Detection and mechanistic studies of multicomponent assembly by fluorescence resonance energy transfer

Article abstract of DOI:10.1021/ja994397m

The kinetics and thermodynamics of multicomponent assembly in organic solution are investigated using fluorescence resonance energy transfer (FRET). Calix[4]arenes functionalized with ureas on their wider rims dimerize in organic solution, via intermolecu

Full text of DOI:10.1021/ja994397m

7876
J. Am. Chem. Soc. 2000, 122, 7876-7882
Detection and Mechanistic Studies of Multicomponent Assembly by
Fluorescence Resonance Energy Transfer
Ronald K. Castellano,†,‡ Stephen L. Craig,† Colin Nuckolls,† and Julius Rebek, Jr.*,†
Contribution from the The Skaggs Institute for Chemical Biology and the Department of Chemistry,
The Scripps Research Institute, La Jolla, California 92037, and the Department of Chemistry,
Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
ReceiVed December 16, 1999
Abstract: The kinetics and thermodynamics of multicomponent assembly in organic solution are investigated
using fluorescence resonance energy transfer (FRET). Calix[4]arenes functionalized with ureas on their wider
rims dimerize in organic solution, via intermolecular hydrogen bonding, to form encapsulation complexes.
When one calixarene component is outfitted with a donor fluorophore and the other with an acceptor,
dimerization brings the chromophores within 20 Å, a distance suitable for efficient energy transfer to occur
between them. Excitation of the donor results in two colors of emitted light: one fluorescence band at the
donor emission wavelength, and a second at the acceptor emission wavelength signaling the noncovalent union
of three speciessdonor, acceptor, and small-molecule guest. By monitoring these wavelengths, assembly and
dissociation processes are observed in real time at nanomolar concentrations. Rate and association constants
for assembly processes are determined for the first time, and they reveal unexpected contributions from the
structure and concentration of the monomer. Finally, the combination of FRET and the molecular recognition
capabilities of the encapsulation complexes provides a sensitive and specific method for small-molecule sensing.
Introduction
tion has been determined. The encapsulation complexes have
been studied by NMR,1-4 circular dichroism (CD) spectroscopy,3b
electrospray mass spectrometry,5 and X-ray crystallography,2b
and of those only NMR and CD offer a glimpse of the reversible
solution dynamics of the assemblies. Even so, there is a limited
knowledge of the rates and equilibria of dimerization because
at millimolar concentrations the assembly process is too rapid5
and too complete6 to be fully quantified; only dissociation rates
have been measured.2c,e
We report here the use of fluorescence resonance energy
transfer (FRET) to investigate multicomponent, hydrogen-
bonded assemblies in organic solution. FRET7 is widely used
to study dynamic processes in biological systems, including
conformational changes8 and self-assembly,9 at low concentra-
tions and on rapid time scales. The sensitivity of the technique
Calixarenes functionalized with ureas on their wider rims
form discrete dimers in organic solution (Figure 1).1,2 These
capsular structures are held together by a seam of 16 hydrogen
bonds, and their interiors are large enough to accommodate
molecular guests. The assemblies are capable of a variety of
chemical tasks. For example, the presence of a suitable guest
is a necessary condition for assembly and the host capsule is
selective toward the size, shape, and chemical functionality of
the encapsulated guest.1 The direction of the hydrogen bond
seam imparts chirality into capsules formed from nonidentical
monomers, and chirality can be transferred from substituents
on the ureas to the capsule interior.3 The dimer is also the basis
of a platform for reversible polymeric materials that exhibit a
wide range of structural, physical, and dynamic properties.4
The dynamics and thermodynamics of dimerization are central
to the function of the assemblies, but little quantitative informa-
(4) (a) Castellano, R. K.; Rudkevich, D. M.; Rebek, J., Jr. Proc. Natl.
Acad. Sci. U.S.A. 1997, 94, 7132-7137. (b) Castellano, R. K.; Rebek, J.,
Jr. J. Am. Chem. Soc. 1998, 120, 3657-3663. (c) Castellano, R. K.;
Nuckolls, C.; Eichhorn, S. H.; Wood, M. R.; Lovinger, A. J.; Rebek, J., Jr.
Angew. Chem., Int. Ed. Engl. 1999, 38, 2603-2606. (d) Castellano, R. K.;
Nuckolls, C.; Rebek, J., Jr. Polym. News 2000, 25, 44-52. (e) Nuckolls,
C.; Castellano, R. K.; Rebek, J., Jr. Polym. Prepr. 2000, 41, 930-931. (f)
Castellano, R. K.; Clark, R.; Craig, S. L.; Nuckolls, C.; Rebek, J., Jr.,
submitted for publication.
(5) Schalley, C. A.; Castellano, R. K.; Brody, M. S.; Rudkevich, D. M.;
Siuzdak, G.; Rebek, J., Jr. J. Am. Chem. Soc. 1999, 121, 4568-4579.
(6) Calixarenes featuring aryl ureas have been shown to be exclusively
dimeric at the dilution limit of the 600 MHz NMR spectrometers72 µM
in CDCl3 (295K): Castellano, R. K.; Rebek, J., Jr., unpublished work.
(7) For reviews on FRET theory and applications, see: (a) Wu, P.; Brand,
L. Anal. Biochem. 1994, 218, 1-13. (b) Clegg, R. M. Curr. Opin.
Biotechnol. 1995, 6, 103-110. (c) Selvin, P. R. Methods Enzymol. 1995,
246, 300-334. (d) Yang, M.; Millar, D. P. Methods Enzymol. 1997, 278,
417-444. (e) Szo¨llo˜si, J.; Damjanovich, S.; Ma´tyus, L. Cytometry 1998,
34, 159-179. (f) Tsien, R. Y.; Miyawaki, A. Science 1998, 280, 1954-
1955.
† The Scripps Research Institute.
‡ Massachusetts Institute of Technology.
* To whom correspondence should be addressed.
(1) (a) Shimizu, K. D.; Rebek, J., Jr. Proc. Natl. Acad. Sci. U.S.A. 1995,
92, 12403-12407. (b) Hamann, B. C.; Shimizu, K. D.; Rebek, J., Jr. Angew.
Chem., Int. Ed. Engl. 1996, 35, 1326-1329. (c) Castellano, R. K.;
Rudkevich, D. M.; Rebek, J., Jr. J. Am. Chem. Soc. 1996, 41, 10002-
10003.
(2) (a) Mogck, O.; Bo¨hmer, V.; Vogt, W. Tetrahedron 1996, 52, 8489-
8496. (b) Mogck, O.; Paulus, E. F.; Bo¨hmer, V.; Thondorf, I.; Vogt, W.
Chem. Commun. 1996, 2533-2534. (c) Mogck, O.; Pons, M.; Bo¨hmer, V.;
Vogt, W. J. Am. Chem. Soc. 1997, 119, 5706-5712. (d) Bo¨hmer, V.;
Mogck, O.; Pons, M.; Paulus, E. F. Reversible Dimerization of Tetraureas
Derived From Calix[4]arenes. In NMR in Supramolecular Chemistry; M.
Pons, ed.; NATO ASI Ser., Ser. C, Vol. 526; Kluwer Academic Publish-
ers: Dordrecht, The Netherlands, 1999; pp 45-60. (e) Vysotsky, M. O.;
Thondorf, I.; Bo¨hmer, V. Angew. Chem., Int. Ed. Engl. 2000, 39, 1264-
1267.
(3) (a) Castellano, R. K.; Kim, B. H.; Rebek, J., Jr. J. Am. Chem. Soc.
1997, 119, 12671-12672. (b) Castellano, R. K.; Nuckolls, C.; Rebek, J.,
Jr. J. Am. Chem. Soc. 1999, 121, 11156-11163.
(8) For a recent example, see: Ha, T.; Zhuang, X.; Kim, H. D.; Orr, J.
W.; Williamson, J. R.; Chu, S. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 9077-
9082 and references therein.
10.1021/ja994397m CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/08/2000

Multicomponent Assembly by FRET
J. Am. Chem. Soc., Vol. 122, No. 33, 2000 7877
Figure 1. A donor chromophore (D) and an acceptor chromophore (A), in the presence of a guest, are brought within FRET distance upon the
dimeric assembly of calix[4]arene tetraureas. Some groups have been omitted for clarity. (alk ) C10H21).
allows detection of interacting species at nanomolar concentra-
tions. At these low concentrations, the individual (“free”)
components as well as the assemblies can be observed directly,
assembly/dissociation processes can be observed in real time,
and equilibria can be determined that are not accessible at the
millimolar concentrations that are required for NMR analysis.
The experimental design is described in Figure 1. Calixarene
monomers are tagged with donor (D) and acceptor (A) fluoro-
phores, and dimerization (A‚D) brings them within 20 Å of one
another.10 The fluorophores are chosen so that the emission
spectrum of D overlaps the absorption spectrum of A, and
nonradiative energy transfer occurs between them upon irradia-
tion (hν).11 Because A is also a fluorophore, the energy transfer
process can be monitored at both the emission maximum of D
(hν′) and that of A (hν′′). Energy transfer is extremely efficient,
so that a fluorescence signal at hν′′ indicates the noncovalent
union of three speciessdonor, acceptor, and guest.
Figure 2. The donor (coumarin 2) and the acceptor (coumarin 343)
dyes can be easily modified for connection to the calixarene scaffold.
We report here the kinetics and thermodynamics of calixarene
dimerization for several capsules in different solvents. The
mechanistic studies provide insight into the structure and
thermodynamics of monomers as well as the dimeric assemblies.
These experiments greatly expand the concentration range at
which assembly can be observed, and we find that calixarene
concentration has an unexpected effect on the dissociation rates
of the dimers. Finally, we show that the combination of
fluorescence and the molecular recognition properties of the
dimers holds promise for specific small-molecule sensing.
workers in their light-harvesting dendrimers.12 Donor-tagged
tetrasulfonyl urea 1D and tetraaryl urea 2D were prepared in
one step through the PyBOP-mediated coupling of modified
coumarin 343, 6, to the appropriate monoacid precursor, either
4 or 5 (Scheme 1). Through similar strategies, the aryl urea
ethylene amine derivative 8 was prepared and coupled directly
to the acid function of coumarin 2 (Scheme 2) to afford 2A.
Control Experiments. The fluorescence spectra of simple
tert-butyl control compounds 9D and 9A (Figure 3) confirm
that their absorption and emission maxima in chloroform and
benzene are essentially identical to the parent coumarin
chromophores12c and the tetraurea derivatives (Figure 2)sthe
amide connectivity does not significantly change the electronic
properties of the dyes.13 Dilution studies performed with
equimolar mixtures of 9D and 9A establish the concentration
range wherein energy transfer no longer occurs between the free
dyes in solution (Figure 4a). Considerable FRET is observed
at 5 µM in chloroform, evidenced by the pronounced acceptor
Results and Discussion
Preparation of Dye-Labeled Calixarenes. For these studies,
modified coumarin laser dyesscoumarin 2 as the donor and
coumarin 343 as the acceptor (Figure 2)swere connected to
the appropriate calixarene platforms via amide linkages using
the synthetic procedures outlined in Schemes 1 and 2. This
FRET pair has recently been employed by Fre´chet and co-
(9) For a recent example, see: Gelfand, C. A.; Plum, G. E.; Mielewczyk,
S.; Remeta, D. P.; Breslauer, K. J. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
6113-6118.
(10) Within the presumed Fo¨rster distance of the FRET dye pair (typically
<80 Å). See ref 7a.
(11) For energy transfer in other neutral, hydrogen-bonded assemblies,
see: (a) Ward, M. D. Chem. Soc. ReV. 1997, 26, 365-375 and references
therein. (b) Springs, S. L.; Gosztola, D.; Wasielewski, M. R.; Kra´l, V.;
Andrievsky, A.; Sessler, J. L. J. Am. Chem. Soc. 1999, 121, 2281-2289
and references therein.
(12) (a) Gilat, S. L.; Adronov, A.; Fre´chet, J. M. J. Angew. Chem., Int.
Ed. Engl. 1999, 38, 1422-1427. (b) Gilat, S. L.; Adronov, A.; Fre´chet, J.
M. J. J. Org. Chem. 1999, 64, 7474-7484. (c) Adronov, A.; Gilat, S. L.;
Fre´chet, J. M. J.; Ohta, K.; Neuwahl, F. V. R.; Fleming, G. R. J. Am. Chem.
Soc. 2000, 122, 1175-1185.
(13) N-alkylation of coumarin 2 does cause a decrease in molar
absorptivity. See ref 12c for details.

7878 J. Am. Chem. Soc., Vol. 122, No. 33, 2000
Castellano et al.
Scheme 1. Synthesis of the Donor-Tagged Tetraureas 1D and 2D (alk ) C10H21)
Scheme 2. Synthesis of the Acceptor-Tagged Aryl Urea 2A (alk ) C10H21)
emission signal at 465 nm. This band essentially disappears upon
10-fold dilution and only a small shoulder remainssirradiation
of the donor at these concentrations gives only donor emission.14
In related experiments, the donor control 9D is combined with
the aryl urea acceptor 2A, 1:1 in chloroform (Figure 4b). At
the three highest concentrations investigated, 2A is dimeric, and
likely not aggregating with the tert-butyl donor. The results of
the experiment again suggest that the two coumarin dyes form
an extremely efficient FRET pair. These dilution experiments
have important consequences for the assembly of donor- and
acceptor-labeled calixarene tetraureassan appreciable FRET
signal arising from their combination at concentrations below
∼500 nM must indicate an aggregation event (“intra-assembly”
FRET). Moreover, at higher concentrations some contribution
from “inter-assembly” FRET can be expected.
“Intra-Assembly” Energy Transfer. Calixarenes 1D and
2A, featuring sulfonyl ureas and aryl ureas, respectively, have
been shown to form exclusively heterodimeric pairs in
solution,3a,4b so that the combination of these two species was
expected to give a large excess of the desired FRET complex.
Figure 5a shows the change in the emission spectrum with time
upon mixing 100 nM each of 1D and 2A in chloroform. Initially,
the donor and acceptor are isolated from each other in solution,
and only emission from the donor is observed. Over the course
of hours, the heterodimer 1D‚2A forms, energy transfer occurs,
and the acceptor emission increases until essentially quantitative
energy transfer is observed. That the FRET signal depends on
assembly is evident from (a) the slow onset; energy transfer
between free monomers would be observed immediately,
(b) the concentration; energy transfer between the “free” dyes
is precluded based on the control dilution experiments, and (c)
the insensitivity of the emission intensity to added (excess)
acceptor.
Finally, by titrating the dimer solution with a solvent that
can effectively compete for the hydrogen bonds, such as
dimethyl sulfoxide (DMSO), this process can be reversed,
restoring the donor emission (Figure 5b).15 Unlike the associa-
tion processes, which take minutes to hours to reach equilibrium,
“denaturation” is complete within seconds of adding DMSO.
The spectra pass through the same isosbestic point in both the
assembly and dissociation sequences, underscoring the revers-
ibility of the process.16
(14) Throughout all of these experiments ca. 10% of the acceptor
emission signal arises from its direct excitation.
(15) Control experiments determined the inherent sensitivities of the dyes
to DMSO. The emission intensity of both the donor and acceptor controls
gradually increased upon titration (at 1% DMSO, 14% increase for the
donor, 10% increase for the acceptor). See the Supporting Information for
details.
Figure 3. Various tagged and untagged calixarenes used in the control
experiments. (alk ) C10H21).

Multicomponent Assembly by FRET
J. Am. Chem. Soc., Vol. 122, No. 33, 2000 7879
Figure 4. Dilution experiments in chloroform establish the concentration threshold for FRET between the “free” dyes in solution: (a) with 1:1
mixtures of 9D and 9A; (b) with 1:1 mixtures of 9D and 2A.
Figure 5. (a) FRET occurs upon combination of 1D and 2A at 100 nM in chloroform and the emission of the acceptor grows in as 1D‚2A formssa
process that takes hours. (b) Added DMSO instantaneously reverses the assembly of 1D‚2A in CHCl3, as indicated by restoration of the donor
emission signal.
Rate and Equilibrium Constants. The association rate
constant for assembly of 1D and 2A, kass, was measured by
using an excess of 1D and simultaneously monitoring the donor
and acceptor emission intensities (Figure 6). The observed
pseudo-first-order kinetics yield an apparent rate constant for
dimer formation (Figure 6, inset). The apparent rate constant
reflects not only the desired second-order rate constant, but also
the concentrations of free monomers 2A and 1D in solution.
Accurate association rate constants were obtained from studies
at low concentrations (10 nM) of 2A where essentially all of
the calixarene exists as free monomer. Rate constants for
association in both chloroform and benzene are reported in Table
1.
Dissociation rates were obtained from two similar experi-
ments. The dissociation rate constant (kdiss) for 2‚2, for example,
came from measuring either the appearance of the FRET signal
upon mixing 2A‚2A and 2D‚2D or the disappearance of the
FRET signal upon mixing 2A‚2D (in equilibrium with 2A‚2A
and 2D‚2D) with untagged 2U‚2U (Figure 3). The dissociation
constant of 1‚2 was obtained from the disappearance of FRET
upon mixing 1D‚2A with an excess of untagged 1U. The
Figure 6. Change in fluorescence intensity with time of the donor
emission maximum (415 nm, squares) and acceptor intensity maximum
(455 nm, circles) upon combination of 2A (100 nM) and 1D (250 nM)
in benzene at 25 °C. Inset: Pseudo-first-order kinetic treatment of the
data.
association and dissociation rate constants for 1D‚2A were
combined to give the equilibration association constant KA for
the assembly.
(16) For DMSO denaturation experiments monitored by FRET in a
biological context, see: Garzon-Rodriguez, W.; Sepulveda-Becerra, M.;
Milton, S.; Glabe, C. G. J. Biol. Chem. 1997, 272, 21037-21044.

7880 J. Am. Chem. Soc., Vol. 122, No. 33, 2000
Castellano et al.
give a value of ∼0.03 s-1 for kdiss of 2U‚2U (-C10 chains only
on the lower rim) at 4 mMsan order of magnitude slower than
previously reported. Second, the amide linker in 2A and 2D
imparts additional stability by hydrogen bonding to adjacent
lower-rim oxygens. This is confirmed by the rate of guest
benzene exchange in the dimer 8‚8: by NMR we measured a
rate constant of 0.003 s-1, which is another order of magnitude
slower than that for the analogous dimer 2U‚2U, and closer to
the value measured by FRET. Finally, the concentration of 2
plays a role. The dissociation rate, measured by FRET kinetics,
of 2A‚2D varies with the concentration of added 2U, increasing
almost an order of magnitude from 100 nM (6 × 10-4 s-1) to
1 mM (3 × 10-3 s-1), where the rate is equivalent to the NMR
results on 8‚8. At millimolar concentrations, free 2 is apparently
present in sufficient concentration to accelerate the dissociation
of the dimer.20 The combination of these three factors explains
the disparity between the dynamic NMR measurements and
those in the present study.
The association and dissociation rates in benzene reveal
unique behavior for the sulfonyl urea 1. Although its homodimer
is the least stable of the three capsules investigated, it dissociates
more slowly than aryl urea homodimer 2‚2. Moreover, it is very
slow to assemble; its kass is more than 2 orders of magnitude
lower than that of the aryl urea and 6 orders of magnitude lower
than the diffusion rate (∼109 M-1 s-1). These data point to a
stable sulfonyl urea monomer. Reinhoudt and co-workers were
the first to show that calixarene 1,3-bisureas could stabilize the
pinched cone conformation through formation of strong in-
tramolecular hydrogen bonds.18 Modeling of the sulfonyl
tetraurea monomer reveals that intramolecular hydrogen bonding
is likely between neighboring ureas through participation of the
sulfonyl urea oxygens.
Small-Molecule Sensing. The assembly process shown in
Figure 1 only occurs in the presence of a suitable guest molecule,
and the solvent itself fills that role in the work described above.
When the solvent is not a viable guest, then the FRET system
may be a sensitive method for small-molecule detection in which
the analytes remain chemically unmodified.21 We have designed
such a system using heterodimers of calixarenes functionalized
with aryl ureas 2D and amino acid-derived ureas 3A.4 These
components form a strongly associating dimer in chloroform
(KA ≈ 2 × 107 M-1), but in p-xylene both NMR and FRET
studies indicate no assembly. The presence of a good guest
should trigger capsule formation and a concomitant FRET
signal. The results, shown in Figure 7, are promising; in p-xylene
energy transfer is observed only in the presence of added guest,
in this case 3-methylcyclopentanone. When 2D and 3A are
present in 150 nM concentrations, the threshold concentration
for energy transfer is ca. 0.1 mM 3-methylcyclopentanone and
assembly is 50% complete at ca. 4 mM added guest. The
Table 1. Rate and Association Constants for the Assembly and
Dissociation of Calixarene Tetraureas in Chloroform and Benzene
as Measured by FRET at Concentrations of 0.05-0.5 µM
a
a
a
solvent
kass (M-1 s-1
)
kdiss (s-1
)
KA (M-1
)
b
b
b
1‚1
1‚2
2‚2
1‚1
1‚2
2‚2
CHCl3
CHCl3
CHCl3
C6H6
C6H6
C6H6
-
-
-
2.7b × 103
1.5 × 104
<50
2.3 × 10-5
6.4 × 10-3
5.4 × 10-5
5 × 10-6
1.2 × 108
2.4 × 106
<106
3 × 103
6 × 108
2.4 × 104
6 × 10-4
4 × 107
a Uncertainties in kdiss are (15%, and in kass and KA are (40%.
b Compound 1 does not appear to form a well-defined dimer in CHCl3
by NMR,3 and its behavior in this work is consistent with a species
that is effectively monomeric at these concentrations.
The association constants of 2‚2 were derived from the
concentration dependencies of heterodimerization rates. It was
assumed that the rate at which 1D is converted to 1D‚2A is
proportional to the concentration of free monomer 2A in
solution. When 1D was mixed with 2A, therefore, the apparent
pseudo-first-order rate constant was divided by the value of kass
from the low-concentration studies to give the concentration of
free 2A. The concentration of free 2A and calculated concentra-
tion of dimeric 2A‚2A ([total 2A - free 2A]/2) were converted
into an association constant. Experiments across a range of
concentrations and stoichiometries were internally consistent and
agreed with approximate association constants derived from the
amount of FRET observed as a function of concentration of
2A‚2D. The combination of KA and kdiss was used to calculate
kass, and a range of experiments gave the assembly parameters
in chloroform and benzene that are reported in Table 1.
The measured association constants (KA values) are largely
consistent with previous observations. The KA for 2‚2 in
chloroform, 2.4 × 106 M-1, is higher than the lower limit set
by NMR6 and in excellent agreement with recent quasi-elastic
light scattering results.17 The KA of 1‚2 is ∼50× greater than
that of 2‚2, which agrees qualitatively with the preference for
heterodimerization observed by NMR.3a,4b In all cases, the KA
values are higher and dissociation rates are slower in benzene
than in the more polar chloroform, and the kinetic and
equilibrium measurements are self-consistent.
The dissociation constants, however, are much slower than
expected. Our kdiss value for the aryl urea homodimer 2‚2 in
benzene (6 × 10-4 s-1) is almost 3 orders of magnitude slower
than that found by Bo¨hmer and co-workers using dynamic NMR
techniques (0.26 s-1) for a structurally similar calixarene dimer.2c
We attribute the dramatic difference in rates to the additive
contribution of three factors: steric congestion at the calixarene
lower rim, hydrogen bonding at the lower rim, and the difference
in concentration between the studies.
First, the solubilizing groups on the calixarene lower rim that
are used in the present work (-C10 alkyl chains) are more bulky
than the methyl and -C5 chains used in the previous dynamic
NMR study. These chains may inhibit the calixarene monomer
from forming a “pinched cone”18 as it dissociates, due to steric
congestion on the lower rim. This conjecture is supported by
the recent observation of very slow dissociation dynamics in
tetraurea calixarene dimers that are structurally prevented from
forming the pinched cone.19 Our own conventional NMR kinetic
measurements of heterodimer formation and benzene exchange
association constant for the ternary complex is ∼3 × 109 M-2
,
which is considerably higher than that of the assembly in
benzene (∼106 M-2)sconsistent with our observation that
3-methylcyclopentanone is a good guest for the capsule interior.
The sensitivity and specificity of the fluorescence signal is
advantageous for more complex sensing schemes based on
(20) Even low concentrations of hydrogen-bonding groups have a
dramatic effect on the rates of the assembly processes, e.g., dissociation of
1D‚2A in the presence of only 0.3% DMSO occurs in seconds vs hours.
(21) For a review of fluorescence-based sensing, see: Prasanna de Silva,
A.; Nimal Gunaratne, H. Q.; Gunnlaugsson, T.; Huxley, A. J. M.; McCoy,
C. P.; Rademacher, J. T.; Rice, T. E. Chem. ReV. 1997, 97, 1515-1566.
See also (a) Song, X.; Nolan, J.; Swanson, B. I. J. Am. Chem. Soc. 1998,
120, 11514-11515. (b) Chen, C.-T.; Wagner, H.; Still, W. C. Science 1998,
279, 851-853. (c) Rothman, J. H.; Still, W. C. Bioorg. Med. Chem. Lett.
1999, 9, 509-512.
(17) Lomakin, A.; Benedek, G. B.; Castellano, R. K.; Nuckolls, C.;
Rebek, J., Jr., unpublished work.
(18) Scheerder, J.; Vreekamp, R. H.; Engbersen, J. F. J.; Verboom, W.;
van Duynhoven, J. P. M.; Reinhoudt, D. N. J. Org. Chem. 1996, 61, 3476-
3481.
(19) Brody, M. S.; Rebek, J., Jr., unpublished work.

Multicomponent Assembly by FRET
J. Am. Chem. Soc., Vol. 122, No. 33, 2000 7881
further purification. All other reagents were purchased from either
Aldrich or Fluka and used as received. HCl (g) was bubbled through
H2SO4 prior to use. The preparations of tetraureas 1U,3a 2U,5 and 55
have been described in the literature.
General PyBOP Coupling Procedure. The calixarene monoacid
(1 equiv) is dissolved in dimethylformamide (DMF) and treated with
PyBOP (1.2 equiv) and NEt3 (2-10 equivs.). To this is then added the
amine (1-1.2 equiv). The coupling reaction is monitored by thin-layer
chromatography (TLC) and is generally complete within 1 h. The
solvent is removed in vacuo and the residue is dissolved in CH2Cl2.
After the organic phase is wased with 1 M HCl, 1 M NaOH (except
for 1D), and brine, it is dried over MgSO4, filtered, and evaporated to
give the crude product.
Preparation of Modified Coumarin 2 (6). To coumarin 2 (0.40 g,
1.8 mmol) was added CH3CN (20 mL), tert-butyl N-[4-(bromomethyl)-
benzyl]carbamate23 (0.37 g, 1.2 mmol), and K2CO3 (0.43 g, 3.1 mmol).
The mixture was heated to reflux for 36 h. After this time, the mixture
was filtered and the filtrate was concentrated to dryness. Addition of
MeOH resulted in the precipitation of remaining coumarin 2. This solid
was filtered off and the filtrate was concentrated and purified by SiO2
chromatography (gradient elution, CH2Cl2 to 30:1 CH2Cl2/MeOH). The
desired product was slightly less polar than the starting amine: yield
0.29 g (55%). 1H NMR (DMF-d7) δ 7.60 (s, 1H), 7.35 (d, 2H, J ) 8.0
Hz), 7.24 (d, 2H, J ) 8.0 Hz), 7.24 (m, 1H), 7.02 (s, 1H), 6.19 (s,
1H), 4.33 (s, 2H), 4.22 (d, 2H, J ) 6.2 Hz), 3.14 (q, 2H, J ) 6.9 Hz),
2.44 (s, 3H), 2.43 (s, 3H), 1.40 (s, 9H), 1.10 (t, 3H, J ) 6.9 Hz). 13C
NMR (DMF-d7) δ 160.80, 156.51, 154.07, 153.56, 152.88, 139.56,
137.65, 129.71, 128.42, 127.45, 127.36, 114.79, 112.25, 108.98, 78.12,
55.32, 47.31, 43.78, 28.17, 18.30, 17.97, 11.80. IR (thin film) 3349,
2975, 1712, 1613, 1512, 1390, 1267, 1248, 1170 cm-1. High-resolution
mass spectrometry (HRMS) (FAB; M+‚) calcd for C26H32N2O4: 436.2362.
Found 436.2368.
Figure 7. Addition of 3-methylcyclopentanone (equiv) to an equimolar
mixture of 2D and 3A in p-xylene (150 nM). A growing FRET signal
indicates assembly nucleation as more guest is added.
combinations of capsules and dyes. Current research efforts are
directed along these lines.
Conclusions
FRET techniques, although widely used in biology, have been
largely underutilized in molecular recognition studies using
synthetic receptors.19b,c They are particularly useful when
association constants are high and compound availability is low.
The energy transfer methodology is chemically benignsthe dyes
do not participate in binding, nor are they modified by the
assembly process; they act as noninvasive “reporters” of the
recognition event. The sensitivity of the technique has revealed
behavior and thermodynamic information not accessible at
higher concentrations, and its specificity promises greater use
in small-molecule sensing.
The protected amine (0.022 g, 0.050 mmol) was then dissolved in
dioxane (10 mL) and treated with HCl (g). After 30 min, the solution
was sparged with N2 and concentrated to dryness. The hygroscopic,
off-white powder obtained was used without further purification
(quant.). HRMS (FAB; M + H+) calcd for C21H25N2O2: 337.1916.
Found 337.1920.
Experimental Section
Sulfonyl Urea Donor (1D). The calixarene monoacid 424 (0.10 g,
0.13 mmol) was combined with the coumarin amine hydrochloride 6
(0.028 g, 0.16 mmol), PyBOP (0.036 g, 0.16 mmol), and NEt3 (80 µL,
1.3 mmol) in DMF (10 mL). The initial oily product obtained was
Fluorescence Measurements. Measurements were performed on an
SLM-AMINCO 8100 spectrofluorometer equipped with a 450 W xenon
lamp, an MC200 single-grating excitation monochromator (BP 4 nm),
and two single-grating MC200 emission monochromators (T-optics
geometry, BP 8 nm). Left and right polarizers were set to the magic
angle, 54.7°. Fluorescence is presented in arbitrary units calculated as
the ratio of observed emission intensity divided by an internal reference.
All measurements were performed at 25.0 ( 0.1 °C. The dyes used
showed no susceptibility to photobleaching. Spectrophotometric grade
chloroform (stabilized with amylenes) and benzene were purchased from
Aldrich and used as received.
1
triturated with MeOH to furnish a white powder (0.081 g, 69%). H
NMR (DMF-d7) δ 10.40 (s, 1H), 10.35 (s, 1H), 10.30 (s, 2H), 8.65 (s,
1H), 8.60 (s, 1H), 8.53 (s, 2H), 8.45 (t, 1H, J ) 6.0 Hz), 7.98-7.94
(m, 8H), 7.58 (s, 1H), 7.49-7.46 (m, 8H), 7.35 (d, 2H, J ) 8.0 Hz),
7.24 (d, 2H, J ) 8.0 Hz), 6.99 (s, 1H), 6.76 (s, 2H), 6.75 (s, 2H), 6.62
(s, 4H), 6.13 (s, 1H), 4.54 (m, 4H), 4.39 (d, 2H, J ) 13.4 Hz), 4.32 (s,
2H), 4.30 (d, 2H, J ) 13.1 Hz), 3.83-3.72 (m, 6H), 3.12 (q, 2H, J )
6.8 Hz), 3.05 (d, 2H, J ) 13.1 Hz), 3.04 (d, 2H, J ) 13.6 Hz), 2.47-
2.45 (m, 12H), 2.42 (s, 3H), 2.39 (s, 3H), 1.81 (m, 2H), 1.73 (m, 4H),
1.36-1.26 (m, 42H), 1.09 (t, 3H, J ) 7.0 Hz), 0.88-0.85 (m, 9H).
13C NMR (DMF-d7) δ 169.27, 160.74, 154.49, 153.97, 153.46, 152.87,
152.61, 152.31, 149.98, 144.38, 138.31, 138.21, 135.47, 135.19, 134.63,
134.55, 133.31, 132.85, 129.79, 129.76, 128.58, 128.16, 127.43, 127.18,
120.02, 119.92, 119.76, 114.87, 112.33, 109.09, 75.73, 75.18, 74.64,
55.40, 47.33, 42.21, 32.16, 31.77, 31.51, 31.25, 26.48, 26.30, 22.84,
21.01, 13.98, 11.77. IR (thin film) 3340, 2924, 2853, 1715, 1694, 1613,
1552, 1464, 1346, 1216, 1164, 1091, 1056, 895 cm-1. Low-resolution
mass spectra (LRMS) electron spray ionization ((ESI); M + H+) calcd
for C113H141N10O19S4: 2071. Found 2071.
Aryl Urea Donor (2D). The calixarene monoacid 55 (0.054 g, 0.036
mmol) was combined with the coumarin amine hydrochloride 6 (0.015
g, 0.036 mmol), PyBOP (0.023 g, 0.043 mmol), and NEt3 (30 µL, 0.22
mmol) in DMF (5 mL). The crude product was purified by chroma-
tography on SiO2 (15:1 CH2Cl2/MeOH) to give a white powder (0.032
g, 49%). 1H NMR (DMF-d7) δ 8.63 (t, 1H, J ) 6.1 Hz), 8.51 (s, 1H),
8.50 (s, 1H), 8.48 (s, 1H), 8.44 (s, 1H), 8.35 (s, 2H), 8.32 (s, 2H), 7.60
Representative Kinetics Measurement. Benzene solutions of 100
nM 1D and 250 nM 2A were equilibrated at 25 °C for 15 min. From
each solution 1.5 mL was added to a 1 cm quartz cuvette (Hellma).
An emission spectrum of the initial reactant mixture was taken. The
fluorescence values at the donor (415 nm) and acceptor (455 nm)
emission maxima were simultaneously recorded every 90 s using the
two emission monochromators. Fluorescence signals were integrated
for 10 s, and the excitation signal was closed between acquisitions.
An emission spectrum was recorded following the experiment.
1
General. H NMR (600 MHz) and 13C NMR (151 MHz) spectra
were recorded on a Bruker DRX-600 spectrometer. Infrared spectra
were recorded on a Perkin-Elmer Paragon 1000PC FT-IR spectrometer.
Matrix-assisted laser desorption/ionization (MALDI) FTMS experiments
were performed on an IonSpec FTMS mass spectrometer. The fast atom
bombardment (FAB) positive ion mass spectra were obtained on a VG
ZAB-VSE double-focusing high-resolution mass spectrometer. Dichlo-
romethane (CH2Cl2) was passed through columns of activated aluminum
oxide as described by Grubbs and co-workers prior to use.22 Coumarin
2 and coumarin 343 were purchased from Acros and used without
(22) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518-1520.
(23) Niu, J.; Lawrence, D. S. J. Biol. Chem. 1997, 272, 1493-1499.
(24) See the Supporting Information for synthetic details.

7882 J. Am. Chem. Soc., Vol. 122, No. 33, 2000
Castellano et al.
(s, 1H), 7.43-7.38 (m, 6H), 7.32-7.29 (m, 6H), 7.11 (d, 2H, J ) 8.5
Hz), 7.10 (d, 2H, J ) 8.5 Hz), 7.08-7.06 (m, 8H), 7.02 (s, 1H), 6.76
(s, 2H), 6.75 (s, 2H), 6.18 (s, 1H), 4.71 (s, 2H), 4.65 (d, 2H, J ) 6.0
Hz), 4.49 (d, 2H, J ) 13.4 Hz), 4.43 (d, 2H, J ) 13.1 Hz), 4.34 (s,
2H), 3.91 (t, 2H, J ) 7.5 Hz), 3.86-3.78 (m, 4H), 3.21-3.16 (m, 4H),
3.14 (q, 2H, J ) 7.0 Hz), 2.45 (s, 3H), 2.42 (s, 3H), 2.26 (s, 6H), 2.25
(s, 6H), 1.91 (m, 2H), 1.79 (m, 4H), 1.41-1.30 (m, 42H), 1.10 (t, 3H,
J ) 6.5 Hz), 0.90-0.87 (m, 9H). 13C NMR (DMF-d7) δ 169.84, 160.77,
154.00, 153.47, 153.29, 153.26, 153.22, 152.90, 152.40, 151.86, 151.20,
138.44, 138.33, 138.29, 138.25, 137.86, 135.82, 134.93, 134.87, 134.67,
134.61, 134.21, 130.99, 130.94, 130.76, 129.77, 129.46, 129.37, 128.68,
127.43, 127.12, 119.36, 119.16, 118.98, 118.91, 118.49, 118.47, 118.29,
114.90, 112.37, 109.15, 75.83, 75.12, 74.78, 55.51, 47.28, 42.25, 32.19,
31.86, 31.57, 30.41, 30.22, 30.14, 30.09, 29.99, 29.67, 26.54, 26.41,
22.87, 20.23, 20.21, 18.33, 17.99, 14.01, 11.78. IR (thin film) 3341,
2924, 2854, 1715, 1667, 1634, 1607, 1555, 1515, 1476, 1417, 1315,
Acceptor Aryl Urea (2A). The calixarene amine hydrochloride 8
(0.060 g, 0.039 mmol) was combined with coumarin 343 (0.011 g,
0.039 mmol), PyBOP (0.024 g, 0.047 mmol), and NEt3 (11 µL, 0.078
mmol) in DMF (5 mL). The crude product was purified by chroma-
tography on SiO2 (10:1 CH2Cl2/MeOH) to give a bright yellow powder
1
(0.046 g, 65%). H NMR (600 MHz, DMF-d7) δ 9.04 (t, 1H, J ) 5.8
Hz), 8.59 (s, 1H), 8.58 (m, 2H), 8.52 (s, 1H), 8.50 (s, 2H), 8.49 (s,
1H), 8.40 (t, 1H, J ) 5.4 Hz), 8.38 (s, 2H), 7.39 (d, 2H, J ) 8.1 Hz),
7.38 (d, 2H, J ) 8.2 Hz), 7.31 (d, 4H, J ) 8.3 Hz), 7.25 (s, 1H), 7.09
(d, 2H, J ) 7.9 Hz), 7.08 (d, 2H, J ) 8.0 Hz), 7.05 (d, 4H, J ) 8.3
Hz), 7.02 (m, 4H), 6.76 (m, 4H), 4.61 (s, 2H), 4.51 (d, 2H, J ) 13.5
Hz), 4.44 (d, 2H, J ) 13.1 Hz), 4.02 (m, 2H), 3.95 (m, 2H), 3.88 (m,
2H), 3.70 (m, 2H), 3.63 (m, 2H), 3.37 (m, 4H), 3.22 (d, 2H, J ) 13.7
Hz), 3.16 (d, 2H, J ) 13.3 Hz), 2.80 (m, 4H), 2.26 (s, 3H), 2.25 (s,
3H), 2.24 (s, 6H), 1.97-1.86 (m, 10H), 1.42-1.27 (m, 42H), 0.88-
0.86 (m, 9H). 13C NMR (DMF-d7) δ 170.01, 163.65, 153.23, 152.97,
152.39, 151.80, 151.26, 148.66, 147.98, 138.32, 138.26, 135.73, 135.05,
134.86, 134.67, 134.59, 134.51, 134.41, 130.91, 130.87, 130.74, 129.42,
129.36, 127.60, 120.19, 119.37, 119.20, 119.06, 118.42, 118.29, 108.76,
108.14, 105.39, 75.88, 75.16, 74.78, 50.15, 49.63, 39.76, 39.31, 32.18,
31.86, 31.61, 30.41, 27.42, 26.55, 26.46, 22.86, 21.19, 20.22, 14.00.
IR (thin film) 3339, 2924, 2853, 1666, 1606, 1552, 1515, 1476, 1311,
1210 cm-1. HRMS (FAB; M + Cs+) calcd for C110H137N11O12Cs:
1936.9503. Found: 1936.9618.
1212 cm-1. HRMS (MALDI-FTMS; M + Na+) calcd for C113H140
-
N10O11Na: 1836.0601. Found: 1836.0584.
tert-Butoxycarbonyl (BOC)-Protected Tetraaryl Urea (7). The
calixarene acid 5 (0.30 g, 0.20 mmol) was combined with mono-BOC
ethylenediamine (38 µL, 0.24 mmol), PyBOP (0.13 g, 0.24 mmol),
and NEt3 (56 mL, 0.40 mmol) in DMF (10 mL). The crude product,
following trituration with MeOH, was further purified by precipitation
from THF with MeOH to give a white powder (0.20 g, 61%). 1H NMR
(DMF-d7) δ 8.49 (s, 2H), 8.48 (s, 1H), 8.44 (s, 1H), 8.39 (t, 1H, J )
5.7 Hz), 8.31 (d, 4H, J ) 8.3 Hz), 7.39 (d, 4H, J ) 8.2 Hz), 7.29 (d,
4H, J ) 8.3 Hz), 7.11-7.06 (m, 12H), 6.98 (t, 1H, J ) 5.8 Hz), 6.76
(m, 2H), 6.75 (m, 2H), 4.62 (s, 2H), 4.51 (d, 2H, J ) 13.5 Hz), 4.45
(d, 2H, J ) 13.1 Hz), 4.03 (m, 2H), 3.95 (m, 2H), 3.90 (m, 2H), 3.51
(m, 2H), 3.30 (m, 2H), 3.25 (d, 2H, J ) 13.7 Hz), 3.18 (d, 2H, J )
13.3 Hz), 2.26 (s, 6H), 2.25 (s, 6H), 1.89 (m, 6H), 1.47-1.29 (m, 42H),
1.43 (s, 9H), 0.91-0.88 (m, 9H). 13C NMR (DMF-d7) δ 170.04, 156.47,
153.28, 153.24, 153.21, 152.65, 151.82, 151.15, 138.32, 138.28, 138.22,
135.91, 134.87, 134.84, 134.68, 134.60, 134.15, 130.98, 130.93, 130.74,
129.45, 129.35, 119.33, 119.13, 118.99, 118.95, 118.48, 118.27, 78.08,
75.95, 75.04, 74.82, 40.50, 39.51, 32.17, 32.15, 31.86, 31.63, 30.34,
30.18, 30.12, 30.11, 29.99, 29.94, 28.22, 26.50, 26.46, 22.85, 20.22,
20.20, 14.00. IR (thin film) 3351, 2924, 2854, 1715, 1667, 1607, 1552,
1514, 1476, 1316, 1212 cm-1. HRMS (MALDI-FTMS; M + Na+)
calcd for C99H132N10O11Na: 1659.9975. Found: 1659.9948.
Acknowledgment. We are grateful to the Skaggs Research
Foundation, the National Institutes of Health (NIH), and the
National Aeronautics and Space Administration (NASA) for
financial support. R.K.C. thanks the American Chemical Society,
Division of Organic Chemistry, and Schering-Plough for a
graduate research fellowship (1998-1999). S.L.C. and C.N.
thank the NIH for postdoctoral fellowships. We are indebted
to Professor David P. Millar (TSRI) and his staff for the use of
their equipment and for insightful discussions. We also thank
Professor Jamie R. Williamson (TSRI) for helpful advice.
Supporting Information Available: Supporting Informa-
tion Available: Full synthetic details including the characteriza-
tion of all precursors and final compounds including 3A, 4, 9D,
and 9A. Also DMSO titration experiments performed with the
control samples (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
Deprotection (8). The protected amine 7 (0.10 g, 0.061 mmol) was
dissolved in dioxane (15 mL) and treated with HCl (g). After 30 min,
the solution was sparged with N2 and concentrated to dryness. The
oily solid obtained was used without further purification (quant.). HRMS
(FAB; M + Cs+) calcd for C94H124N10O9Cs: 1669.8607. Found:
1669.8709.
JA994397M