of inter-
molecular side chain interactions (···).
E-mail: j.kilburn@qmul.ac.uk
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201001861: Dilution ITC data
1
for 5, 6, 8, 10, 11 and 13–16, H NMR dilution data for 14 and
13C NMR spectra for compounds 3, 4, and 6–16.
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
12387

The development of well-defined, self-assembled systems
is an important goal with potential for creating functional
supramolecular systems, and the use of guanidinium–
optimal, giving the corresponding acid in reasonable yield
(70%). Amide coupling with methyl 4-(aminomethyl)ben-
zoate, followed by N-deprotection, then furnished 4, carbo-
diimide-mediated coupling of which with carbamoyl-activi-
ated thiourea 1 or 2 (see below), installed Cbz-protected
carboxyACHTUNGTRENNUNGlate binding to drive self-assembly is exemplified by
Schmuck and co-workersꢁ recent studies of polymer and
vesicle formation by self-assembling (zwitterionic) guanidi-
nium–carboxylates.[5] In order to develop functional applica-
tion of the efficient host–guest association depicted in Fig-
ure 1a, we have prepared a series of pyridyl guanidinium re-
ceptors, extended in structure to incorporate a tethered car-
boxylate group. The success of the simple guanidinium–car-
boxylate motif, Figure 1a, in competitive polar solvents
(H2O/DMSO) suggests that the proposed dimerisation of
pyridyl guanidinium–carboxylates, Figure 1b, could be
strengthened by (designed) complementary, intermolecular,
non-covalent interactions between adjacent tethering
strands (X), and would enable us to develop a simple supra-
molecular assembly driven primarily by the guanidinium–
carboxylate binding, in which the energetic contribution of
specific intermolecular binding interactions (X–X) can be
determined, and ultimately quantified in aqueous solvents.[6]
In particular, the detailed characterisation of amino acid
side-chain–side-chain interactions is essential for under-
standing the factors controlling stability of protein secon-
dary structural features (e.g., a-helices, b-sheets),[7] and for
allowing prediction of protein folding from knowledge of
the primary sequence alone, which remains a significant
challenge.[8] Additionally, intermolecular b-sheet interactions
are important for control of protein quaternary structure,
and are, for example, also involved in the protein aggrega-
tion implicated in neurodegenerative disease states.[9] Syn-
thetic b-sheet mimics have been used to explore inter-strand
recognition and sequence selectivity, largely through NMR
studies,[10] however, quantification of inter-stand interactions
remains key for the development of accurate models of mo-
lecular recognition events in formation of secondary b-struc-
ture.[11] Herein we describe a proof-of-concept study of the
self-assembly of pyridyl guanidinium–carboxylates, of the
general structure shown in Figure 1b, using dilution microca-
lorimetry to establish thermodynamic parameters of dimer
dissociation. The strength of dimerisation provides a mea-
sure of intermolecular interactions between the variable
component of the system (X), antiparallel alkyl, aryl or pep-
tide strands, relative to each other within the context of the
model complex.[12]
guanidine functionality.[14] Lastly, ester hydrolysis and Cbz-
removal gave guanidinium–carboxylates 5 and 6 (Scheme 1),
each incorporating aromatic tethers between the carboxylate
Scheme 1. Synthesis of b-alanine-, and 4-(aminomethyl)benzoate-derived
pyridyl guanidinium–carboxylates. i) Me3SiOK, THF, RT, 3–24 h; ii)
methyl 4-(aminomethyl)benzoate hydrochloride, iPr2EtN, PyBOP,
CH2Cl2, RT, 12 h; iii) 20% v/v CF3CO2H in CH2Cl2, RT, 2 h, then K2CO3
(aq); iv) 2, Et3N, EDC, CH2Cl2, RT, 24 h; v) H2 (1 atm), 10% w/w Pd/C,
CH2Cl2/MeOH, RT, 24 h; vi) 1, Et3N, EDC, CH2Cl2, RT, 48–72 h; vii)
LiBr, Et3N, 2% v/v H2O in CH3CN, RT, 24 h; viii) b-alanine methyl ester
hydrochloride, iPr2EtN, EDC, HOBt, CH2Cl2/DMF 4:1, RT, 72 h;
ix) 20% v/v CF3CO2H in CH2Cl2, RT, 5 h.
and guanidinium groups, potentially providing stabilisation
of dimeric self-assembly via p-stacking interactions. An ana-
logue, 8, with simple ethylene tether, was prepared similarly;
LiBr-mediated hydrolysis[15] of 3 was high yielding, although
partial N-deprotection took place upon work-up to give
mixed (separable) -NHBoc (32%) and -NBoc2 (57%) prod-
ucts, the former of which was subsequently utilised for cou-
pling of its free acid moiety with b-alanine methyl ester,
giving 7 after N-deprotection. Subsequent coupling with thi-
ourea 1, ester hydrolysis and Cbz removal, were straightfor-
ward (Scheme 1).
Results and Discussion
Synthesis: We chose to prepare pyridyl guanidinium–
carboxyACHTUNGTRENNUNGlates via two-directional functionalisation of a 6-ami-
nomethyl-2-carboxylate-substituted core pyridine. Partial re-
duction of diethyl pyridine-2,6-dicarboxylate, followed by
mesylation and nucleophilic substitution with tert-butyl imi-
nodicarboxylate, furnished this central motif, 3.[13]
Saponification of 3 was somewhat low yielding, under var-
ious conditions, and Me3SiOK-mediated hydrolysis proved
In order to undertake a comparative study of dimerisa-
tion, and ultimately to evaluate the strength of specific
amino acid side-chain–side-chain interactions, we used an
12388
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12387 – 12397

Dimeric Self-Assembly
FULL PAPER
identical synthetic strategy to prepare pyridyl guanidinium
carboxylates with a number of l-valine (10, 11 and 13,
Scheme 2) or l-leucine (14–16, see below)[16] residues incor-
to oligomer dissociation, which decreased non-linearly as
the concentration of injected molecule increased in the cell.
In each case, the thermal profile which resulted from dilu-
tion was consistent with homodimer dissociation, and non-
linear regression fitting to a monomer-dimer model enabled
estimation of the dimerisation constant (Kdim) and enthalpy
of dimerisation (DHdim).[20] Thermodynamic parameters for
dimeric self-assembly of pyridyl guanidinium–carboxylates,
in dimethyl sulfoxide (DMSO) and DMSO/H2O mixtures,
are reported in Table 1, although caution should be exer-
cised when interpreting the entropy–enthalpy balance in
cases of weak dimerisation since empirical studies of weak
affinity binding using ITC have shown that values of Ka
(and therefore DG) are reliable, but accuracy of DH (and
therefore TDS) may be strongly dependent upon the error
in sample concentration.[21]
Comparison of data for 5 and 6 indicates that dimerisa-
tion is only slightly enhanced by incorporation of the addi-
tional amide hydrogen bond donor in 6, both in DMSO (en-
tries 1 and 3) and 10% H2O/DMSO (entries 2 and 4). Di-
merisation of both 5 and 6 is significantly reduced by addi-
tion of only 10% (by volume) H2O to the DMSO solution,
but notably for 6 the reduction in dimerisation is due to a
much reduced entropic contribution, in contrast to the
simple guanidinium–acetate (host–guest) association shown
in Figure 1a, which is entropically driven in H2O/DMSO
mixtures. X-ray crystal structures were obtained for both 5
and 6 (Figure 2)[22] and reveal two possible dimerisation ar-
chitectures, both involving guanidinium–carboxylate salt
bridges. The crystal structure for 6 is very close to that en-
visaged for the self-assembled dimer (Figure 1b) and in-
volves a p–p stacking interaction[23] and a guanidinium–car-
boxylate interaction which involves just one of the carboxyl-
ate oxygens, participating in three hydrogen bonds [N5···O1
Scheme 2. Synthesis of l-valine-derived pyridyl guanidinium–carboxy-
lates. i) LiBr, Et3N, 2% v/v H2O in CH3CN, RT, 24 h; ii) H-l-Val-
OMe·HCl, iPr2EtN, EDC, HOBt, CH2Cl2/DMF 50:1, 72 h; iii) 20% v/v
CF3CO2H in CH2Cl2, RT, 5 h; iv) 2, Et3N, EDC, CH2Cl2, RT, 72 h; v)
Me3SiOK, THF, RT, 12–24 h; vi) H2 (1 atm), 10% w/w Pd/C, MeOH, RT,
6–48 h; vii) 1, Et3N, EDC, CH2Cl2, RT, 48–72 h; viii) 20% v/v CF3CO2H
in CH2Cl2, RT, 3 h, then K2CO3 (aq); ix) H-l-Val-l-Val-OMe, iPr2EtN,
PyBOP, CH2Cl2, RT, 12 h.
2.770 (3), N3···O1 3.180 (3) and N1···O1 3.084 (3) ꢂ].[24]
A
bridging water molecule stabilises the dimer via two hydro-
gen-bonding interactions, with the other oxygen of the
bound carboxylate [O1W···O2 2.734 (3) ꢂ] and pendant
amide [N6···O1W 2.884 (3) ꢂ].[25] Whilst accepting that the
solid-state structure is not a reliable indication of solution-
phase structure, it is interesting to note that similar involve-
ment of solvent bridging in the solution-phase dimerisation,
when water is present, could account for an increase in en-
thalpic contribution, and significant reduction of entropic
contribution, to dimerisation of 6 in 10% H2O/DMSO.
Compound 5 adopts a (solid state) dimer involving a “twist-
ed” guanidinium–carboxylate ion-pair interaction, clearly
energetically preferred to the interaction observed for 6 in
the absence of the pendant amide, emphasising that more
than one dimerisation architecture may be favoured, at least
in the solid-state, dependent upon small structural changes.
Structure 8, incorporating a short, flexible, b-alanine-de-
rived tether, demonstrated only weak dimerisation in
DMSO (Kdim =26mꢀ1, entry 5), suggesting that an intramo-
lecular guanidinium–carboxylate interaction is preferred if
conformationally allowed, as in this example. Compounds
10, 11 and 13, incorporating l-valine residues in the tether-
porated in the tethering region where intermolecular inter-
actions are anticipated upon self-assembly. Hence, 6-amino-
methyl-2-carboxylate-substituted pyridine 3 was in each case
2-C-functionalised via hydrolysis followed by amide cou-
pling, and 6-N-functionalised through Boc removal and
amine coupling with thiourea 1 or 2. Final ester hydrolysis
and guanidine deprotection steps furnished the target struc-
tures.
Dimerisation studies: The self-assembly of pyridyl guanidi-
nium–carboxylates 5, 6, 8, 10, 11 and 13–16 was investigated
using dilution isothermal titration calorimetry (ITC),[17–19] in
which sequential injection of aliquots of a solution of one of
these into a calorimeter cell initially containing only solvent,
resulted in a series of endothermic heat signals attributable
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12389

J. D. Kilburn et al.
Table 1. Thermodynamic data for the dimerisation of pyridyl guanidinium–carboxylates, determined from dilu-
sociation model. Comparison of
dimerisation of 11 and 13 in
DMSO (entries 9 and 11) indi-
cates that increased amino acid
content has led to reduced
tion ITC at 258C.[a]
Entry
Solvent
Kdim [mꢀ1
6622
(ꢁ2105)
(ꢁ31)
11876
10% H2O/DMSO 532
(ꢁ23)
DMSO
26(ꢁ4)
2409
(ꢁ197)
]
DGdim [kJmolꢀ1
]
ꢀDHdim [kJmolꢀ1
]
TDSdim [kJmolꢀ1
]
1
2
3
4
5
6
7
5
5
6
6
8
DMSO
10% H2O/DMSO 148
DMSO
N
ꢀ21.8
ꢀ12.4
8.8
4.8
(ꢁ0.2)
10.0
(ꢁ0.4)
11.5
(ꢁ0.1)
6.4
(ꢁ0.3)
16.4
(ꢁ0.2)
5.6
(ꢁ0.1)
(ꢁ0.5)
13.0
7.6
13.2
4.1
1.7
2.9
4.2
–
G
ACHTUNGTRENNUNG
(ꢁ1692) ꢀ23.2
ACHTUNGTRENNUNG
strength
of
dimerisation
R
ꢀ15.6
ꢀ8.1
ꢀ19.3
ꢀ9.8
–
ACHTUNGTRENNUNG
(DDG=8.7 kJmolꢀ1), and the
relative dimerisation strength of
l-leucine analogues 14 and 15
mirrors this (entries 13 and 17,
DDG=7.8 kJmolꢀ1). Increasing
the length of the peptide com-
ponent can increase the entrop-
ic penalty associated with dimer
formation, and in addition may
introduce additional intramo-
lecular interactions in the mon-
omeric species, although intro-
duction of a third amino acid
component does not have any
further effect on the overall
Gibbs free energy of dimerisa-
tion (15 vs 16). The energetic
cost of breaking these upon di-
merisation would lead to lower
dimerisation constants.[26] Im-
portantly, each member of the
structural series 14–16, which
ACHTUNGTRENNUNG
10 DMSO
ACHTUNGTRENNUNG
10 10% H2O/DMSO 54(ꢁ3)
ACHTUNGTRENNUNG
8[b]
9
10 20% H2O/DMSO
11 DMSO
–
–
400
–
N
ꢀ14.8
11.1
–
G
3.7
–
10[b]
11
12[c]
13
14
15
16
17
18
19
20
21
11 10% H2O/DMSO
13 DMSO
13 20% H2O/DMSO
14 DMSO
14 10% H2O/DMSO 68(ꢁ6)
14 20% H2O/DMSO 47
(ꢁ10)
14 50% H2O/DMSO 55(ꢁ8)
15 DMSO 38
(ꢁ11)
15 20% H2O/DMSO 25(ꢁ3)
15 50% H2O/DMSO 50
(ꢁ10)
16 DMSO
71(ꢁ2)
16 20% H2O/DMSO 42(ꢁ2)
–
12(ꢁ2)
ꢀ6.1
–
7.9
–
G
ꢀ1.8
–
–
877
N
ꢀ16.8
ꢀ10.5
ꢀ9.6
ꢀ9.9
ꢀ9.0
ꢀ7.9
ꢀ9.7
ꢀ10.6
ꢀ9.3
16.1
9.4
(ꢁ0.2)
3.5
(ꢁ0.2)
1.5
(ꢁ0.1)
2.1
(ꢁ0.2)
4.4
(ꢁ0.1)
1.0
(ꢁ0.1)
16.5
(ꢁ0.1)
14.5
(ꢁ0.2)
A
0.7
1.1
6.1
8.4
6.9
3.5
8.7
ꢀ5.9
ꢀ5.2
ACHTUNGTRENNUNG
E
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
T
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
[a] Calorimetric data were collected in duplicate except for entry 17, and are corrected for heat of mixing of
blank solvent, carried out under identical conditions. Kdim and DHdim are derived from non-linear regression
analysis of calorimetric data in terms of a monomer-dimer equilibrium model.[20] DGdim and TDSdim were calcu-
lated using DG=ꢀRTlnK=DHꢀTDS. All raw calorimetric data, and fits of corrected data to the dimer-disso-
ciation model, are provided in the ESI. [b] No evidence of dimer dissociation was observed in the dilution ITC
experiment. [c] A thermogram consistent with oligomer dissociation was obtained but could not be fitted to
the dimer-dissociation model.
ing region (entries 6–12) generally formed weaker dimers in
comparison with compounds 5 and 6, in identical solvent. In
the case of 10, diminution of binding upon switch of solvent
from DMSO to 10% H2O/DMSO, is of similar magnitude
to that seen for 5 and 6 (DDG=~7–10 kJmolꢀ1) and no
measurable self-assembly of 10 in 20% H2O/DMSO, or of
11 in 10% H2O/DMSO, took place (entries 8 and 10). Dilu-
tion ITC of dipeptide derivative 13 in DMSO was very
weak (Kdim =12mꢀ1, entry 11), and in 20% H2O/DMSO the
thermal profile obtained could not be fitted to a dimer-dis-
incorporates l-leucine (1–3 residues) in the tethering region,
underwent dimerisation which, although weak in the pres-
ence of water (<100mꢀ1), was maintained with increasing
water content, up to 50% H2O/DMSO for both 14 and 15.
Unfortunately 16 proved insoluble in solvent mixtures con-
taining >20% H2O. The sustained dimerisation of 14–16 in
increasingly aqueous media presumably relies primarily
upon intermolecular hydrophobic interactions, such as can
be anticipated between l-leucine side-chains, although the
corresponding energetic gain as the water content is in-
Figure 2. X-ray crystal structures of two-fold symmetric dimers of a) 6 and b) 5. Partial numbering for clarity.[22]
12390
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12387 – 12397

Dimeric Self-Assembly
FULL PAPER
creased must be balanced by the increasingly competitive
nature of the solvent towards ion-pair and hydrogen bond
formation.[27] As already stated, the values obtained for the
entropic and enthalpic contributions to the overall dimerisa-
tion free energy must be treated with caution when the di-
merisation is weak and so changes to the enthalpy-entropy
balance in the series of compounds 14–16 cannot be inter-
preted with confidence, although it is notable that the re-
sults obtained for all thermodynamic parameters were close-
ly reproducible in repeat experiments.
We have also determined the dimerisation constant for
1
self-assembly of 14 using H NMR dilution experiments in
[D6]DMSO,[28] since concentration dependence of the chemi-
cal shift of two protons was observed (Figure 3). An upfield
shift of amide proton Ha, but downfield shift of a guanidini-
um NH proton, is observed as the concentration of 14 in-
creases. Ha is presumably involved in intramolecular hydro-
gen-bonding in the monomeric species, and disruption of
this intramolecular interaction upon dimerisation, and poor
alignment of Ha with the carboxylate “guest” (as a conse-
quence of a displaced ion-pair interaction similar to that de-
scribed for 6, or an alternative dimer structure) fails to
“compensate” the chemical shift which consequently dem-
onstrates an overall movement upfield. The guanidinium
NH on the other hand, is probably involved in intermolecu-
lar hydrogen bonding to carboxylate in the dimer, and gives
the expected downfield shift at higher concentration. The
observed chemical shifts were fitted to a dimer equilibrium
model,[29] and the dimerisation constant obtained, Kdim
=
696mꢀ1 (ꢁ15%) is consistent with that determined using
calorimetry (Table 1, entry 13). Thus, the dimerisation con-
stant determined by microcalorimetry is substantiated by
NMR experiments. We also investigated whether NMR
could be used to provide useful NOE data and information
about the structure of the dimers in solution, but in practice
no useful data was obtained because of ambiguity in assign-
ing any observed NOE to intra- or intermolecular contacts.
Conclusion
Figure 3. Concentration dependence of the chemical shifts of an amide
proton (Ha) and
a guanidinium proton (NH), from guanidinium–
We have carried out a preliminary study of the self-assembly
of pyridyl guanidinium–carboxylates and demonstrated that
dimerisation of these structures occurs in competitive media
(up to 50% H2O in DMSO). These zwitterionic structures
are simple to prepare and the tethering region is readily
varied. The potential for incorporation, in this region, of
functional groups which assist dimerisation via p-stacking or
hydrophobic interactions has been demonstrated, the latter
leading to sustained dimerisation strength under increasingly
aqueous conditions such that the extent of our study in
H2O/DMSO mixtures was determined by substrate solubility
and not the limit of self-assembly. Although a detailed dis-
cussion of the entropy–enthalpy balance cannot be reliably
substantiated by thermodynamic parameters determined
from dilution ITC when dimerisation is weak, the technique
permits accurate calculation of the dimerisation constant
carboxylate 14, in [D6]DMSO at 400 MHz and 298 K.
AHCTUNGTRENNUNG
under these conditions and is therefore a valuable tool for
characterisation of individual intermolecular binding inter-
actions within the context of an overall weakly bound supra-
1
molecular assembly. Successful application of H NMR dilu-
tion spectroscopy for study of self-assembly in one case, cor-
roborates the use of dilution ITC for characterisation of the
dimerisation phenomenon and, importantly, also highlights
the value of the latter technique for application to mono-
mer–dimer equilibria which exist outside the concentration
1
range of H NMR spectroscopy. The dimerisation constants
of pyridyl guanidinium–carboxylates described herein are
lower than the previously reported association constants for
(host–guest) carboxylate recognition by pyridyl guanidinium
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12391

J. D. Kilburn et al.
software[29] and the error calculated on the basis of the average for each
dimerisation constant measured.
receptors, and for related zwitterionic guanidinium–carboxy-
ACHTUNGTRENNUNGlate systems, suggesting that we have not optimised the co-
Synthesis: The synthesis and full characterisation of thiourea 1 has been
reported by us previously.[2e] Benzyl (benzylcarbamothioyl)carbamate (2)
was prepared using the published procedure.[14]
operativity of possible intermolecular interactions in our
structures. Obtaining detailed structural information about
the dimers in solution has not proved possible for the struc-
tures we have investigated so far, but the X-ray crystal struc-
ture of 6 indicates that the anticipated dimerisation architec-
ture (Figure 1b) is plausible. We are currently addressing
design of the carboxylate binding pocket in order to gener-
ate improved dimeric, self-assemblies as plausible models of
b-sheet structure, in which the pyridyl guanidinium–carbox-
ylate region acts as a non-covalent “hairpin” linking antipar-
allel peptide strands and potentially enabling future, con-
text-specific characterisation of intermolecular amino acid
side-chain - side-chain interactions.
Methyl-4-[({[6-(aminomethyl)pyridin-2-yl]carbonyl}amino)methyl] ben-
zoate (4): To a stirred solution of ethyl 6-[[bis(tert-butoxycarbonyl)ami-
no]methyl]pyridine-2-carboxylate (3) (780 mg, 2.05 mmol) in THF
(10 mL) at room temperature, was added Me3SiOK (316 mg, 2.64 mmol)
in a single portion and the resulting mixture stirred for 12 h before cool-
ing to 08C and addition of H2O (50 mL) and CH2Cl2 (50 mL). The mix-
ture was adjusted to pH ~6 by dropwise addition of citric acid (0.1m
aqueous solution) at 08C and the biphasic mixture stirred vigorously at
08C for 1 h before separation and extraction of the aqueous phase with
EtOAc (3ꢃ50 mL). The combined organic phase was dried over MgSO4
and concentrated in vacuo to yield the carboxylic acid, 17, as a white
solid (540 mg, 70%), which was not purified further.
To a stirred suspension of 17 (500 mg, 1.4 mmol) and methyl 4-(amino-
methyl)benzoate hydrochloride (340 mg, 1.18 mmol) in CH2Cl2 (15 mL),
was added iPr2NEt (1.22 mL, 7 mmol) followed by PyBop (875 mg,
1.68 mmol). The resulting mixture was stirred at room temperature for
12 h, before addition of KHSO4 (25 mL of a 1m aqueous solution), sepa-
ration of the organic phase, and washing with K2CO3 (25 mL of a saturat-
ed aqueous solution) and brine (25 mL). The organic phase was dried
over MgSO4 and concentrated in vacuo. Purification by column chroma-
tography (SiO2 eluted with petrol/EtOAc 85:15) gave the amide coupled
Experimental Section
General techniques: Reagents and solvents were obtained from commer-
cial suppliers and if necessary dried and distilled before use. THF was
freshly distilled from sodium/benzophenone under argon. Toluene was
distilled from sodium under argon. Dichloromethane, acetonitrile and
triethylamine were freshly distilled from CaH2. N,N-Dimethylformamide
was distilled from CaH2 and stored over 4 ꢂ molecular sieves. Reactions
requiring a dry atmosphere were conducted in oven dried glassware
under nitrogen. Petrol refers to the fraction boiling between 40 and
608C.
product, 18, as
a white solid (685 mg, 98%). This amide (240 mg,
0.48 mmol) was treated with trifluoroacetic acid (10 mL of a 20% v/v so-
lution in CH2Cl2) at room temperature and the mixture stirred for 2 h.
CH2Cl2 was removed in vacuo and the residue washed with toluene (3ꢃ
10 mL), removal of the azeptrope with TFA was made in vacuo each
time. The crude oil was taken into CH2Cl2 (25 mL) and washed with
K2CO3 (25 mL of a saturated aqueous solution) before drying over
MgSO4 and concentration in vacuo, to give the title compound 4 as a col-
ourless oil without further purification (134 mg, 94%). 1H NMR
(300 MHz, CDCl3): d=8.61 (brs, 1H), 8.05 (d, J=8.0 Hz, 1H), 7.93 (d,
J=8.4 Hz, 2H), 7.76 (t, J=7.7 Hz, 1H), 7.37 (m, 4H), 4.66 (d, J=5.9 Hz,
2H), 3.99–3.85 (m, 2H), 3.85 ppm (s, 3H); 13C NMR (75 MHz, CDCl3):
d=166.81 (s), 164.51 (s), 160.12 (s), 149.07 (s), 143.69 (s), 138.01 (d),
129.91 (d), 129.18 (s), 127.51 (d), 124.11 (d), 120.73 (d), 52.07 (q), 47.03
(t), 43.00 ppm (t); IR (oil): n˜ = 3350 (w, br), 1716 (s), 1666 (s), 1522 (s),
1108 cmꢀ1 (m); MS (ES+): m/z (%): 300 (100) [M+H]+.
1H and 13C NMR spectra were recorded on Bruker AV300, AM300 or
DPX400 spectrometers. 1H chemical shifts are reported as values in ppm
referenced to residual solvent. The following abbreviations are used to
denote multiplicity and may be compounded: s=singlet, d=doublet, t=
triplet, q=quartet. 13C spectra were proton decoupled and referenced to
solvent. Signals are reported as s, d, t, q, depending on the number of di-
rectly attached protons (0, 1, 2, 3, respectively), this being determined by
DEPT experiments. Infrared spectra were recorded either as neat solids
or as oils on a Bio-Rad Golden Gate ATR FT-IR spectrometer fitted
with an ATR accessory. Absorptions are given in wavenumbers (cmꢀ1
)
and the following abbreviations used to denote peak intensities: s=
strong, m=medium, w=weak and/or br (broad). Low-resolution mass
spectra were recorded on a Micromass platform single quadrupole mass
spectrometer in methanol or acetonitrile. Accurate mass spectra were re-
corded on a double focusing mass spectrometer. Melting points were de-
termined in open capillary tubes using a Gallenkamp Electrothermal
melting point apparatus and are uncorrected.
Guanidinium–carboxylate 5: To a stirred mixture of amine 4 (244 mg,
0.82 mmol) and benzyl (benzylcarbamothioyl)carbamate (2) (295 mg,
0.98 mmol) in CH2Cl2 (15 mL), was added Et3N (0.34 mL, 2.46 mmol) fol-
lowed by EDC·HCl (472 mg, 2.46 mmol). The mixture was stirred at
room temperature for 24 h, before addition of KHSO4 (25 mL of a 1m
aqueous solution), separation of the organic phase, and washing with
K2CO3 (25 mL of a saturated aqueous solution) and brine (25 mL). The
organic phase was dried over MgSO4 and concentrated in vacuo. Purifica-
tion by column chromatography (SiO2 eluted with 1% MeOH in
CH2Cl2) gave the Cbz-guanidine product, 19, as a white solid (372 mg,
80%). To a stirred solution of 19 (165 mg, 0.29 mmol) in THF (10 mL) at
room temperature, was added Me3SiOK (56 mg, 0.44 mmol) in a single
portion and the resulting mixture stirred for 12 h before cooling to 08C
and addition of H2O (20 mL) and CH2Cl2 (20 mL). The mixture was ad-
justed to pH ~7 by dropwise addition of citric acid (0.1m aqueous solu-
tion) at 08C before separation and extraction of the aqueous phase with
CH2Cl2 (3ꢃ50 mL). The combined organic phase was dried over MgSO4
and concentrated in vacuo. Purification by column chromatography (SiO2
eluted with 3% MeOH in CH2Cl2) gave carboxylic acid product 20 as a
white solid (87 mg, 54%). A stirred solution of 20 (100 mg, 0.18 mmol) in
MeOH (5 mL) was treated with 10% Pd/C (19 mg, 0.18 mmol) and
stirred under H2 (1 atm) at room temperature for 24 h before filtration
through Celite, washing of the filter-cake with MeOH (10 mL), and con-
centration of the filtrate in vacuo. Recrystallisation from CH2Cl2/MeOH
gave the title compound 5 as a white solid (37 mg, 50%). Mp. 302–
3048C; 1H NMR (400 MHz, [D6]DMSO): d=11.0 (brs, 1H), 9.61 (brs,
Isothermal calorimetry dilution experiments: All data from ITC dilution
experiments are provided in the ESI. All experiments were performed in
duplicate using a VP-ITC isothermal titration calorimeter from MicroCal
at 258C. In a typical experiment sequential injections of concentrated
substrate solution (29ꢃ10 mL of a 3–40 mm solution) were made, at inter-
vals of 262 s, into the stirred calorimeter cell (1.6 mL volume) initially
containing solvent alone. Resulting data were corrected for heat of
mixing of solvent, carried out separately under identical conditions, and
analysed by non-linear regression using a corrected version of MicroCal
Origin 7.0 software.[20]
1H NMR dilution experiments: All data from 1H NMR dilution experi-
ments are provided in the Supporting Information. All experiments were
conducted on a Bruker DPX400 spectrometer at 298 K. [D6]DMSO of
commercial grade was used. 1H NMR spectra were recorded for a series
of fifteen samples of increasing concentration of compound 14, each of
approx. 5% incremental increase in dimer, from 0.13 up to 40.6 mm.
Each of these samples was prepared directly in a single dilution, from
40.6 mm stock solution. Changes in chemical shift of proton signals, as a
function of sample concentration, were analysed using NMRDil_Dimer
12392
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12387 – 12397

Dimeric Self-Assembly
FULL PAPER
1H), 8.0–7.92 (m, 2H), 7.51 (d, J=7.5 Hz, 1H), 7.43 (d, J=7.2 Hz, 2H),
7.30–7.19 (m, 5H), 6.73 (d, J=5.5 Hz, 2H), 4.65 (d, J=5.1 Hz, 2H), 4.42
(d, J=5.5 Hz, 2H), 4.34 ppm (d, J=6.1 Hz, 2H); 13C NMR (100.5 MHz,
[D6]DMSO): d=140.59 (d), 130.89 (d), 130.54 (d), 129.34 (d), 129.06 (d),
127.30 (d), 126.73 (d), 122.76 (d), 47.49 (t), 45.96 (t), 44.52 ppm (t); IR
(solid): n˜ = 3252 (br, w), 1667 (s), 1545 (m), 1375 cmꢀ1 (s); MS (ES+):
m/z (%): 418 (100) [M+H]+; HRMS (ES+): m/z: calcd for C23H24N5O3:
418.1879; found: 418.1875 [M+H]+.
1H), 7.63 (dd, J=7.7, 1.1 Hz, 1H), 4.31 (q, J=5.7 Hz, 2H), 3.60 (s, 3H),
3.56 (apparent q, J=6.4 Hz, 2H), 2.63 ppm (t, J=6.8 Hz, 2H); 13C NMR
(75 MHz, [D6]DMSO): d=171.67 (s), 163.50 (s), 152.27 (s), 148.79 (s),
138.94 (d), 124.86 (d), 121.06 (d), 51.47 (q), 42.14 (t), 35.05 (t), 33.72 ppm
(t); MS (ES+): m/z (%): 238 (100) [M]+; HRMS (ES+): m/z: calcd for
C11H16N3O3: 238.1186; found: 238.1187 [M]+.
Guanidinium–carboxylate 8: To
a stirred suspension of 7·TFA salt
(230 mg, ~1.04 mmol) in CH2Cl2 (40 mL) was added Et3N (500 mL,
3.12 mmol) followed by EDC·HCl (480 mg, 2.6 mmol) and thiourea 1
(456 mg, 1.23 mmol), each in a single portion. The resulting mixture was
stirred for 72 h at room temperature before addition of KHSO4 (50 mL
of a 1m aqueous solution), separation of the organic phase, and washing
with brine (50 mL). The organic phase was dried over MgSO4 and con-
centrated in vacuo. Purification by column chromatography (SiO2 eluted
with 1% MeOH in CH2Cl2 ! 5% MeOH in CH2Cl2) gave the Cbz-gua-
nidine product, 25, as a white solid (500 mg, 77%). A stirred solution of
25 (450 mg, 0.8 mmol) in THF (30 mL) was added to a suspension of
Me3SiOK (100 mg, 0.8 mmol) in THF (10 mL) and the resulting mixture
stirred at room temperature for 16 h before cooling to 08C and addition
of H2O (50 mL) and CH2Cl2 (50 mL). The mixture was adjusted to pH
~6 by dropwise addition of citric acid (0.1m aqueous solution) at 08C
and the resulting biphasic mixture stirred vigorously at 08C for 1 h
before separation of the organic phase, drying over MgSO4 and concen-
trated in vacuo. Purification by column chromatography (SiO2 eluted
with 2% MeOH in CH2Cl2 ! 5% MeOH in CH2Cl2) gave the carboxylic
acid, 26, as a white solid (300 mg, 75%). A stirred solution of 26 (330 mg,
0.6 mmol) in MeOH (50 mL) was treated with 10% Pd/C (60 mg) and
stirred under H2 (1 atm) at room temperature for 48 h before filtration
through Celite. Concentration of the filtrate in vacuo, followed by precip-
itation of the residue from Et2O/MeOH, gave the title compound 8 as a
white solid (200 mg, 78%). M. p. 218–2208C; 1H NMR (300 MHz,
[D4]MeOD): d=8.00–7.85 (m, 2H), 7.47 (d, J=8.0 Hz, 1H), 7.24–6.86
(m, 5H), 4.50 (brs, 2H), 4.29 (brs, 2H), 3.57–3.47 (m, 4H), 2.68 (t, J=
5.8 Hz, 2H), 2.42 ppm (m, 2H); 13C NMR (75 MHz, [D6]DMSO): d=
178.45 (s), 170.32 (s), 162.43 (s), 156.08 (s), 152.12 (s), 148.98 (s), 139.35
(s), 138.75 (d), 128.00 (d), 127.03 (d), 126.46 (d), 123.85 (d), 119.92 (d),
44.64 (t), 42.12 (t), 38.42 (t), 36.91 (t), 36.12 (t), 34.59 ppm (t); IR (solid):
n˜ = 3286 (w), 1644 (s), 1567 (s), 1537 (s), 697 (m), 607 cmꢀ1 (m); MS
(ES+): m/z (%): 427 (100) [M+H]+; HRMS (ES+): m/z: calcd for
C21H27N6O4: 427.2088; found: 427.2077 [M+H]+.
Guanidinium–carboxylate 6: To a stirred mixture of amine 4 (127 mg,
0.42 mmol) and thiourea 1 (187 mg, 0.5 mmol) in CH2Cl2 (4 mL), was
added Et3N (0.18 mL, 1.26 mmol) followed by EDC·HCl (242 mg,
1.26 mmol). The mixture was stirred at room temperature for 48 h,
before addition of KHSO4 (20 mL of a 1m aqueous solution), separation
of the organic phase, and washing with Na2CO3 (20 mL of a saturated
aqueous solution) and brine (20 mL). The organic phase was dried over
MgSO4 and concentrated in vacuo. Purification by column chromatogra-
phy (SiO2 eluted with 1% MeOH in CH2Cl2 ! 5% MeOH in CH2Cl2)
gave the Cbz-guanidine product, 21, as a white solid (235 mg, 88%). To a
stirred solution of 21 (168 mg, 0.26 mmol) in THF (15 mL) at room tem-
perature, was added Me3SiOK (51 mg, 0.4 mmol) in a single portion and
the resulting mixture stirred for 24 h before cooling to 08C and addition
of H2O (20 mL) and CH2Cl2 (20 mL). The mixture was adjusted to pH
~7 by dropwise addition of citric acid (0.1m aqueous solution) at 08C
before separation and extraction of the aqueous phase with CH2Cl2 (3ꢃ
20 mL). The combined organic phase was dried over MgSO4 and concen-
trated in vacuo. Purification by column chromatography (SiO2 eluted
with 5% MeOH in CH2Cl2) gave the carboxylic acid product, 22, as a
white solid (121 mg, 75%). A stirred solution of 22 (200 mg, 0.32 mmol)
in MeOH/CH2Cl2 (5 mL) was treated with 10% Pd/C (34 mg) and stirred
under H2 (1 atm) at room temperature for 24 h before filtration through
Celite, washing of the filter-cake with MeOH (15 mL), and concentration
of the filtrate in vacuo. Precipitation from Et2O/MeOH gave the title
compound 6 as a white solid (134 mg, 86%). M.p. 168–1708C; 1H NMR
(300 MHz, [D4]MeOD): d=8.04 (d, J=7.7 Hz, 1H), 7.93 (t, J=7.7 Hz,
1H), 7.82 (d, J=8.0 Hz, 2H), 7.49 (d, J=7.7 Hz, 1H), 7.25–7.12 (m, 7H),
4.56 (s, 2H), 4.54 (s, 2H), 4.25 (s, 2H), 3.44 (t, J=5.9 Hz, 2H), 2.41 ppm
(m, 2H); 13C NMR (100.5 MHz, [D6]DMSO): d=171.91 (s), 170.29 (s),
164.23 (s), 156.86 (s), 155.15 (s), 149.32 (s), 140.16 (s), 139.23 (s), 138.42
(d), 136.77 (s), 128.67 (d), 128.25 (d), 127.20 (d), 126.72 (d), 125.26 (d),
124.42 (d), 120.57 (d), 45.31 (t), 42.29 (t), 42.19 (t), 37.74 (t), 34.86 ppm
(t); IR (solid): n˜
=
3206 (m), 1667 (s), 1515 (s), 1297 cmꢀ1 (s); MS
(ES+): m/z (%): 489 (100) [M+H]+; HRMS
C26H29N6O4: 489.2250; found: 489.2237 [M+H]+.
AHCTUNGTRENNUNG
(ES+): m/z: calcd for
(S)-Methyl
2-(6-(aminomethyl)picolinamido)-3-methylbutanoate·TFA
salt (9). To a stirred solution of carboxylic acid 23a (100 mg, 0.36 mmol)
in CH2Cl2 (25 mL) was added iPr2NEt (0.2 mL, 1.26 mmol) followed by
EDC·HCl (165 mg, 0.9 mmol) and HOBt (146 mg, 1.08 mmol). To the re-
sulting mixture was then added a solution of l-valine methyl ester hydro-
chloride (120 mg, 0.72 mmol) in CH2Cl2/DMF (26 mL of a 25:1 mixture),
before stirring at room temperature for 72 h. CH2Cl2 (40 mL) and
KHSO4 (40 mL of a 1m aqueous solution) were then added before sepa-
ration of the organic layer, washing with Na2CO3 (30 mL of a saturated
aqueous solution) and brine (30 mL), drying over MgSO4 and concentra-
tion in vacuo. Purification by column chromatography (SiO2 eluted with
CH2Cl2/petrol 1:1 ! 2% MeOH in CH2Cl2) gave the amide coupled
product, 27, as a pale yellow oil (80 mg, 50%). A solution of 27 (56 mg,
0.15 mmol) in CH2Cl2 (10 mL) was treated with trifluoroacetic acid
(2 mL of a 20% v/v solution in CH2Cl2) at room temperature and the
mixture stirred for 5 h. CH2Cl2 was removed in vacuo and the residue
washed with toluene (3ꢃ2 mL), removal of the azeptrope with TFA was
made in vacuo each time to give the title salt 9 as a yellow oil (30 mg,
~70%). 13C NMR (75 MHz, CDCl3): d=174.34 (s), 164.34 (s), 151.04 (s),
149.26 (s), 139.09 (d), 125.31 (d), 122.95 (d), 58.47 (d), 52.62 (q), 43.38
(t), 31.16 (d), 19.22 (q), 18.57 ppm (q); IR (oil): n˜ = 2961 (w), 1668 (m),
1594 cmꢀ1 (m); MS (ES+): m/z (%): 266 (100) [M]+; HRMS (ES+): m/z:
calcd for C13H20N3O3: 266.1499; found: 266.1495 [M]+.
Methyl 3-(6-(aminomethyl)picolinamido)propanoate·TFA salt (7). To a
stirred solution of ethyl ester 3 (1.8 g, 5 mmol) in CH3CN (20 mL con-
taining 2% (v/v) of H2O), was added Et3N (2.2 mL, 15.8 mmol) followed
by LiBr (4.6 g, 52.9 mmol). The resulting mixture was stirred vigorously
at room temperature for 24 h before addition of H2O (100 mL) followed
by EtOAc (100 mL). The mixture was adjusted to pH 2–3 by dropwise
addition of HCl (2m aqueous) and the organic phase then separated and
washed with H2O (50 mL), before drying over MgSO4 and concentration
in vacuo. Purification by column chromatography (SiO2 eluted with 5%
MeOH in CH2Cl2) gave separable -NHBoc (23a, 404 mg, 32%) and
-NBoc2 (23b, 1.0 g, 57%) products, each as a white solid. To a stirred so-
lution of 23a (3.4 g, 9.7 mmol) in CH2Cl2 (50 mL) was added iPr2NEt
(5.95 mL, 34.2 mmol) followed by EDC·HCl (4.4 g, 24.3 mmol) and
HOBt (3.9 g, 29.1 mmol). To the resulting mixture was then added a solu-
tion of b-alanine methyl ester hydrochloride (2.6 g, 19.4 mmol) in
CH2Cl2/DMF (25 mL of a 4:1 mixture), before stirring at room tempera-
ture for 72 h. Solvents were then removed in vacuo. Purification by
column chromatography (SiO2 eluted with CH2Cl2/MeOH 20:1) gave the
amide product, 24, as a white gummy solid (2.3 g, 48%). A solution of 24
(500 mg, 1.14 mmol) in CH2Cl2 (50 mL) was treated with trifluoroacetic
acid (5 mL of a 20% v/v solution in CH2Cl2) at room temperature and
the mixture stirred for 5 h. CH2Cl2 was removed in vacuo and the residue
washed with toluene (3ꢃ2 mL), removal of the azeptrope with TFA was
made in vacuo each time to give the title salt 7 as a white gummy solid
(0.5 g, ~72%). 1H NMR (300 MHz, [D6]DMSO): d=9.19 (t, J=6.2 Hz,
1H), 8.40 (brs, 3H), 8.04 (t, J=7.7 Hz, 1H), 7.98 (dd, J=7.7, 1.1 Hz,
Guanidinium–carboxylate 10: To
a stirred suspension of 9·TFA salt
(40 mg, ~0.15 mmol) in CH2Cl2 (15 mL) was added Et3N (0.1 mL,
0.7 mmol) followed by EDC·HCl (55 mg, 0.2 mmol) and benzyl (benzyl-
carbamothioyl)carbamate (2) (67 mg, 0.2 mmol), each in a single portion.
The resulting mixture was stirred for 72 h at room temperature before re-
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12393

J. D. Kilburn et al.
moval of solvents in vacuo. Purification by column chromatography
(SiO2 eluted with CH2Cl2 ! 3% MeOH in CH2Cl2) gave the Cbz-guani-
dine product, 28, as a white solid (1.35 g, 75%). A stirred solution of 28
(350 mg, 0.66 mmol) in THF (25 mL) was added to a suspension of
Me3SiOK (85 mg, 0.66 mmol) in THF (25 mL) and the resulting mixture
stirred at room temperature for 16 h before cooling to 08C and addition
of H2O (50 mL) and CH2Cl2 (50 mL). The mixture was adjusted to pH
~6 by dropwise addition of citric acid (0.1m aqueous solution) at 08C
and the resulting biphasic mixture stirred vigorously at 08C for 1 h
before separation of the organic phase, drying over MgSO4 and concen-
trated in vacuo. Purification by column chromatography (SiO2 eluted
with 2% MeOH in CH2Cl2 ! 5% MeOH in CH2Cl2) gave the carboxylic
acid, 29, as a white solid (310 mg, 71%). A stirred solution of 29 (300 mg,
0.6 mmol) in MeOH (30 mL) was treated with 10% Pd/C (64 mg) and
stirred under H2 (1 atm) at room temperature for 48 h before filtration
through Celite. Concentration of the filtrate in vacuo, followed by precip-
itation of the residue from Et2O/MeOH, gave the title compound 10 as a
white solid (110 mg, 50%). M.p. 260–2628C; 1H NMR (300 MHz,
[D4]MeOD): d=7.96 (d, J=8.0 Hz, 1H), 7.89 (tt, J=7.5, 2.0 Hz, 1H),
7.43 (d, J=7.5 Hz, 1H), 7.26–7.18 (m, 5H), 4.60 (s, 2H), 4.42 (s, 2H),
4.35 (brd, J=4.5 Hz, 1H), 2.20 (m, 1H), 0.93 (d, J=7.0 Hz, 3H),
0.90 ppm (d, J=7.0 Hz, 3H); 13C NMR (75 MHz, [D4]MeOD): d=177.94
(s), 165.15 (s), 158.42 (s), 156.27 (s), 150.87 (s), 139.92 (d), 137.64 (s),
129.86 (d), 128.94 (d), 128.30 (d), 125.58 (d), 122.15 (d), 61.50 (d), 46.64
(t), 46.16 (t), 33.50 (d), 20.23 (q), 1858 ppm (q); IR (solid): n˜ = 3188 (br,
w), 1651 (s), 1393 cmꢀ1 (m); MS (ES+): m/z (%): 384 (100) [M+H]+;
HRMS (ES+): m/z: calcd for C20H26N5O3: 384.2030; found: 384.2036
[M+H]+.
azeptrope with TFA was made in vacuo each time. The crude oil was
taken into CH2Cl2 (50 mL) and washed with K2CO3 (50 mL of a saturat-
ed aqueous solution) before drying over MgSO4 and concentration in
vacuo, to give the amine product 32 as a colourless oil without further
purification (886 mg, 75%). To a stirred mixture of amine 32 (1.2 g,
6.63 mmol) and thiourea 1 (2.95 g, 7.96 mmol) in CH2Cl2 (50 mL), was
added Et3N (2.7 mL, 19.9 mmol) followed by EDC·HCl (3.82 g,
19.9 mmol). The mixture was stirred at room temperature for 48 h,
before addition of KHSO4 (50 mL of a 1m aqueous solution), separation
of the organic phase, and washing with K2CO3 (50 mL of a saturated
aqueous solution) and brine (50 mL). The organic phase was dried over
MgSO4 and concentrated in vacuo. Purification by column chromatogra-
phy (SiO2 eluted with 1% MeOH in CH2Cl2) gave the Cbz-guanidine
product, 33, as a white solid (1.57 g, 46%). To a stirred solution of 33
(1.54 g, 3.06 mmol) in THF (20 mL) at room temperature, was added
Me3SiOK (590 mg, 4.6 mmol) in a single portion and the resulting mix-
ture stirred for 12 h before cooling to 08C and addition of H2O (25 mL)
and CH2Cl2 (25 mL). The mixture was adjusted to pH ~6 by dropwise ad-
dition of citric acid (0.1m aqueous solution) at 08C and the biphasic mix-
ture stirred vigorously at 08C for 1 h before separation and extraction of
the aqueous phase with CH2Cl2 (3ꢃ100 mL). The combined organic
phase was dried over MgSO4 and concentrated in vacuo. Purification by
column chromatography (SiO2 eluted with 7% MeOH in CH2Cl2) gave
the title compound, 12, as an off-white solid (1.35 g, 90%). M.p. 79–
1
818C; H NMR (400 MHz, [D6]DMSO): d=9.0 (brs, 1H), 8.48 (brs, 1H),
7.91 (m, 2H), 7.47 (t, J=4.5 Hz, 1H), 7.33–7.17 (m, 9H), 4.91 (m, 2H),
4.56 (m, 2H), 4.27 (d, J=4.5 Hz, 2H), 3.49 ppm (m, 2H); 13C NMR
(100.5 MHz, [D6]DMSO): d=180.45 (s), 170.29 (s), 166.06 (s), 162.81 (s),
159.86 (s), 139.31 (s), 138.11 (d), 137.88 (s), 128.22 (d), 128.21 (d), 127.66
(d), 127.42 (d), 127.17 (d), 126.70 (d), 122.91 (d), 65.49 (t), 42.06 (t),
Guanidinium–carboxylate 11: To
(460 mg, ~1.75 mmol) in CH2Cl2 (30 mL) was added Et3N (0.74 mL,
5.25 mmol) followed by EDC·HCl (840 mg, 4.4 mmol) thiourea
a stirred suspension of 9·TFA salt
37.31 (t), 35.12 ppm (t); IR (solid): n˜
= 3269 (br, m), 1589 (m),
1
1381 cmꢀ1 (m); MS (ES+): m/z (%): 490 (100) [M+H]+; HRMS (ES+):
m/z: calcd for C26H28N5O5: 490.2085; found: 490.2074 [M+H]+.
(650 mg, 1.75 mmol), each in a single portion. The resulting mixture was
stirred for 72 h at room temperature before addition of KHSO4 (50 mL
of a 1m aqueous solution), separation of the organic layer, washing with
brine (50 mL), drying over MgSO4 and concentration in vacuo. Purifica-
tion by column chromatography (SiO2 eluted with 1% MeOH in CH2Cl2
! 3% MeOH in CH2Cl2) gave the Cbz-guanidine product, 30, as a white
solid (500 mg, 50%). A stirred solution of 30 (470 mg, 0.8 mmol) in THF
(30 mL) was added to a suspension of Me3SiOK (133 mg, 1.04 mmol) in
THF (10 mL) and the resulting mixture stirred at room temperature for
16 h before cooling to 08C and addition of H2O (50 mL) and CH2Cl2
(50 mL). The mixture was adjusted to pH ~6 by dropwise addition of
citric acid (0.1m aqueous solution) at 08C and the resulting biphasic mix-
ture stirred vigorously at 08C for 1 h before separation of the organic
phase, drying over MgSO4 and concentrated in vacuo. Purification by
Guanidinium–carboxylate 13: Boc-l-Val-l-Val-OMe (141 mg, 0.43 mmol)
was treated with trifluoroacetic acid (5 mL of a 20% v/v solution in
CH2Cl2) at room temperature and the mixture stirred for 3 h. CH2Cl2 was
removed in vacuo and the residue washed with toluene (3ꢃ10 mL), re-
moval of the azeptrope with TFA was made in vacuo each time. The
crude oil was taken into CH2Cl2 (10 mL) and washed with K2CO3 (10 mL
of a saturated aqueous solution) before drying over MgSO4 and concen-
tration in vacuo, to give the amine dipeptide product as a colourless oil
which was used directly (98 mg, ~98%). To a mixture of this amine prod-
uct (98 mg, 0.43 mmol) and carboxylic acid 12 (208 mg, 0.43 mmol) in
CH2Cl2 (20 mL) was added iPr2NEt (0.37 mL, 2.14 mmol) and PyBop
(266 mg, 0.51 mmol), and the resulting mixture stirred at room tempera-
ture for 12 h before addition of KHSO4 (25 mL of a 1m aqueous solu-
tion), separation of the organic phase, and washing with K2CO3 (25 mL
of a saturated aqueous solution) and brine (25 mL). The organic phase
was dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2 eluted with EtOAc/MeCN 85:15) gave the amide
product, 34, as a white solid (288 mg, 96%). To a stirred solution of 34
(183 mg, 0.26 mmol) in THF (10 mL) at room temperature, was added
Me3SiOK (50 mg, 0.39 mmol) in a single portion and the resulting mix-
ture stirred for 12 h before cooling to 08C and addition of H2O (20 mL)
and CH2Cl2 (20 mL). The mixture was adjusted to pH ~6 by dropwise ad-
dition of citric acid (0.1m aqueous solution) at 08C and the biphasic mix-
ture stirred vigorously at 08C for 1 h before separation and extraction of
the aqueous phase with CH2Cl2 (3ꢃ50 mL). The combined organic phase
was dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2 eluted with 10% MeOH in CH2Cl2) gave the car-
boxylic acid product, 35, as a white solid (96 mg, 54%). A stirred solution
of 35 (163 mg, 0.24 mmol) in MeOH (10 mL) was treated with 10% Pd/C
(25 mg) and stirred under H2 (1 atm) at room temperature for 6 h before
filtration through Celite. Concentration of the filtrate in vacuo, washing
of the filter cake with MeOH (20 mL) and precipitation of the residue
from Et2O/MeOH, gave the title compound 13 as a white solid (100 mg,
column chromatography (SiO2 eluted with 3% MeOH in CH2Cl2
!
10% MeOH in CH2Cl2) gave the carboxylic acid, 31, as a white solid
(250 mg, 60%). A stirred solution of 31 (240 mg, 0.41 mmol) in MeOH
(30 mL) was treated with 10% Pd/C (43 mg) and stirred under H2 (1
atm) at room temperature for 48 h before filtration through Celite. Con-
centration of the filtrate in vacuo, followed by precipitation of the resi-
due from Et2O/MeOH, gave the title compound 11 as a white solid
(120 mg, 55%). M.p. 242–2448C; 1H NMR (400 MHz, [D4]MeOD): d=
8.00 (dd, J=6.4, 1.1 Hz, 1H), 7.95 (t, J=7.5 Hz, 1H), 7.50 (dd, J=7.5,
1.1 Hz, 1H), 7.29–7.17 (m, 5H), 4.59 (s, 2H), 4.40 (d, J=4.9 Hz, 1H),
4.31 (s, 2H), 3.55 (m, 2H), 2.48 (t, J=6.0 Hz, 2H), 2.17 (m, 1H), 0.88 (d,
J=7.1 Hz, 3H), 0.85 ppm (t, J=7.1 Hz, 3H);13C NMR (100.5 MHz,
[D4]MeOD): d=173.02 (s), 165.18 (s), 158.41 (s), 156.13 (s), 150.83 (s),
139.93 (d), 139.83 (s), 129.53 (d), 128.59 (d), 128.24 (d), 125.67 (d), 122.20
(d), 61.53 (d), 46.56 (t), 44.19 (t), 39.19 (t), 36.00 (t), 33.50 (d), 20.24 (q),
18.59 ppm (q); IR (solid): n˜ = 3208 (w, br), 2957 (s), 2927 (s), 2858 (w),
2361 (s), 2342 (m), 1728 cmꢀ1 (s); MS (ES+): m/z (%): 455 (100)
[M+H]+; HRMS (ES+): m/z: calcd for C23H31N6O4: 455.2401; found:
455.2389 [M+H]+.
6-((3-(3-(Benzylamino)-3-oxopropyl)-2-((benzyloxy)carbonyl)
guanidi-
no)methyl)picolinic acid (12): Ethyl ester 3 (2.5 g, 3.81 mmol) was treated
with trifluoroacetic acid (30 mL of a 20% v/v solution in CH2Cl2) at
room temperature and the mixture stirred for 3 h. CH2Cl2 was removed
in vacuo and the residue washed with toluene (3ꢃ20 mL), removal of the
76%). m.p. 228–2308C. IR (solid): 3387 (br, w), 1647 (m), 1016 (m) cmꢀ1
.
1H NMR (400 MHz, [D4]MeOD): d =8.04 (d, J=7.5 Hz, 1H), 7.98 (t, J=
7.5 Hz, 1H), 7.53 (d, J=7.5 Hz, 1H), 7.24–7.17 (m, 5H), 4.60 (s, 2H),
12394
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12387 – 12397

Dimeric Self-Assembly
FULL PAPER
4.53 (d, J=7.0 Hz, 1H), 4.35 (s, 2H), 4.16 (d, J=5.5 Hz, 1H), 3.55–3.51
(m, 2H), 2.62 (t, J=6.5 Hz, 2H), 2.22 (m, 1H), 2.12 (m, 1H), 1.02 (d, J=
6.5 Hz, 3H), 0.97 (d, J=7.0 Hz, 3H), 0.93–0.91 ppm (m, 6H); 13C NMR
(100.5 MHz, [D4]MeOD): d=182.22 (s), 178.21 (s), 173.18 (s), 172.64 (s),
165.78 (s), 158.35 (s), 156.25 (s), 140.01 (d), 139.85 (s), 129.52 (d), 128.57
(d), 128.22 (d), 125.92 (d), 122.41 (d), 61.95 (d), 60.35 (d), 46.56 (t), 44.19
(t), 39.25 (t), 35.97 (t), 33.02 (d), 32.48 (d), 20.21 (q), 20.01 (q), 18.73 (q),
18.67 ppm (q); MS (ES+): m/z (%): 554 (82) [M+H]+; HRMS (ES+):
m/z: calcd for C28H40N7O5: 554.3085; found: 554.3073 [M+H]+.
each time. The crude oil was taken into CH2Cl2 (15 mL) and washed with
K2CO3 (20 mL of a saturated aqueous solution) before drying over
MgSO4 and concentration in vacuo, to give the amine product, 39, as a
colourless oil which was used directly (744 mg, 95%). To a stirred solu-
tion of 39 (744 mg, 1.9 mmol) in CH2Cl2 (70 mL) was added Et3N
(0.8 mL, 5.7 mmol) followed by EDC·HCl (1.1 g, 5.7 mmol) and thiourea
1 (846 mg, 2.28 mmol), each in a single portion. The resulting mixture
was stirred for 12 h at room temperature before addition of KHSO4
(50 mL of a 1m aqueous solution), separation of the organic phase, and
washing with K2CO3 (50 mL of a saturated aqueous solution) and brine
(50 mL). The organic phase was dried over MgSO4 and concentrated in
vacuo. Purification by column chromatography (SiO2 eluted with 2%
MeOH in CH2Cl2) gave the Cbz-guanidine product, 40, as a white solid
(838 mg, 60%). A stirred solution of 40 (635 mg, 0.87 mmol) in THF
(30 mL) was added to a suspension of Me3SiOK (168 mg, 1.31 mmol) in
THF (25 mL) and the resulting mixture stirred at room temperature for
12 h before cooling to 08C and addition of H2O (25 mL) and CH2Cl2
(25 mL). The mixture was adjusted to pH ~6 by dropwise addition of
citric acid (0.1m aqueous solution) at 08C and the resulting biphasic mix-
ture stirred vigorously at 08C for 1 h before separation of the organic
phase, drying over MgSO4 and concentrated in vacuo. Purification by
column chromatography (SiO2 eluted with 7% MeOH in CH2Cl2) gave
the carboxylic acid, 41, as a white solid (493 mg, 80%). A stirred solution
of 41 (277 mg, 0.39 mmol) in MeOH (10 mL) was treated with 10% Pd/C
(41 mg) and stirred under H2 (1 atm) at room temperature for 6 h before
filtration through Celite. Concentration of the filtrate in vacuo, and pu-
rification by column chromatography (SiO2 eluted with 5% ! 12%
MeOH (saturated with NH3) in CH2Cl2) followed by precipitation from
Et2O/MeOH, gave the title compound 15 as a white solid (180 mg, 79%).
Guanidinium–carboxylate 14: To
a mixture of H-l-Leu-OMe·HCl
(167 mg, 0.92 mmol) and carboxylic acid 12 (300 mg, 0.61 mmol) in
CH2Cl2 (20 mL) was added iPr2NEt (0.54 mL, 3.1 mmol) and PyBop
(385 mg, 0.74 mmol), and the resulting mixture stirred at room tempera-
ture for 12 h before addition of KHSO4 (25 mL of a 1m aqueous solu-
tion), separation of the organic phase, and washing with K2CO3 (25 mL
of a saturated aqueous solution) and brine (25 mL). The organic phase
was dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2 eluted with EtOAc/MeCN 3:1) gave the amide
product, 36, as a white solid (340 mg, 90%). To a stirred solution of 36
(273 mg, 0.44 mmol) in THF (10 mL) at room temperature, was added
Me3SiOK (85 mg, 0.66 mmol) in a single portion and the resulting mix-
ture stirred for 12 h before cooling to 08C and addition of H2O (20 mL)
and CH2Cl2 (20 mL). The mixture was adjusted to pH ~6 by dropwise ad-
dition of citric acid (0.1m aqueous solution) at 08C and the biphasic mix-
ture stirred vigorously at 08C for 1 h before separation and extraction of
the aqueous phase with CH2Cl2 (3ꢃ50 mL). The combined organic phase
was dried over MgSO4 and concentrated in vacuo. Purification by column
chromatography (SiO2 eluted with 5% MeOH in CH2Cl2) gave the car-
boxylic acid product, 37, as a white solid (240 mg, 90%). A stirred solu-
tion of 37 (150 mg, 0.25 mmol) in MeOH (10 mL) was treated with 10%
Pd/C (27 mg) and stirred under H2 (1 atm) at room temperature for 6 h
before filtration through Celite. Concentration of the filtrate in vacuo,
washing of the filter cake with MeOH (20 mL) and purification by
column chromatography (SiO2 eluted with 5 ! 12% MeOH (saturated
with NH3) in CH2Cl2) followed by precipitation from Et2O/MeOH, gave
the title compound 14 as a white solid (94 mg, 80%). M.p. 181–1838C;
1
M.p. 170–1728C; [a]2D5 = ꢀ26.88 (c = 0.01, MeOH); H NMR (400 MHz,
[D4]MeOD): d=8.02 (dd, J=7.5, 1.0 Hz, 1H), 7.98 (t, J=7.5 Hz, 1H),
7.53 (dd, J=7.5, 1.0 Hz, 1H), 7.25–7.17 (m, 5H), 4.57 (m, 3H), 4.35 (s,
2H), 4.29 (m, 1H), 3.65–3.52 (m, 2H), 2.61 (t, J=6.6 Hz, 2H), 1.78–1.57
(m, 6H), 0.97 (d, J=6.5 Hz, 3H), 0.96 (d, J=6.5 Hz, 3H), 0.91 (d, J=
6.5 Hz, 3H), 0.897 ppm (d, J=6.5 Hz, 3H); 13C NMR (100.5 MHz,
[D4]MeOD): d=179.70 (s), 173.55 (s), 173.26 (s), 165.87 (s), 158.22 (s),
155.98 (s), 150.54 (s), 139.99 (d), 139.81 (s), 129.51 (d), 128.57 (d), 128.21
(d), 125.94 (d), 122.45 (d), 55.23 (d), 53.69 (d), 46.66 (t), 44.25 (t), 43.09
(t), 42.71 (t), 39.31 (t), 36.14 (t), 26.25 (d), 26.18 (d), 23.69 (q), 23.54 (q),
22.41 (q), 22.22 ppm (q); IR (solid): 3274 (br, w), 1634 (s), 1519 (s),
1384 cmꢀ1 (m); MS (ES+): m/z (%): 582 (100) [M+H]+; HRMS (ES+):
m/z: calcd for C30H44N7O5: 582.3398; found: 582.3392 [M+H]+.
1
[a]2D5 = ꢀ2.98 (c = 0.005, MeOH); H NMR (400 MHz, [D4]MeOD): d=
7.99–7.91 (m, 2H), 7.49 (m, 1H), 7.26–7.17 (m, 5H), 4.58–4.56 (m, 3H),
4.35 (s, 2H), 3.56 (m, 2H), 2.60 (t, J=6.5 Hz, 2H), 1.80–1.66 (m, 3H),
0.98 (d, J=6.6 Hz, 3H), 0.96 ppm (d, J=6.0 Hz, 3H); 13C NMR
(100.5 MHz, [D4]MeOD): d=182.26 (s), 179.42 (s), 173.06 (s), 164.99 (s),
158.34 (s), 155.90 (s), 150.80 (s), 139.84 (d), 129.52 (d), 128.58 (d), 128.22
(d), 125.54 (d), 122.09 (d), 55.18 (d), 46.57 (t), 44.21 (t), 44.08 (t), 39.21
(t), 36.09 (t), 26.39 (d), 23.75 (q), 22.69 ppm (q); IR (solid): n˜ = 3232 (br,
w), 1651 (s), 1538 (s), 1392 cmꢀ1 (s); MS (ES+): m/z (%): 469 (100)
[M+H]+; HRMS (ES+): m/z: calcd for C24H33N6O4: 469.2558; found:
469.2562 [M+H]+.
Guanidinium–carboxylate 16. Boc-l-Leu-l-Leu-l-Leu-OMe (350 mg,
0.74 mmol) was treated with TFA (15 mL of a 20% v/v solution in
CH2Cl2) at room temperature and the mixture stirred for 3 h. CH2Cl2 was
removed in vacuo and the residue washed with toluene (3ꢃ10 mL), re-
moval of the azeptrope with TFA was made in vacuo each time. The
crude oil was taken into CH2Cl2 (15 mL) and washed with K2CO3 (20 mL
of a saturated aqueous solution) before drying over MgSO4 and concen-
tration in vacuo, to give the amine dipeptide product as a colourless oil
which was used directly (190 mg, ~68%). To a mixture of this tripeptide
(185 mg, 0.5 mmol) and carboxylic acid 12 (244 mg, 0.5 mmol) in CH2Cl2
(20 mL) was added iPr2NEt (0.43 mL, 2.5 mmol) and PyBop (312 mg,
0.6 mmol), and the resulting mixture stirred at room temperature for 12 h
before addition of KHSO4 (20 mL of a 1m aqueous solution), separation
of the organic phase, and washing with K2CO3 (20 mL of a saturated
aqueous solution). The organic phase was dried over MgSO4 and concen-
trated in vacuo. Purification by column chromatography (SiO2 eluted
with EtOAc/MeCN 85:15) gave the amide product, 42, as a white solid
(322 mg, 76%). To a stirred solution of 42 (308 mg, 0.36 mmol) in THF
(15 mL) at room temperature, was added Me3SiOK (71 mg, 0.55 mmol)
in a single portion and the resulting mixture stirred for 12 h before cool-
ing to 08C and addition of H2O (25 mL) and CH2Cl2 (25 mL). The mix-
ture was adjusted to pH ~6 by dropwise addition of citric acid (0.1m
aqueous solution) at 08C and the biphasic mixture stirred vigorously at
08C for 1 h before separation and extraction of the aqueous phase with
CH2Cl2 (3ꢃ100 mL). The combined organic phase was dried over MgSO4
and concentrated in vacuo. Purification by column chromatography (SiO2
eluted with 7% MeOH in CH2Cl2) gave the carboxylic acid product, 43,
Guanidinium–carboxylate
15.
Boc-l-Leu-l-Leu-OMe
(760 mg,
2.12 mmol) was treated with trifluoroacetic acid (15 mL of a 20% v/v so-
lution in CH2Cl2) at room temperature and the mixture stirred for 2 h.
CH2Cl2 was removed in vacuo and the residue washed with toluene (3ꢃ
10 mL), removal of the azeptrope with TFA was made in vacuo each
time. The crude oil was taken into CH2Cl2 (15 mL) and washed with
K2CO3 (20 mL of a saturated aqueous solution) before drying over
MgSO4 and concentration in vacuo, to give the amine dipeptide product
as a colourless oil which was used directly (529 mg, ~97%). To a mixture
of this dipeptide (527 mg, 2.1 mmol) and carboxylic acid 17 (534 mg,
2.12 mmol) in CH2Cl2 (50 mL) was added iPr2NEt (1.84 mL, 10.6 mmol)
and PyBop (1.32 g, 2.54 mmol), and the resulting mixture stirred at room
temperature for 12 h before addition of KHSO4 (25 mL of a 1m aqueous
solution), separation of the organic phase, and washing with K2CO3
(25 mL of a saturated aqueous solution) and brine (25 mL). The organic
phase was dried over MgSO4 and concentrated in vacuo. Purification by
column chromatography (SiO2 eluted with EtOAc/MeCN 85:15) gave the
amide product, 38, as a white solid (876 mg, 84%). This product, 38,
(980 mg, 1.99 mmol) was treated with trifluoroacetic acid (20 mL of a
20% v/v solution in CH2Cl2) at room temperature and the mixture stirred
for 2 h. CH2Cl2 was removed in vacuo and the residue washed with tolu-
ene (3ꢃ10 mL), removal of the azeptrope with TFA was made in vacuo
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12395

J. D. Kilburn et al.
as
a
white solid (294 mg, 97%). A stirred solution of 43 (208 mg,
[5] a) G. Grçger, V. Stepanenko, F. Wurthner, C. Schmuck, Chem.
Commun. 2009, 698–700; b) T. Rehm, C. Schmuck, Chem.
Commun. 2008, 801–813; c) T. Rehm, V. Stepanenko, X. Zhang, F.
Wurthner, F. Grohn, K. Klein, C. Schmuck, Org. Lett. 2008, 10,
1469–1472; d) C. Schmuck, J. Dudaczek, Eur. J. Org. Chem. 2007,
3326–3330.
[6] Quantification of non-covalent interactions (e.g., aromatic edge-face
and hydrogen-bonding interactions) and cooperative effects have
been extensively investigated by using synthetic supramolecular sys-
tems in non-polar solvents. For a review, see: a) S. L. Cockroft,
C. A. Hunter, Chem. Soc. Rev. 2007, 36, 172–188. See also b) S. L.
Cockroft, C. A. Hunter, Chem. Commun. 2009, 3961–3962; c) A.
Camara-Campos, D. Musumeci, C. A. Hunter, S. Turega, J. Am.
Chem. Soc. 2009, 131, 18518–18524.
[7] For statistical and theoretical approaches to the quantification of
amino acid side-chain–side-chain interactions in proteins, see: a) K.
Berka, R. Laskowski, K. E. Riley, P. Hobza, J. Vondrasek, J. Chem.
Theory Comput. 2009, 5, 982–992; b) G. Faure, A. Bornot, A. G. de
Brevern, Biochimie 2008, 90, 626–639; c) H. M. Fooks, A. C. R.
Martin, D. N. Woolfson, R. B. Sessions, E. G. Hutchinson, J. Mol.
Biol. 2006, 356, 32–44.
[8] a) J. U. Bowie, Nature 2005, 438, 581–589; b) G. Helles, J. R. Soc.
Interface 2008, 5, 387–396; c) C. D. Snow, E. J. Sorin, Y. M. Rhee,
V. S. Pande, Annu. Rev. Biophys. Biomol. Struct. 2005, 34, 43–69;
d) R. L. Baldwin, J. Mol. Biol. 2007, 371, 283–301.
0.25 mmol) in MeOH (10 mL) was treated with 10% Pd/C (27 mg) and
stirred under H2 (1 atm) at room temperature for 6 h before filtration
through Celite and washing of the filter cake with MeOH (20 mL). Con-
centration of the filtrate in vacuo and purification by column chromatog-
raphy (SiO2 eluted with 5 ! 12% MeOH (saturated with NH3) in
CH2Cl2) followed by precipitation from Et2O/MeOH, gave the title com-
pound 16 as a white solid (125 mg, 72%). M.p. 188–1908C; [a]D25
=
1
ꢀ45.18 (c = 0.01, MeOH); H NMR (400 MHz, [D4]MeOD): d=8.04 (d,
J=7.0 Hz, 1H), 7.97 (t, J=7.7 Hz, 1H), 7.55 (d, J=7.0 Hz, 1H), 7.26–
7.20 (m, 5H), 4.58 (s, 2H), 4.40 (m, 1H), 4.35 (s, 2H), 4.30 (m, 1H), 3.60
(m, 2H), 3.35 (s, 1H), 2.62 (t, J=6.0 Hz, 2H), 1.72–1.61 (m, 7H), 1.55
(m, 1H), 1.00 (d, J=6.5 Hz, 3H), 0.98 (d, J=6.5 Hz, 3H), 0.92–0.91 (m,
9H), 0.86 ppm (d, J=6.0 Hz, 3H); 13C NMR (100.5 MHz, [D4]MeOD):
d=179.53 (s), 174.36 (s), 173.33 (s), 165.77 (s), 158.32 (s), 156.16 (s),
150.35 (s), 140.00 (d), 139.84 (s), 129.52 (d), 128.57 (d), 128.21 (d), 125.99
(d), 122.44 (d), 54.86 (d), 53.80 (d), 52.90 (d), 46.65 (t), 44.24 (t), 43.78
(t), 42.90 (t), 41.30 (t), 39.28 (t), 36.20 (t), 26.19 (d), 26.13 (d), 25.92 (d),
23.81 (q), 23.62 (q), 23.44 (q), 22.66 (q), 22.10 (q), 22.04 ppm (q); IR
(solid): n˜ = 3272 (br, w), 1642 (s), 1520 (s), 1385 cmꢀ1 (m); MS (ES+):
m/z (%): 695 (100) [M+H]+; HRMS (ES+): m/z: calcd for C36H55N8O6:
695.4239; found: 695.4235 [M+H]+.
[9] A. Tahiri-Alaoui, M. Bouchard, J. Zurdo, W. James, Protein Sci.
2003, 12, 600–608, and references therein.
Acknowledgements
[10] a) S. Levin, J. S. Nowick, Org. Lett. 2009, 11, 1003–1006; b) J. S.
Nowick, Acc. Chem. Res. 2008, 41, 1319–1330; c) R. M. Hughes,
M. L. Waters, Curr. Opin. Struct. Biol. 2006, 16, 514–524; d) J. S.
Nowick, O. Khakshoor, Curr. Opin. Chem. Biol. 2008, 12, 722–729,
and references therein.
We gratefully acknowledge Prof. Alan Cooper, WestCHEM, University
of Glasgow, for provision of corrective files for the MicroCal Origin for
ITC dissociation model, independent validation of data analysis and very
helpful discussion. We thank the Higher Education Commision of Paki-
stan (MIA under the split PhD program), HEFCE (BFT under the ORS
awards scheme) and the University of Southampton for support of this
work.
[11] For empirical approaches to the quantification of amino acid side-
chain–side-chain interactions in folded peptides, see: a) E. B.
Hadley, A. M. Witek, F. Freire, A. J. Peoples, S. H. Gellman, Angew.
Chem. 2007, 119, 7186–7189; Angew. Chem. Int. Ed. 2007, 46, 7056–
7059; b) S. T. Phillips, G. Piersanti, P. A. Bartlett, Proc. Natl. Acad.
Sci. USA 2005, 102, 13737–13742; c) C. D. Tatko, M. L. Waters, Pro-
tein Sci. 2004, 13, 2515–2522; d) C. D. Tatko, M. L. Waters, J. Am.
Chem. Soc. 2004, 126, 2028–2034; e) C. D. Tatko, M. L. Waters, Pro-
tein Sci. 2003, 12, 2443–2452; f) C. D. Tatko, M. L. Waters, J. Am.
Chem. Soc. 2002, 124, 9372–9373; g) M. S. Searle, S. R. Griffiths-
Jones, H. Skinner-Smith, J. Am. Chem. Soc. 1999, 121, 11615–11620;
h) C. A. Blasie, J. M. Berg, Biochemistry 1997, 36, 6218–6222.
[12] Quantification must be relative, within the context of this particular
model, since individual contributions to binding are not independent
of each other but determined by the overall structure of the com-
plex, see: a) C. T. Calderone, D. H. Williams, J. Am. Chem. Soc.
2001, 123, 6262–6267; b) A. Cooper, Curr. Opin. Chem. Biol. 1999,
3, 557–563.
[1] For reviews, see: a) P. Blondeau, M. Segura, R. Perez-Fernandez, J.
de Mendoza, Chem. Soc. Rev. 2007, 36, 198–210; b) R. J. Fitzmaur-
ice, G. M. Kyne, D. Douheret, J. D. Kilburn, J. Chem. Soc. Perkin
Trans. 1 2002, 841–864.
[2] For recent examples, see: a) X. Wang, G. Sarycheva, B. D. Koivisto,
A. H. Mckie, F. Ho, Org. Lett. 2008, 10, 297–300; b) V. D. Jadhav, E.
Herdtweck, F. P. Schmidtchen, Chem. Eur. J. 2008, 14, 6098–6107;
c) V. D. Jadhav, F. P. Schmidtchen, J. Org. Chem. 2008, 73, 1077–
1087; d) D. Moiani, C. Cavallotti, A. Famulari, C. Schmuck, Chem.
Eur. J. 2008, 14, 5207–5219; e) R. J. Fitzmaurice, F. Gaggini, N. Sri-
nivasan, J. D. Kilburn, Org. Biomol. Chem. 2007, 5, 1706–1714;
f) M. Mazik, H. Cavga, J. Org. Chem. 2007, 72, 831–838; g) C.
Schmuck, V. Bickert, J. Org. Chem. 2007, 72, 6832–6839; h) C.
Schmuck, E. Dudaczek, Tetrahedron Lett. 2005, 46, 7101–7105; i) C.
Schmuck, M. Schwegmann, J. Am. Chem. Soc. 2005, 127, 3373–
3379; j) S. L. Wiskur, J. L. Lavigne, A. Metzger, S. L. Tobey, V.
Lynch, E. V. Anslyn, Chem. Eur. J. 2004, 10, 3792–3804; k) A.
[13] T. Storr, B. R. Cameron, R. A. Gossage, H. Yee, R. T. Skerlj, M. C.
Darkes, S. P. Fricker, G. J. Bridger, N. A. Davies, M. T. Wilson, K. P.
Maresca, J. Zubieta, Eur. J. Inorg. Chem. 2005, 2685–2697.
[14] B. R. Linton, A. J. Carr, B. P. Orner, A. D. Hamilton, J. Org. Chem.
2000, 65, 1566–1568.
ˇ
Spꢄth, B. Kçnig, Tetrahedron 2010, 66, 1859–1873; l) T. Smejkal, D.
[15] S. Mattsson, M. Dahlstrom, S. Karlsson, Tetrahedron Lett. 2007, 48,
2497–2499.
Gribkov, J. Geier, M. Keller, B. Breit, Chem. Eur. J. 2010, 16, 2470–
2478.
[16] Preferred pairing of hydrophobic residues valine, isoleucine and leu-
cine in parallel b-sheets has been determined through statistical
analysis, see: reference [7c].
[17] a) A. Lovatt, A. Cooper, P. Camilleri, Eur. Biophys. J. 1996, 24,
354–357; b) D. McPhail, A. Cooper, J. Chem. Soc. Faraday Trans.
1997, 93, 2283–2289; c) P. R. Stoesser, S. J. Gill, J. Phys. Chem. 1967,
71, 564–567.
[18] For recent examples of the physical characterisation of protein ho-
modimerisation using dilution ITC, see: a) J. R. Alford, S. C. Kwok,
J. N. Roberts, D. S. Wuttke, B. S. Kendrick, J. F. Carpenter, T. W.
Randolph, J. Pharm. Sci. 2008, 97, 3035–3050; b) M. Bello, G.
Perez-Hernandez, D. A. Fernandez-Velasco, R. Arreguin-Espinosa,
[3] a) S. Bartoli, T. Mahmood, A. Malik, S. Dixon, J. D. Kilburn, Org.
Biomol. Chem. 2008, 6, 2340–2345; b) C. Schmuck, L. Geiger, J.
Am. Chem. Soc. 2005, 127, 10486–10487; c) C. Schmuck, S. Graup-
ner, Tetrahedron Lett. 2005, 46, 1295–1298; d) C. Schmuck, L. Her-
nandez-Folgado, Org. Biomol. Chem. 2007, 5, 2390–2394; e) C.
Schmuck, U. Machon, Chem. Eur. J. 2005, 11, 1109–1118; f) K. B.
Jensen, T. M. Braxmeier, M. Demarcus, J. G. Frey, J. D. Kilburn,
Chem. Eur. J. 2002, 8, 1300–1309; g) J. Shepherd, T. Gale, K. B.
Jensen, J. D. Kilburn, Chem. Eur. J. 2006, 12, 713–720; h) C.
Schmuck, Coord. Chem. Rev. 2006, 250, 3053–3067.
[4] C. A. Hunter, D. H. Purvis, Angew. Chem. 1992, 104, 779–782;
Angew. Chem. Int. Ed. Engl. 1992, 31, 792–795.
12396
www.chemeurj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2010, 16, 12387 – 12397

Dimeric Self-Assembly
FULL PAPER
E. Garcia-Hernandez, Proteins Struct. Funct. Bioinf. 2008, 70, 1475–
1487.
M. S. Kaucher, J. T. Davis, J. Am. Chem. Soc. 2003, 125, 15140–
15150.
[19] For a recent example of dilution ITC characterisation of a synthetic
ferrocene-b-cyclodextrin dimer, see: J. M. Casas-Solvas, E. Ortiz-
Salmeron, I. Fernandez, L. Garcia-Fuentes, F. Santoyo-Gonzalez, A.
Vargas-Berenguel, Chem. Eur. J. 2009, 15, 8146–8162.
[20] ITC dilution data were analysed in terms of a monomer-dimer equi-
librium model using an updated and corrected (June 2008) version
of the MicroCal Origin dissociation option (available at: http://
www.chem.gla.ac.uk/staff/alanc/service1.htm Notes). Initial injection
data points were removed before data analysis. Data analysis
method and sample derived parameters (Table 1, entry 17) were in-
dependently validated using the original in-house routines from the
[26] For an example of “tweezer” receptors which demonstrate a binding
profile for peptidic guests, in which the energetic cost of receptor
unfolding results in weaker binding but higher selectivity upon in-
creased receptor complexity, see: E. Botana, S. Ongeri, R. Arienzo,
M. Demarcus, J. G. Frey, U. Piarulli, D. Potenza, C. Gennari, J. D.
Kilburn, Chem. Commun. 2001, 1358–1359.
[27] This postulated enthalpy-entropy compensation is well known, see
for example: a) A. Cooper, C. M. Johnson, J. H. Lakey, M. Noll-
mann, Biophys. Chem. 2001, 93, 215–230; b) D. M. Ford, J. Am.
Chem. Soc. 2005, 127, 16167–16170; c) C. A. Hunter, S. Tomas,
Chem. Biol. 2003, 10, 1023–1032; d) D. H. Williams, D. P. OꢁBrien,
A. M. Sandercock, E. Stephens, J. Mol. Biol. 2004, 340, 373–383.
[28] Calculation of required sample concentrations for the range 90–
10% dimer was made for each of the pyridyl guanidinium–carboxy-
lates 5, 6, 8, 10, 11 and 13–16 in each of the solvent systems in which
a dimerisation constant had been determined using calorimetry.
Only for dimerisation of compound 14 in DMSO did a concentra-
Cooper/Glasgow group (ref.ACTHNUGRTENUNG[17a,b]), slightly modified (in Glasgow)
to permit omission of initial pre-injections from the analysis.
[21] W. B. Turnbull, A. H. Daranas, J. Am. Chem. Soc. 2003, 125, 14859–
14866.
[22] For a detailed, comparative discussion of the dimeric solid-state
structures of 5 and 6, see: M. I. Ashiq, I. Hussain, S. Dixon, M. E.
Light, J. D. Kilburn, Acta Crystallogr. Sect. C 2010, 66, 455–458.
[23] Centroid (C2 > C7) to centroid (C2* > C7*) separation=3.662 ꢂ,
with the perpendicular distance between the least squares planes
through each benzene ring (C2 > C7, C2* > C7*) being 3.372 ꢂ;
*[ꢀ1ꢀx, ꢀy, ꢀz].
1
tion range suitable for H NMR dilution study (40.6–0.13 mm) corre-
spond to sufficient coverage of the monomer - dimer equilibrium
(approx. 85% dimer at the higher concentration and 10% dimer at
the lower, based upon determination of Kdim =877mꢀ1 by using calo-
rimetry).
[24] A similar motif, in which a single carbonyl oxygen is bound by both
donor NHs of a thiourea, has been observed in the solid-state (see:
G. M. Kyne, M. E. Light, M. B. Hursthouse, J. de Mendoza, J. D. Kil-
burn, J. Chem. Soc. Perkin Trans. 1 2001, 1258–1263) and also pre-
dicted by molecular modelling for a related macrocyclic receptor
(see: A. Ragusa, S. Rossi, J. M. Hayes, M. Stein, J. D. Kilburn,
Chem. Eur. J. 2005, 11, 5674–5688).
[29] 1H NMR dilution data were fitted to a dimer dissociation isotherm
using NMRDil_Dimer software kindly provided by Professor C. A.
Hunter, University of Sheffield: A. P. Bisson, C. A. Hunter, J. C.
Morales, K. Young, Chem. Eur. J. 1998, 4, 845–851. All data from
NMR dilution experiments are provided in the Supporting Informa-
tion.
[25] a) R. P. Bonar-Law, J. K. M. Sanders, J. Am. Chem. Soc. 1995, 117,
259–271; b) F. W. Kotch, V. Sidorov, Y.-F. Lam, K. J. Kayser, H. Li,
Received: July 1, 2010
Published online: September 20, 2010
Chem. Eur. J. 2010, 16, 12387 – 12397
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
12397