Protons as the triggers to regulate hydrogen-bonding receptors

Article abstract of DOI:10.1021/ol0170915

(equation presented) The protonation of alkylamines in two novel receptors results in intramolecular host-guest associations between the resulting ammonium ions and crown ether macrocycles. These interactions result in conformational changes of the receptors and prevent them from acting as hydrogen bond complements for uracil and carboxylate guest species.

Full text of DOI:10.1021/ol0170915

ORGANIC
LETTERS
2002
Vol. 4, No. 6
881-884
Protons as the Triggers to Regulate
Hydrogen-Bonding Receptors
,†
Mohammad H. Al-Sayah and Neil R. Branda*
Department of Chemistry, UniVersity of Alberta, Edmonton, Alberta, Canada T6G 2G2
nbranda@sfu.ca
Received November 21, 2001
ABSTRACT
The protonation of alkylamines in two novel receptors results in intramolecular host−guest associations between the resulting ammonium
ions and crown ether macrocycles. These interactions result in conformational changes of the receptors and prevent them from acting as
hydrogen bond complements for uracil and carboxylate guest species.
It is remarkable how the presence or absence of mere
hydrogen atoms can influence how molecular systems as
complicated as those in Nature operate. It is well accepted,
however, that the harnessing of structural changes that
accompany variations in the pH of the environment is one
of the most versatile methods to regulate the activity of
natural receptors.1 This regulation is often accomplished not
by interfering directly with the recognition event but by
reversibly protonating a location distant from the active
binding site to induce a conformational change in the receptor
and to subsequently affect its affinity for its substrate.2 The
induced structural change can either enhance (positive
allostery) or diminish (negative allostery) the receptor’s
binding efficacy.
negative allosteric cofactor to inhibit substrate recognition.
The operation of the two receptors is designed around the
same principle: the protonation of a functional group that
is not an integral member of the substrate-specific molecular
recognition motif creates a self-complementary species that
folds in on itself with the concomitant distortion of the
hydrogen bond site. The only difference between the two
receptors is in what substrates they target. Receptor 1 presents
a two-point donor-donor hydrogen-bonding urea appropriate
for binding anionic carboxylate guests (Scheme 1). Receptor
2 possesses the three-point triaminotriazine donor-acceptor-
donor site better suited to bind imides (Scheme 2). Both
Despite Nature’s frequent use of this regulatory strategy,
there are few synthetic systems that have borrowed from its
principles.3 Here we report two novel examples of artificial
hydrogen-bonding receptors that use a single proton as a
Scheme 1
† Current address: Department of Chemistry, 8888 University Drive,
Simon Fraser University, Burnaby, British Columbia, Canada V5A 1S6.
(1) Scrimgeour, K. G. Chemistry and Control of Enzyme Reactions;
Acedemic Press Inc.: New York, 1977.
(2) (a) Schetz, J. A.; Sibley, D. R. J. Pharmacol. Exp. Ther. 2000, 296,
359. (b) Traynelis, S. F.; Burgess, M. F.; Zheng, F.; Lyuboslavsky, P.;
Powers, J. L. J. Neurosci. 1998, 18, 6163.
10.1021/ol0170915 CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/22/2002

The intramolecular association between the ammonium
group and the crown ether macrocycle in 1‚H+ is easily di-
agnosed by two-dimensional NMR spectroscopy (TROESY)
which shows correlation peaks between the methylene CH2
protons of the ethylamino chain and those of the crown ether.
These correlation peaks are absent in the TROESY spectrum
of the neutral form of the receptor, 1, suggesting that
protonation forces the receptor to turn in on itself, although
the presence of intermolecular association between the cation
of one receptor and the macrocycle of another cannot be
completely ruled out with only this evidence. However, the
electrospray mass spectrum of 1‚H+ exhibits a major peak
corresponding to the molecular mass of the monomer 1‚H+
as well a minor peak corresponding to the dimer (1‚H+)2.
The binding efficacy of receptor 1 to its substrate class
Scheme 2
1
was evaluated by analyzing the H NMR spectral changes
that occur upon titrating DMSO-d6 solutions of 1 with
tetrabutylammonium (TBA) acetate. The significant down-
field shift (∆δ > 2.5 ppm) observed for the signals
corresponding to the urea N-H of 1 when it is treated with
TBA acetate is indicative of effective hydrogen bonding even
in this competitive solvent (Figure 1). After 2-3 equiv of
brandish the two components responsible for the allosteric
event, a primary amine and a crown ether. In both receptors,
1 equiv of an acid converts the former functional group into
an ammonium cation which is a well-tested guest for the
crown ether macrocycle. It is the creation of this host-guest
pair that induces the conformational change and forces the
receptors’ C-N bonds to rotate into a nonproductive
geometry, thus shutting down the receptors. Deprotonation
destroys the ammonium-crown ether partnership and reac-
tivates the receptors.4
Urea 1 was prepared in two steps by treating N-[2-(4-
aminophenyl)ethyl]-2,2,2-trifluoroacetamide5 with triphos-
gene, followed by 4-aminobenzo-18-crown-6 as outlined in
Scheme 3. The primary amine was deprotected with 10%
Figure 1. Observed chemical shifts (δ) of the urea N-H
resonances in the 1H NMR spectra of receptor 1 when titrated with
aliquots of TBA acetate ([) and when 1‚H+ is titrated with TBA
acetate (O) in DMSO-d6. [1] and [1‚H+] ) 2 × 10-3 M, [TBA
acetate] ) 2 × 10-2 M.
Scheme 3
acetate have been added, the neutral form of the receptor
reaches saturation. Data sets from all titration experiments
correlate well with the curves calculated using a 1:1 binding
model6 and give an average association constant (Ka) value
(3) (a) Cordova, E.; Bissell, R. A.; Spencer, N.; Ashton, P.; Stoddart, J.
F.; Kaifer, A. E. J. Org. Chem. 1993, 58, 6550. (b) Martinez-Diaz, M.-V.;
Spencer, N.; Stoddart, J. F. Angew. Chem., Int. Ed. Engl. 1997, 36, 1904.
(c) Nakatsuji, Y.; Kobayashi, H.; Okahara, M. J. Chem. Soc., Chem.
Commun. 1983, 801.
(4) We have previously shown how metal ions can be used as allosteric
cofactors to bring about similar structural changes: (a) Al-Sayah, M. H.;
Branda, N. R. Chem. Commun. 2002, 178. (b) Al-Sayah, M. H.; Branda,
N. R. Angew. Chem., Int. Ed. Engl. 2000, 39, 945.
(5) Desaubry, L.; Johnson, R. Bioorg. Med. Chem. Lett. 1997, 7, 123.
(6) The 1H NMR titration data were analyzed using Christopher A.
Hunter’s 1:1 complexation model program (Department of Chemistry,
University of Sheffield, UK).
K2CO3/MeOH, affording the final receptor 1 as a white solid.
The ammonium form, 1‚H+, was generated as its tetrafluoro-
borate salt by the careful addition of a methanolic solution
of HBF4 (1.2 equiv) to 1, followed by drying under vacuum.
882
Org. Lett., Vol. 4, No. 6, 2002

of 4400 ( 1100 M-1 over three experiments. This value is
large when compared to those reported for the complexation
between alkylurea hosts and carboxylate guests which are
typically in the range of 50-100 M-1 in this solvent.7,8 We
attribute this discrepancy to the fact that the reported values
of Ka are for dialkylureas which possess significantly less
acidic N-H protons than their diaryl counterparts. The
greater acidity of these hydrogen bond donor groups in
receptor 1 explain the enhanced binding observed in our
studies. Hamilton invokes a similar explanation to account
for the increase in the binding of carboxylate guests by the
more acidic thiourea receptors (pKa ∼ 21) than the urea
analogues (pKa ∼ 27).8 Receptor 1 must, therefore, be treated
as a diphenylurea derivative (pKa ) 19.5)9a instead of a
diethylurea derivative (pKa ) 26).9b As a consequence, the
urea in 1 will act as a better hydrogen bond donor and the
larger value of Ka is not unexpected.8
Scheme 4
Repeating the titration experiments but replacing the active
receptor with the preformed complex 1‚H+ generated data
that could not be fit to any of the binding models. The
insignificant shifting (∆δ < 0.1 ppm) of the resonances
corresponding to the N-H protons upon the addition of a
DMSO-d6 solution TBA acetate signifies that the two
components only associate to a very small extent, at least in
the early phase of the titration experiment. However, as the
experiment progresses and 2-3 equiv of TBA acetate are
added to the solution, the receptor seems to be reactivated,
as evident by the sudden downfield shifting of the N-H
signal. Clearly the hydrogen-bonding recognition site for the
acetate anion is reconstructed in response to the excess guest
species.
The association constant for the binding of receptor 2 to
imide guests was estimated by titrating CD2Cl2 solutions of
n-butyluracil with the receptor while monitoring the signals
for the protons of the guest species in the 1H NMR spectra.
In this case, the signal for the N-H proton of N-butyluracil
could not be followed because it consistently disappeared
after about 0.5 equiv of the receptor was added, although
the significant downfield shifting (∆δ > 2 ppm) that occurred
before the signal completely vanished clearly indicates
effective host-guest association through hydrogen bonding.
The disappearance of the N-H signal can be explained by
the occurrence of rapid proton exchange between the N-H
of the uracil substrate and the alkylamine of receptor 2 as
they are expected to be compatible in an acid-base reaction.
(The acidities of the ethylammonium cation (pKa ) 10.8)
and N-pentyluracil (pKa ) 9.98) are similar.)10 This argument
is supported by a simple experiment. When 1 equiv of
n-butylamine was added to a CD2Cl2 solution of N-butyl-
uracil, the signal corresponding to the N-H proton of the
latter species immediately disappeared. The presence of
n-butylamine, however, had no observable effect on the
chemical shifts of the methine C-H protons of N-butyluracil.
The latter observation is significant because the small (∆δ
∼ 0.1 ppm) but still meaningful downfield shifting of signals
corresponding to the C-H protons of the uracil provide a
means to measure the association constant which was
estimated to be 570 ( 100 M-1 as an average of three runs
(Figure 2). The data fit well to a 1:1 binding model.
1
At saturation (>10 equiv of TBA acetate), the H NMR
spectrum shows that the chemical shift of the N-H signal
and the aromatic region of 1‚H+ are similar to that of the
neutral receptor 1, also in the presence of excess guest. This
implies that a host-guest complex (1‚H+)(TBA acetate)n has
adopted a similar geometry to that of (1)(TBA acetate) where
the ammonium is solvated by the excess acetate instead of
forming an intramolecular interaction with the macrocycle.
This does not, however, rule out the presence of dimers and
oligomers at lower concentrations of acetate. In fact, the
electrospray mass spectrometry studies of the protonated
form (1‚H+)(TBA acetate)n showed peaks corresponding to
both the monomer and the dimer in a 9:1 ratio. When more
TBA acetate was added, this ratio remained unchanged even
in the presence of up to 20 equiv of acetate.
The protonated form of the triazine receptor, 2‚H+, did
not show this substrate competition. Receptor 2 was prepared
as outlined in Scheme 4 starting with cyanuric chloride and
the same two amines as were used to prepare urea 1. The
ammonium form 2‚H+ was prepared as its tetrafluoroborate
salt by using HBF4 in an identical fashion as already
described for 1‚H+.
This association constant corresponds well with that
measured for the binding of N-butyluracil to diphenyltriazine
(Ka ) 487 and 508 M-1 when the signals for the N-H and
C-H protons were followed, respectively). The fact that the
signals for the N-H could be monitored throughout this last
titration experiment supports our claim that the N-H in
N-butyluracil is exchanging with the NH2 of receptor 2 in
an acid-base reaction. The association constant for the
protonated form of the receptor, 2‚H+, could only be
estimated to be below 10 M-1 due to the shallow curve which
(7) Kelly, T. R.; Kim, M. H. J. Am. Chem. Soc. 1994, 116, 7072.
(8) Fan, E.; Arman, A. V.; Kincaid, S.; Hamilton, A. D. J. Am. Chem.
Soc. 1993, 115, 369.
(9) (a) Bordwell, F. G.; Algrim, D. J.; Harrelson, J. A., Jr. J. Am. Chem.
Soc. 1988, 110, 5903. (b) Valter, B.; Terekhova, M. I.; Petrov, E. S.;
Stehlicek, J.; Sebenda, J. Collect. Czech. Chem. Commun. 1985, 50, 840.
(10) Hine, J.; Hwang, J. S.; Balasubramanian, V. J. Org. Chem. 1988,
53, 5145.
Org. Lett., Vol. 4, No. 6, 2002
883

excess TBA acetate. Clearly, 2‚H+ is not reactivated by the
presence of its substrate. We attribute this absence of
substrate competition to three factors. First, because CD2Cl2
does not have the same ability to solvate ions as does DMSO,
the unimolecular complexation equilibrium will lie heavily
toward the intramolecularly associated ammonium ion and
the crown ether. Second, the N-butyluracil is a neutral
substrate and, therefore, cannot effectively solvate the
ammonium ion even when it is in excess. Finally, the
rotational barrier imposed on the two exocyclic C-N bonds
on the triaminotriazine heterocycle in structures such as
receptor 2 is smaller than that of the two C-N bonds in
urea derivatives (such as found in receptor 1). In fact, urea
derivatives most often prefer to exist in their trans-trans
conformations.11 This is a geometry that is satisfied in the
neutral form of the receptor (1) and not in its protonated
state (1‚H+). The other conformations of ureas exhibit little,
if any, affinity for carboxylate guests.
Figure 2. Observed chemical shift (δ) of the C-H signal of
1
N-butyluracil in the H NMR spectra when titrated with aliquots
of receptor 2 ([) and when titrated with 2‚H+ (O) in CD2Cl2.
[N-butyluracil] ) 2 × 10-3 M, [2] and [2‚H+] ) 2 × 10-2 M.
In conclusion, urea 1 and triazine 2 represent novel
examples of proton-regulated receptors that are only active
at high pH. The conformational changes that accompany the
lowering of the pH govern the activity of the receptors.
could not be fit to any binding model (Figure 2). The small
movement of the C-H resonance of N-butyluracil (∆δ <
0.01 ppm) can be attributed to the hydrogen bonding of
N-butyluracil to the substantially less effective donor-
acceptor hydrogen bond sites in 2‚H+. Despite the existence
of two of these sites in the protonated compound, the reduced
inherent binding affinity of recognition surfaces, presenting
only two hydrogen bonds couples with the fact that these
sites are projecting into the bulky N-butyl groups and
prevents the effective docking of the imide substrate. It is
the combination of these factors that are responsible for the
substantially reduced efficacy of the protonated receptor.
The addition of excess N-butyluracil to receptor 2‚H+ did
not result in any increase in the extent of the changes in the
1H NMR spectra as was seen when 1‚H+ was treated with
Acknowledgment. This work was supported by the
Natural Sciences and Engineering Research Council of
Canada and by the University of Alberta.
Supporting Information Available: Experimental pro-
1
cedures for the preparation of compounds 1 and 2 and H
NMR spectra for 1, (1‚H+), 2, and (2‚H+). This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0170915
(11) Haushalter, A.; Lau, J.; Roberts, J. D. J. Am. Chem. Soc. 1996,
118, 8891.
884
Org. Lett., Vol. 4, No. 6, 2002