A designed receptor for pH-switchable ion binding in water

Article abstract of DOI:10.1021/ja058126p

Molecules with conditional (switchable) properties are of considerable fundamental interest and are potentially useful for a broad range of applications, including chemical sensing. We have prepared a novel receptor, derived from the peracylation of cyclo

Full text of DOI:10.1021/ja058126p

Published on Web 02/02/2006
A Designed Receptor for pH-Switchable Ion Binding in Water
Stephen G. Tajc† and Benjamin L. Miller*,†,‡
Department of Biochemistry and Biophysics, Department of Dermatology, and the Center for Future Health,
UniVersity of Rochester, Rochester, New York 14642
Received December 20, 2005; E-mail: benjamin_miller@futurehealth.rochester.edu
The design and synthesis of molecules with conditional or
“switchable” properties is of considerable practical and fundamental
interest. Fundamentally, such systems serve as models of Nature’s
strategies for regulating cellular processes.1 Potential applications
of “molecular switches” include drug delivery systems, optical
devices, and sensors. While the majority of studies on switchable
receptors have focused on photochemical switching, changes in
other environmental properties, including temperature and pH, are
also known. These include cyclodextrin-conjugated poly(ꢀ-lysine)
inclusion complexes,2 ion recognition-mediated rotaxane switching,3
pH-switchable polymeric capsules derived from calixarenes,4 and
pH-switchable ligands for transition metal complexation.5
Figure 1. H-bonded closed conformation of 2 predicted by molecular
mechanics.
Figure 2. Extracted rows (δt2 ) 6.8 ppm) from 2D 1H-1H NOESY spectra
of 2 at pH* 9.50 (100 mM bicarbonate/carbonate buffer) and 2 at pH*
3.17 in D2O.
In the context of our work on cycloalkane-based receptors for
carbohydrates and other metabolites,6 we began exploring the
structures and binding properties of derivatives of cyclohexane
1,3,5-cis-trimethanol (1). In particular, we were intrigued by
molecular mechanics conformational searches7 of trityrosine deriva-
tive 2, which suggested that this receptor could access a low-energy
“closed” conformation in its neutral state, mediated by hydrogen
bonding between tyrosine -OH and amino groups (Figure 1). In
contrast, 2 would be expected to exist as a complex mixture of
“open” conformers as either a cation or an anion. Literature values8
for the tyrosine amino and phenolic -OH pKa’s are 9.21 and 10.46,
respectively. Therefore, we anticipated that 2 might serve as a pH-
switchable receptor, adopting the closed conformation only in the
range of 9.2 < pH < 10.5. While phenol-amine H-bonded
complexes are well-known,9 they have been characterized almost
exclusively in the solid state. Therefore, we set out to test whether
the predicted conformation-switching behavior of 2 could indeed
be observed in solution, and if it translated to pH-dependent
differences in ion binding ability.
capable of achieving the same H-bonded closed structure proposed
for 2, it was also synthesized as a control. With 2 and 3 in hand,
we next set out to examine their solution conformational behavior.
We anticipated that 2D NOE (NOESY) spectroscopy would be
particularly valuable for examining pH-dependent changes in the
solution conformation of 2. Indeed, NOESY spectra of 2 in D2O
at pH* ) 3.17 and in a sodium bicarbonate/carbonate buffer (100
mM in D2O; pH* ) 9.40) showed significant differences in the
number, intensity, and sign of proton NOEs. In particular, NOESY
cross-peaks between aromatic and cyclohexyl protons are visible
in the pH* 9.4 spectrum, and not in the pH* 3.17 spectrum. The
change in sign of NOESY cross-peaks is also noteworthy. The
intensity and sign of the NOE is a function of ωτc, where ω is the
magnetic field strength and τc is the correlation time. Since “small”
organic molecules tend to have values of τc that place them near
the zero-crossing point for the theoretical NOE curve, their
enhancements can be positive, zero, or negative.10 In this case, the
sign change from negative NOE (higher τc) at pH* 3.17 to positive
(lower τc) is consistent with a conformational change from extended
to compact. Differences in the solution conformational behavior
of compounds are also reflected in their longitudinal or spin-lattice
relaxation (T1) times.11 T1 measurement of 2 at pH* ) 3.17 and
9.40 showed considerable differences in the relaxation rates of
several protons. These pH-dependent differences are most pro-
Synthesis of 2 was accomplished in a straightforward three-step
reduction/esterification/deprotection sequence from cyclohexane
1,3,5-tricarboxylic acid, with final purification by preparative re-
verse-phase HPLC. Since triphenylalanine derivative 3 would not be
† Department of Biochemistry and Biophysics.
‡ Department of Dermatology.
9
2532
J. AM. CHEM. SOC. 2006, 128, 2532-2533
10.1021/ja058126p CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Table 1. ITC Results for Ion Binding to 2 in AMPSO, pH 9.50
shift and line broadening of cyclohexyl protons, consistent with
ion binding occurring inside the cavity formed by the closed
conformation of 2.
1
ion
K (M-
)
∆H (cal)
∆S (cal)
fluoride
chloride
bromide
sulfate
(very weak)
490 ( 54
680 ( 92
0
In conclusion, we have employed NMR spectroscopy to dem-
onstrate that the cyclohexane-based receptor 2 undergoes a pH-
dependent conformational change, consistent with molecular me-
chanics-based predictions. UV-vis and calorimetric titrations
confirm that 2 can act as a receptor for ions at high pH, but not at
low pH. In combination, these data suggest that the tyrosine OH-
amine H-bond is a useful structural element for the design of pH-
switchable receptors. Compounds such as 2 may also serve as useful
models for naturally occurring phenol-amine H-bonds. Hydrogen
bonds to tyrosine hydroxyl protons are common structural motifs
in proteins and often manifest as changes in pKa. RNAse A is a
prototypical example, with several buried/H-bonded tyrosine resi-
dues exhibiting pKa’s > 11.5.14 Since new methods for character-
izing such interactions will be particularly useful, an immediate
goal will be to extend our recently reported method for pKa
determination by ITC to tyrosines.15
870 ( 50
720 ( 44
15.2
33.8
nitrate
0
zinc(OTf)2
Cd(OTf)2
1410 ( 170
1900 ( 150
-2230 ( 96
-2880 ( 70
7.1
5.5
nounced for the pendant tyrosines, consistent with the majority of
the change in conformational bias occurring in this portion of 2. In
contrast, neither 3 nor tyrosine methyl ester showed pH-dependent
changes in T1 values, supporting the idea that changes in the
relaxation behavior of 2 are due to a conformational shift, and not
simply due to variations in dissolved ion concentrations or to the
change in ionization state.
We next examined the ability of 2 to function as a receptor for
ions.12 In addition to constituting a fundamentally interesting
problem in itself, we anticipated that ion binding would serve as
an efficient structural reporter for the conformation (open or closed)
of 2. Molecular modeling of the closed conformer suggested that
smaller ions would likely fit into the cavity formed by the
cyclohexane core and tyrosine side chains, while larger ions, such
as nitrate, would likely not fit as well.
An initial screen was conducted in 96-well plates, monitoring
changes in UV-vis absorbance as a function of added inorganic
salt. Several species (Ce4+, Ag+, and Sn2+) caused immediate
precipitation of 2 at pH 9.5. However, both Zn2+ and Cd2+ induced
changes in UV-vis absorbance without causing obvious precipita-
tion, warranting further exploration. We also examined anion
binding to 2 by UV-vis titration. Chloride and bromide anions,
present from the dissociation of the corresponding tetrabutylam-
monium salts, bind to 2 at pH 9.50 with association constants (Ka)
of 680 ( 146 (bromide) and 550 ( 112 (chloride). Binding to
tetrabutylammonium nitrate or fluoride was not observed. In
addition, there was no binding observed to 2 at pH 7.0, to 3 at pH
9.50, or to tyrosine methyl ester at pH 9.50.
Isothermal titration calorimetry (ITC) experiments confirmed that
both chloride and bromide bind to 2 at pH 9.5 (Table 1), with
association constants consistent with those determined by UV-
vis titration. While ITC showed some interaction between 2 and
fluoride, the small amount of heat evolved precluded extraction of
thermodynamic data. No interaction between 2 and nitrate or sulfate
was detectable at pH 9.5. As in the UV-vis experiments, no binding
of fluoride, chloride, or bromide was observed to 2 at pH < 9.5.
As the putative binding cavity of 2 is electron-rich, the observation
that Cl- and Br- bind was somewhat surprising. ITC thermograms
for bromide and chloride at pH 9.50 were endothermic, character-
ized by an unfavorable enthalpic term (∆H) and balanced by
positive entropy (∆S). This suggests that anion binding may be
driven not by specific interactions within the receptor cavity, but
rather by the disruption of solvent within the highly hydrophobic
structure. This is consistent with other authors’ analysis of entropy-
driven binding to hydrophobic receptors.13
Acknowledgment. The authors thank the NIH-NIGMS for
support of this work (5-R01-GM-062825). S.G.T. was supported
via an NIH training grant in dermatology (T32AR007472).
Supporting Information Available: Synthetic procedures and
characterization of compounds 2 and 3, ITC, UV and NMR titration
experiments, Job plots, NOESY spectra, and 1H T1 measurements (29
pages). This material is available free of charge via the Internet at http://
pubs.acs.org.
References
(1) Schnarr, N. A.; Kennan, A. J. Org. Lett. 2005, 7, 395.
(2) Choi, H. S.; Huh, K. M.; Ooya, T.; Yui, N. J. Am. Chem. Soc. 2003, 125,
6350.
(3) Vignon, S. A.; Jarrosson, T.; Iijima, T.; Tseng, H.-R.; Sanders, J. K. M.;
Stoddart, J. F. J. Am. Chem. Soc. 2004, 126, 9884.
(4) Xu, H.; Stampp, S. P.; Rudkevich, D. M. Org. Lett. 2003, 5, 4583.
(5) Gunnlaugsson, T.; Leonard, J. P.; Senechal, K.; Harte, A. J. J. Am. Chem.
Soc. 2003, 125, 12062.
(6) (a) Hubbard, R. D.; Horner, S. R.; Miller, B. L. J. Am. Chem. Soc. 2001,
123, 5810. (b) Hubbard, R. D.; Miller, B. L. Tetrahedron 2003, 59, 8143.
(7) Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton,
M.; Caufield, C.; Chang, G.; Hendrickson, T. Still, W. C. J. Comput.
Chem. 1990, 11, 440 (Maestro v. 4.1.012/Macromodel v. 7.2, Schrodinger,
Inc.).
(8) (a) Nozaki, Y.; Tanford, C. J. Am. Chem. Soc. 1967, 89, 736. (b) Dawson,
R. M. C.; Elliott, D. C.; Elliott, W. H.; Jones, K. M. Data for Biochemical
Research, 3rd ed.; Oxford Science Publications: New York, 1986; pp
1-31.
(9) (a) Barry, J. E.; Finkelstein, M.; Ross, S. J. Org. Chem. 1984, 49, 1669.
(b) Majerz, I.; Malarski, Z.; Sobczyk, L. Chem. Phys. Lett. 1997, 274,
361. (c) Fun, H.-K.; Usman, A.; Chantrapromma, S.; Osman, J.; Ong,
L.-H.; Tilley, D. R.; Ishibashi, Y. Solid State Commun. 2003, 127, 677.
(10) Neuhaus, D.; Williamson, M. The Nuclear OVerhauser Effect in Structural
and Conformational Analyais; VCH Publishers: New York, 1989; pp 30-
39.
(11) For selected examples, see: (a) Holzberg, A.; Holdt, H.-J.; Kleinpeter,
E. Org. Biomol. Chem. 2004, 2, 1691. (b) Bleich, H. E.; Cutnell, J. D.;
Day, A. R.; Freer, R. J.; Glasel, J. A.; McKelvy, J. F. Proc. Natl. Acad.
Sci. U.S.A. 1976, 73, 2589.
(12) For lead references on ion-binding macrocyclic organic receptors, see:
(a) Custelcean, R.; Delmau, L. H.; Moyer, B. A.; Sessler, J. L.; Cho, W.-
S.; Gross, D.; Bates, G. W.; Brooks, S. J.; Light, M. E.; Gale, P. A. Angew.
Chem., Int. Ed. 2005, 44, 2537. (b) Nabeshima, T.; Saiki, T.; Iwabuchi,
J.; Akine, S. J. Am. Chem. Soc. 2005, 127, 5507. (c) Izatt, R. M.; Pawlak,
K.; Bradshaw, J. S.; Bruening, R. L. Chem. ReV. 1995, 95, 2529. (d) Best,
M. D.; Tobey, S. L.; Anslyn, E. V. Coord. Chem. ReV. 2003, 240, 3. (e)
Arduini, A.; Casnati, A.; Pochini, A.; Ungaro, R. Curr. Opin. Chem. Biol.
1997, 1, 467-474.
In contrast, ITC measurements of cation binding to 2 at pH 9.5
showed exothermic binding, driven by favorable enthalpy. As with
anion binding, no cation binding was observed for 2 at pH < 9.5.
Confirming that none of our NMR or ion-binding results were
affected by potential receptor aggregation, the ITC of 2 into buffer
alone (both at pH 5.27 and 9.5) showed essentially that no heat
evolved. Job plot analyses of 2 with Zn2+ confirmed the 1:1 binding
model indicated by ITC data. Finally, an NMR titration of zinc
triflate into 2 at pH 9.5 showed Zn2+-dependent changes in chemical
(13) Leavitt, S.; Freire, E. Curr. Opin. Struct. Biol. 2001, 11, 560-566.
(14) (a) Tanford, C.; Hauenstein, J. D.; Rands, D. G. J. Am. Chem. Soc. 1955,
77, 6409. (b) Laurents, D. V.; Huyghues-Despointes, B. M. P.; Bruix,
M.; Thurlkill, R. L.; Schell, D.; Newsom, S.; Grimsley, G. R.; Shaw, K.
L.; Trevin˜o, S.; Rico, M.; Briggs, J. M.; Antosiewicz, J. M.; Scholtz, J.
M.; Pace, C. N. J. Mol. Biol. 2003, 325, 1077.
(15) Tajc, S. G.; Tolbert, B. S.; Basavappa, R.; Miller, B. L. J. Am. Chem.
Soc. 2004, 126, 10508.
JA058126P
9
J. AM. CHEM. SOC. VOL. 128, NO. 8, 2006 2533