Hydrophobic and hydrophilic yoctowells as receptors in water

Article abstract of DOI:10.1021/ja053816e

Hydrophilic yoctowells (volume = 10^-24 L) with OEG walls are introduced, which entrap tobramycin, a tetraamino trisaccharide, in water with a binding constant of 10⁷ M^-1 but do not interact with cellobiose. This is in cont

Full text of DOI:10.1021/ja053816e

Published on Web 02/02/2006
Hydrophobic and Hydrophilic Yoctowells as Receptors in Water
Sidhanath Bhosale, Sheshanath Bhosale, Tianyu Wang, Marta Kopaczynska, and
Jurgen-Hinrich Fuhrhop*
Institut fu¨r Chemie/Organische Chemie, Takustr. 3, D-14195 Berlin, Germany
Received June 10, 2005; E-mail: fuhrhop@chemie.fu-berlin.de
Hydrophobic cavities in molecular monolayers on aminated silica
particles (Figure 1a) were prepared earlier by a two-step self-
assembly of a porphyrin, 1a, and a diamido bolaamphiphile, 2.1,2
These form-stable gaps were designated as “yoctowells” for their
volume of a few yoctoliters (yL, 10-24 L).3 Yoctowells are useful
as model systems for the binding of small molecules to proteins in
water since their size and inner surface area are similar and can be
manipulated. The most characteristic property lies in their power
to induce the formation of well-filling “nanocrystals” in dilute (10-1
M) aqueous solutions of cyclic and rigid edge amphiphiles, such
as cellobiose or tyrosine,3,4 as well as to allow sorting of fitting
molecules.2
The hydrophobic yoctowells act, in general, as size- and
stereoselective kinetic traps for various solutes in water and provide
unique means to study water-soluble molecules in confined systems.
To check on our hypothesis of “hydrophobic kinetic trapping” and
also to apply the yoctowells as 3D crown ethers5 for single polar
molecules, we replaced the hydrophobic oligomethylene walls made
of 2 with functional tetraethyleneglycol (TEG, 3; Figure 1b) or
diglycinyl triamide (4; Figure 4) walls. The same two-step self-
assembly of flat-lying single molecules of porphyrin 1a first,
followed by upright-standing bolas 3 or 4, was applied to aminated
silica particles6 as described earlier.1-4 The form-stability of the
new wells was proven by size-selective fluorescence quenching
experiments; the too large manganese porphyrinate 1d did not
quench the fluorescence of the bottom porphyrin, whereas the fitting
manganese porphyrins 1b,c did so quantitatively.
The half-time of the fluorescence quenching with fitting metal-
loporphyrins 1b,c was around 1000 s in the hydrophobic wells and
dropped to 10-50 s with the more hydrophilic TEG walls (Figure
2a). Hydration water on the TEG walls obviously keeps adsorbed
porphyrins much more mobile than water on hydrophobic walls.
Furthermore, and more important, the hydrated TEG 3 wall also
totally prevented the formation of cellobiose nanocrystals after 3
days. Fluorescence quenching quinones reached the bottom por-
phyrins immediately after addition, whereas the hydrophobic 2 wells
were clogged up by nanocrystals formed under the same conditions
(Figure 2b).The new hydrophilic TEG yoctowells can be seen as
three-dimensional crown ethers and should tightly bind to oligoam-
ines. We added the flexible oligoamines spermine, polylysine (Mwav
) 300 000), and the rigid tricyclic tetraamine tobramycin 5 in water
at pH 7-8 or, in few cases, at pH 9-10. No differences were
observed. Above pH 10-11, the silica particles lost their smooth-
ness rapidly, and the wells were not form-stable any more. The
intact particles at pH 7-8 were then centrifuged, re-dispersed in
water, and titrated with a 10-3 M solution of naphthoquinone
2-sulfonate. Spermine blocked the well partially in water and much
more efficiently in ethanol at concentrations of ∼3 × 10-3 M-1
and pH 7-8. This corresponds to an equilibrium constant K of
about 103 M-1. Most of the spermine molecules floated in the
solvent volume. This disappointingly weak binding was probably
Figure 1. (a) Hydrophobic yoctowells are made of oligomethylene chains
in the center, terminal secondary amides, and two different headgroups for
fixation on aminated silica particles and solubilization. (b) Hydrophilic
yoctowells contain a hydrophilic triethyleneglycol (TEG) chain in the center
between the secondary amide groups.
caused by coiling in water, which buries some amino groups inside
the molecule. Solutions of polylysine (Mw 300 000) or tobramycin
5 were more efficient. A concentration of 1.2 × 10-6 M with respect
to lysine monomers or tobramycin was sufficient to block the
passage of the quinone into the TEG bola (3) yoctowells quanti-
tatively. One molecule of polylysine blocked about 20 yoctowells.
The rest presumably stretched over the TEG surface of the particles.
We could not determine how many tobramycin molecules were on
average entrapped within the wells. Assuming the case of a 1:1
complex between 5 and the yoctowells made of 3 (Figure 3), the
minimal binding constant is K ) 107 M-1. In the hydrophobic bola
(2) wells (Figure 1a), containing only flexible TEG headgroups on
the outer surface, neither spermine nor polylysine or tobramycin
had any measurable blocking effect in water at the concentrations
9
2156
J. AM. CHEM. SOC. 2006, 128, 2156-2157
10.1021/ja053816e CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
Figure 2. (a) Time course of the fluorescence quenching of bottom
porphyrin 1a by a fitting Mn(III) porphyrin TPPS 1b in hydrophilic (3,
blue) and hydrophobic (2, green) yoctowells. 1b reaches bottom 1a about
100 times faster in the hydrophilic wells. (b) Quenching of the fluorescence
with naphthoquinone-3-sulfonate after treatment with a 0.1 M cellobiose
in the same yoctowells. Cellobiose has no blocking effect in the polar wells
(fluorescence totally quenched, blue), whereas the hydrophobic wells are
blocked (green).
Figure 4. Fluorescence spectra (excitation at 420 nm) of centrifuged and
resuspended silica particles in ethanol (light blue) and ultrafiltered and
resuspended particles in water (dark blue), and a model of the yoctowell
entrapping two fluorescein triglycine (6) molecules.
the “thermodynamic entrapment” in hydrophilic wells, which occurs
in seconds. Ammonium-TEG and amide-amide hydrogen bonding
are probably the major binding forces. A comparison of -NH2-
and -NH3+- binding was not possible because the silica particles
did not retain their smoothness at pH values above 10-11.
Figure 3. Quenching of the fluorescence of the bottom porphyrin 1a in
hydrophilic TEG yoctowells (the blank spectrum is in gray) made of 3 after
addition of p-naphtho-1,4-quinone 2-sulfonate in the absence (orange) and
presence (blue) of 1.2 × 10-6 M tobramycin 5 in water.
which were used with the hydrophilic well (3). Only spermine in
ethanol had a small blocking effect (30% in 2, 90% in 3).We finally
tested for amide hydrogen bonding in yoctowells applying the
triglycine TEG bola 4 as wall material and the triglycine derivative
6 of fluorescein as a guest molecule. This arrangement allowed,
for the first time, a counting of the number of entrapped molecules
in each well by fluorescence comparisons of the bottom porphyrin
1a and the guest 6. The particles were dissolved in water or ethanol,
treated with 10-6 M of triglycinyl fluorescein 6, and ultrafiltered
in the case of an aqueous solution or centrifuged in the case of
ethanol solutions.2 The fluorescence spectra were obtained with an
excitation wavelength of either 420 nm (Soret band of the
porphyrin) or 495 nm (fluorescein). Comparisons consistently
pointed to a fluorescein:porphyrin molecular ratio of 2:1 or two
fluorescein-triglycinyl 6 molecules within each well corresponding
to a binding constant K > 1013 M-2 (Figure 4).
Acknowledgment. Financial support by the European TMR
research network “Carbohydrate Recognition of Nucleic Acids
(Carbona)”, by the Deutsche Forschungsgemeinschaft (SFB 348
“Mesoscopic Systems”, Graduiertenkolleg “Hydrogen Bonds”), the
Fonds der Deutschen Chemischen Industrie, and by the FNK of
the Free University is gratefully acknowledged.
Supporting Information Available: Syntheses of 3-4 and sper-
mine and polylysine titrations. This material is available free of charge
via the Internet at http://pubs.acs.org.
References
(1) Li, G.; Bhosale, S. V.; Wang, T.; Hackbarth, S.; Roeder, B.; Siggel, U.;
Fuhrhop, J.-H. J. Am. Chem. Soc. 2003, 125, 10693-10702.
(2) Bhosale, Sh.; Bhosale, S.; Wang, T.; Li, G.; Siggel, U.; Fuhrhop, J.-H. J.
Am. Chem. Soc. 2004, 126, 13111-13118.
(3) Bhosale, Sh.; Li, G.; Li, F.; Wang, T.; Ludwig, R.; Emmler, T.;
Buntkowsky, G.; Fuhrhop, J.-H. Chem. Commun. 2005, 28, 3559-3561.
(4) Li, G.; Doblhofer, K.; Fuhrhop, J.-H. Angew. Chem. 2002, 114, 2855-
2859; Angew. Chem., Int. Ed. 2002, 41, 2730-2734.
The form-stability of the yoctowells in water is thus guaranteed
by the two parallel amide hydrogen bond chains, even if the walls
made of 3 are strongly hydrated or contain the additional amide
bonds of 4. The slow “kinetic entrapment” in hydrophobic wells,
taking at least 8 h to be complete, is a very different process from
(5) Cram, D. J.; Cram J. M. In Container Molecules and Their Guests;
Stoddart, J. F., Ed.; RSC: Cambridge, 1994.
(6) van Blaaderen, A.; Vrij, J. J. Colloid Interface Sci. 1993, 156, 1-12.
JA053816E
9
J. AM. CHEM. SOC. VOL. 128, NO. 7, 2006 2157