The role of the amino acid-derived side chain in the preorganization of c2-symmetric pseudopeptides: Effect on sn2 macrocyclization reactions

Article abstract of DOI:10.1021/jo4022309

A family of pseudopeptidic macrocycles containing non-natural amino acids have been synthesized. The macrocyclization reaction has been studied experimentally and computationally, demonstrating the key role of both the amino acid side chain and the catalytic bromide anion. The bromide anion acts as an external template assisting the folding of the open-chain precursor in a proper conformation. Computations revealed that in the presence of the anion, the effect of the side chain on the energy barrier for the macrocyclization is very small. However, the effect on the conformational equilibria of the open-chain precursors is very important. Overall, the stabilization of those conformation(s) in which the two reactive ends of the open-chain intermediate are located at short distances from each other with the correct orientation is the critical parameter defining the success of the macrocyclization. The best yield was found for the compound containing cyclohexylalanine, for which the computationally-predicted most stable conformer in the presence of Br- has a proper preorganization for cyclization. The remarkable agreement obtained between experiments and theory reveals that the computational approach here considered can be of great utility for the prediction of the behavior of other related systems and for the design of appropriate synthetic routes to new macrocyclic compounds.

Full text of DOI:10.1021/jo4022309

Article
pubs.acs.org/joc
The Role of the Amino Acid-Derived Side Chain in the
Preorganization of C2‑Symmetric Pseudopeptides: Effect on SN2
Macrocyclization Reactions
Vicente Martí-Centelles,†,‡ M. Isabel Burguete,† Carlos Cativiela,*,‡ and Santiago V. Luis*,†
†Departamento de Química Inorgan
́
ica y Organ
́
ica, ESTCE, Universitat Jaume I, E-12071 Castellon
́
, Spain
‡Departamento de Química Organ
́
ica, Instituto de Síntesis Química y Catal
́
isis Homogenea (ISQCH), CSIC-Universidad de
́
Zaragoza, E-50009 Zaragoza, Spain
S
* Supporting Information
ABSTRACT: A family of pseudopeptidic macrocycles containing non-natural amino acids
have been synthesized. The macrocyclization reaction has been studied experimentally and
computationally, demonstrating the key role of both the amino acid side chain and the
catalytic bromide anion. The bromide anion acts as an external template assisting the
folding of the open-chain precursor in a proper conformation. Computations revealed that
in the presence of the anion, the effect of the side chain on the energy barrier for the
macrocyclization is very small. However, the effect on the conformational equilibria of the
open-chain precursors is very important. Overall, the stabilization of those
conformation(s) in which the two reactive ends of the open-chain intermediate are
located at short distances from each other with the correct orientation is the critical
parameter defining the success of the macrocyclization. The best yield was found for the
compound containing cyclohexylalanine, for which the computationally-predicted most
stable conformer in the presence of Br− has a proper preorganization for cyclization. The
remarkable agreement obtained between experiments and theory reveals that the computational approach here considered can be
of great utility for the prediction of the behavior of other related systems and for the design of appropriate synthetic routes to
new macrocyclic compounds.
cyclization reaction competes with oligomerization/polymer-
ization processes, with the latter ones usually being favored
unless high-dilution conditions are employed (Figure 1).13,14
Thus, a detailed conformational analysis of both the expected
macrocycle and the corresponding open-chain precursor can be
a powerful tool to predict the potential outcome of a given
macrocyclization.13,15 As a matter of fact, macrocyclization
reactions can be employed as a test to analyze conformational
preferences on different systems.16 This analysis, for instance,
has been of importance in understanding polymerization
processes involving bifunctional monomers. A second area in
which this concept has been critical is the study of efficient
approaches for the preparation of cyclopeptides and cyclic
pseudopeptides, according to the importance of such structures
for biomedical applications.9,17,18
INTRODUCTION
■
The concept of preorganization is central to many, if not most,
of the chemical functionalities that nature has developed. This
is nicely illustrated in the case of enzymes.1 Their unsurpassed
catalytic activities depend on the presence in the active site of
different functional groups belonging to amino acid moieties
often situated very far away in the peptidic sequence.2
Preorganization of such groups involves their proper location
at the right distances and with the right orientation. When
smaller molecules are considered, both in nature and for abiotic
systems, preorganization is often approached through the
preparation of cyclic structures.3−5 The loss of conformational
freedom associated with the cycle can provide a mechanism to
fix the corresponding functional groups with the proper relative
disposition.
The generation of many macrocyclic structures is not a
simple matter, however, macrocyclization is favored when the
corresponding open-chain precursors are themselves preor-
ganized into a suitable folded conformation.4−8 Different
experimental evidences have been gathered both in the field
of low-molecular-weight macrocycles and in the field of cyclic
polymers, revealing that the distance between the two reactive
ends in the open-chain intermediate experiencing the ring-
closing reaction is a critical parameter determining the rate of
the process.9−12 In the absence of such preorganization, the
In recent years, we have been involved in the preparation of
different pseudopeptidic macrocyclic structures based on [1 +
1] or [2 + 2] cyclization processes, as the resulting families of
compounds have been shown to display interesting proper-
ties.19−21 These embrace the preparation of fluorescent sensors,
leading to the development of intracellular pH sensors,
Received: October 15, 2013
Published: December 16, 2013
© 2013 American Chemical Society
559
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
Figure 1. Competition between macrocyclization and oligomerization reactions.
minimalistic molecular machines, and new materials based on
their self-assembling properties.
The successful synthesis of those macrocycles, which are
based on a common C2-symmetric scaffold (Scheme 1), has
According to the former, we have designed and prepared a
series of new C2-symmetric pseudopeptides by considering a
variety of potential amino acid side chains, and we have studied
their behavior in [1 + 1] macrocyclization reactions based on
SN2 processes. The combination of experimental results and
computational studies has been shown to be essential for the
proper design of the selected systems and for an understanding
of the effects observed, in particular the influence of the
different structural features on the appropriate preorganization
of the open-chain precursors. These results provide new
insights into the preparation of macrocyclic peptides and
related compounds, an important area of work because of the
biomedical and pharmaceutical relevance of many of such
structures. The difficulties associated with their synthesis often
represent a limiting factor. The amino acid involved in the
macrocyclization step is a key design vector, and understanding
the specific effects of different side chains provides important
parameters for this design.29−31
Scheme 1. [1 + 1] Macrocyclic Structures Obtained from C2-
Symmetric Open-Chain Pseudopeptides
RESULTS AND DISCUSSION
■
The general structure of the pseudopeptidic macrocycles
obtained by [1 + 1] cyclization processes contains three main
elements of diversity: the central diamine spacer present in the
open-chain starting pseudopeptide, the second spacer [usually a
1,3- or 1,4-bis(methylene)aryl fragment], and the component
amino acid (Figure 2).
been carried out by favoring the correct folded conformation of
the final open-chain intermediates in different ways. For simple
[1 + 1] processes based on SN2 reactions, the coexistence of
intramolecular hydrogen bonds (particularly for compounds
with shorter central spacers) and solvophobic effects plays an
essential role,22 while halide anions can act as efficient kinetic
templates.23−26 As, for the cases studied, halide anions are the
leaving groups, this factor can act even in the absence of any
added anion. In the case of [2 + 2] macrocyclizations based on
reductive amination processes, the proper preorganization can
be achieved through correct matching of the configurations of
the chiral centers in the central spacer and the amino acid
fragment or through the use of specifically designed organic
anionic templates.19
As reaching a preorganization suitable for cyclization involves
achieving a U-folded conformation, it should be expected that
the nature of the side chain of the component amino acid
fragments would have an important effect in this regard. Many
examples in the literature have revealed that the presence of
specific amino acids or amino acid sequences is a key feature for
peptides to adopt folded or helical arrangements.16 For this
purpose, the introduction of non-natural amino acids in the
sequence is a common approach to modify the usual
conformational trends and to force the presence of folding
motifs.27 Non-natural amino acids, on the other hand, can be
synthesized through a variety of methods, allowing a large
diversity of side chains to be introduced at the α and β
positions to favor the appropriate conformational constrains.28
Figure 2. Structural variables in [1 + 1] macrocyclic compounds.
In order to simplify the study and highlight the role of the
amino acid side chain, several constraints were considered.
First, the alkylating agent to produce the cyclization from the
open-chain compounds was fixed to be 1,3-bis(bromomethyl)-
benzene. This is an standard alkylating agent used in previous
experiments and thus allows an easy integration of results.22,24
For the diamine spacer we considered exclusively aliphatic
chains [−(CH2)n−]. Although [1 + 1] cyclization reactions
have also been studied using diamines containing aromatic
fragments,24 the higher conformational flexibility of the
aliphatic spacers could allow clearer observation of the
contribution of the side chain to the conformational equilibria
of the resulting open-chain pseudopeptides. In order to fix
other constraints and to select the most appropriate materials
for our work, various computational studies were initially
560
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
carried out. In this regard, the structures considered for these
studies are shown in Chart 1.
Chart 1. Pseudopeptidic Compounds Considered in This
Work: the Constituent Amino Acids are (a) Phe, (b) Val, (c)
Ala, (d) Aib, (e) Ac6c, (f) Cha, (g) Ile, (h) Leu, (i) Gly; PG
= Protecting Group
Figure 4. N(amine)···C(halide)−C(benzene ring)−X dihedral angle
for the reactant interaction complex: (a) favorable orientation of 180°;
(b) favorable orientation of 160°; (c) unfavorable orientation of 80°.
Dark-red atoms, Br; green atoms, C; blue atom, N; blue plane,
C(halide)−C(benzene ring)−X plane; yellow plane, N(amine)···
C(halide)−C(benzene ring) plane.
Thus, reactive conformations must have low relative energy,
a short N−C distance, a N···C−X angle in the 150−180° range,
and a N···C−C−X dihedral angle ranging from 160° to 200° or
from −200° to −160°. Accordingly, reactive conformations
must be located in very limited regions of the overall
conformational space. The ideal situation should be that in
which the most stable conformers are located within the gray
squares indicated in Figures 6 and 7 for all of the parameters
considered.
This methodology can be illustrated for the case of the
reaction intermediate 3f derived from cyclohexylalanine. Figure
5 shows the geometries of the most stable conformers
calculated for 3f in (a) the absence and (b) the presence of
the bromide anion.
Conformational Studies. As considered above, the
conformational behavior of the open-chain structures 3 has a
direct influence on their reactivity. If the nitrogen atom of the
reactive amine group has a proper orientation with regard to
the carbon−halide group and is located a short distance away,
the macrocyclization is favored. Otherwise, oligomerization
takes place preferentially. Thus, the end-to-end distance
averaged over all possible conformations must be analyzed to
predict its reactivity to form a macrocyclic structure.32,33 To
analyze these intermediates, a conformational search was
carried out using the Spartan’08 software to obtain the most
stable conformer for each structure. This calculation was
repeated at least three times until no variation of the
Boltzmann-averaged nitrogen−carbon distance was observed.
According to our previous results on the role of anions as
kinetic templates,24 the simulation was carried out for each
amino acid in the absence and in the presence of the bromide
anion (Figure 3).
Figure 5. Representations of the MMFF-computed most stable
conformers for the reaction intermediate 3f in (a) the absence and (b)
the presence of bromide anion.
The results of the MMFF34 Monte Carlo conformational
searches for 3f in the absence of any added anion are presented
in Figure 6. This provides a simple way to correlate the
different structural parameters determined in the simulation of
the reaction intermediate. According to the computational
searches, the most stable conformers (red points in Figure 6)
have short N···C distances but do not have dihedral angles that
provide the correct orientation of the N atom relative to the
C−Br bond. The calculated angle is not unique and varies from
−100° to 175°, with a value near 0° for the most stable
conformation (see Figure 6c).
Figure 3. Schematic representation of the open-chain intermediates to
be modeled in (a) the absence and (b) the presence of halide anion
(denoted by the orange sphere).
The most important parameters to analyze in the optimized
structures for reaction intermediates 3 obtained from the
conformational searches are the N(amine)···C(halide) distance,
the N(amine)···C(halide)−X angle, and the N(amine)···
C(halide)−C(benzene ring)−X dihedral angle (where X
denotes the halide). It seems clear, as has been shown for
other systems, that a shorter N(amine)···C(halide) distance
leads to a bigger probability for the reaction between the
nitrogen atom and the carbon−halide, as the local concen-
tration increases. However, it is important to take into
consideration the fact that the presence of a short distance
must be accompanied by a proper orientation of the two
functional groups (Figure 4).
The same simulation in the presence of the bromide anion
(Figure 7) shows that the preferred conformation is completely
different (Figure 5b), and only one stable conformation is
present. This conformer (red dot in Figure 7) has the
appropriate dihedral angle combined with a short N···C
distance, which nicely correspond with the experimental results
described below.
Thus, the presence of a Br− anion acting as a kinetic template
seems to be of paramount importance for the success of the
macrocyclization. However, the side chain of the original amino
561
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
Figure 6. Representation of the conformational space for the reaction intermediate 3f in the absence of any anion.
Figure 7. Representation of the conformational space for the reaction intermediate 3f in the presence of Br− anion.
acid also plays a critical role. This is illustrated by the data
gathered in Figure 8, representing the Boltzmann-averaged C···
N distances calculated for the intermediates 3 bearing different
side chains. As can be observed, very large variations in this
distance were calculated, and the effect of the anion can also be
significantly different.
Experimental Kinetic Measurements. In order to carry
out an experimental study of the corresponding macro-
cyclization processes, the required bis(amino amide)s 1 were
synthesized as summarized in Scheme 1. All of the starting N-
Boc-protected amino acids were commercially available. Their
in situ activation with BOP followed by reaction with
ethylenediamine afforded the N-Boc-protected open-chain
compounds 1-Boc.35 After chromatographic purification, the
Boc protecting groups were removed with trifluoroacetic acid
to provide the expected derivatives 1 in good yields. Specific
reaction conditions needed to be optimized for each particular
case in view of the variety of amino acids considered. In
particular, the reaction times required for quaternary amino
acids were significantly longer because of their lower
reactivities.
Figure 8. Representation of the average C···N distance for the reaction
intermediates 3. The amino acids are (left to right) Ac6c, Aib, Ala,
Cha, Gly, Ile, Leu, Phe, and Val.
For the study of the macrocyclization reactions of open-chain
pseudopeptides 1, dry acetonitrile (ACN) was selected as the
562
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
●
○
Figure 9. Kinetic curves for the formation of macrocycles (a) 2c and (b) 2d obtained in the presence ( ) or absence ( ) of bromide ion under
otherwise identical conditions.
solvent, taking into consideration that its polarity seems to
induce a favorable folding of the intermediates.22 However, this
precluded the use of the Gly derivative because of its low
solubility in this solvent.
therefore, this interaction should also exist in the transition
state (TS).
On the basis of these results, the same experimental protocol
was used to analyze the kinetics of the different macro-
cyclization reactions. Each reaction was carried out in an NMR
tube, and each bis(amino amide) was allowed to react with 1
equiv of 1,3-bis(bromomethyl)benzene. In all cases, ACN-d3
was used as the solvent, the temperature was kept at 25 °C, and
0.5 equiv of bromide anion was added at the beginning of the
reaction. DIPEA (10 equiv) was used as the base and
tetrakis(trimethylsilyl)silane (0.556 equiv) as the internal
Initial kinetic experiments for the [1 + 1] macrocyclization
were carried out with compounds 1d (Aib) and 1c (Ala). They
were selected as models to compare the behavior of the
derivative from a nonhindered natural amino acid (Ala) with
that of the one resulting from the related quaternary non-
natural amino acid (Aib). Macrocyclization experiments were
performed using DIPEA (10 equiv), in the absence of any
added anion or after the addition of 0.5 equiv of Br− at the
beginning of the reaction. The progress of the reaction was
followed by 1H NMR spectroscopy. DIPEA was selected as the
base because it is a tertiary amine that is strong enough to avoid
protonation of the amine groups of the different chemical
species involved in the macrocyclization but not reactive toward
any of the species participating in the process.24 The results are
represented in Figure 9. It can be seen that in agreement with
the initial hypotheses, the bromide anion has a clear positive
effect on the reaction rate and the final yield.
The interactions of both the open-chain and macrocyclic
compounds with the bromide anion was confirmed by means of
1H NMR experiments. The addition of increasing quantities of
bromide anion produced a downfield shift of the signal of the
amide group in both cases (Figure 10). This agrees with the
existence of a hydrogen-bonding interaction of the amide with
the bromide anion in the reactants and in the products, and
1
standard. H NMR spectra were registered directly from the
reaction mixture at different reaction times. The reactions were
monitored for ca. 300 h for all of the compounds except those
containing quaternary amino acids (Aib and Ac6c), for which
monitoring was extended to ca. 800 h.
1
From the H NMR spectra it was possible to obtain the
concentrations of the macrocycle and the starting materials and
detect the formation of byproducts. Figure 11 shows the
1
overlapped H NMR spectra for the process leading to 2f. The
increasing signals correspond to the macrocyclic product; the
signal that disappears at 4.4 ppm is assigned to the benzylic
protons of 1,3-bis(bromomethyl)benzene, and the one
disappearing at around 7 ppm corresponds to the amide
proton of the bis(amino amide). The signals of DIPEA move
downfield, thereby confirming its protonation as the reaction
takes place.
The formation of oligomeric/polymeric species was con-
firmed by mass spectrometry experiments. With ESI-MS it was
possible to identify the starting material, the macrocyclic
product, and some oligomers. Although the number of possible
oligomers was high (see Figure S-1 in the Supporting
Information) and their concentrations were always very low,
the MS experiments allowed unambiguous detection and
identification of several of them (see Figure S-2 in the
Supporting Information).
Figure 12 gathers the different kinetic curves obtained from
the 1H NMR experiments. From these curves, a detailed kinetic
analysis was carried out in order to determine the main kinetic
parameters for each specific macrocyclization reaction.
To avoid overparametrization of the model, it was assumed
that the kinetic constants for the SN2 reactions that afford open-
chain compounds (i.e., the first reaction step and the
oligomerization/polymerization steps) are essentially identical
(i.e., k1 = kp).24,36 This provided a simple kinetic model to
which the experimental kinetic data could be fitted. In this way,
the corresponding kinetic parameters could be calculated.
Figure 13a displays the expected time variation of the
concentrations of the different species involved for 2f obtained
Figure 10. Variation of the amide signal in the 1H NMR spectra of (a)
2d and (b) 1d with the addition of TBABr in ACN-d3 at 25 °C.
563
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
1
Figure 11. Overlapped H NMR spectra for the kinetic experiment leading to the formation of macrocycle 2f.
Figure 12. Kinetic curves obtained from 1H NMR experiments for the
formation of the different macrocycles 2 (7.5 mM) under the same
reaction conditions (0.5 equiv of bromide anion, deuterated
acetonitrile, 25 °C, 10 equiv of DIPEA as the base). For clarity,
only the data for the first 160 h of reaction have been displayed.
according to this kinetic model using the calculated kinetic
parameters [curve A, 1f; curve B, 1,3-bis(bromomethyl)-
benzene; curve C, 3f; curve D, 2f]. Complete overlap of the
experimental and calculated curves corresponding to the
starting materials was observed (Figure 13b). It can be seen
that the concentrations of the intermediate and the oligomers
are very low for the whole time span.
Table 1 summarizes the kinetic constants obtained from the
Figure 13. (a) Time variation of the concentrations of the different
species modeled according to the simplified kinetic model for 2f.
Dotted lines: curve A, 1f; curve B, 1,3-bis(bromomethyl)benzene.
Dashed line: curve C, reaction intermediate 3f. Solid line: curve D,
macrocyclic product 2f. Complete overlap of the curves corresponding
to the starting materials is observed. (b) Time variation of the
concentration of macrocyclic compound 2f using the optimized
conditions. The black solid curve was obtained from the calculated k1
and k2 values, and the red dashed lines represent the standard
deviations. Blue open circles show the experimental data.
corresponding fits using the Mathematica software and the
developed kinetic model.37 The associated errors have been
included for all cases. Moreover, the Eyring equation, k = (kBT/
h) exp(−ΔG⧧/RT),38,39 was employed to calculate the Gibbs
free energy barrier between the reactant interaction complex
and the transition state. The kinetic constants k1 associated with
the first reaction step for the tertiary amino acids Ala, Cha, and
Leu are ca. 25 M−1 h−1, indicating that the substitution of one
H atom in the β-carbon has a minor influence on the this
kinetic constant. Interestingly, Phe, which contains a phenyl
ring that favors the formation of aggregates and should decrease
its reactivity, shows a smaller k1 of 18 M−1 h−1.40,41 The
substitution of two H atoms on the β-carbon has a small but
perceptible influence on the kinetic constant of the first
reaction step. Thus, Val and Ile have k1 = 17 and 21 M−1 h−1,
respectively. In sharp contrast, an additional substitution at the
α-carbon atom has a dramatic influence on k1, and the k1 values
for the quaternary amino acids are consistently lower by factors
of 7−10. Similar k1 values were obtained for the two quaternary
amino acids, indicating again that substitution at the β-carbon
has a minor influence. A decrease of 34% in k1, similar to that
observed in going from Ala to Val (33%), was observed.
564
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
a
Table 1. Kinetic Parameters Obtained for the Different Cyclization Reactions from Open-Chain Pseudopeptides 1
b
1
Aaa
k1 (M−1 h−1
)
k2 (h−1
)
ΔG⧧1 (kcal mol−1
)
ΔG⧧2 (kcal mol−1
)
ΔΔG⧧ (kcal mol−1
)
1a
1b
1c
1d
1e
1f
Phe
Val
17.54 0.01
16.5 1.3
24.8 1.0
3.65 0.11
2.40 0.12
24.9 2.5
21.42 1.5
25.0 1.0
1.53 0.14
1.94 0.20
20.60 0.01
20.64 0.05
20.40 0.02
21.54 0.02
21.78 0.03
20.40 0.06
20.49 0.05
20.40 0.02
22.05 0.05
21.91 0.06
22.22 0.01
23.40 0.01
23.70 0.02
21.61 0.01
21.81 0.05
22.13 0.02
1.45 0.06
1.27 0.11
1.82 0.03
1.86 0.03
1.92 0.05
1.21 0.07
1.32 0.10
1.73 0.04
Ala
1.145 0.001
0.158 0.004
0.094 0.003
3.211 0.001
2.29 0.20
Aib
Ac6c
Cha
Ile
1g
1h
Leu
1.33 0.04
a
Compound 1i (Aaa = Gly) was found to yield only oligomers; the macrocycle could not be isolated from the reaction mixture.42 This compound
b
has low solubility under the reaction conditions. ΔΔG⧧ = ΔG2⧧ − ΔG1⧧.
Figure 14. IRC profiles for the macrocyclization reaction steps of pseudopeptides 1 with 1,3-bis(bromomethyl)benzene in acetonitrile in the absence
or presence of bromide computed at the B3LYP/6-31G* level of theory: (a) first reaction step; (b) second reaction step.
However, the values of the kinetic constant k2 associated with
the macrocyclization step for the quaternary amino acids
(0.10−0.15 h−1) are 8−30 times lower than those for the
tertiary amino acids (3.2−1.2 h−1). This provides evidence that
the substituent has a more pronounced effect on the
macrocyclization step. Interestingly, this effect is not directly
related to the volume of the substituent.
For this system (Figure 1), the yield obtained in the
macrocyclization reaction must be related to the k2/k1 ratio. As
the first step is common for oligomerization and cyclization,
this ratio determines the relative rates at which macrocycles and
oligomers are formed. Consequently, the simplest way to
analyze our results is through the corresponding values of
ΔΔG⧧ = ΔG2⧧ − ΔG⧧1 gathered in Table 1. The yields are
higher for systems displaying smaller values of ΔΔG⧧.
It can be seen that the cyclohexylalanine derivative (1f) has
the lowest difference between the energy barriers for the
macrocyclization and the first reaction. On the contrary, the
compounds derived from quaternary amino acids (1d and 1e)
are the least favorable for macrocyclization and also present the
highest energy barriers for both reaction steps. Although large
variations were observed among the other compounds, no clear
structural trends could be detected. Thus, the less hindered Ala
derivative (1c) is less favorable for cyclization than the sterically
more demanding Val derivative (1b) (ΔΔG⧧ = 1.82 and 1.27
kcal mol−1, respectively). This is essentially associated with the
difference in the energy barriers for the cyclization step, for
which the conformational effects are critical, suggesting a
complex effect of the side chains beyond their steric volume. As
can be seen, except for the quaternary amino acid derivatives,
differences in the energy barrier for k1 are very small (<0.25
kcal mol−1) while the differences for the second step are more
significant (up to 0.60 kcal mol−1).
The data from the kinetic studies were also useful for
determining the reaction parameters for the syntheses of the
different macrocycles on a synthetic scale.22,43 In all cases, the
yields obtained were in excellent agreement with the results
from the kinetic analysis. However, for the quaternary amino
acid derivatives, the formation of larger amounts of side
products led to more difficult purification processes, signifi-
cantly decreasing the final isolated yields.
Ab Initio Calculations. No direct correlations could be
found between the experimental data and the parameters
obtained by molecular mechanics calculations (i.e., the average
C···N distance for the reaction intermediate 3). Accordingly, an
in silico study of these processes was carried out by means of
reaction profile calculations. A kinetic analysis involves
calculating the transition state that connects reactants and
products and the energy barrier between them.24 For this
purpose, we selected the systems derived from the amino acids
Aib, Ala, and Gly. Their sizes still allow the use of high-level
calculations. On the other hand, they represent a well-defined
family of derivatives for which the substitution at the α-carbon
increases steadily.
Both reaction steps were modeled with the presence of
added bromide anion as a catalytic species taken into account.
The structures of the reactants, complexes of interaction of
reactants (RICs), transition states (TSs), complexes of
interaction of products (PICs), and products were optimized
to obtain the corresponding reaction profiles.43,44
An initial potential energy surface (PES) calculation at the
PM3 level of theory allowed the approximate TS geometry as
565
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
well as those for the other stationary points to be found (see
Figures S-3 and S-4 in the Supporting Information for the
transformation of 1c). These structures were then fully
optimized at the B3LYP/6-31G* level of theory in the presence
of solvent (acetonitrile) using the self-consistent reaction field
(SCRF) theory, which performs polarized continuum model
(PCM) calculations.45−47 From these optimized TSs, intrinsic
reaction coordinate (IRC) calculations were completed in order
to obtain the whole reaction profile.48 The last geometries at
both ends of the IRC path were optimized in order to obtain
the corresponding optimized geometries of the completely true
RIC and PIC minima (Figure 14).
The optimized geometries of the transition states for 1c, 1d,
and 1i for the first and second reaction steps in the absence of
anion are represented in Figures 15a,c,e and 16a,c,e,
Figure 16. TSs for the macrocyclization step of the reaction of 1 with
1,3-bis(bromomethyl)benzene in acetonitrile in (a, c, e) the absence
and (b, d, f) the presence of bromide, calculated B3LYP/6-31G* level
of theory: (a) 1d Aib; (b) 1d Aib + Br−; (c) 1c Ala; (d) 1c Ala + Br−;
(e) 1i Gly; (f) 1i Gly + Br−.
reactive amine is observed only for the first step, but the anion
forms additional H-bonds with the different amine and amide
groups. In the first reaction step, three H-bonds between the
bromide and the hydrogen atoms of one amide and the two
amines are observed. In the macrocyclization step, however,
four H-bonds are observed with the bromide, involving the two
amide hydrogen atoms and one hydrogen atom of each amino
group. This preorganizes the system and neutralizes the
increment on the partial charge on the amine group as it
becomes an ammonium group as the reaction advances. The
overall result is a decrease in the energy barrier between the
reactants and products.
In order to analyze the influence of the side chain of the
constituent amino acid and to compare the computational data
with the kinetic results previously obtained, the RIC−TS Gibbs
free energy barriers, entropy changes, and enthalpy barriers
were calculated. Table 2 summarizes the results obtained for
Figure 15. TSs for the first step of the reaction of 1 with 1,3-
bis(bromomethyl)benzene in acetonitrile in (a, c, e) the absence and
(b, d, f) the presence of bromide, calculated at the B3LYP/6-31G*
level of theory: (a) 1d Aib; (b) 1d Aib + Br−; (c) 1c Ala; (d) 1c Ala +
Br−; (e) 1i Gly; (f) 1i Gly + Br−.
respectively. In all cases, an intramolecular H-bond between
the oxygen atom of a carbonyl group and one hydrogen atom
of the reactive amine is observed. This preorganizes the
reaction intermediate and helps to reduce the positive partial
charge evolving on the amine group during its transformation
into an ammonium group. Moreover, for the geometries for the
first reaction step, the higher flexibility of the open-chain
compounds allows for an additional H-bond between the N
atom of the nonreacting amine and the second hydrogen atom
of the reacting amine, which also stabilizes the developing
ammonium group. In the case of the macrocyclization step, this
second stabilizing H-bond is not possible.
Table 2. RIC−TS Gibbs Free Energy Barriers (kcal mol−1)
for Both Reaction Steps
Aaa
no catalyst Br− catalyst
ΔGRIC−TS for first reaction step
ΔGRIC−TS for macrocyclization step
ΔΔGRIC−TS
Aib
Ala
Gly
Aib
Ala
Gly
Aib
Ala
Gly
19.66
15.38
15.08
19.41
17.25
17.19
−0.25
1.87
17.64
15.26
14.68
15.22
15.12
15.13
−2.42
−0.14
0.45
In contrast, in the presence of bromide (Figures 15b,d,e and
16b,d,e) the intramolecular H-bond between the oxygen atom
of the carbonyl group and one of the hydrogen atoms of the
2.11
566
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
both reaction steps for 1d (Aib), 1c (Ala), and 1i (Gly). It can
be observed that the anion significantly reduces the calculated
ΔGRIC−TS energy barriers in acetonitrile. The corresponding
calculated entropy and enthalpy barriers (see Tables S1−S3 in
the Supporting Information) are presented in Figure 17.
3, and the corresponding Boltzmann distributions of the RIC
(% abundance) are shown in Table 4. The amino acid side
Table 3. Relative Gibbs Free Energies of the RIC Relative to
the Most Stable Calculated Conformer for Both Reaction
Steps at 25 °C (B3LYP/6-31G*, Acetonitrile)
ΔG (kcal mol−1
)
reaction step
first step
Aaa
no catalyst
Br− catalyst
Aib
Ala
Gly
Aib
Ala
Gly
8.21
6.03
5.48
9.89
8.01
7.17
0.466
0.446
1.58
1.92
1.26
1.79
macrocyclization step
Table 4. Boltzmann Distributions of the RIC (%) for Both
Reaction Steps at 25 °C (B3LYP/6-31G*, Acetonitrile)
Boltzmann population (%)
reaction step
first step
Aaa
no catalyst
Br− catalyst
Aib
Ala
Gly
Aib
Ala
Gly
0.00829
0.0731
0.1264
0.00154
0.0101
0.0233
19.06
19.43
6.25
4.45
8.64
5.06
Figure 17. Enthalpy vs entropy plots (T = 298.15 K) for both reaction
steps in acetonitrile, calculated at the B3LYP/6-31G* level of theory:
Aib, red; Ala, blue; Gly, green.
macrocyclization step
The calculated entropy change for going from the RIC to the
TS is negative in all cases, as expected in an SN2 reaction. From
the ΔSRIC−TS values it must be pointed out that in the case of
the Br−-catalyzed reaction in acetonitrile, the value of ΔSRIC−TS
for the macrocyclization step is more negative than that for first
reaction step. Therefore, the use of low temperatures will favor
the formation of the macrocycle as long as the negative
contribution of the enthalpy to the RIC−TS free energy barrier
will be reduced. Regarding the enthalpy change from RIC to
TS, the lowest ΔHRIC−TS value was obtained for the
macrocyclization step catalyzed by bromide in acetonitrile
(diamonds in Figure 17), in good agreement with our initial
selection of working hypotheses.
The values show that the energy barrier for the first reaction
step increases significantly with the steric hindrance of the
amino acid side chain in the absence and presence of Br−. The
same trend is observed for the second step in the absence of
bromide. On the contrary, the calculations suggest that the
amino acid side chain has essentially no effect on the
macrocyclization energy barrier in the presence of the anion.
Thus, the data in Table 2 explain the catalytic effect observed
experimentally for the presence of the bromide anion but do
not allow a simple explanation of the effects detected for the
changes in the side chain.
In this regard, the existence of different conformers for each
open-chain intermediate and their populations at equilibrium
needs to be considered. According to the data obtained from
the MMFF calculations, the structures of such conformers and
their relative energies are strongly dependent on the nature of
the side chain. In such equilibria, we can consider that a reactive
conformation must have the appropriate geometry (RIC) to
react to afford the products. Thus, the low-energy conformers
obtained from the Monte Carlo conformational searches at the
MMFF level of theory were optimized at the B3LYP/6-31G*
level of theory in acetonitrile. The relative energies of the RIC
relative to that of the most stable conformer are shown in Table
chain has a direct influence on the Boltzmann distribution,
although a simple correlation with the size does not exist,
particularly in the presence of Br−. In all cases, the RIC is less
stable than the most stable conformer, but this unfavorable
energy difference is much less important in the presence of Br−.
Therefore, the effect of the anion is double: it reduces the
RIC−TS energy barrier and increases the population of reactive
RIC molecules.
Thus, in the presence of the catalytic anion, the amino acid
side chain has an effect on both the conformation and the
ΔGRIC−TS barrier for the first reaction step, while for the critical
macrocyclization step the changes in the side chain are reflected
only in changes in the conformational equilibria of the open-
chain precursor. This highlights the importance of the amino
acid side chain on the preorganization of the open-chain
precursor with the appropriate folded conformation (RIC).
These results are in nice agreement with our previous studies
with 2,6-bis(aminomethyl)pyridine, in which the side chain of
the constituent amino acid had a minor role as long as the
pyridine spacer itself facilitated the folding of the open-chain
precursor to a proper conformation.24 As could be expected,
other factors associated with the side chain besides the steric
volume, such as the hydrophobic/hydrophilic character or the
possibility of participating in supramolecular interactions, can
be of importance for determining the corresponding conforma-
tional equilibria of the open-chain intermediates 3.
Taking into account both the calculated energy barriers and
the Boltzman partition function of the RIC, the corresponding
reaction kinetics for the formation of the macrocyclic
compound can be predicted, as shown in Figure 18.24,37 As
can be seen, there is an excellent match between the trends
detected by the calculations and the ones observed
experimentally (Figure 12).
567
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
apparatus in positive mode at 20 and 45 V to improve the resolution of
high-mass polymers.
NMR Kinetic Experiments. The macrocyclization reactions were
1
followed by H NMR spectroscopy. Tetrakis(trimethylsilyl)silane was
used as an internal standard to reference the peaks and the integrals of
the different signals of the reactants and products. This allowed the
quantification of the signals of the product of the macrocyclization
reaction and those corresponding to the starting reactants. It was
checked in an isothermal experiment at 25 °C, using a constant gain of
the NMR receiver, that the integral of the internal standard was not
disturbed under the reaction conditions. The different experiments
were performed directly in NMR tubes at 25 °C with 1.00 mL of
deuterated acetonitrile. In this way the macrocyclization reaction could
be monitored using very small amounts of reactants.
Figure 18. Simulation of the macrocyclization reaction kinetics from
the B3LYP/6-31G*-calculated energy barriers in acetonitrile. [1]0 =
[1,3-bis(bromomethyl)benzene]0 = 7.450 mM and [Br−]0 = 3.725
mM at 25 °C in acetonitrile.
All of the reactants, bis(amino amide) (1.723 mg, 0.00747 mmol for
compound 1d), TBABr (1.206 mg, 0.00374 mmol), DIPEA (12.81 μL,
0.07482 mmol), and internal standard tetrakis(trimethylsilyl)silane
(0.1333 mg, 0.004156 mmol), were placed in a vial, and then
deuterated acetonitrile (1.00 mL) was added. The reaction mixture
was stirred and heated until a clear solution was obtained. Then 1,3-
bis(bromomethyl)benzene was weighed in a second vial (1.974 mg,
0.00747 mmol). The reaction mixture was transferred quantitatively to
the NMR tube; this was inserted into the NMR probe and allowed to
equilibrate. Then the lock and shim were adjusted, and a reference 1H
NMR was acquired. After that, the contents of the NMR tube were
quantitatively poured into the vial containing 1,3-bis(bromomethyl)-
benzene and immediately put back to the NMR tube. The lock and
CONCLUSIONS
■
Both experimental results and theoretical calculations show that
the amino acid side chain plays a critical role in the
preorganization of pseudopeptides and therefore in their
macrocyclization tendency. An increase in the number of
substituents on the α carbon atom (quaternary amino acids)
decreases the macrocyclization reaction rate significantly. The
overall observed effect for the change in the side chain of the
constituent amino acid is not directly related to its size but
rather involves the modification of the conformational
preferences of the open-chain intermediate 3.
1
shim were adjusted again. The H NMR spectra were acquired at
different times until the end of the reaction.
Computational Details. All of the structures were minimized
using Gaussian 09.49 Calculations were performed in acetonitrile using
the SCRF methodology. The true nature of the minima was
demonstrated by a normal mode of vibration analysis. For TSs, one
single negative eigenvalue was found, corresponding to the reaction
path involving the breaking and forming bonds. For the other
structures, all of the eigenvalues were positive. Monte Carlo
conformational searches using the MMFF force field were carried
out with the Spartan’08 software.50
Among the different compounds studied, compound 1d
derived from cyclohexylalanine is the one that cyclizes more
efficiently. The calculations revealed that in this case the most
stable conformation in the presence of the bromide anion
approximates very efficiently, with the correct orientation, the
two reactive ends of the reaction intermediate. Thus, the
selection of the appropriate amino acid can play a key role in
determining the success of a macrocyclization in peptidic or
pseudopeptidic structures.
Synthesis of Compounds. Bis(amino amine)s 1a, 1b, 1c, 1g, 1h,
and 1i and macrocycles 2a, 2b, 2c, 2g, and 2h were prepared using the
conditions previously reported.22,43
General Procedure for the Synthesis of Boc-Protected Bis(amino
amide)s: Synthesis of tert-Butyl N-(1-{[2-(2-{[(tert-Butoxy)carbonyl]-
amino}-2-methylpropanamido)ethyl]carbamoyl}-1-methylethyl)-
carbamate (1d-Boc). Boc-Aib-OH (1.455 g, 7.160 mmol) was placed
in a 250 mL round-bottom flask, and dry acetonitrile (100 mL) was
added. BOP (2.802 g, 8.596 mmol) was then added to the reaction
mixture, followed by DIPEA (1.497 mL, 8.596 mmol). The reaction
was stirred for 30 min at rt under a nitrogen atmosphere, and then
ethylenediamine (239 μL, 3.580 mmol) was added. The reaction was
kept for 1 day at rt, and then the solvent was eliminated under reduced
pressure. The obtained oil was dissolved in dichloromethane (50 mL)
and washed with a 5% solution of KHSO4 (3 × 20 mL), a 5% solution
of NaHCO3 (3 × 20 mL), and water (3 × 20 mL). The organic layer
was dried with MgSO4, and the solvent was evaporated under reduced
pressure to obtain a white foam. The product was purified by column
chromatography on flash silica gel [3 cm diameter, 15 cm height;
mobile phases: CH2Cl2/MeOH 20:1 to 10:1 v/v (mL/mL)]. The pure
product was collected in fractions 3−5 as a white solid (2.682 g, 6.229
mmol, 87% yield). Mp 234−235 °C; IR (ATR) 3330, 2979, 2930,
1688, 1659, 1545, 1518, 1449, 1388, 1364 cm−1; 1H NMR (500 MHz,
DMSO-d6) δ 1.25 (6H, s), 1.34 (9H, s), 3.05 (2H, s), 6.79 (1H, s),
7.51 (1H, s); 13C NMR (126 MHz, DMSO-d6) δ 25.6, 28.6, 39.3, 56.0,
78.3, 154.6, 174.9; HRMS (ESI-TOF)+ calcd for C20H38N4O6 (M +
Na)+ 453.2689, found 453.2685. Anal. Calcd for C20H38N4O6: C,
55.79; H, 8.90; N, 13.01. Found: C, 55.44; H, 8.68; N, 12.89.
Synthesis of tert-Butyl N-[1-({2-[(1-{[(tert-Butoxy)carbonyl]-
amino}cyclohexyl)formamido]ethyl}carbamoyl)cyclohexyl]-
carbamate (1e-Boc). This compound was prepared using the general
procedure described above as a white solid (1.350 g, 2.64 mmol, 26%
The excellent agreement between the experimental and
computational results highlights the potential of the appropriate
use of computational modeling of open-chain intermediates to
predict their tendencies to cyclize. For this purpose, the
corresponding energy barriers are not the only parameters to be
considered, and the associated conformational equilibria of the
open-chain intermediates have to be considered in detail.
Therefore, this methodology can be efficiently used in the
design of new macrocyclic peptidic and pseudopeptidic
structures and the most appropriate synthetic routes for their
preparation.
EXPERIMENTAL SECTION
■
1
The NMR experiments were carried out at 500 or 300 MHz for H
and 125 or 75 MHz for 13C. Chemical shifts are reported in parts per
million from tetramethylsilane with the solvent resonance as the
internal standard. FT-IR spectra were recorded using an instrument
with a MIRacle Single Reflection ATR Diamond/ZnSe accessory.
Microanalyses were performed on an elemental analyzer. Mass spectra
were recorded with a Q-TOF instrument. Rotatory power was
determined with a digital polarimeter (Na: 589 nm). Melting points
were measured using a standard apparatus and are uncorrected.
ESI-MS Experiments. The crude reaction product in acetonitrile,
which contained some precipitates because of the poor solubility of the
oligomers in acetonitrile, was solubilized in a DMSO/methanol 1:1
(v/v) mixture and diluted to the appropriate concentration for the
mass spectrometer. The samples were measured in the ESI-MS
568
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Article
yield). Mp 179−180 °C; IR (ATR) 3320, 2980, 2934, 2852, 1686,
1644, 1542, 1515, 1453, 1390, 1365 cm−1; 1H NMR (500 MHz,
DMSO-d6) δ 1.11−1.19 (2H, m), 1.35 (18H, s), 1.36−1.51 (12H, m),
1.52−1.65 (4H, m), 1.67−1.96 (4H, m), 3.04 (4H, s), 6.54 (2H, s),
7.39 (2H, s); 13C NMR (126 MHz, DMSO-d6) δ 21.1, 25.1, 28.2, 31.8,
38.8, 58.4, 77.9, 154.3, 174.7; HRMS (ESI-TOF)+ calcd for
C26H46N4O6 (M + Na)+ 533.3315, found 533.3321. Anal. Calcd for
C26H46N4O6: C, 61.15; H, 9.08; N, 10.97. Found: C, 60.77; H, 9.04;
N, 10.90.
reaction mixture was heated until all of the reactants were dissolved.
Then 1,3-bis(bromomethyl)benzene (114.6 mg, 0.434 mmol) was
weighed and dissolved into a small quantity of dry acetonitrile and
placed into the reaction flask. The reaction mixture was stirred for 3
weeks at 25 °C under a nitrogen atmosphere. The solvent was
removed under reduced pressure to obtain the crude reaction as a
foam. The foam was dissolved in CHCl3 (25 mL) and washed with
water (3 × 25 mL). The organic layer was dried with anhydrous
magnesium sulfate and evaporated to dryness. Purification of this
product was carried out by column chromatography with flash silica
gel (50 g). CH2Cl2/MeOH mixtures (from 10:0 to 10:2 v/v)
containing a few drops of aqueous ammonia to prevent the retention
of amines in the column were used as the mobile phase. Macrocycle 2d
was obtained as a white solid (15 mg, 0.045 mmol, 10% yield). Mp
187−188 °C; IR (ATR) 3357, 3301, 2981, 2945, 1645, 1506, 1480,
1441, 1378, 1361 cm−1; 1H NMR (500 MHz, CDCl3) δ 1.38 (12H, s),
1.61 (2H, s), 3.37 (4H, s), 3.72 (4H, s), 7.08 (2H, d, J = 6.7 Hz),
7.22−7.26 (1H, m), 7.57 (1H, s), 7.92 (2H, s); 13C NMR (75 MHz,
CDCl3) δ 25.9, 39.7, 48.3, 59.4, 126.4, 128.7, 129.0, 141.6, 177.5;
HRMS (ESI-TOF)+ calcd for C18H28N4O2 (M + H)+ 333.2291, found
333.2294. Anal. Calcd for C18H28N4O2: C, 65.03; H, 8.49; N, 16.85.
Found: C, 64.72; H, 8.62; N, 16.57.
Synthesis of Dispiro[cyclohexane-1,4′-[3,6,9,12]tetraazabicyclo-
[12.3.1]octadecane-11′,1″-cyclohexane]-1′(17′),14′(18′),15′-triene-
5′,10′-dione (2e). This compound was prepared using the general
procedure described above. The macrocycle was obtained as a white
solid (18 mg, 0.0436 mmol, 9% yield). Mp 192−193 °C; IR (ATR)
3319, 2918, 2856, 1662, 1498, 1479, 1458, 1445, 1427, 1354 cm−1; 1H
NMR (500 MHz, CDCl3) δ 1.06−1.25 (19H, m), 1.41−1.55 (31H,
m), 1.64−1.74 (19H, m), 3.17 (12H, s), 3.42 (12H, s), 6.92 (6H, d, J
= 7.4 Hz), 7.06−7.11 (3H, m), 7.33 (3H, s), 7.66 (6H, s); 13C NMR
(75 MHz, CDCl3) δ 21.7, 25.1, 32.1, 40.0, 47.7, 61.9, 126.7, 128.8,
129.9, 141.7, 177.8; HRMS (ESI-TOF)+ calcd for C24H36N4O2 (M +
H)+ 413.2917, found 413.2913. Anal. Calcd for C24H36N4O2: C, 69.87;
H, 8.80; N, 13.58. Found: C, 69.59; H, 8.88; N, 13.63.
Synthesis of (4S,11S)-4,11-Bis(cyclohexylmethyl)-3,6,9,12-
tetraazabicyclo[12.3.1]octadeca-1(18),14,16-triene-5,10-dione (2f).
This compound was prepared using the general procedure described
above. The macrocycle was obtained as a white solid (114 mg, 0.2432
mmol, 45% yield). Mp 199−200 °C; [α]2D5 = +8.75 (c = 0.01, CHCl3);
IR (ATR) 3344, 3306, 2921, 2846, 1654, 1520, 1447, 1362, 1323
cm−1; 1H NMR (500 MHz, CDCl3) δ 0.89−1.02 (4H, m), 1.10−1.30
(6H, m), 1.31−1.46 (4H, m), 1.62−1.82 (12H, m), 1.83 (2H, s),
2.99−3.08 (4H, m), 3.21 (2H, dd, J = 4.3, 9.3 Hz), 3.52 (2H, d, J =
13.1 Hz), 4.05 (2H, d, J = 13.1 Hz), 7.08 (2H, dd, J = 1.6, 7.5 Hz),
7.23 (3H, dd, J = 5.7, 9.3 Hz), 7.30 (1H, s); 13C NMR (75 MHz,
CDCl3) δ 26.3, 26.5, 26.6, 32.8, 34.2, 34.8, 39.6, 42.3, 54.8, 62.2, 127.9,
128.8, 129.5, 140.6, 176.6; HRMS (ESI-TOF)+ calcd for C28H44N4O2
(M + H)+ 469.3543, found 469.3549. Anal. Calcd for C28H44N4O2: C,
71.76; H, 9.46; N, 11.95. Found: C, 71.38; H, 9.40; N, 11.87.
Synthesis of tert-Butyl N-[(1S)-1-({2-[(2S)-2-{[(tert-Butoxy)-
carbonyl]amino}-3-cyclohexylpropanamido]ethyl}carbamoyl)-2-
cyclohexylethyl]carbamate (1f-Boc). This compound was prepared
using the general procedure described above as a white solid (1.732 g,
3.056 mmol, 44% yield). Mp 170−171 °C; [α]2D5 = +56.40 (c = 0.01,
CHCl3); IR (ATR) 3340, 2923, 2852, 1680, 1661, 1520, 1446, 1392,
1
1368 cm−1; H NMR (500 MHz, DMSO-d6) δ 0.75−0.90 (4H, m),
1.03−1.17 (6H, m), 1.19−1.26 (2H, m), 1.32−1.39 (22H, m), 1.53−
1.65 (8H, m), 1.71 (2H, d, J = 12.6 Hz), 2.95−3.15 (4H, m), 3.90
(2H, dd, J = 8.7, 14.6 Hz), 6.73 (2H, d, J = 8.2 Hz), 7.77 (2H, s); 13C
NMR (75 MHz, DMSO-d6) δ 26.1, 26.3, 26.5, 28.6, 32.3, 33.7, 34.1,
38.8, 52.5, 78.4, 155.8, 173.2; HRMS (ESI-TOF)+ calcd for
C30H54N4O6 (M + Na)+ 589.3941, found 589.3940. Anal. Calcd for
C30H54N4O6: C, 63.57; H, 9.60; N, 9.89. Found: C, 63.29; H, 9.31; N,
10.12.
General Procedure for the Deprotection of Boc-Protected
Bis(amino amide)s: Synthesis of 2-Amino-N-[2-(2-amino-2-
methylpropanamido)ethyl]-2-methylpropanamide (1d). Com-
pound 1d-Boc (679 mg, 1.577 mmol) was dissolved in the minimum
quantity of dichloromethane (80 mL) by reflux heating, and then TFA
(10 mL) was added. The reaction was stirred under a nitrogen
atmosphere for 1 h, and then the solvent was removed under vacuum.
Water (10 mL) was added, and the mixture was basicified to pH 12
with solid NaOH and extracted with CHCl3 (3 × 25 mL). The organic
layers were combined and dried with anhydrous MgSO4. The solvent
was evaporated under reduced pressure. The product was obtained as
a white solid (241 mg, 1.046 mmol, 66% yield). Mp 114−115 °C; IR
(ATR) 3384, 3283, 2972, 2934, 1630, 1519, 1442, 1375, 1357 cm−1;
1H NMR (500 MHz, CDCl3) δ 1.32 (6H, s), 1.42 (2H, s), 3.35 (2H,
dd, J = 2.5, 3.1 Hz), 7.90 (1H, s); 13C NMR (126 MHz, CDCl3) δ
29.2, 39.6, 54.7, 178.5; HRMS (ESI-TOF)+ calcd for C10H22N4O2 (M
+ H)+ 231.1821, found 231.1815. Anal. Calcd for C10H22N4O2: C,
52.15; H, 9.63; N, 24.33. Found: C, 52.04; H, 9.26; N, 23.99.
Synthesis of 1-Amino-N-{2-[(1-aminocyclohexyl)formamido]-
ethyl}cyclohexane-1-carboxamide (1e). This compound was pre-
pared using the general procedure described above as a white solid
(410 mg, 1.321 mmol, 72% yield). Mp 149−150 °C; IR (ATR) 3338,
3284, 2921, 2851, 1649, 1598, 1444, 1361 cm−1; 1H NMR (500 MHz,
CDCl3) δ 1.21−1.34 (6H, m), 1.35−1.44 (8H, m), 1.56−1.63 (6H,
m), 1.88−1.97 (4H, m), 3.30−3.35 (4H, m), 8.05 (2H, s); 13C NMR
(126 MHz, CDCl3) δ 21.3, 25.3, 34.7, 39.6, 57.3, 178.9; HRMS (ESI-
TOF)+ calcd for C16H30N4O2 (M + H)+ 311.2447, found 311.2438.
Anal. Calcd for C16H30N4O2: C, 61.90; H, 9.74; N, 18.05. Found: C,
61.58; H, 9.56; N, 17.86.
Synthesis of (2S)-2-Amino-N-{2-[(2S)-2-amino-3-cyclohexyl-
propanamido]ethyl}3-cycloexylropanamide (1f). This compound
was prepared using the general procedure described above as a
white solid (837 mg, 1.477 mmol, 55% yield). Mp 131−132 °C; [α]D25
= −31.34 (c = 0.01, CHCl3); IR (ATR) 3330, 3301, 2979, 2930, 1688,
1659, 1546, 1518, 1449, 1388, 1364 cm−1; 1H NMR (500 MHz,
CDCl3) δ 0.84−1.01 (4H, m), 1.09−1.30 (8H, m), 1.33−1.46 (6H,
m), 1.59−1.78 (12H, m), 3.37 (6H, s), 7.64 (2H, s); 13C NMR (126
MHz, CDCl3) δ 26.2, 26.5, 26.6, 32.2, 34.3, 34.4, 39.6, 42.9, 52.9,
176.9; HRMS (ESI-TOF)+ calcd for C20H38N4O2 (M + H)+ 367.3073,
found 367.3066. Anal. Calcd for C20H38N4O2: C, 65.54; H, 10.45; N,
15.29. Found: C, 65.40; H, 10.06; N, 15.15.
ASSOCIATED CONTENT
■
S
* Supporting Information
1H and 13C NMR spectra as well as mass spectra for
compounds 1-Boc, 1, and 2; 2D NMR spectra for compounds
2; and optimized XYZ geometries. This material is available free
of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Authors
*E-mail: luiss@qio.uji.es.
*E-mail: cativiela@unizar.es.
Notes
General Procedure for the Macrocyclization of Bis(amino
amide)s: Synthesis of 4,4,11,11-Tetramethyl-3,6,9,12-tetraazabi-
cyclo[12.3.1]octadeca-1(18),14,16-triene-5,10-dione (2d). Com-
pound 1d (100 mg, 0.434 mmol), TBABr (70.0 mg, 0.217 mmol),
and DIPEA (743 mL, 4.343 mmol) were placed in a 250 mL round-
bottom flask, and then dry acetonitrile (120 mL) was added. The
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
S.V.L. acknowledges GV (PROMETEO/2012/020) and
MINECO (CTQ2012-38543-C03-01). C.C. acknowledges
569
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570

The Journal of Organic Chemistry
Ministerio de Ciencia e Innovacion
Article
(30) Wang, D.; Song, S.; Tian, Y.; Xu, Y.; Miao, Z.; Zhang, A. J. Nat.
Prod. 2013, 76, 974−978.
(31) Fluxa, V. S.; Reymond, J.-L. Bioorg. Med. Chem. 2009, 17, 1018−
1025.
́
−FEDER (CTQ2010-
17436) and Gobierno de Aragon−FSE (Research Group
́
E40). V.M.-C. thanks MICINN for personal financial support
(FPU Fellowship AP2007-02562). The support of the SCIC of
the UJI for the different instrumental techniques is acknowl-
edged.
(32) Flory, P. J.; Semlyen, J. A. J. Am. Chem. Soc. 1966, 88, 3209−
3212.
(33) Winnik, M. A. Acc. Chem. Res. 1985, 18, 73−79.
(34) Halgren, T. A. J. Comput. Chem. 1996, 17, 490−519.
(35) Gutte, B. Peptides: Synthesis, Structures, and Applications; Elsevier
Science: London, 1995.
REFERENCES
■
(1) Wessjohann, L. A.; Ruijter, E.; Garcia-Rivera, D.; Brandt, W. Mol.
Diversity 2005, 9, 171−186.
(36) Martí-Centelles, V.; Burguete, M. I.; Luis, S. V. Sci. World J.
2012, No. 748251.
(2) Wessjohann, L. A.; Ruijter, E. Strategies for Total and Diversity-
Oriented Synthesis of Natural Product(-Like) Macrocycles. In Natural
Product Synthesis I; Springer: Berlin, 2005; pp 137−184.
(3) Kubik, S. Chem. Soc. Rev. 2009, 38, 585−605.
(4) Kubik, S. Cyclopeptide Derived Synthetic Receptors. In Artificial
Receptors for Chemical Sensors; Mirsky, V. M., Yatsimirsky, A. K., Eds.;
Wiley-VCH: Weinheim, Germany, 2010; pp 135−167.
(5) Kubik, S. Synthetic Peptide-Based Receptors. In Supramolecular
Chemistry: From Molecules to Nanomaterials; Gale, P. A., Steed, J. W.,
Eds.; John Wiley & Sons: Chichester, U.K., 2012; Vol. 3, pp 1179−
1203.
(37) Mathematica Software; Wolfram Research, Inc.: Champaign, IL,
2008.
(38) Laidler, K. J.; King, M. C. J. Phys. Chem. 1983, 87, 2657−2664.
(39) Petersson, G. A. Theor. Chem. Acc. 2000, 103, 190−195.
(40) Fujita, Y.; Fujita, S.; Okada, Y.; Chiba, K. Org. Lett. 2013, 15,
1155−1157.
(41) Becerril, J.; Bolte, M.; Burguete, M. I.; Escorihuela, J.; Galindo,
F.; Luis, S. V. CrystEngComm 2010, 12, 1722−1725.
(42) Becerril, J.; Burguete, M. I.; Escuder, B.; Galindo, F.; Gavara, R.;
Miravet, J. F.; Luis, S. V.; Peris, G. Chem.Eur. J. 2004, 10, 3879−
3890.
(6) Blankenstein, J.; Zhu, J. Eur. J. Org. Chem. 2005, 1949−1964.
(7) Wessjohann, L. A.; Ruijter, E. Mol. Diversity 2005, 9, 159−169.
(8) Bazzicalupi, C.; Bencini, A.; Bianchi, A.; Giorgi, C.; Paoletti, P.;
(43) Laidler, K. J.; King, M. C. J. Phys. Chem. 1983, 87, 2657−2664.
(44) Truhlar, D. G.; Hase, W. L.; Hynes, J. T. J. Phys. Chem. 1983, 87,
2664−2682.
(45) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.
(46) Tomasi, J.; Persico, M. Chem. Rev. 1994, 94, 2027−2094.
(47) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3094.
Valtancoli, B.; Fusi, V.; Garcia-Espanta, B. A.; Llinares, J. M. J.
̃
Heterocycl. Chem. 2001, 38, 1273−1279.
(9) White, C. J.; Yudin, A. K. Nat. Chem. 2011, 3, 509−524.
(10) Jiang, S.; Li, Z.; Ding, K.; Roller, P. P. Curr. Org. Chem. 2008, 12,
1502−1542.
(48) Fukui, K. Acc. Chem. Res. 1981, 14, 363−368.
(49) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,
B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.
P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;
Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,
T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.;
Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,
K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;
Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,
N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;
Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;
Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;
Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;
(11) Davies, J. S. J. Pept. Sci. 2003, 9, 471−501.
(12) Lambert, J. N.; Mitchell, J. P.; Roberts, K. D. J. Chem. Soc., Perkin
Trans. 1 2001, 471−484.
(13) Laurent, B. A.; Grayson, S. M. Chem. Soc. Rev. 2009, 38, 2202−
2213.
(14) Winnik, M. A. Chem. Rev. 1981, 81, 491−524.
(15) Ten Cate, A. T.; Kooijman, H.; Spek, A. L.; Sijbesma, R. P.;
Meijer, E. W. J. Am. Chem. Soc. 2004, 126, 3801−3808.
(16) Huang, F.; Nau, W. M. Angew. Chem., Int. Ed. 2003, 42, 2269−
2272.
(17) Marsault, E.; Peterson, M. L. J. Med. Chem. 2011, 54, 1961−
2004.
(18) Gibson, S. E.; Lecci, C. Angew. Chem., Int. Ed. 2006, 45, 1364−
1377.
(19) Luis, S. V.; Alfonso, I. Acc. Chem. Res. 2013, DOI: 10.1021/
̈
Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.;
Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09,
revision B.01; Gaussian, Inc.: Wallingford, CT, 2009.
(50) Deppmeier, B. J.; Driessen, A. J.; Hehre, T. S.; Hehre, W. J.;
Johnson, J. A.; Klunzinger, P. E.; Leonard, J. M.; Pham, I. N.; Pietro,
W. J.; Jianguo, Y. Spartan’08, build 132 (Mar 27, 2009); Wavefunction,
Inc.: Irvine, CA, 2009.
ar400085p.
(20) Martí-Centelles, V.; Burguete, M. I.; Galindo, F.; Izquierdo, M.
A.; Kumar, D. K.; White, A. J. P.; Luis, S. V.; Vilar, R. J. Org. Chem.
2012, 77, 490−500.
(21) Alfonso, I.; Burguete, M. I.; Galindo, F.; Luis, S. V.; Vigara, L. J.
Org. Chem. 2007, 72, 7947−7956.
(22) Becerril, J.; Bolte, M.; Burguete, M. I.; Galindo, F.; García-
Espana, E.; Luis, S. V.; Miravet, J. F. J. Am. Chem. Soc. 2003, 125,
̃
6677−6686.
́
(23) Ramos, S.; Alcalde, E.; Doddi, G.; Mencarelli, P.; Perez-García,
L. J. Org. Chem. 2002, 67, 8463−8468.
(24) Martí-Centelles, V.; Burguete, M. I.; Luis, S. V. Chem.Eur. J.
2012, 18, 2409−2422.
(25) Vilar, R. Angew. Chem., Int. Ed. 2003, 42, 1460−1477.
́
(26) Alcalde, E.; Ramos, S.; Perez-García, L. Org. Lett. 1999, 1,
1035−1038.
(27) Revilla-Lop
Cativiela, C.; Aleman
(28) Solanas, C. N.; de la Torre, B. G.; Fernan
Santiveri, C. M.; Jimenez, M. A. N.; Rivas, L.; Jimen
́
ez, G.; Warren, J. G.; Torras, J.; Jimen
, C. Biopolymers: Pept. Sci. 2012, 98, 98−110.
dez-Reyes, M. A.;
ez, A. I.; Andreu,
́
ez, A. I.;
́
́
́
́
D.; Cativiela, C. J. Med. Chem. 2010, 53, 4119−4129.
(29) Frost, J. R.; Vitali, F.; Jacob, N. T.; Brown, M. D.; Fasan, R.
ChemBioChem 2013, 14, 147−160.
570
dx.doi.org/10.1021/jo4022309 | J. Org. Chem. 2014, 79, 559−570