Remote conformational control of a molecular switch via methylation and deprotonation

Article abstract of DOI:10.1039/c4ob01991a

Exacting control over conformation in response to an external stimulus is the central focus of molecular switching. Here we describe the synthesis of a series of diphenylacetylene-based molecular switches, and examine their response to covalent modificati

Full text of DOI:10.1039/c4ob01991a

Organic &
Biomolecular Chemistry
View Article Online
View Journal
PAPER
Remote conformational control of a molecular
switch via methylation and deprotonation†
Cite this: DOI: 10.1039/c4ob01991a
Peter C. Knipe, Ian M. Jones, Sam Thompson* and Andrew D. Hamilton*
Exacting control over conformation in response to an external stimulus is the central focus of molecular
switching. Here we describe the synthesis of a series of diphenylacetylene-based molecular switches,
and examine their response to covalent modification and deprotonation at remote phenolic positions.
A complex interplay between multiple intramolecular hydrogen bond donors and acceptors determines
the global conformation.
Received 18th September 2014,
Accepted 8th October 2014
DOI: 10.1039/c4ob01991a
www.rsc.org/obc
determinant of conformation is an intramolecular H-bond
between the methyl benzoate carbonyl and one of the two
Introduction
The control of molecular switching has the potential to amide NH’s. The H-bond acceptor is predictably biased toward
achieve targeted drug delivery, and is the basis of a new gene- the most electron-deficient H-bond donor. Structurally similar
ration of molecular sensors.1–4 For example, encapsulating acetylene-based compounds have previously been deployed as
drug cargo in a pH-responsive molecular container provides a specific anion receptors,15 and as components of switchable
vehicle to deliver therapy specifically to the acidic environment helical foldamers.16,17
that surrounds solid tumours, potentially lessening the dele-
Building on our previous work we set out to incorporate
terious off-target effects of traditional chemotherapies.5,6 two phenolic sites at the periphery of the scaffold, which are
Increasing the scope of stimuli for molecular switches is there- connected through a hydrogen-bonding network to a proximal
fore an important endeavour, and will facilitate their deploy- benzamide N–H. Each of these remote positions may be de-
ment in an expanded set of applications.
protonated or methylated before studying the global confor-
The post-translational modification of macromolecules, mational change (Fig. 1).
such as proteins and nucleic acids, exerts an additional layer
of conformational regulation.7–9 Two such reversible events are
Results and discussion
protonation and methylation. The latter has attracted particu-
lar attention, and is central to normal genetic regulation as
well as maladies such as cancer, diabetes, and Alzheimer’s
disease.7,10 The reversible protonation of proteins is respon-
sible for modulating regulatory processes, but is also impli-
cated as an aggravating factor in amyloid disease.11
The synthesis of switch compound 2 was achieved in 73%
yield upon treatment of aniline 118 with 2,5-dimethoxybenzoyl
chloride. The effects of methylation state on the conformation
Inspired by these systems, we were curious to explore the
effects of deprotonation and methylation on the conformation-
al behaviour of a molecular switch. Specifically we wished to
provide insights into the effect these modifications have on
hydrogen-bonded networks. We have previously employed a
bis-benzamido diphenylacetylene (DPA) system to examine the
effects of anions, Brønsted-, and Lewis acids on internal
conformational dynamics (Fig. 1).12–14 In these systems the key
Department of Chemistry, Chemistry Research Laboratory, University of Oxford,
Mansfield Road, Oxford OX1 3TA, UK. E-mail: sam.thompson@chem.ox.ac.uk,
andrew.hamilton@chem.ox.ac.uk; Tel: +44 (0) 1865 275978
†Electronic supplementary information (ESI) available: 1H & 13C NMR – includ-
ing spectra, HRMS, IR. Crystallographic data (excluding structure factors) for 2, 3
and 5 have been deposited. CCDC 1017646–1017648. For ESI and crystallo-
graphic data in CIF or other electronic format see DOI: 10.1039/c4ob01991a
Fig. 1 Study outline: conformational change upon methylation and
deprotonation.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
Scheme 1 The synthesis of dimethoxy, monomethoxy and dihydroxy
switch compounds 2, 3 and 4: (i) 2,5-dimethoxybenzoyl chloride (2 eq.),
pyridine, DMAP, CH2Cl2, room temperature, 30 min, 73%; (ii) BBr3·OEt2
(1.5 eq.), −30 °C, 15 min, 94%; (iii) BBr3·OEt2 (5 eq.), room temperature,
30 min, 71%.
Fig. 3 (a) X-ray crystal structure of 2, showing a NH⋯O hydrogen bond
with the ortho-methoxy substituent; (b) computed lowest energy con-
formation of 2. Bond lengths (in Å) are indicated in red.
Fig. 4 X-ray crystal structure of 3; (a) side view, distances (in Å) indi-
cated in red; (b) edge view, Cortho–C–C–O dihedral angle indicated in
red.
bonding from this amide to the ester. This postulate is further
borne out by the X-ray crystal structure of 2, and by its calcu-
lated lowest energy conformation (Fig. 3).21
The same analysis was applied to the monomethoxy com-
pound 3, and suggested a significant conformational change.
A large upfield shift in the benzamide NH relative to
dimethoxy compound 2 indicated a 4 : 3 ratio in favour of the
conformation in which the unsubstituted benzamide is
engaged in a hydrogen bond (ε = 0.57). The increased propen-
sity for hydrogen bonding to the substituted benzamide (rela-
Fig. 2 Generic switch, 0% and 100% control molecule structures and
NMR assay of conformation. R = H, CH3. The benzamide proton exam-
ined by 1H NMR is highlighted in pink.
of the DPA were initially investigated by sequential chemical tive to 2) is likely a consequence of hydrogen bonding between
demethylation of dimethoxybenzamide 2 with BBr3·OEt2. At the ortho-hydroxy donor and amide carbonyl acceptor, enhan-
cryogenic temperatures mono-demethylation to give 3 pro- cing the donor ability of the amide N–H in a cooperative
ceeded in 94% yield and with complete regioselectivity for the manner.22 X-ray crystallographic analysis provides evidence for
2-methoxy group. Conversely, performing the reaction at room this strong O–H⋯O hydrogen bond (Fig. 4a), and supports the
temperature with a greater excess of boron tribromide, dihy- slight solution-phase preference for the unsubstituted benz-
droxy phenol 4 was obtained as the sole product in 71% yield amide engaging in a hydrogen bond with the ester. The struc-
(Scheme 1).
ture also exhibits the out-of-plane twisting of the benzamide
The solution-phase conformation of each switch molecule engaged in hydrogen bonding characteristic of these systems
was examined by comparing the 1H NMR chemical shift of the (Fig. 4b).13
benzamide NH (δsw) with those of two control molecules, one
Lastly the solution-state conformational preference of dihy-
of which is incapable of forming an intramolecular hydrogen droxy switch compound 4 was examined.‡ As with the removal
bond (0% control, δ0) and one which is assumed to solely of one methyl group to form 3, removal of a second methyl
adopt a conformation in which the intramolecular hydrogen group elicited an upfield shift in the unsubstituted benzamide
bond exists (100% control, δ100, Fig. 2).18–20 A measure of the proton, consistent with a weaker intramolecular hydrogen
1
position of the conformational equilibrium, ε, is then obtained bonding interaction with the ester. H NMR analysis indicated
by employing eqn (1). Application of this assay to dimethoxy a 2 : 1 preference for the conformer in which the substituted
compound 2 indicates that hydrogen bonding occurs predomi- benzamide is engaged in a hydrogen bond (ε = 0.32). This
nantly between the ester and the unsubstituted benzamide
(a ratio of 11 : 2, ε = 0.85). This result is consistent with the
formation of an NH⋯O hydrogen bond between the ortho-
‡Due to low solubility in CDCl3, modified control molecules were employed, see
methoxy substituent and amide, effectively blocking hydrogen
ESI.†
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
observation suggests that the meta-hydroxyl group – known to
be electron-withdrawing in benzoic acid model systems23 – is
increasing the acidity of the corresponding amide NH, causing
it to be preferred over the unsubstituted benzamide. Interest-
ingly, both 3 and 4 exhibited strong solvent-dependent switch-
ing behaviour: when placed in 9 : 1 CDCl3–d6-DMSO (v/v), their
preference towards hydrogen bonding to the unsubstituted
benzamide was increased dramatically from the values
observed in pure CDCl3, to 18 : 1 and 17 : 1 respectively (ε3
=
0.95; ε4 = 0.94). Dimethoxy compound 2, on the other hand,
exhibited virtually no change (8 : 1 preference, ε2 = 0.89). This
dramatic difference is likely due to increased solvation of the
hydroxyl groups in 3 and 4, breaking the key structural
OH⋯OvC hydrogen bond.
Scheme 2 (a) Deprotonation of 3 to examine the solid-state confor-
mational preference of the corresponding phenolate 5. (b) Computed
lowest energy conformer of 5 (counter-ion not modelled). (c) X-Ray
crystal structure of 5. Distances (in Å) indicated in red.
We proceeded to investigate the dynamic behaviour of this
system upon treatment with base. Dihydroxy switch 4 was
unsuitable for this study, since quantitative assaying was con-
founded by its poor solubility. Therefore studies were carried
out solely on 3. We postulated that base would deprotonate the
acidic phenol, giving a phenolate that would hydrogen bond to
the adjacent amide, blocking it from bonding to the ester. An
NMR titration experiment was carried out by adding DBU in
small portions to CDCl3 solutions of 3 and its corresponding
control compounds in turn. As expected the unsubstituted
benzamide NH shifted markedly downfield, consistent with
its increased involvement in H-bonding. The second amide
proton was immediately broadened to the extent that it could
no longer be observed after the addition of 0.3 eq. DBU, con-
sistent with its undergoing rapid exchange in the presence of
the base. Quantitative analysis of switching as a function of
base was not possible, since after the addition of ca. 1 eq. DBU
the switch NH chemical shift moved downfield beyond the
shift of the 100% control, rendering eqn (1) inapplicable.
Nonetheless, the significant, dose-dependent downfield shift
of the benzamide proton provides compelling evidence of
switching behaviour. To confirm the preferred conformation
of the phenolate, n-Bu4N+OH− was added to a solution of 3 in
anhydrous dichloromethane to generate tetrabutylammonium
salt 5 (Scheme 2a). Diffraction quality crystals were obtained
by recrystallization from acetonitrile, and the salt 5 was
observed to adopt the predicted conformation, with the
phenolate H-bonding to the adjacent amide, thus blocking it
from hydrogen bonding to the methyl ester (Scheme 2c). This
structure is similar to 2, wherein the ortho-methyl ether partici-
pates in an analogous intramolecular H-bond. The computed
lowest energy conformation is in good agreement with this
structure; the alternative rotamer in which the hydrogen bond
is formed between the ester and substituted benzamide lies
more than 12 kcal mol−1 (K ≈ 6.8 × 108) higher in energy
(Scheme 2b).
Fig. 5 1H NMR titration (600 MHz) of 3 (3.2 mM) with DBU (1,8-diaza-
bicycloundec-7-ene) in CDCl3.
Fig. 6 A comparison of the fluorescence spectra 2 (blue), 3 (red), and 4
(black).
When the system is conformationally unlocked by mono-
demethylation to form 3, λmax redshifts to 515 nm (Fig. 6).
This is a significant change, and may be a consequence of the
availability of the phenol OH to participate in an alternative
ESIPT reaction with the adjacent amide, generating a transient
excited enol-phenoxide tautomer.25 Upon removing the second
methyl group to generate 4 λmax remains the same – consistent
with the same mechanism of fluorescence being in operation –
but its intensity is quenched. This is in agreement with the
Lastly, we investigated the fluorescence properties of the
three methylation states – 2, 3 and 4 – to determine whether
they could be characterized by a change in emission spectra.
Compound 2 has a single emission band at λmax = 410 nm,
which is likely due to excited-state intramolecular proton transfer
(ESIPT), an effect previously observed for salicylamide.24
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.

View Article Online
Paper
Organic & Biomolecular Chemistry
conformational changes indicated by NMR analysis, where 4 2 (167 mg, 73%) as a viscous yellow oil that solidified upon
exists primarily with the 2,5-dihydroxybenzamide engaged in a standing; Rf 0.45 (9 : 1 chloroform–ethyl acetate); δH (400 MHz,
hydrogen bond with the ester. In this conformation, intramole- CDCl3) 10.67 (s, 1H), 9.50 (s, 1H), 8.42 (d, J 8.4, 1H), 8.38 (dd,
cular collisional quenching of the excited state is favoured by J 8.4, 0.7, 1H), 8.08 (dd, J 8.0, 1.1, 1H), 8.03 (s, 1H), 7.96 (d,
the proximity of the ester, and a reduction in fluorescence J 7.1, 2H), 7.86 (d, J 3.2, 1H), 7.69 (dd, J 7.8, 1.1, 1H), 7.57
quantum yield is therefore expected. Thus, both the position (td, J 7.6, 1.3, 1H), 7.48 (d, J 8.0, 2H), 7.42 (t, J 7.9, 2H), 7.09 (d,
and intensity of fluorescence peaks provide a semi-quantitative J 3.3, 1H), 6.96 (d, J 9.0, 1H), 3.88 (s, 3H), 3.70 (s, 3H), 3.43
metric for molecular conformation.
(s, 3H); δC (126 MHz, CDCl3) 167.1, 165.6, 163.4, 154.4, 151.9,
140.7, 140.4, 135.8, 133.8, 132.4, 131.6, 131.0, 130.6, 128.8,
128.4, 128.2, 123.4, 123.0, 120.3, 115.9, 115.7, 115.5, 114.1,
102.9, 101.5, 86.9, 57.4, 56.0, 52.2; HRMS (ESI): found
535.1863; C32H27N2O6 [M + H]+ requires 535.1791.
Conclusions
Seeking to expand the scope of stimulus-responsive molecular
switching we have developed a system to observe how methyl-
ation and deprotonation can affect the interplay of H-bond
donors and acceptors and lead to a global conformation
change. Using molecular mechanics models, X-ray crystallo-
graphy and 1H NMR, we have observed three conformational
states that are dependent on the degree of methylation. These
states also have definitive fluorescence characteristics, which
enable their rapid identification. Whilst methylation and
demethylation are static states with relatively long lifetimes,
protonation and deprotonation are dynamic, fluxional pro-
cesses. We have demonstrated that similar conformational
changes can be brought about both by methylation and depro-
tonation. These stimuli provide a novel means by which to
control the global conformation of molecular switches, and
should inform future developments in the field.
Methyl 2-((2-benzamido-6-(2-hydroxy-5-methoxybenzamido)-
phenyl)ethynyl)benzoate (3). Boron tribromide (1
M in
dichloromethane, 140 μL, 0.14 mmol) was added dropwise to a
stirred −30 °C solution of 2 (50 mg, 0.094 mmol) in dichloro-
methane (4 mL). After 15 min the reaction was quenched by
the cautious addition of water (5 mL) and diluted with dichloro-
methane (10 mL). The layers were separated, and the aqueous
layer was extracted with dichloromethane (3 × 5 mL). The com-
bined organic extracts were dried over anhydrous magnesium
sulfate, filtered and concentrated in vacuo. The crude product
was purified by flash column chromatography (silica gel,
dichloromethane–ethyl acetate, 50 : 1) to afford 3 (46 mg, 94%)
as a white powder. δH (CDCl3, 400 MHz): 11.27 (s, 1H), 9.27
(s, 1H), 9.18 (s, 1H), 8.40 (dd, J 8.3, 0.5, 1H), 8.18 (dd, J 8.3, 0.6,
1H), 8.09 (dd, J 7.9, 1.0, 1H), 7.97 (d, J 7.1, 2H), 7.70 (dd, J 7.8,
0.9, 1H), 7.60–7.52 (m, 2H), 7.51–7.44 (m, 4H), 7.25 (d, J 2.8,
1H), 7.07 (dd, J 9.1, 2.9, 1H), 6.99 (d, J 9.0, 1H), 3.70 (s, 3H),
3.50 (s, 3H); δC (CDCl3, 101 MHz): 168.4, 166.5, 165.4, 155.8,
151.9, 140.4, 138.8, 135.3, 133.3, 132.5, 131.8, 131.0, 130.9,
129.1, 128.5, 127.8, 122.8, 121.0, 119.4, 115.9, 115.7, 115.2,
111.3, 103.1, 103.0, 85.6, 56.0, 52.2; HRMS (ESI): found
521.1591; C31H25N2O6 [M + H]+ requires 521.1634.
Experimental details
The synthesis of 1 has previously been reported.26 For full
experimental details describing the synthesis of the control
molecules, please refer to the ESI.†
Methyl 2-((2-benzamido-6-(2,5-dihydroxybenzamido)phenyl)-
ethynyl)benzoate (4). Boron tribromide (1 M in dichloro-
methane, 0.20 mL, 0.20 mmol) was added dropwise to a
Synthetic procedures
Methyl 2-((2-benzamido-6-(2,5-dimethoxybenzamido)phenyl)- stirred RT solution of 2 (20 mg, 0.037 mmol) in dichloro-
ethynyl)benzoate (2). Oxalyl chloride (0.15 mL) was added methane (2 mL). After 2 h the reaction mixture was quenched
dropwise over 1 min to a solution of 2,5-dimethoxybenzoic by the cautious addition of water (5 mL) and diluted with
acid (157 mg, 0.43 mmol) in dichloromethane (5 mL) and N,N- dichloromethane (5 mL). The layers were separated and the
dimethylformamide (2 drops). The reaction was stirred for aqueous layer was extracted exhaustively with 3 : 1 chloroform–
40 min. After this period the solution was concentrated isopropanol (v/v, 6 × 5 mL). The combined organic extracts
in vacuo and then re-suspended in dichloromethane (2 mL). were dried over anhydrous magnesium sulfate, filtered and
In another flask, methyl 2-((2-amino-6-benzamidophenyl)- concentrated in vacuo. The crude residue was purified by
ethynyl)benzoate 1 (160 mg, 0.43 mmol) and 4-dimethyl- flash column chromatography (silica gel, dichloromethane–
aminopyridine (ca. 2 mg) were added to a mixture of dichloro- ethyl acetate, 4 : 1→1 : 1) to afford 4 (13.5 mg, 71%) as a white
methane (2 mL) and pyridine (0.24 mL, 0.65 mmol). This powder. δH (400 MHz, CDCl3) 11.37 (s, 1H), 9.08 (s, 1H), 9.02
solution was stirred for 0.5 h before the solution of acid (s, 1H), 8.39 (d, J 8.3, 1H), 8.13 (d, J 8.4, 1H), 8.07 (d, J 7.6, 1H),
chloride in dichloromethane was added to the amine over 7.97 (d, J 7.4, 2H), 7.66 (d, J 2.8, 1H), 7.59 (d, J 7.0, 2H), 7.56 (d,
3 min. The reaction was stirred for 30 min after which time J 7.3, 1H), 7.48 (quint., J 8.5, 7.9, 4H), 7.08 (dd, J 9.0, 2.8, 1H),
the solution was diluted with dichloromethane and the 6.93 (d, J 9.0, 1H), 3.82 (s, 3H); δH (75 MHz, CDCl3) 168.7, 167.8,
organic layer washed with 2 M HCl and brine, dried over anhy- 165.7, 156.5, 150.9, 140.0, 138.7, 135.1, 133.1, 133.0, 132.3, 131.2,
drous magnesium sulfate, filtered and concentrated in vacuo. 131.0, 130.3, 129.2, 129.0, 127.4, 123.3, 120.5, 119.0, 116.7,
The crude residue was purified by flash column chromato- 115.7, 115.3, 114.3, 110.2, 103.5, 103.1, 85.8, 77.4, 57.1; HRMS
graphy (silica gel, dichloromethane–ethyl acetate, 20 : 1) to give (ESI): found 507.1461; C30H23N2O6 [M + H]+ requires 507.1478.
Org. Biomol. Chem.
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Tetra-n-butylammonium 2-((3-benzamido-2-((2-(methoxycarb-
onyl)phenyl)ethynyl)phenyl)carbamoyl)-4-methoxyphenolate (5).
A solution of tetrabutylammonium hydroxide (1.0 M in metha-
nol, 10 μL, 0.01 mmol) was added dropwise to a RT solution of
3 (5 mg) in anhydrous dichloromethane (0.5 mL) under argon.
The solution immediately changed from colourless to a bright
yellow colour. After 30 min the reaction vessel was connected
via tubing to a reservoir containing hexane (∼3 mL) under
argon to facilitate vapour diffusion crystallization. After 72 h
only yellow oily film was observed, which was taken up
in acetonitrile (0.5 mL) and immediately formed yellow
crystalline plates, which were used for X-ray crystallographic
analysis. The yield was not determined. MP: 165–167 °C
(acetonitrile).
3 B. Valeur and I. Leray, Coord. Chem. Rev., 2000, 205, 3–40.
4 Molecular Switches, ed. B. L. Feringa and W. R. Browne,
Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2nd edn,
2011.
5 A. V. Samoshin, I. S. Veselov, V. a. Chertkov, A. a.
Yaroslavov, G. V. Grishina, N. M. Samoshina and
V. V. Samoshin, Tetrahedron Lett., 2013, 54, 5600–5604.
6 H. Karanth and R. S. R. Murthy, J. Pharm. Pharmacol., 2007,
59, 469–483.
7 E. L. Greer and Y. Shi, Nat. Rev. Genet., 2012, 13, 343–357.
8 D. Gioeli and B. M. Paschal, Mol. Cell. Endocrinol., 2012,
352, 70–78.
9 E. T. Yukl and C. M. Wilmot, Curr. Opin. Chem. Biol., 2012,
16, 54–59.
10 D. Y. Lee, C. Teyssier, B. D. Strahl and M. R. Stallcup,
Endocr. Rev., 2005, 26, 147–170.
11 A. Schönichen, B. A. Webb, M. P. Jacobson and
D. L. Barber, Annu. Rev. Biophys. Biomol. Struct., 2013, 42,
289–314.
12 I. M. Jones and A. D. Hamilton, Angew. Chem., Int. Ed.,
2011, 50, 4597–4600.
13 I. M. Jones, H. Lingard and A. D. Hamilton, Angew. Chem.,
Int. Ed., 2011, 50, 12569–12571.
14 P. C. Knipe, H. Lingard, I. M. Jones, S. Thompson and
A. D. Hamilton, Org. Biomol. Chem., 2014, 12, 7937–7941.
15 J. Raker and T. E. Glass, J. Org. Chem., 2002, 67, 6113–6116.
16 J. Suk, V. R. Naidu, X. Liu, M. S. Lah and K. Jeong, J. Am.
Chem. Soc., 2011, 133, 13938–13941.
Conformational analysis
In general, the position of the conformational equilibrium
exhibited by switch molecules 2, 3 and 4 was examined by com-
parison of the position of the 1H NMR signal corresponding to
the unsubstituted benzamide NH with its position in the 0%
and 100% control molecules. By applying eqn (1), a value ε is
obtained whose value reflects the position of the equilibrium.
δsw ꢀ δ0
ε ¼
ð1Þ
δ100 ꢀ δ0
In extremis, ε assumes values of zero where the switch does
not hydrogen bond to the benzamide, and unity where it exclu-
sively does so.
DBU-mediated switching
17 Y. Haketa, Y. Bando, K. Takaishi, M. Uchiyama,
A. Muranaka, M. Naito, H. Shibaguchi, T. Kawai and
H. Maeda, Angew. Chem., Int. Ed., 2012, 51, 7967–7971.
18 I. M. Jones and A. D. Hamilton, Org. Lett., 2010, 12, 3651–
3653.
A solution of switch 3 (1.0 mg, 0.0019 mmol) in CDCl3
(600 μL) was placed in an NMR tube, and an initial H NMR
1
spectrum acquired. A solution of 1,8-diazabicycloundec-7-ene
(0.05 M in CDCl3) was subsequently added in portions (10 ×
8 μL, 2 × 16 μL, 1 × 50 μL). After each addition the NMR tube
was shaken vigorously, and an additional 1H NMR spectrum
was acquired. The precise concentration of DBU at each stage
was determined by integration of the DBU CH2 multiplet at
∼2.5 ppm relative to the switch compound’s CH3 signal at
∼3.7 ppm. Repeating the experiment and analysis described
above for the control compounds and plotting the position of
the benzamide NH vs. equivalents of base generated the data
presented in Fig. 5.
19 I. M. Jones, P. C. Knipe, T. Michaelos, S. Thompson and
A. D. Hamilton, Molecules, 2014, 19, 11316–11332.
20 Calculation21 gave the hydrogen-bonded conformations of
the 100% controls relating to switches 2, 3 and 4 to be 5.0,
4.8 and 4.3 kcal mol−1 lower in energy than the non-hydro-
gen-bonded conformers (see ESI† for full method). These
values correspond to hydrogen-bonded populations of
greater than 99.9% at room temperature in each case.
21 Molecular Operating Environment (MOE), Chemical Comput-
ing Group, Inc., Montreal, QC, Canada, 2013.
22 Ion pairing may cause quantitative differences in the
strengths of hydrogen bonds within the network. However,
since the switching ratios are calculated relative to controls
in which ion pairing is equally as likely to exist, this effect
can be neglected to a first approximation.
Acknowledgements
The authors would like to thank The University of Oxford for
financial support, and Dr Michael Takase for assistance with
X-ray crystallography.
23 C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91,
165–195.
24 J. Catalan, F. Toribio and A. U. Acuna, J. Phys. Chem., 1982,
86, 303–306.
Notes and references
1 G. T. Spence, C. Chan, F. Szemes and P. D. Beer, Dalton 25 T. Nishiya, S. Yamauchi, N. Hirota, M. Baba and
Trans., 2012, 41, 13474–13485. I. Hanazaki, J. Phys. Chem., 1986, 90, 5730–5735.
2 Y. B. Zheng, B. Kiraly and T. J. Huang, Nanomedicine, 2010, 26 I. M. Jones and A. D. Hamilton, Org. Lett., 2010, 12, 3651–
5, 1309–1312.
3653.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem.