Quantification of the effect of conformational restriction on supramolecular effective molarities

Article abstract of DOI:10.1021/ja310221t

The association constants for a family of 96 closely related zinc porphyrin-pyridine ligand complexes have been measured in two different solvents, toluene and 1,1,2,2-tetrachloroethane (TCE). The zinc porphyrin receptors are equipped with phenol side arms, which can form intramolecular H-bonds with ester or amide side arms on the pyridine ligands. These association constants were used to construct 64 chemical double mutant cycles, which measure the free energy contributions of intramolecular H-bonding interactions to the overall stability of the complexes. Measurement of association constants for the corresponding intermolecular H-bonding interactions allowed determination of the effective molarities (EM) for the intramolecular interactions. Comparison of ligands that feature amide H-bond acceptors and ester H-bonds at identical sites on the ligand framework show that the values of EM are practically identical. Similarly, the values of EM are practically identical in toluene and in TCE. However, comparison of two ligand series that differ by one degree of torsional freedom shows that the values of EM for the flexible ligands are an order of magnitude lower than for the corresponding rigid ligands. This observation holds for a range of different supramolecular architectures with different degrees of receptor-ligand complementarity and suggests that in general the cost of freezing a rotor in supramolecular complexes is of the order of 5 kJ/mol.

Full text of DOI:10.1021/ja310221t

Article
pubs.acs.org/JACS
Quantification of the Effect of Conformational Restriction
on Supramolecular Effective Molarities
Harry Adams, Elena Chekmeneva, Christopher A. Hunter,* Maria Cristina Misuraca, Cristina Navarro,
and Simon M. Turega
Department of Chemistry, University of Sheffield, Sheffield S3 7HF, United Kingdom
S
* Supporting Information
ABSTRACT: The association constants for a family of 96 closely
related zinc porphyrin−pyridine ligand complexes have been
measured in two different solvents, toluene and 1,1,2,2-tetrachloro-
ethane (TCE). The zinc porphyrin receptors are equipped with
phenol side arms, which can form intramolecular H-bonds with
ester or amide side arms on the pyridine ligands. These association
constants were used to construct 64 chemical double mutant
cycles, which measure the free energy contributions of intra-
molecular H-bonding interactions to the overall stability of the
complexes. Measurement of association constants for the corresponding intermolecular H-bonding interactions allowed determination
of the effective molarities (EM) for the intramolecular interactions. Comparison of ligands that feature amide H-bond acceptors and ester
H-bonds at identical sites on the ligand framework show that the values of EM are practically identical. Similarly, the values of EM are
practically identical in toluene and in TCE. However, comparison of two ligand series that differ by one degree of torsional freedom
shows that the values of EM for the flexible ligands are an order of magnitude lower than for the corresponding rigid ligands. This
observation holds for a range of different supramolecular architectures with different degrees of receptor−ligand complementarity and
suggests that in general the cost of freezing a rotor in supramolecular complexes is of the order of 5 kJ/mol.
flexibility lead to lower values of EM, and the thermodynamic
advantage of freezing out a rotor, in the formation of an
intramolecular covalent bond has been estimated as 5−6 kJ/mol.5
Qualitatively similar results are obtained for non-covalent bond
formation, which leads to the strategy of preorganization in order
to maximize the stability of intermolecular complexes.
However, the variation in EM values for the formation of non-
covalent bonds (supramolecular EM) is much smaller than
observed for the formation of covalent bonds (covalent EM).6
Increasing the length of the linker between the two non-covalent
binding sites leads to a relatively small decrease in EM, and the
effect of freezing out a rotor in the linker has been estimated as
0.5−5 kJ/mol.7 The ability to make reasonable estimates of the
likely value of EM for non-covalent interactions would be of
tremendous utility in supramolecular design, where fully
preorganized systems can be difficult to obtain and the interactions
are sufficiently weak that small changes in EM can have a dramatic
effect on the efficiency of the assembly process.8 Here we describe
a quantitative investigation of the influence of conformational
flexibility on the magnitude of supramolecular effective molarities.
INTRODUCTION
■
Molecular assemblies that constitute the functional elements of
biological and synthetic systems are controlled by the interplay
of multiple weak non-covalent interactions.1 Cooperation
between these interactions leads to robust structures with
well-defined properties, but the nature of this cooperativity and
the intrinsic properties of individual interactions can be difficult
to dissect from the study of complex systems. Reliable design of
new synthetic molecular assemblies will rely on the ability to
make accurate predictions of the thermodynamic properties of
multiple weak interactions. There has been considerable
progress in estimating the contributions of specific functional
group contacts to the stability of intermolecular complexes, but
understanding the interplay of multiple interactions remains a
challenge.2
Intramolecular interactions are more favorable than inter-
molecular interactions due to the unfavorable entropy
associated with bimolecular processes. This effect is generally
quantified by the effective molarity (EM), which is defined as the
ratio of the intramolecular rate or equilibrium constant to the
corresponding intermolecular rate or equilibrium constant. The
relationship between EM and molecular structure has been
thoroughly investigated for covalent bond formation and the
values vary over many orders of magnitude.3 In general, increasing
the length of the linker between the two reactive ends in a
covalent cyclization process decreases the rate and equilibrium
constant for the cyclization reaction.4 The flexibility of the linker is
also important. In general, linkers with more conformational
APPROACH
■
We have been using complexes formed between zinc porphyrins
and pyridine ligands to make a systematic quantitative investigation
of the properties of intramolecular non-covalent interactions.9 The
Received: October 19, 2012
Published: January 29, 2013
© 2013 American Chemical Society
1853
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
Figure 1. Stepwise equilibria in the formation of a porphyrin−pyridine complex containing an intramolecular H-bond. K0 is the intermolecular
association constant for formation of the zinc−nitrogen interaction. KrefEM is the equilibrium constant for formation of the intramolecular H-bond.
Kref is the equilibrium constant for formation of the corresponding intermolecular H-bond. EM is the effective molarity for the intramolecular
interaction.
idea is outlined in Figure 1. Formation of a zinc pyridine
coordination bond leads to formation of an intermolecular
complex and peripheral groups on the porphyrin and ligand can
then make intramolecular interactions. Strictly speaking the
H-bond shown in Figure 1 is an intermolecular interaction, but
we will use the term “intramolecular” to describe non-covalent
interactions between functional groups within a molecule or within
a supramolecular complex. The formation of an intramolecular
interaction depends on the intrinsic strength of the interaction
(Kref, which is the association constant for formation of the
corresponding intermolecular interaction) and the EM for the
cyclization process.10 If the product KrefEM > 1, then the overall
stability of the complex will be enhanced by the presence of the H-
bonding groups, and the increase in the measured association
constant can be used to determine the value of EM.
functional group interactions to the free energy change of
complexation are additive, the DMC provides a straightforward
method for quantifying secondary effects of the chemical mutations
and dissecting out the free energy contribution from the
intramolecular H-bond to the stability of complex A, ΔΔG° (eq 1):
ΔΔG° = ΔG°A − ΔG°B − ΔG°C + ΔG°D
(1)
This free energy can then be used to determine the value of EM,
provided the association constant, Kref, for the corresponding
intermolecular process can be determined.
We have used this approach with a family of closely related
porphyrin and ligand systems to investigate the effects of
changing the solvent, changing the functional groups involved
in H-bond formation, and geometric complementarity on
supramolecular effective molarities. The results suggest that EM
is relatively insensitive to all of these parameters, with values
falling in the range 1−1000 mM.9 One of the striking features
of the results is that changes in geometry have a relatively small
impact on EM, unless the ligand is simply too short to span the
H-bonding and zinc binding sites. However, the ligand systems
studied to date are all relatively flexible, and here we report the
effects of reducing conformational flexibility on supramolecular
effective molarities.
In practice, there are a number of control experiments that
are required to dissect the intramolecular equilibrium constant,
KrefEM, from the measurement of the overall stability of the
complex, and we have formalized the experiment as a chemical
double mutant cycle (DMC).11 Figure 2 illustrates the DMC
Figure 3 illustrates two new ligand families designed to
quantify the impact of the conformational restriction on
Figure 2. Chemical double mutant cycle (DMC) for measurement of
the free energy contribution of an intramolecular H-bond to the
stability of complex A.
Figure 3. Flexible (top) and rigid (bottom) ligands. The nitrogen
(blue) binds to the zinc of the porphyrin, and the carbonyl oxygen
(red) can make intramolecular H-bonds with the porphyrin phenol
groups. The torsion angle that is restricted on going from the flexible
to the rigid ligands is highlighted.
experiment for measurement of an intramolecular H-bond in a
porphyrin−ligand complex. The contribution of the H-bond to
the overall stability of complex A can be estimated by measuring
the stability of similar complexes where the H-bonding group on
the porphyrin or ligand has been removed. However, these
chemical mutations could also affect the zinc−nitrogen interaction
or alter additional secondary interactions that contribute to the
stability of complex A. Assuming that the contributions of pairwise
binding affinity. These ligands will be referred to as “flexible”
and “rigid” for the purposes of discussion. The key carbonyl
oxygen that can make H-bonds with the porphyrin phenol
groups is highlighted in red and is located at the same position
on the ligand framework in both ligand families. We have
shown previously that the other ester carbonyl in the flexible ligand
1854
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
framework does not make H-bonds to any of the porphyrins
studied here.9 Comparison of EM values for related flexible and
rigid ligand systems will provide a measure of the impact of
conformational flexibility on binding affinity. In addition to the
restriction of conformational flexibility in the rigid ligands, there are
likely to be some differences in the distributions of conformations
accessible to the different ligand families. The results will therefore
be perturbed by variations in geometric complementarity between
the porphyrin and ligand frameworks, but by studying a number of
different supramolecular architectures that vary in complementarity,
we hope that any important underlying preferences will emerge.
Figures 4 and 5 show the porphyrins and ligands used in this
work. The porphyrins vary in the location of the phenol H-bond
Figure 5. Porphyrin receptors, P1a−P4a (R = OH) and P1b−P4b
(R = OMe).
Coupling of the acid chlorides with N,N-diethyl-2-hydroxyaceta-
mide gave amides L2e and L3e, and coupling with ethyl glycolate
gave esters L2f and L3f in reasonable yields (Scheme 1).9
Scheme 1. Synthesis of Flexible Ligands
Figure 4. Amide ligands, L2e, L3e, L7e, and L8e; ester ligands, L2f,
L3f, L7f, and L8f; and control ligands with no H-bonding groups, L2b,
L3b, L7c, and L8c.
donor sites around the periphery. The ligands vary in flexibility (as
illustrated in Figure 3), in the polarity of the H-bond acceptor
group (ester or amide), and in the number of H-bond acceptor sites
(one or two). Comparison of the one-armed and the corresponding
two-armed ligands provides two independent measurements of the
same intramolecular H-bond, provided there are no conformational
problems in forming the doubly H-bonded complex. The
experiments were carried out in two different solvents, toluene
and 1,1,2,2-tetrachloroethane (TCE), which modulates the intrinsic
strength of the H-bond interaction. This set of porphyrins, ligands,
and solvents can be used to measure EM for 64 different supra-
molecular systems, and comparison of the flexible and rigid ligand
results provides a clear picture of the effect of conformational
flexibility on binding affinity.
The rigid ester ligands L7f and L8f were synthesized using
palladium-catalyzed Suzuki−Morita cross-coupling of 3-ethox-
ycarbonylphenylboronic acid with 3-bromopyridine and 3,5-
bromopyridine, respectively (Scheme 2).12 The rigid amide
ligands were synthesized in two steps. Suzuki−Morita cross-
coupling of 3-carboxyphenylboronic acid with 3-bromopyridine and
3,5-bromopyridine gave carboxylic acids 3 and 4, respectively
(Scheme 2). Each acid was converted to the acid chloride and
coupled with diethylamine to give ligands L7e and L8e (Scheme 2).
Control ligands L7c and L8c were synthesized from 3-bromo-
pyridine and 3,5-bromopyridine by Suzuki−Morita coupling with
3-methylphenylboronic acid (Scheme 2).
High-Throughput Titration Analysis of Binding. The
association constants for the 96 different complexes formed
by all pairwise combinations of the 12 ligands and 8 zinc
porphyrins were measured using UV/vis absorption titrations
in toluene and in TCE. The porphyrin Soret band undergoes a
large shift on complexation of the zinc with a pyridine ligand,
and this provides a convenient spectroscopic probe to monitor
RESULTS AND DISCUSSION
■
Synthesis. Synthesis of the porphyrin receptors and control
ligands L2b and L3b was carried out as described previously.9 The
ligands with flexible linkers were synthesized from the
corresponding carboxylic acids 1 and 2 via the acid chlorides.
1855
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
than one technique, the results are the same within error. For
example, the P1a·L7f association constant measured by UV/vis
absorption titrations was (1.8 0.1) × 106 M−1, compared with
(2.3 0.8) × 106 M−1 measured by ITC.
Scheme 2. Synthesis of Rigid Ligands
Figure 6 compares the association constants measured in
toluene with the results in TCE. As we have found previously in
related systems, the complexes are more stable by 1−2 orders of
magnitude in the less polar solvent, toluene.9 There is a reasonable
correlation between the two data sets, indicating that the complexes
that are more stable in toluene are generally more stable in TCE,
but the scatter in Figure 6 shows that the relative stabilities of the
complexes are altered by changes in solvation.
DMC Analysis of Intramolecular H-Bonding. The data
in Tables 1 and 2 are illustrated graphically in Figure 7, with the
complexes organized and colored according to their role in the
DMC. The complexes that can make intramolecular H-bonds
(blue) are generally more stable than the complexes that cannot,
and the increase in stability is larger for the amide ligands (pale
blue) than for the ester ligands (dark blue). The free energy
contributions due to intramolecular H-bonding interactions,
ΔΔG°, were determined using the data in Tables 1 and 2 in eq
1, and the results are presented in Tables 3−6.
In toluene, 14 of the 16 amide complexes make detectable
H-bonding interactions with free energy contributions of up to
21 kJ/mol to the overall stability of the complex. The ester
H-bonds are significantly weaker and contribute less than 10 kJ/mol
to the overall stability of the complexes in all cases. In TCE, the free
energy contributions from H-bonding are reduced somewhat,
because this solvent is more polar, and there is a larger desolvation
penalty for H-bond formation.
Figure 8 shows the relationship between the free energy
contributions due to H-bonding in the one-armed and two-armed
ligands. An assumption of the DMC analysis is that free energy
contributions from individual intermolecular interactions
are additive. The two-armed ligands are symmetrical, so if this
assumption is correct, then the value of ΔΔG° for the two-
armed ligands should be double the value for the corresponding
one-armed ligand. Figure 8 shows that in general this is indeed
the case. There is one outlier in Figure 8, the P1a·L7e/L8e
complex in toluene. This is the complex that makes the
strongest H-bonding interactions, and it is possible that there is a
conformational issue that prevents formation of two optimal
interactions simultaneously. However, additive behavior is observed
for this complex in TCE.
binding. Titrations were carried out using automated protocols
on a UV/vis plate reader, providing a convenient method for
collecting a large amount of quantitative data.
In most cases, the data fit well to a 1:1 binding isotherm, and
the association constants are reported in Tables 1 and 2 for
toluene and TCE, respectively. However, some of the
complexes are too stable allow accurate measurement of
association constants using UV/vis absorption spectroscopy
(K > 106 M−1). In these cases, fluorescence spectroscopy or
isothermal titration calorimetry (ITC) was used to measure the
association constant. Again the data fit well to a 1:1 binding
isotherm, and the results are included in Tables 1 and 2. In
cases where association constants could be measured by more
Table 1. Association Constants (K, M−1) for the Formation of 1:1 Complexes in Toluene at 298 K (with Percentage Errors)
porphyrin
ligand
P1a
P2a
P3a
P4a
P1b
P2b
P3b
P4b
L2e 9.6 × 104 (10%) 1.4 × 104 (7%)
L3e 2.8 × 106 (40%) 4.0 × 104 (8%)
6.6 × 104 (3%)
3.2 × 103 (20%) 2.2 × 103 (40%) 1.5 × 103 (8%) 3.6 × 103 (10%) 2.4 × 103 (20%)
2.9 × 105 (30%) 2.3 × 103 (40%) 2.0 × 103 (30%) 8.9 × 102 (6%) 3.0 × 103 (3%) 2.2 × 103 (5%)
L7e 1.3 × 106 (20%) 1.8 × 105 (20%) 7.4 × 105 (8%)
2.6 × 104 (10%) 7.2 × 103 (1%) 5.6 × 103 (4%) 9.3 × 103 (2%) 9.6 × 103 (5%)
a
a
a
L8e 2.6 × 107 (20%) 2.0 × 106 (40%) 8.2 × 106 (10%) 9.7 × 104 (30%) 7.4 × 103 (7%) 5.1 × 103 (20%) 9.7 × 103 (9%) 1.1 × 104 (6%)
L2f
L3f
L7f
L8f
6.5 × 103 (30%) 3.4 × 103 (1%)
1.2 × 104 (5%) 2.4 × 103 (30%) 8.5 × 103 (20%) 2.3 × 103 (10%) 2.2 × 103 (10%) 1.1 × 103 (20%) 3.6 × 103 (10%) 2.3 × 103 (20%)
4.1 × 104 (20%) 1.5 × 104 (20%) 2.0 × 104 (5%) 1.1 × 104 (9%) 7.5 × 103 (1%) 6.3 × 103 (1%) 1.1 × 104 (1%) 1.0 × 104 (5%)
7.4 × 103 (4%)
3.0 × 103 (7%) 2.3 × 103 (40%) 1.6 × 103 (10%) 3.5 × 103 (6%) 2.9 × 103 (20%)
2.7 × 105 (7%)
1.3 × 104 (5%)
5.1 × 103 (2%)
4.3 × 103 (5%)
2.8 × 104 (10%) 1.6 × 104 (6%) 9.3 × 103 (3%) 6.6 × 103 (3%) 1.6 × 104 (4%) 1.4 × 104 (5%)
L2b 3.6 × 103 (8%)
L3b 3.2 × 103 (1%)
8.5 × 103 (1%)
6.9 × 103 (9%)
5.1 × 103 (6%) 3.9 × 103 (20%) 3.3 × 103 (6%) 6.2 × 103 (30%) 5.4 × 103 (4%)
5.1 × 103 (40%) 3.8 × 103 (20%) 2.6 × 103 (10%) 6.0 × 103 (20%) 4.5 × 103 (20%)
L7c 8.0 × 103 (30%) 1.3 × 104 (3%)
1.8 × 104 (10%) 1.3 × 104 (1%) 8.8 × 103 (1%) 7.3 × 103 (4%) 1.3 × 104 (20%) 1.1 × 104 (5%)
2.5 × 104 (10%) 1.9 × 104 (4%) 1.4 × 104 (7%) 9.0 × 103 (10%) 2.4 × 104 (4%) 1.8 × 104 (10%)
L8c 1.0 × 104 (3%)
1.4 × 104 (7%)
a
Measured using ITC.
1856
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
Table 2. Association Constants (K, M−1) for the Formation of 1:1 Complexes in TCE at 298 K (with Percentage Errors)
porphyrin
ligand
P1a
P2a
P3a
P4a
P1b
P2b
P3b
P4b
L2e 1.2 × 104 (8%)
L3e 7.0 × 104 (3%)
1.3 × 103 (3%)
6.7 × 102 (1%)
2.9 × 103 (3%)
3.1 × 103 (3%)
4.1 × 104 (4%)
3.6 × 102 (10%) 4.0 × 102 (30%) 2.3 × 102 (4%)
3.8 × 102 (8%)
4.0 × 102 (3%)
1.6 × 102 (4%)
2.4 × 103 (3%)
2.3 × 103 (9%)
4.0 × 102 (5%)
1.3 × 102 (6%)
7.3 × 101 (5%)
7.4 × 101 (30%) 9.0 × 101 (10%)
L7e 1.1 × 105 (30%) 1.5 × 104 (1%)
2.0 × 103 (20%) 1.5 × 103 (3%)
1.9 × 103 (5%)
1.7 × 103 (4%)
1.1 × 103 (6%)
a
L8e 3.7 × 106 (20%) 3.1 × 104 (10%) 2.8 × 105 (1%)
1.1 × 103 (20%) 8.0 × 102 (10%) 1.1 × 103 (9%)
L2f
L3f
L7f
L8f
1.5 × 103 (5%)
7.9 × 102 (4%)
1.0 × 103 (3%)
3.9 × 102 (5%)
9.1 × 101 (8%)
2.2 × 102 (9%)
7.5 × 101 (3%)
4.1 × 102 (10%) 4.4 × 102 (7%)
1.1 × 103 (30%) 2.3 × 102 (4%)
1.6 × 104 (10%) 3.9 × 103 (3%)
4.1 × 102 (10%) 1.1 × 102 (9%)
1.2 × 102 (7%)
2.3 × 103 (4%)
8.7 × 101 (9%)
2.0 × 103 (5%)
3.3 × 103 (3%)
3.3 × 103 (6%)
2.3 × 103 (4%)
1.9 × 103 (10%) 1.3 × 103 (20%) 1.2 × 103 (8%)
2.1 × 103 (10%) 1.8 × 103 (6%)
8.8 × 104 (5%)
L2b 1.5 × 103 (30%) 1.8 × 103 (6%)
2.7 × 103 (3%)
1.6 × 103 (10%) 1.6 × 103 (4%)
1.7 × 103 (20%) 1.3 × 103 (30%) 1.1 × 103 (9%)
9.0 × 102 (9%)
4.0 × 102 (20%) 3.2 × 102 (6%)
1.2 × 103 (8%)
1.1 × 103 (20%)
L3b 7.0 × 102 (10%) 6.7 × 102 (20%) 7.1 × 102 (7%)
4.7 × 102 (4%)
3.2 × 103 (9%)
4.6 × 103 (4%)
3.7 × 102 (10%) 3.5 × 102 (6%)
L7c 4.1 × 103 (2%)
4.5 × 103 (4%)
4.0 × 103 (2%)
5.5 × 103 (4%)
3.0 × 103 (1%)
3.2 × 103 (3%)
2.4 × 103 (4%)
2.4 × 103 (2%)
3.3 × 103 (2%)
3.6 × 103 (3%)
2.8 × 103 (7%)
3.4 × 103 (9%)
L8c 5.1 × 103 (10%) 5.4 × 103 (4%)
a
Measured using fluorescence spectroscopy.
Table 3. Free Energy Contributions from Amide−Phenol
H-Bonds at 298 K in Toluene (ΔΔG°, kJ/mol) Determined
a
Using the Chemical Double Mutant Cycle in Figure 2
ligand
porphyrin
L2e
L3e
L7e
L8e
P1a
P2a
P3a
P4a
−10
−4
−6
−18
−8
−11
0
−13
−7
−10
−2
−21
−14
−17
−5
−1
a
Average error over the data set 1 kJ/mol. Complexes that do not
make detectable H-bonds are in italics (ΔΔG° > −2 kJ/mol).
all cases, the H-bonds formed by the rigid ligands are more
favorable than those formed by the flexible ligands with
differences of up to 6 kJ/mol. These differences are a measure
of the effects of preorganization of the ligand framework on
binding affinity. However, the variations in ΔΔG° contain
contributions that vary with the solvent, with the functional
groups involved in the H-bond, and with the geometric
complementarity of the ligand−porphyrin architecture. To
account for the contributions due to solvent and H-bond
Figure 6. Comparison of the 1:1 association constants (log K/M−1)
for formation of porphyrin−ligand complexes in TCE with the
corresponding values measured in toluene. Data for the complexes that
can make intramolecular H-bonds are shown in dark gray, and data for
control complexes are shown in pale gray. The line corresponds to log
K(TCE) = log K(toluene).
Figure 9 compares the values of ΔΔG° for rigid ligands with
the values measured for the corresponding flexible ligands. In
Figure 7. Association constants (log K/M−1) measured in (a) toluene and (b) TCE. (c) Schematic representation of the chemical DMC used to
extract information on the magnitude of the intramolecular H-bond interaction between A and D in the complex formed between a zinc porphyrin
(P) and a pyridine ligand (L). Data for the amide ligand−hydroxyporphyrin complexes are shown in blue, amide ligand−methoxyporphyrin
complexes in green, ester ligand−hydroxyporphyrin complexes in dark blue, ester ligand−methoxyporphyrin complexes in dark green, control
ligand−hydroxyporphyrin complexes in yellow, and control ligand−methoxyporphyrin complexes in red.
1857
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
Table 4. Free Energy Contributions from Amide−Phenol
H-Bonds at 298 K in TCE (ΔΔG°, kJ/mol) Determined
a
Using the Chemical Double Mutant Cycle in Figure 2
ligand
porphyrin
L2e
L3e
L7e
L8e
P1a
P2a
P3a
P4a
−8
−3
−4
1
−14
−4
−8
−9
−4
−7
−1
−19
−7
−13
−1
−1
a
Average error over the data set 1 kJ/mol. Complexes that do not
make detectable H-bonds are in italics (ΔΔG° > −2 kJ/mol).
Table 5. Free Energy Contributions from Ester−Phenol
H-Bonds at 298 K in Toluene (ΔΔG°, kJ/mol) Determined
Figure 9. Total free energy contribution due to intramolecular
H-bonding for rigid ligands, ΔΔG°(rigid), compared with data for the
corresponding flexible ligands, ΔΔG°(flexible), in toluene (gray) and
TCE (black). The line corresponds to ΔΔG°(flexible) = ΔΔG°(rigid).
a
Using the Chemical Double Mutant Cycle in Figure 2
ligand
porphyrin
L2f
L3f
L7f
L8f
used are shown in Figure 10. Association constants for the
interaction of p-cresol with aliphatic and aromatic esters and
P1a
P2a
P3a
P4a
−3
−1
−1
0
−5
−1
−2
0
−4
−1
−1
0
−9
−1
−1
0
a
Average error over the data set 1 kJ/mol. Complexes that do not
make detectable H-bonds are in italics (ΔΔG° > −2 kJ/mol).
Table 6. Free Energy Contributions from Ester−Phenol
H-Bonds at 298 K in TCE (ΔΔG°, kJ/mol) Determined
a
Using the Chemical Double Mutant Cycle in Figure 2
ligand
porphyrin
L2f
L3f
L7f
L8f
P1a
P2a
P3a
P4a
−3
−1
−1
1
−5
−1
−1
0
−4
0
−9
0
Figure 10. Compounds used to quantify intermolecular H-bond
interactions.
0
−1
0
0
a
Average error over the data set 1 kJ/mol. Complexes that do not
amides were measured using 1H NMR titrations in toluene and
in TCE, and the results are shown in Table 7.
make detectable H-bonds are in italics (ΔΔG° > −2 kJ/mol).
The results are consistent with the observations made for the
intramolecular interactions in the porphyrin−ligand complexes.
The amide−phenol H-bonds are stronger than the ester−
phenol H-bonds, and the interactions are marginally weaker in
TCE than in toluene. There are also substituent effects: the
aliphatic amide and ester groups of the flexible ligands are
slightly better H-bond acceptors than the aromatic groups on
the rigid ligands. The ester complexes are not sufficiently stable
for accurate measurement of small differences, but there is a
two-fold difference between the stabilities of the aromatic and
aliphatic amide−phenol complexes. Table 7 compares the
measured association constants with the values predicted using
literature H-bond parameters in eq 2,2e
−RTln Kcalc = −(αD − αS)(β − βS) + 6 kJ/mol
Figure 8. Total free energy contribution due to intramolecular H-bonding
for ligands with two identical side arms, ΔΔG°(2), compared with data for
the corresponding one-armed ligands, ΔΔG°(1), in toluene (gray) and
TCE (black). The line corresponds to ΔΔG°(2) = 2ΔΔG°(1).
(2)
A
where Kcalc is the intermolecular association constant at T = 298
K, αD and βA are the H-bond parameters of the H-bond donor
(D) and H-bond acceptor (A), αS and βS are the H-bond donor
and acceptor parameters of the solvent, and the constant of
6 kJ/mol was experimentally determined in carbon tetra-
chloride solution but is assumed to apply to all organic solvents.
There is good agreement in Table 7, and this gives some
confidence that the small association constants measured for
the ester complexes are reliable.
strength, it is necessary to compare the effective molarities
(EM) for the intramolecular interactions.
In order to determine the values of EM for the intra-
molecular H-bonds, association constants for the corresponding
intermolecular interactions, Kref, were measured. The compounds
1858
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
Table 7. Association Constants (K/M−1) for the Formation
1
of H-Bonded Complexes at 298 K Measured by H NMR
a
Titrations (Kexpt) and Estimated Using Eq 2 (Kcalc
)
complex
solvent
α
β
αS
βS
Kexpt
Kcalc
5·6
5·7
5·8
5·9
5·6
5·7
5·8
5·9
toluene
toluene
toluene
toluene
TCE
3.8
3.8
3.8
3.8
3.8
3.8
3.8
3.8
5.4
5.1
8.5
7.9
5.4
5.1
8.5
7.9
1.0
1.0
1.0
1.0
2.0
2.0
2.0
2.0
2.2
2.2
2.2
2.2
1.3
1.3
1.3
1.3
3
3
1
1
3
2
86 20
110
54
2
33
2
1
1
1
3
2
TCE
2
1
TCE
22
11
16
11
TCE
a
H-bond parameters from ref 13.
Figure 11. Crystal structure of the P1a·L8f complex showing the fully
bound state with two H-bonds between the ester substituents on the
ligand and the trans-related meso-phenol groups on the porphyrin.
Hydrogen atoms not involved in H-bonding, and two DCM molecules
are omitted for clarity. One of the ester carbonyl oxygen atoms is
disordered over two sites that are very close in space.
The observed association constant for the formation of a
zinc porphyrin−pyridine complex containing an intramolecular
H-bond can be described in terms of K0, the zinc−nitrogen
interaction, and KrefEM, the equilibrium constant for the
formation of the intramolecular H-bond (Figure 1).9 When
KrefEM ≫ 1, the H-bonded state is fully populated, but if
KrefEM ≤ 1, then partially bound states, where the H-bond is
not formed, must also be considered. The observed association
constant Kobs is the sum of the association constants for all
partially and fully bound states. Where a single intramolecular
H-bond is possible, Kobs is given by eq 3,
In the DMC, any differences in K0 cancel, so the values of
Kref in Table 7 can be used to calculate the values of EM from
the ΔΔG° values in Tables 3−6 as follows. For the one-armed
ligand complexes,
e−ΔΔG°/RT = 1 + 4Kref EM
(7)
Kobs = K0 + K0Kref EM = K0(1 + Kref EM)
(3)
For the rigid two-armed ligand complexes,
e−ΔΔG°/RT = 1 + 8Kref EM + 4(Kref EM)2
For the flexible two-armed ligand complexes,
(8)
For the porphyrin−ligand complexes considered here, there are
multiple H-bonding sites, so a statistical factor that accounts for the
degeneracy of the complex must be included. For the one-armed
ligand complexes, there are four possible H-bonding interactions
that can be formed, and the value of Kobs is given by eq 4,
e−ΔΔG°/RT = 1 + 8Kref EM + 8(Kref EM)2
Solving these equations for EM gives the results reported in
(9)
Tables 8−11.
Kobs = K0(1 + 4Kref EM)
(4)
Table 8. Effective Molarities (EM, mM) for Intramolecular
Amide−Phenol H-Bonds Measured at 298 K in Toluene
a
For the rigid two-armed ligand complexes, we assume that
the value of EM for formation of the first H-bond is the same as
EM for formation of the second H-bond. This is supported by
the additive free energy increments observed for the one-armed
and two-armed ligands (Figure 8). The value of Kobs is
therefore given by eq 5,
ligand
porphyrin
L2e
L3e
L7e
L8e
P1a
P2a
P3a
P4a
140
15
36
b
130
14
27
b
1500
130
430
10
1000
210
400
20
Kobs = K0(1 + 8Kref EM + 4(Kref EM)2)
a
b
(5)
Average error over the data set 50%. No interaction detected.
Table 9. Effective Molarities (EM, mM) for Intramolecular
Amide−Phenol H-Bonds Measured at 298 K in TCE
For the flexible two-armed ligand complexes, the value of
Kobs is given by eq 6.
a
ligand
Kobs = K0(1 + 8Kref EM + 8(Kref EM)2)
(6)
porphyrin
L2e
L3e
L7e
L8e
P1a
P2a
P3a
P4a
240
21
50
b
220
13
47
b
890
98
2000
110
500
b
The statistical factors used for the rigid and flexible two-
armed ligands differ, because ligand flexibility alters the number
of different fully bound complexes that can be formed. Models
show that the flexible ligand can make two H-bonds with both
the cis- and the trans-related meso-phenol substituents on the
porphyrin receptors. In contrast, the rigid ligand can only
interact simultaneously with the trans substituents. This is
confirmed by an X-ray crystal structure of the P1a·L8f complex
(Figure 11).
380
b
a
b
Average error over the data set 50%. No interaction detected.
The values of EM show considerable variation with the
supramolecular architecture of the complex and range from
10 mM to 2 M. Figure 12 compares the values of EM measured
in toluene with the corresponding values measured in TCE.
1859
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
the EM was independent of solvent.9 However, for phosphonate
diester−phenol H-bonds, the value of EM changed by up to an
order of magnitude in different solvents.9 The results presented
here suggest that the behavior of the phosphonate diester ligands
is anomalous, which may be related to their steric bulk.9
Figure 13 compares that values of EM measured for the ester
ligands with the values measured for the amide ligands, and again
there is very good agreement between the two data sets. These
results suggest that EM is a property of the supramolecular
architecture and independent of the solvent and the intrinsic
strength of the H-bonds involved.
Table 10. Effective Molarities (EM, mM) for Intramolecular
Ester−Phenol H-Bonds Measured at 298 K in Toluene
a
ligand
porphyrin
L2f
L3f
L7f
L8f
P1a
P2a
P3a
P4a
170
b
140
b
430
b
790
b
b
38
b
b
b
b
b
b
a
b
Average error over the data set 50%. No interaction detected.
Table 11. Effective molarities (EM, mM) for Intramolecular
Figure 14 compares the values of EM measured for the
flexible ligands with values measured for the corresponding
a
Ester−Phenol H-Bonds Measured at 298 K in TCE
ligand
porphyrin
L2f
L3f
L7f
L8f
P1a
P2a
P3a
P4a
230
b
220
b
760
b
1600
b
b
b
b
b
b
b
b
b
a
b
Average error over the data set 50%. No interaction detected.
Figure 14. Comparison of effective molarities (EM) measured for
formation of intramolecular H-bonds for rigid ligands with the values
measured for the corresponding flexible ligands in toluene (gray) and
in TCE (black). The solid line corresponds to log EM(flexible) = log
EM(rigid), and the dotted line corresponds to log EM(flexible) = log
EM(rigid) − 1.
rigid ligands. Here there are substantial differences. The rigid
ligands give values of EM that are approximately an order of
magnitude higher than the values for the corresponding flexible
ligands. This suggests that cost of restricting the conformational
flexibility conferred by the additional rotor present in the
flexible ligands is one order of magnitude in binding affinity.
Figure 12. Comparison of effective molarities (EM) for formation of
intramolecular H-bonds in toluene with the corresponding values
measured in TCE for ester (black) and amide (gray) ligands. The line
corresponds to log EM(TCE) = log EM(toluene).
CONCLUSION
■
Comparison of the thermodynamic properties of a family of 64
closely related zinc porphyrin−pyridine ligand complexes that
make intramolecular H-bonding interactions has allowed an
evaluation of the effect of conformational flexibility on
cooperativity in supramolecular complexes. Chemical double
mutant cycles were used to measure the free energy contributions
of intramolecular H-bonding interactions in 64 different systems.
The results show that free energy contributions from intra-
molecular H-bonds make an additive contribution to the overall
stability of the complex: the values of ΔΔG° for two-armed ligands
are double the values for the corresponding one-armed ligands.
Comparison of the properties of the corresponding intermo-
lecular H-bonds with the DMC results provides the values of EM
for the intramolecular interactions in these systems. The value of
EM for an ester−phenol H-bond is practically identical to the EM
for the corresponding amide−phenol H-bond embedded in the
same supramolecular architecture. The values of EM are also
independent of the solvent, with similar results obtained in toluene
and in TCE.
Figure 13. Comparison of effective molarities (EM) for formation of
intramolecular H-bonds for ester ligands with the corresponding
values measured for amide ligands in toluene (gray) and in TCE
(black). The line corresponds to log EM(amide) = log EM(ester).
There is extremely good agreement between the two data sets.
We have previously measured a number of ether−phenol
H-bonds and phosphonate diester−phenol H-bonds in the
same solvents using a closely related family of complexes. For ether−
phenol H-bonds the results were similar to those reported here:
However, significant differences in EM are observed for rigid
and flexible ligands. These two families of ligand have identical
1860
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
135.97, 134.38, 129.18, 127.80, 125.84, 125.07, 123.63, 43.33, 39.31,
H-bond donors at identical locations on the ligand framework,
but the flexible ligands have one more torsional degree of
freedom than the rigid ligands. The complexes formed with the
rigid ligands are more stable than the complexes formed with
the flexible ligands, and there is an order of magnitude
difference in the value of EM. Although there are some
conformational differences between the flexible and rigid, and
amide and ester ligands, the results for different supramolecular
architectures with different degrees of receptor−ligand
complementarity are similar, which suggests that the results
have general applicability. These experiments show that the
cost of restricting a rotor in formation of a supramolecular
complex is about 5 kJ/mol, which is comparable to the value
found for intramolecular covalent interactions.
Previous estimates of the cost of resticting a rotor in a non-
covalent complex are as high as 9 kJ/mol, which suggests that
not all of the conformational entropy is frozen out in the
formation of the H-bonded complexes described here, and
higher EMs might be possible in more constrained systems.14
The highest value of EM measured for complexes reported in
this paper is 2 M, and Anderson has shown that it is possible to
obtain an EM 3 orders of magnitude larger in a very highly
organized system.15
14.26, 12.67; MS (ES+) m/z (%) = 255 [M+H+] (100); HRMS (ES+)
calcd for C16H19N2O 255.1497, found 255.1492; FT-IR (thin film) νmax
/
cm−1 2978, 2934, 2874, 1625, 1457, 1428, 1381, 1366, 1319, 1278, 1218,
1102.
Ligand L8e. To 4 (0.297 g, 0.835 mmol) was added toluene (5
mL), SOCl2 (15 mL), and DMF (10 μL). The mixture was refluxed for
1 h, protected by a CaCl2 drying tube. The reaction was allowed to
cool, solvent was removed under reduced-pressure, and the residue
was re-dissolved in DCM (50 mL). To this solution stirring at 0 °C,
protected by a nitrogen atmosphere, was added diethylamine (0.616
mL, 6.68 mmol) in small portions. After 24 h the DCM solution was
washed with 10% NaHCO3(aq) (2 × 20 mL) and brine (20 mL) and
then dried with Na2SO4, the solvent was removed under reduced
pressure, and the residue was purified on silica, eluting with
EtOAc:hexane. The product was isolated as a clear oil: yield 0.325 g
1
(90%); H NMR (250 MHz, CDCl3) δ 8.85 (s, 2H), 8.07 (s, 1H),
7.70−7.66 (m, 4H), 7.55, (dd, 2H, J = 8, J = 8), 7.43 (d, 2H, J = 8),
3.46 (d, 8H, J = 70), 1.22 (d, 12H, J = 30); 13C NMR (62.9 MHz,
CDCl3) δC = 170.74, 147.26, 138.34, 137.96, 136.02, 132.93, 129.27,
27.90, 126.02, 125.25, 43.59, 39.53, 14.46, 13.33; MS (ES+) m/z (%)
= 430, [M+H+] (100); HRMS (ES+) calcd for C27H32N3O2 430.2495,
found 430.2482; FT-IR (thin film) νmax/cm−1 2975, 2937, 2874, 1626,
1472, 1457, 1436, 1384, 1284, 1102.
Ligand L7f. To 3-(ethoxycarbonyl)phenylboronic acid (0.513 g,
2.64 mmol), Pd(0)(PPh3)4 (0.0449g, 0.0387 mmol), and sodium
carbonate (0.342, 3.23 mmol), protected by an argon atmosphere,
were added THF (25 mL), toluene (25 mL), water (1 mL), and 3-
bromopyridine (0.251 mL, 2.58 mmol). The mixture was heated at 90 °C
for 36 h and allowed to cool, and the solvent was removed under reduced
pressure. The solid was re-dissolved in DCM (50 mL), washed with 10%
NaHCO3(aq) (20 mL) and brine (20 mL), and then dried with Na2SO4,
the solvent was removed under reduced pressure, and the residue was
purified on silica, eluting with EtOAc:hexane. The product was isolated as
EXPERIMENTAL SECTION
■
Synthesis. 3-(Pyridin-3-yl)benzoic Acid,12 3. To 3-bromopyridine
(0.190 mL, 2.0 mmol), 3-carboxyphenylboronic acid (0.330 g, 2.00
mmol), and Pd(0)(PPh3)4 (0.12g, 0.1 mmol), protected by an argon
atmosphere, were added 0.4 M sodium carbonate(aq) (10 mL) and
acetonitrile (10 mL). The mixture was heated at 90 °C for 36 h and
allowed to cool. The volume was reduced by 50% under reduced
pressure, washed with DCM (10 mL), and filtered. The product was
precipitated from the aqueous layer with HCl(gas). The product was
purified on silica, eluting with EtOAc:AcOH. The product was isolated
1
a clear oil: yield 0.15 g (27%); H NMR (250 MHz, CDCl3) δ 8.85 (s,
1H), 8.59 (d, 1H, J = 5), 8.23 (s, 1H), 8.05 (d, 1H, J = 8), 7.87 (d, 1H,
J = 8), 7.73 (d, 1H, J = 8), 7.52 (dd, 1H, J = 8, J = 8), 7.35 (dd, 1H, J = 5,
J = 8), 4.38 (q, 2H, J = 7), 1.39 (t, 3H, J = 7); 13C NMR (100.6 MHz,
CDCl3) δC = 166.21, 148.88, 148.26, 138.09, 135.73, 134.42, 131.43,
131.32, 129.14, 129.11, 128.21, 123.62, 61.21, 14.32; MS (ES+) m/z
(%) = 228 [M+H+] (100); HRMS (ES+) calcd for C14H14NO2 228.1025,
found 228.1025; FT-IR (thin film) νmax/cm−1 3037, 2983, 2936,
2904,1718, 1470, 1437, 1369, 1308, 1251, 1114, 1085.
1
as a white solid: yield 0.087 g (22%); H NMR (250 MHz, CDCl3) δ
9.29 (s, 1H), 9.02 (d, 1H, J = 8), 8.92 (d, 1H, J = 6), 8.49 (s, 1H),
8.27−8.22 (m, 2H), 8.10 (d, 1H, J = 8), 7.75 (dd, 1H, J = 8).
3,3′-(Pyridine-3,5-diyl)dibenzoic Acid, 4. To 3,5-dibromopyridine
(0.236 g, 1.0 mmol), 3-carboxyphenylboronic acid (0.330 g, 2.00
mmol), and Pd(0)(PPh3)4 (0.12g, 0.1 mmol), protected by an argon
atmosphere, were added 0.4 M sodium carbonate(aq) (10 mL) and
acetonitrile (10 mL). The mixture was heated at 90 °C for 36 h and
allowed to cool. The volume was reduced by 50% under reduced
pressure, washed with DCM (10 mL), and filtered. The product was
precipitated from the aqueous layer with HCl(gas) and washed with
water (2 mL). The product was isolated as a white solid: yield 0.297 g
Ligand L8f. To 3-(ethoxycarbonyl)phenylboronic acid (0.513 g,
2.64 mmol), Pd(0)(PPh3)4 (0.0449g, 0.0387 mmol), sodium
carbonate (0.342, 3.23 mmol) and 3,5-dibromopyridine (0.306 g,
1.29 mmol), protected by an argon atmosphere, were added THF (25
mL), toluene (25 mL), and water (1 mL). The mixture was heated at
90 °C for 36 h and then allowed to cool, and the solvent was removed
under reduced pressure. The solid was re-dissolved in DCM (50 mL),
washed with 10% NaHCO3(aq) (20 mL) and brine (20 mL), and dried
with Na2SO4, the solvent was removed under reduced pressure, and
the residue was purified on silica, eluting with EtOAc:hexane. The
product was isolated as a white solid: yield 0.211 g (45%); mp = 104−
1
(93%); H NMR (250 MHz, CDCl3) δ 9.14 (s, 2H), 8.84 (s, 1H),
8.43 (s, 2H), 8.20 (d, 2H, J = 8), 8.08 (d, 2H, J = 8), 7.71 (dd, 2H, J =
8); 13C NMR (62.9 MHz, d6-DMSO) δC = 167.48, 141.61, 139.51,
137.93, 135.50, 132.61, 132.28, 130.60, 130.18, 128.84; MS (ES+) m/z
(%) = 320, [M+H+] (100); HRMS (ES+) calcd for C19H14NO4
320.0923, found 320.0921; FT-IR (thin film) νmax/cm−1 3034, 2965,
2858, 1682, 1569, 1463, 1432, 1397, 1303, 1262.
1
106 °C; H NMR (250 MHz, CDCl3) δ 8.88 (s, 2H), 8.33 (s, 2H),
8.12 (d, 2H, J = 6), 8.12 (d, 2H, J = 7), 8.11 (s, 1H), 7.59 (dd, 2H, J =
8, J = 8), 4.43 (q, 4H, J = 7), 1.43 (t, 6H, J = 7); 13C NMR (62.9 MHz,
CDCl3) δC = 166.22, 147.38, 137.82, 135.89, 133.03, 131.55, 131.48,
129.38, 129.27, 128.34, 61.37, 14.37; MS (ES+) m/z (%) = 376,
[M+H+] (80), 382, (100), 425 (90); HRMS (ES+) calcd for C23H22NO4
376.1549, found 376.1560.
Ligand L7e. To 3 (0.087 g, 0.321 mmol) were added toluene
(2 mL), SOCl2 (6 mL) and DMF (10 μL). The mixture was refluxed
for 1 h, protected by a CaCl2 drying tube. The reaction was allowed to
cool, solvent was removed under reduced pressure, and the residue
was re-dissolved in DCM (25 mL). To this solution stirring at 0 °C,
protected by an nitrogen atmosphere, was added diethylamine (0.212 mL,
2.96 mmol) in small portions, After 24 h the DCM solution was washed
with 10% NaHCO3(aq) (2 × 10 mL) and brine (10 mL) and then dried
with Na2SO4, the solvent was removed under reduced pressure, and the
residue was purified on silica, eluting with EtOAc:hexane. The product
Ligand L2e. A mixture of nicotinic acid (1 g, 8.12 mmol), toluene
(10 mL), DMF (10 μL), and thionyl chloride (30 mL) was refluxed for
1 h, protected by a CaCl2 drying tube. The solvent was removed on a
rotary evaporator, and the residue was dissolved in DCM (20 mL).
N,N-Diethyl-2-hydroxyacetamide (1.3 mL, 10.3 mmol) was added in
small portions, and then triethylamine (2.46 mL, 24.4 mmol) was added
dropwise. The solution was allowed to stir 18 h at room temperature.
After dilution with DCM (20 mL), the solution was washed with
aqueous NaHCO3(aq) (10% w/v) (1 × 40 mL) and brine (1 × 40 mL)
1
was isolated as a clear oil: yield 0.060 g (70%); H NMR (250 MHz,
CDCl3) δ 8.86 (s, 1H), 8.63 (s, 1H), 7.90 (d, 1H, J = 8), 7.64−7.50 (m,
3H), 7.43−7.37 (m, 2H), 3.45 (d, 4H, J = 70), 1.22 (d, 6H, J = 30); 13C
NMR (100.6 MHz, CDCl3) δC = 170.76, 148.78, 148.23, 138.17, 138.15,
1861
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
and dried with MgSO4. The solvent was removed on a rotary
evaporator, and the crude product was purified on silica, eluting with
EtOAc:hexane. The product was isolated as a white solid: yield 1.52 g
(79%); mp = 65−66 °C; 1H NMR (250 MHz, CDCl3) δ 9.17 (s, 1H),
8.66 (d, 1H, J = 5), 8.24 (d, 1H, J = 8), 7.28 (dd, 1H, J = 8, J = 5), 4.88
(s, 2H), 3.29 (q, 2H, J = 7), 3.19 (q, 2H, J = 7), 1.13 (t, 3H, J = 7), 1.01
(t, 3H, J = 7); 13C NMR (62.9 MHz, CDCl3) δC = 164.91, 164.91,
153.52, 151.01, 137.23, 125.54, 123.22, 61.91, 40.87, 40.39, 14.06, 12.80;
MS (ES+) m/z (%) = 237 [M+H+] (100), 259 [M+Na+] (20); HRMS
(ES+) calcd for C12H17N2O3 237.1239, found 237.1229; FT-IR (thin
film) νmax/cm−1 3098, 3060, 2980, 2964, 2944, 2903, 1735, 1651, 1472,
1423, 1295, 1269, 1119.
Ligand L3e. A mixture of 3,5-pyridinedicarboxylic acid (1 g, 5.98
mmol), toluene (20 mL), dimethylformamide (20 μL), and thionyl
chloride (30 mL) was refluxed for 1 h, protected by a CaCl2 drying
tube. The solvent was removed on a rotary evaporator, and the residue
was dissolved in dichloromethane (20 mL). N,N-Diethyl-2-hydrox-
yacetamide (1.87 mL, 14.35 mmol) was added in small portions, and
then triethylamine (1.82 mL, 17.9 mmol) was added dropwise. The
solution was allowed to stir 18 h at room temperature. After dilution
with DCM (20 mL), the solution was washed with NaHCO3(aq) (10%
w/v) (40 mL) and brine (40 mL) and dried with MgSO4. The solvent
was removed on a rotary evaporator and the crude product was
purified on silica eluting with EtOAc:MeOH. The product was isolated
as white solid: yield 1.95 g (83%). Mp = 103−104 °C 1H NMR
(250 MHz, CDCl3) δ 9.44 (s, 2H), 9.01 (s, 1H), 5.00 (s, 4H), 3.40 (q,
4H, J = 7), 3.30 (q, 4H, J = 7), 1.26 (t, 6H, J = 7), 1.14 (t, 6H, J = 7);
13C NMR (62.9 MHz, CDCl3) δC = 164.74, 164.19, 154.70, 138.65,
125.57, 62.20, 40.99, 40.55, 14.18, 12.89; MS (ES+) m/z (%) = 394
[M+H+] (100); HRMS (ES+) calcd for C19H28N3O6 394.1978, found
394.1972; FT-IR (thin film) νmax/cm−1 3082, 3037, 2967, 2932, 1725,
1671, 1651, 1465, 1449, 1430, 1234, 1218, 1103, 1042, 1026.
Ligand L7c. To m-tolylboronic acid (0.359 g, 2.64 mmol),
Pd(0)(PPh3)4 (0.0449g, 0.0387 mml), and sodium carbonate (0.342,
3.23 mmol), protected by an argon atmosphere, were added THF (25
mL), toluene (25 mL), water (1 mL), and 3-bromopyridine (0.251
mL, 2.58 mmol). The mixture was heated at 105 °C for 24 h and then
allowed to cool, and the solvent was removed under reduced pressure.
The solid was re-dissolved in DCM (50 mL), washed with 10%
NaHCO3(aq) (20 mL) and brine (20 mL), and then dried with
Na2SO4, the solvent was removed under reduced pressure, and the
residue was purified on silica, eluting with EtOAc:hexane. The product
1
was isolated as a clear oil: yield 0.256 g (59%); H NMR (250 MHz,
CDCl3) δ 8.86 (s, 1H), 8.60 (d, 1H, J = 5), 7.87 (d, 1H, J = 8), 7.40−
7.33 (m, 4H), 7.24 (d, 1H, J = 4), 2.45 (s, 3H); 13C NMR (62.9 MHz,
CDCl3) δC = 148.38, 148.35, 138.77, 137.82, 136.75, 134.38, 129.01,
128.87, 127.92, 124.27, 123.52, 21.55; MS (ES+) m/z (%) = 170 [M
+H+] (100); HRMS (ES+) calcd for C12H12N 170.0790, found
170.0966; FT-IR (thin film) νmax/cm−1 3028, 2955, 2922, 2862, 1610,
1592, 1575, 1474, 1435, 1403, 1338, 1188, 1100, 1022.
Ligand L8c. To m-tolylboronic acid (0.359 g, 2.64 mmol),
Pd(0)(PPh3)4 (0.0449g, 0.0387 mmol), sodium carbonate (0.342 g,
3.23 mmol), and 3,5-dibromopyridine (0.306 g, 1.29 mmol), protected
by an argon atmosphere, were added THF (25 mL), toluene (25 mL),
and water (1 mL). This mixture was heated at 90 °C for 36 h and then
allowed to cool, and the solvent was removed under reduced pressure.
The solid was re-dissolved in DCM (50 mL), washed with 10%
NaHCO3(aq) (20 mL) and brine (20 mL), and dried with Na2SO4, the
solvent was removed under reduced pressure, and the residue was
purified on silica, eluting with EtOAc:hexane. The product was isolated
1
as a white solid: yield 0.082 g (24%); mp = 136−138 °C; H NMR
(250 MHz, CDCl3) δ 8.84 (s, 2H), 8.06 (s, 1H), 7.49−7.39 (m, 6H),
7.28 (d, 2H, J = 8), 2.48 (s, 6H); 13C NMR (62.9 MHz, CDCl3) δC =
146.97, 138.83, 137.80, 136.72, 132.95, 129.05, 128.98, 128,04, 124.40,
21.58; MS (ES+) m/z (%) = 260, [M+H+] (100); HRMS (ES+) calcd
for C19H18N 260.1439, found 260.1438.
Ligand L2f. A mixture of nicotinic acid (1 g, 8.12 mmol), toluene
(10 mL), DMF (10 μL), and thionyl chloride (30 mL) was refluxed for
1 h, protected by a CaCl2 drying tube. The solvent was removed on a
rotary evaporator, and the residue was dissolved in DCM (20 mL).
Ethyl glycolate (0.97 mL, 10.3 mmol) was added in small portions, and
then triethylamine (2.46 mL, 24.4 mmol) was added dropwise. The
solution was allowed to stir 18 h at room temperature. After dilution
with DCM (20 mL), the solution was washed with NaCO3(aq)
(10% w/v) (40 mL) and brine (40 mL) and dried with MgSO4.
The solvent was removed on a rotary evaporator, and the crude
product was purified on silica, eluting with EtOAc:hexane. The product
N,N-Diethyl-4-methylbenzamide, 5. To 4-methylbenzoyl chloride
(5 g, 32.3 mmol) stirring at 0 °C, protected by an N2 atmosphere, was
added diethylamine (16.8 mL, 162 mmol) in small portions. After 24 h
the DCM solution was washed with 10% NaHCO3(aq) (2 × 30 mL)
and brine (30 mL) and then dried with Na2SO4, the solvent was
removed under reduced pressure, and the residue was purified on
silica, eluting with EtOAc:hexane. The product was isolated as a clear
1
oil: yield 4.36 g (71%); H NMR (250 MHz, CDCl3) δ 7.25 (d, 2H,
J = 8), 7.16 (d, 2H, J = 8), 3.38 (d, 4H, J = 30), 2.34 (s, 3H), 1.15 (s,
6H); 13C NMR (100.6 MHz, CDCl3) δC = 171.46, 139.03, 134.38,
138.95, 126.31, 49.30, 39.21, 21.32, 14.23, 12.92; MS (ES+) m/z (%) =
192 [M+H+] (100); HRMS (ES+) calcd for C12H18NO 192.1388, found
192.1390; FT-IR (thin film) νmax/cm−1 2972, 2937, 2879, 1630, 1514, 1466,
1428, 1381, 1365, 1313, 1293, 1214, 1096.
1
was isolated as clear oil: yield 1.27 g (75%); H NMR (250 MHz,
CDCl3) δ 9.14 (s, 1H), 8.67 (d, 1H, J = 5), 8.21 (d, 1H, J = 8), 7.29 (dd,
1H, J = 8, J = 5), 4.76 (s, 2H), 4.12 (q, 2H, J = 7), 1.15 (t, 3H, J = 7);
13C NMR (62.9 MHz, CDCl3) δC = 167.26, 164.53, 153.72, 150.95,
137.16, 125.15, 123.27, 61.44, 61.26, 13.96; MS (ES+) m/z (%) = 210
[M+H+] (100); HRMS (ES+) calcd for C10H12NO4 210.0766, found
210.0768; FT-IR (thin film) νmax/cm−1 3024, 2984, 1732, 1590, 1423,
1383, 1295, 1215, 1124, 1117, 1030, 749.
UV/Visible Absorption and Fluorescence Titrations. UV/vis
titrations were carried out by preparing a 10 mL sample of porphyrin
at known concentration (4−7 μM) in spectroscopic-grade solvent. A
10 mL solution of ligand (12−3900 μM) was prepared using
spectroscopic-grade solvent. To a Hellma quartz 96-well plate was
added 150 μL of porphyrin solution, and the UV/vis absorbance was
recorded at five wavelengths. Aliquots of pyridine solution (3, 6, or 10 μL)
were added successively to the well containing the porphyrin solution,
using the BMG FLUOstar Omega plate reader, and the plate was
equilibrated at 298 K. For absorption experiments, the UV/vis absorbance
was recorded at five wavelengths after each addition. For emission
experiments, the plate was excited at 420 or 430 nM, and the fluorescence
emission was recorded at four wavelengths after each addition. Changes in
absorbance or emission were fit to a 1:1 binding isotherm in Microsoft
Excel to obtain the association constant. Each titration was repeated at
least three times, and the experimental error is quoted as twice the
standard deviation at a precision of one significant figure.
Ligand L3f. A mixture of 3,5-pyridinedicarboxylic acid (1 g, 5.98
mmol), toluene (20 mL), DMF (20 μL), and thionyl chloride (30 mL)
was refluxed for 1 h, protected by a CaCl2 drying tube. The solvent
was removed on a rotary evaporator, and the residue was dissolved in
DCM (20 mL). Ethyl glycolate (1.48 mL, 14.35 mmol) was added in
small portions, and then triethylamine (1.82 mL, 17.9 mmol) was
added dropwise. The solution was allowed to stir for 18 h at room
temperature. After dilution with DCM (20 mL), the solution was
washed with NaHCO3(aq) (10% w/v) (1 × 40 mL) and brine (1 × 40
mL) and dried with MgSO4. The solvent was removed on a rotary
evaporator, and the crude product was purified on silica, eluting with
EtOAc:hexane. The product was isolated as waxy solid: yield 1.74 g
(86%); mp = 41−43 °C; 1H NMR (250 MHz, CDCl3) δ 9.41 (s, 2H),
8.94 (s, 1H), 4.88 (s, 2 × 2H), 4.24 (q, 2 × 2H, J = 7), 1.27 (t, 2 × 3H,
J = 7); 13C NMR (62.9 MHz, CDCl3) δC = 167.09, 163.73, 154.77,
138.53, 125.29, 61.68, 61.57, 14.07; MS (ES+) m/z (%) = 340 [M
+H+] (100); HRMS (ES+) calcd for C15H18NO8 340.1032, found
340.1019; FT-IR (thin film) νmax/cm−1 3069, 3018, 2980, 2938, 1735,
1600, 1430, 1379, 1276, 1205, 1103, 1035, 1013.
ITC Measurments. ITC experiments were performed at 298 K on
a VP-ITC MicroCal titration calorimeter (MicroCal, Inc., North-
ampton, MA). In a typical calorimetric measurement, porphyrin host was
dissolved in HPLC-grade solvents at concentrations of 0.5−50 μM,
and the solution was loaded into the sample cell of the microcalorimeter.
1862
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863

Journal of the American Chemical Society
Article
The ligand guest solutions, 15−20 times more concentrated than the host
solution, were loaded into the injection syringe. The number of injections
was 30−50, and the volumes of the injections were between 7 and 10 μL,
with 30−60 s of duration and 300 s of spacing between the injections.
The dilution experiments were performed for each titration, loading the
guest solution at the same concentration as for the host titration into the
injection syringe, and adding guest to the solvent in the cell. The dilution
data were subtracted from each host titration thermogram. The data
fitting was performed by using ORIGIN (Version 7.0, Microcal, LLC
ITC) and a 1:1 binding isotherm (One Set of Sites model), allowing the
stoichiometry number, the stability constant K, and the binding enthalpy
ΔH values to float. At least two independent measurments were
performed for each host−guest system.
(5) (a) Bruice, T. C.; Lightstone, F. C. Acc. Chem. Res. 1999, 32,
127−136. (b) Bruice, T. C.; Pandit, U. K. J. Am. Chem. Soc. 1960, 82,
5858−5865. (c) Page, M. I.; Jencks, W. P. Proc. Natl. Acad. Sci. U.S.A.
1971, 68, 1678.
(6) (a) Hunter, C. A.; Ihekwaba, N.; Misuraca, M. C.; Segarra-Maset,
M. D.; Turega, S. M. Chem. Commun. 2009, 26, 3964−3966.
(b) Lemonnier, J. F.; Guenee, L.; Bernardinelli, G.; Vigier, J. F.;
Bocquet, B.; Piguet, C. Inorg. Chem. 2010, 49, 1252−1265. (c) Tobey,
S. L.; Anslyn, E. V. J. Am. Chem. Soc. 2003, 125, 10963−10970.
(d) Williams, D. H.; Davies, N. L.; Zerella, R.; Bardsley, B. J. Am.
Chem. Soc. 2004, 126, 2042−2049. (e) Kondratuk, D. V.; Perdigao, L.
M. A.; O’Sullivan, M. C.; Svatek, S.; Smith, G.; O’Shea, J. N.; Beton, P.
H.; Anderson, H. L. Angew. Chem., Int. Ed. 2012, 51, 6696−6699.
(7) (a) Krishnamurthy, V. M.; Semetey, V.; Bracher, P. J.; Shen, N.;
Whitesides, G. M. J. Am. Chem. Soc. 2007, 129, 1312−1320.
(b) Hossain, M. A. S.; Schneider, H. J. Chem.Eur. J. 1999, 5,
1284−1290. (c) ten Cate, A. T.; Kooijman, H.; Spek, A. L.; Sijbesma,
R. P.; Meijer, E. W. J. Am. Chem. Soc. 2004, 126, 3801−3808.
(d) Jiang, W.; Nowosinski, K.; Loew, N. L.; Dzyuba, E. V.; Klautzsch,
F.; Schaefer, A.; Huuskonen, J.; Rissanen, K.; Schalley, C. A. J. Am.
Chem. Soc. 2012, 134, 1860−1868. (e) Misuraca, M. C.; Grecu, T.;
Freixa, Z.; Garavini, V.; Hunter, C. A.; van Leeuwen, P.; Segarra-Maset,
M. D.; Turega, S. M. J. Org. Chem. 2011, 76, 2723−2732. (f) Searle,
M. S.; Williams, D. H. J. Am. Chem. Soc. 1992, 114, 10690−10697.
(g) Searle, M. S.; Williams, D. H.; Gerhard, U. J. Am. Chem. Soc. 1992,
114, 10697−10704. (h) Bohm, H. J. J. Comput.-Aided Mol. Des. 1994,
8, 243−256. (i) Nangreave, J.; Yan, H.; Liu, Y. J. Am. Chem. Soc. 2011,
133, 4490−4497.
(8) (a) Gunasekaran, K.; Nussinov, R. J. Mol. Biol. 2007, 365, 257−
273. (b) Jimenez, R.; Salazar, G.; Baldridge, K. K.; Romesberg, F. E.
Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 92−97. (c) Shi, R.; Munger, C.;
Kalan, L.; Sulea, T.; Wright, G. D.; Cygler, M. Proc. Natl. Acad. Sci.
U.S.A. 2012, 109, 11824−11829. (d) Sohtome, Y.; Nagasawa, K. Chem.
Commun. 2012, 48, 7777−7789. (e) Pearson, R. J.; Kassianidis, E.;
Slawin, A. M. Z.; Philp, D. Chem.Eur. J. 2006, 12, 6829−6840.
(9) (a) Chekmeneva, E.; Hunter, C. A.; Packer, M. J.; Turega, S. M. J.
Am. Chem. Soc. 2008, 130, 17718−17725. (b) Hunter, C. A.; Misuraca,
M. C.; Turega, S. M. J. Am. Chem. Soc. 2011, 133, 582−594.
(c) Hunter, C. A.; Misuraca, M. C.; Turega, S. M. J. Am. Chem. Soc.
2011, 133, 20416−20425. (d) Hunter, C. A.; Misuraca, M. C.; Turega,
S. M. Chem. Sci. 2012, 3, 589−601. (e) Hunter, C. A.; Misuraca, M. C.;
Turega, S. M. Chem. Sci. 2012, 3, 2462−2469. (f) Chekmeneva, E.;
Hunter, C. A.; Misuraca, M. C.; Turega, S. M. Org. Biomol. Chem.
2012, 10, 6022−6031.
NMR Titrations. NMR titrations were carried out by preparing a
2 mL sample of host at known concentration (8−60 mM). Then,
1
0.6 mL of this solution was removed, and a H NMR spectrum was
recorded. A solution of guest (12−1400 mM) was prepared by dissolving
the guest in 1 mL of the host solution, so that the concentration of host
remained constant throughout the titration. Aliquots of guest solution
were added successively to the NMR tube containing the host, and the
NMR spectrum was recorded after each addition. Changes in chemical
shift were fit to a 1:1 binding isotherm in Microsoft Excel. Each titration
was repeated at least three times, and the experimental error is quoted as
twice the standard deviation at a precision of one significant figure.
ASSOCIATED CONTENT
* Supporting Information
■
S
CIF file for the X-ray crystal structure of the P1a·L8f complex.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
c.hunter@sheffield.ac.uk
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the EPSRC and Royal Society for funding and the
National Crystallographic Service at the University of South-
ampton for collecting X-ray data.
(10) (a) Fersht, A. R.; Kirby, A. J. J. Am. Chem. Soc. 1967, 89, 4857.
(b) Chi, X. L.; Guerin, A. J.; Haycock, R. A.; Hunter, C. A.; Sarson, L.
D. Chem. Commun. 1995, 2563−2565. (c) Cacciapaglia, R.; Di
Stefano, S.; Mandolini, L. Acc. Chem. Res. 2004, 37, 113−122.
(d) Ercolani, G.; Piguet, C.; Borkovec, M.; Hamacek, J. J. Phys. Chem.
B 2007, 111, 12195−12203. (e) Sinclair, A. J.; del Amo, V.; Philp, D.
Org. Biomol. Chem. 2009, 7, 3308−3318.
REFERENCES
■
(1) (a) Badjic, J. D.; Nelson, A.; Cantrill, S. J.; Turnbull, W. B.;
Stoddart, J. F. Acc. Chem. Res. 2005, 38, 723−732. (b) Doyle, A. G.;
Jacobsen, E. N. Chem. Rev. 2007, 107, 5713−5743. (c) Kay, E. R.;
Leigh, D. A.; Zerbetto, F. Angew. Chem., Int. Ed. 2007, 46, 72−191.
(d) Lehn, J. M. Angew. Chem., Int. Ed. 1990, 29, 1304−1319. (e) Philp,
D.; Stoddart, J. F. Angew. Chem., Int. Ed. 1996, 35, 1155−1196.
(f) Mulder, A.; Huskens, J.; Reinhoudt, D. N. Org. Biomol. Chem.
2004, 2, 3409−3424. (g) Williams, D. H.; Bardsley, B. Angew. Chem.,
Int. Ed. 1999, 38, 1173−1193. (h) Mammen, M.; Choi, S. K.;
Whitesides, G. M. Angew. Chem., Int. Ed. 1998, 37, 2755−2794.
(i) Takeuchi, M.; Ikeda, M.; Sugasaki, A.; Shinkai, S. Acc. Chem. Res.
2001, 34, 865−873. (j) O’Sullivan, M. C.; Sprafke, J. K.; Kondratuk, D.
V.; Rinfray, C.; Claridge, T. D. W.; Saywell, A.; Blunt, M. O.; O’Shea, J.
N.; Beton, P. H.; Malfois, M.; Anderson, H. L. Nature 2011, 469, 72−
75.
(11) (a) Cockroft, S. L.; Hunter, C. A. Chem. Soc. Rev. 2007, 36,
172−188. (b) Camara-Campos, A.; Musumeci, D.; Hunter, C. A.;
Turega, S. J. Am. Chem. Soc. 2009, 131, 18518−18524.
(12) Gong, Y.; Pauls, H. W. Synlett 2000, 6, 829−831.
(13) (a) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Morris, J. J.;
Taylor, P. J. J Chem. Soc., Perkin Trans. 2 1990, 521−529. (b) Abraham,
M. H.; Platts, J. A. J. Org. Chem. 2001, 66, 3484−3491.
(14) Ercolani, E. Org. Lett. 2005, 7, 803−805.
(15) Hogben, H. J.; Sprafke, J. K.; Hoffmann, M.; Pawlicki, M.;
Anderson, H. L. J. Am. Chem. Soc. 2011, 133, 20962−20969.
(2) (a) Andrews, P. R.; Craik, D. J.; Martin, J. L. J. Med. Chem. 1984,
27, 1648−1657. (b) Free, S. M.; Wilson, J. W. J. Med. Chem. 1964, 7,
395. (c) Gohlke, H.; Klebe, G. Angew. Chem., Int. Ed. 2002, 41, 2645−
2676. (d) Tomic, S.; Nilsson, L.; Wade, R. C. J. Med. Chem. 2000, 43,
1780−1792. (e) Hunter, C. A. Angew. Chem., Int. Ed. 2004, 43, 5310−
5324.
(3) Kirby, A. J. Adv. Phys. Org. Chem. 1980, 17, 183.
(4) Galli, C.; Mandolini, L. Eur. J. Org. Chem. 2000, 3117−3125.
1863
dx.doi.org/10.1021/ja310221t | J. Am. Chem. Soc. 2013, 135, 1853−1863