Molecular discrimination of N-protected amino acid esters by a self-assembled cylindrical capsule: spectroscopic and computational studies.

Article abstract of DOI:10.1021/jo0201171

A self-assembled, cylindrical capsule was used to bind N-alpha-protected amino acid esters. The reversible encapsulation was studied using NMR spectroscopy in deuterated mesitylene solution and by computer-aided molecular modeling. BOC-L-alanine alkyl est

Full text of DOI:10.1021/jo0201171

VOLUME 67, NUMBER 24
NOVEMBER 29, 2002
© Copyright 2002 by the American Chemical Society
Molecu la r Discr im in a tion of N-P r otected Am in o Acid Ester s by a
Self-Assem bled Cylin d r ica l Ca p su le: Sp ectr oscop ic a n d
Com p u ta tion a l Stu d ies
Osamu Hayashida,†,‡ Lubomir Sebo,† and J ulius Rebek, J r.*,†
The Skaggs Institute for Chemical Biology and The Department of Chemistry,
The Scripps Research Institute, MB-26, 10550 North Torrey Pines Road, La J olla, California 92037, and
Institute for Fundamental Research of Organic Chemistry, Kyushu University, Hakozaki, Higashi-ku,
Fukuoka 812-8581, J apan
jrebek@scripps.edu
Received February 20, 2002
A self-assembled, cylindrical capsule was used to bind N-R-protected amino acid esters. The
reversible encapsulation was studied using NMR spectroscopy in deuterated mesitylene solution
and by computer-aided molecular modeling. BOC-L-alanine alkyl esters and BOC-â-alanine alkyl
esters were tested as guests, and the relative binding affinities were established by direct competition
experiments. A good correlation was found between the experimental and calculated relative binding
affinities in these two series. Guests that were slightly longer than the internal dimensions of the
cavity were accommodated by adopting compacted conformations.
In tr od u ction
such as catalysis.8 An example is the resorcin[4]arene-
based cavitand 1 in its vase-shaped C4v conformation. It
forms a cylindrical capsule 1‚1 in apolar organic sol-
vents,3e,9 and we earlier measured its inner space using
rigid guest molecules of well-defined length and shape
(molecular rulers).9 The study revealed that para-
substituted benzanilides about 14.7 Å long and 5.7 Å
thick could be accommodated9 (Figure 1). The current
research was undertaken to answer the following ques-
tions: (1) Will the capsule accommodate longer guests
that can adopt folded conformations? (2) What other
factors, e.g., volume or shape, define an optimum guest?
(3) What can be predicted by modeling of the host/guest
complexes? We proceeded by interrogating a series of
N-protected amino acid esters (2-25) as guest molecules.
The guests could be easily obtained as a large family of
related structures. These are known compounds, some
of which are commercial products. The capsule is notori-
ous for its ability to sequester trace impurities, and only
through synthesis could we have a history of possible
contaminants. The amino group of L-alanine was pro-
tected by tert-butoxycarbonyl (Boc), benzyloxycarbonyl
(Z), and 9-fluorenylmethoxycarbonyl (Fmoc) groups; these
groups are relatively hydrophobic, rigid, and bulky. The
lengths of the guest molecules were defined by their ester
Self-assembled capsules1 are current vehicles for ex-
ploring how molecules fit together.2,3 These capsules, held
together by hydrogen bonds or metal-ligand coordina-
tion,4-6 are capable of providing well-defined inner spaces
that more or less completely surround targets and detain
them. The host/guest interactions are themselves molec-
ular recognition phenomena7 and offer other possibilities
† The Scripps Research Institute.
‡ Kyushu University.
(1) Rebek, J ., J r. Acc. Chem. Res. 1999, 32, 278-286.
(2) (a) Wyler, R.; de Mendoza, J .; Rebek, J ., J r. Angew. Chem., Int.
Ed. 1993, 32, 1699. (b) Meissner, R. S.; Rebek, J ., J r.; de Mendoza, J .
Science 1995, 270, 1485. (c) Grotzfeld, R. M.; Branda, N.; Rebek, J .,
J r. Science 1996, 271, 487. (d) Szabo, T.; O’Leary, B. M.; Rebek, J ., J r.
Angew. Chem., Int. Ed. 1999, 37, 3410. (e) Martin, T.; Obst, U.; Rebek,
J ., J r. Science 1998, 281, 1842. (f) Hof, F.; Nuckolls, C.; Craig, S. L.;
Martin, T.; Rebek, J ., J r. J . Am. Chem. Soc. 2000, 122, 10991.
(3) (a) Shimizu, K. D.; Rebek, J ., J r. Proc. Natl. Acad. Sci. U.S.A.
1995, 92, 12403. (b) Scheerder, J .; van Duynhoven, J . P. M.; Engbersen,
J . F. J .; Reinhoudt, D. N.; Angew. Chem., Int. Ed. Engl. 1996, 35, 1090.
(c) Vysotsky, M. O.; Thondorf, I.; Bo¨hmer, V.; Angew. Chem., Int. Ed.
2000, 39, 1264. (d) MacGillivary, L. R.; Atwood, J . L.; Nature 1997,
389, 469. (e) Heinz, T.; Rudkevich, D. M.; Rebek, J ., J r. Nature 1998,
394, 764-766.
(4) (a) Zhong, Z.; Ikeda, A.; Ayabe, M.; Shinkai, S.; Sakamoto, S.;
Yamaguchi, K. J . Org. Chem. 2001, 66, 1002. (b) Fox, O. D.; Dalley,
N. K.; Harrison, R. G. J . Am. Chem. Soc. 1998, 120, 7111.
(5) Fujita, M.; Umemoto, K.; Yoshizawa, M.; Fujita, N.; Kuzukawa,
T.; Biradha, K. Chem. Commun. 2001, 509-518 and references therein.
(6) Parac, T. N.; Caulder, D. L.; Raymond, K. N. J . Am. Chem. Soc.
1998, 120, 8003.
(8) (a) Kang, J .; Rebek, J ., J r. Nature 1997, 385, 50. (b) Ito, H.;
Kusukawa, T.; Fujita, M. Chem. Lett. 2000, 598.
(7) Catellano, R. K.; Kim, B. H.; Rebek, J ., J r. J . Am. Chem. Soc.
1997, 119, 12671.
(9) Ko¨rner, S. K.; Tucci, F. C.; Rudkevich, D. M.; Heinz, T.; Rebek,
J ., J r. Chem. Eur. J . 2000, 6, 187.
10.1021/jo0201171 CCC: $22.00 © 2002 American Chemical Society
Published on Web 05/14/2002
J . Org. Chem. 2002, 67, 8291-8298
8291

Hayashida et al.
F IGURE 1. (Top) structure of 1 and the energy-minimized (Amber force field) presentation of self-assembled cylindrical capsule
(1‚1); the long alkyl chains and CH hydrogen atoms are omitted for clarity. (Bottom) the cartoon representation of the capsule.
The dimension of internal space estimated experimentally is also shown.
CHART 1
groups: methyl, ethyl, n-propyl, n-butyl, n-pentyl, n-
hexyl, tert-butyl, cyclopentyl, cyclohexyl, cycloheptyl, and
benzyl (2-12). Other amino acid esters: Boc-â-alanine,
Boc-glycine, Boc-L-valine, and Boc-L-phenylalanine were
prepared to probe additional specific molecular discrimi-
nation feature of the capsule (13-25) (Chart 1).
chemical shifts (δ, ppm) and the change in the chemical
shifts (∆δ, ppm) for encapsulated guest molecules and δ
for capsule N-H protons are summarized in Table 1.
Among the series of Boc-L-alanine n-alkyl esters (2-7),
the shortest guest 2, with a molecular length of 10.9 Å,11
was not encapsulated (Figure 2a), but the slightly longer
guests (3 and 4) were. (Figure 2b,c). The large upfield
shifts of the guest Boc and terminal alkyl groups (∆δ )
-4.15 to -4.30, -4.40 to -4.43 ppm, respectively) place
them near the ends of capsule 1‚1, while the methyl
protons of the amino acid side chain showed relatively
Resu lts a n d Discu ssion
NMR Sp ectr oscop ic Stu d y. Encapsulation of guests
1
in mesitylene-d12 at 296 K was determined by H NMR
spectroscopy as previously described (Figure 2).10 Selected
(11) From molecular modeling by AMBER* force field (Macromodel/
Maestro); Mohamadi, F.; Richhards, N. G.; Guida, W. C.; Liskamp,
R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C.
J . Comput. Chem. 1990, 11, 440.
(10) (a) Heinz, T.; Rudkevich, D. M.; Rebek, J ., J r. Angew. Chem.,
Int. Ed. 1999, 38, 1136-1139. (b) Tucci, F. C., Rudkevich, D. M.; Rebek,
J ., J r. J . Am. Chem. Soc. 1999, 121, 4928.
8292 J . Org. Chem., Vol. 67, No. 24, 2002

Self-Assembled Cylindrical Capsule
F IGURE 2. Downfield and upfield portions of 1H NMR spectra (600 MHz, 296 K) of 1 in mesitylene-d12 in the presence of 2 (a),
3 (b), 4 (c), 5 (d), 6 (e), 7 (f), 9 (g), 10 (h), 11 (i), 13 (j), 14 (k), 15 (l), 16 (m), and 17 (n).
small upfield shifts (∆δ ) -0.94 to -0.97 ppm) (Table
1). These locate them around the seam of hydrogen bonds
in the middle of the capsule (Figure 2b, c). Intense
intermolecular NOE contacts were observed between
encapsulated guest Boc protons and the arene protons
of capsule 1‚1, supporting these geometrical arrange-
ments. Moreover, nonequivalent N-H signals for the two
halves of the capsule were observed in the downfield
region of the spectra (Figure 2b,c), indicating that the
tumbling of the encapsulated guest molecules is re-
stricted in the cavity, at least on the NMR time scale.
Moreover, n-butyl and n-pentyl esters (5 and 6) were also
encapsulated by 1‚1, even though their molecular lengths
(14.6 and 15.9 Å, respectively, in their extended confor-
mation) are equivalent to or even exceed the dimensions
of 1‚1. In all cases the terminal methyl resonances of the
esters were the highest upfield signals.12 The guest
molecules must reduce their lengths to accommodate
themselves, but their contractions are not through folding
as shown schematically in Figure 2d,e,m. Such folding
would place methylene groups in the most shielded
regions of the hosts. Rather, the contortions must take
place elsewhere in the guest’s structure. The ∆δ values
for Boc protons of 6 and 5 were larger than those of 3
and 4, while the observed ∆δ values for terminal methyl
(12) Shivanyuk, A.; Rebek, J ., J r. Proc. Natl. Acad. Sci. USA 2001,
98, 7662.
J . Org. Chem, Vol. 67, No. 24, 2002 8293

Hayashida et al.
TABLE 1. Selected Ch em ica l Sh ifts (δ, p p m ) a n d th e
Ch a n ge in Ch em ica l Sh ifts (∆δ, p p m ) for En ca p su la ted
Gu est Molecu les a n d δ for Ca p su le
δ (∆δ) for encapsulated species
terminal methyl methyl protons
protons of
ester group
of amino acid δ for capsule
guest Boc protons
side chain
imide protons
2
n.d.a
3
4
5
6
7
8
9
-2.71 (-4.15) -3.43 (-4.40)
-2.86 (-4.30) -3.67 (-4.43)
-2.94 (-4.38) -3.39 (-4.22)
-2.99 (-4.43) -3.45 (-4.23)
n.d.a
0.17 (-0.97)
0.20 (-0.94)
0.16 (-0.99)
0.24 (-0.80)
9.93, 9.82
9.93, 9.75
9.92, 9.62
9.88, 9.50
F IGURE 3. Upfield portion of 1H NMR spectrum for the
competition experiment with 1 (2.0mM) between 14 (5.0 mM)
-2.76 (-4.20) -2.84 (-4.15)
0.27 (-0.88)
0.15 (-0.97)
0.23 (-0.95)
9.89, 9.72
9.88, 9.67
9.81, 9.56
and 15 (5.0 mM) in mesitylene-d12
.
-2.92 (-4.36)
-
-
10 -2.98 (-4.43)
11
12
n.d.a
n.d.a
13 -2.64 (-4.05) -1.32 (-4.60)
14 -2.85 (-4.26) -3.58 (-4.57)
15 -2.93 (-4.35) -3.87 (-4.65)
16 -2.95 (-4.37)
-
-
-
-
10.12, 10.07
10.11, 10.07
10.03, 9.99
9.93, 9.76
17
n.d.a
18 -2.73 (-4.16) -3.57 (-4.50)
-
-
10.02, 10.17
10.00, 9.85
19 -2.71 (-4.14) -2.85 (-4.16)
20
21
22
23
24
25
n.d.a
n.d.a
n.d.a
n.d.a
-
-1.27 (-4.57) -0.47 (-1.55)
9.89, 9.82
n.d.a
a
No detectable signals of encapsulated guests.
proton of 6 and 5 were smaller than those of 3 and 4
(Table 1). Again, their positions near the capule’s ends
effect the shifts. Neither the longest ester guest 7 nor
benzyl ester 12 were encapsulated; only solvent-filled
capsules10 were detected by 1H NMR spectroscopy (for 7,
Figure 2f; for 12, not shown). Encapsulation with con-
formational change of the guest molecules is a reasonable
expectatition: The cyclic analogues (9 and 10) were also
encapsulated (Figure 2g,h), and they resemble the tightly
folded conformers of the longer guests mentioned above.
However, guest 11, bearing a slightly larger seven-
membered ring was not encapsulated (Figure 2i). Similar
molecular discrimination behavior was observed for the
homologous series of Boc-â-alanine esters (13-17) (Fig-
ure 2j-n). Boc-glycine esters 18 and 19, lacking the
methyl group from 3 and 8,13 were also encapsulated,
while more “stocky” amino acids such as 20, 21, 22, and
23 were not. The Fmoc-protecting group of L-alanine
methyl ester (25) was too large to be accommodated.
Relative binding affinities of the guests to the capsule
were examined by 1H NMR competition experiments;
[1‚1] ) 1.0 mM, [G1] ) [G2] ) 5.0 mM. Integration of
the corresponding peaks of the encapsulated guests after
equilibrium was attained gave a direct readout of the
relative binding affinities. For example, the 1H NMR
spectrum for the competition experiment between 14 and
15 was shown in Figure 3. Through this method the
relative binding affinities of all of the guests (Ki/K14)
were determined and they are summarized in Figure 4.
The Boc-L-alanine n-alkyl esters follow the sequence:
4:5:3:6 ) 1.00:0.24:0.03:0.001, while Boc-â-alanine esters,
14:15:13:16 ) 1.00:0.43:0.05:0.01. In short, the relative
F IGURE 4. Relative binding affinities of N-R-protected amino
acid esters to capsule 1‚1.
F IGURE 5. A space-filling representation of the empty
cylindrical capsule (left), sliced in XY plane (middle), showing
the capsule cavity (right).
binding affinities are subject to the length of the guest
molecules. In case of the encapsulation with 6 and 16,
relatively upfield shifted δ values for the imide reso-
nances of the hosts suggested loosely held dimeric
capsules. Among these guests, Boc-â-alanine ethyl ester
14 with a molecular length of about 13.5 Å in the
extended conformation fits the inner space of capsule 1‚1
best. The cyclic esters proved better guests than their
open-chain counterparts (e.g., 9 > 6). Apparently, the
binding affinity is also effected by preorganization of the
guest.
Com p u ta tion a l Stu d y. The structure of the empty
cylindrical capsule 1‚1 was built and minimized using
program Macromodel/Maestro11 (AMBER* force-field,
(13) Boc-protected L-alanine tert-butyl ester (8) and Z-protected
L-alanine methyl ester (24) can be also encapsulated.
8294 J . Org. Chem., Vol. 67, No. 24, 2002

Self-Assembled Cylindrical Capsule
F IGURE 6. Representative lowest-energy structures of the complexes between the cylindrical capsule and guests (2, 4, 7, and
14)) obtained from MCMM conformational search.
vacuum, user defined cutoff). The inner space of the
capsule was examined using the program GRASP.14 The
cavity was generated by rolling a sphere of 1.4 Å around
the inner surface. Its estimated size is 16.5 × 7.1 Å
comprising a volume of ∼420 Å3 (Figure 5). The cavity
has a (roughly) cylindrical shape tapered at both ends
with a distance of ∼16.5 Å, tip to tip. The tapered ends
are best suited for complements that are thin, e.g.,
terminal acetylenes. The thick BOC groups must be
lodged at some distance from these ends. Accordingly,
the effective space for guests bearing this terminus is
considerably shortened. Given the limits imposed by the
BOC group, the longer derivatives cannot fit without
some buckling along their length.
The size and the molecular volume of the encapsulated
guests provide an indication of the capsule binding
preferences. Acceptable guest volume ranges from 180
Å3 to 240 Å3. The length of the encapsulated guests is
between 12 and 16 Å in the extended lowest-energy
conformation. The flexible guests are compacted upon
encapsulation, causing changes in the guests’ lengths, but
not their volume. At the same time, the cavity of the
cylindrical capsule can adjust its shape and volume. All
components of the encapsulation complex must be care-
fully analyzed in comparisons.
The initial structures of the cylindrical capsule with
encapsulated guests were built and their local energy-
minima were found using program Macromodel/Maestro
(AMBER* force-field, vacuum, user defined cutoff). To
find the global energy minima of the molecular com-
plexes, Monte Carlo multiple minimum (MCMM) con-
formational searches, as implemented in Macromodel/
Maestro, were used (3000 steps, AMBER* force-field,
vacuum, user defined cutoff). In the case of guests 11,
12, 22, 23, and 25, no global energy-minimum structure
of the complex could be obtained because the MCMM
conformational search provided only structures with
guest located outside the cylindrical capsule. The repre-
sentative lowest-energy structures are shown in Fig-
ure 6.
encapsulation are summarized in Table 2. The cavity
volume of the corresponding complexes were also calcu-
lated and summarized in Table 2 together with Packing
Coefficient15 (PC). The length of the capsule was found
to be nearly constant (17.7 Å ( 1%) in all complexes
examined. On the other hand, the capsule width reflected
extremes in the guest size and led to changes in cavity
volume. In the case of the small guests (13, 14, and 18),
the relatively flexible walls of the capsule collapsed,
probably in order to eliminate the residual empty space
in the cavity (Figure 6). Another stabilizing factor for
such a collapsed capsule structure may be a formation
of new intermolecular hydrogen bonds between the guest
and the capsule. The result is a reduced volume of the
cavity (-13% for 13 and 14, -18% for 18) and increased
PC. Therefore the cylindrical capsule can be seen as a
container structure with rigid ends and a flexible center.
The PC values of the five best guests (14, 15, 4, 19, and
9) range from 51 to 55%.
When a large flexible guest is encapsulated, its struc-
ture adopts the shape of the cavity. Often an energetically
less favorable, but more compact gauche, conformation
of the alkyl chain is preferred over the most stable trans
conformation in the folded guest structure. A 20% length
reduction (3.2 Å) was estimated for the largest encapsu-
lated guest 16. The length of the folded encapsulated
guests ranges from 11.7 to 13.5 Å.
The lowest-energy structures obtained from MCMM
conformational search were further refined and the
energies of the complexes were estimated including a GB/
SA solvation model for CHCl3 in a final minimization
(Macromodel/Maestro, AMBER* force-field, extended
cutoff. Although this rough estimate neglects any en-
tropic changes during the encapsulation, the structure-
affinity relationship obtained may have value for the
interpretation of the experimental data and permit
prediction of the binding affinity for new guests.
In the series of Boc-L-alanine n-alkyl esters 2-7 the
highest relative binding affinities were estimated for
guest 4; this is in agreement with experimental data. The
same is true for the Boc-â-alanine series 13-17, where
both the calculation and the experiment identified 14 as
the best guest (Figure 7).
The volumes and length of free guests in the extended
conformation, the length of encapsulated guests in the
folded conformation, and change in the length upon
(14) Nicholls, A.; Sharp, K. A.; Honig, B. Proteins 1991, 11, 281.
(15) Mecozzi, S.; Rebek, J . J r. Chem. Eur. J . 1998, 4, 1016.
J . Org. Chem, Vol. 67, No. 24, 2002 8295

Hayashida et al.
TABLE 2. Str u ctu r a l Ch a n ges u p on En ca p su la tion
guest volumeb cavity volumec cavity volume packing coefficient free guest length folded guest length guest length change
guesta
(VG), Å3
(VC), Å3
change,d
%
(PC) () VG/VC), %
(L0), Å
(Lf), Å
(Lf - L0),e Å (%)
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
180
194
208
222
236
250
222
228
241
255
246
180
194
208
222
237
180
208
208
222
246
261
204
263
416
411
408
417
424
437
426
415
415
f
-1
-2
-3
-1
1
4
1
-1
-1
f
43
47
51
53
56
57
52
55
58
f
10.9
12.1
13.4
14.6
15.9
17.1
12.2
13.3
13.9
13.9
14.9
12.2
13.5
14.7
16.0
17.1
12.3
12.3
10.3
11.5
12.9
12.9
13.0
15.3
10.9
12.1
12.5
12.3
13.5
13.9
12.0
12.8
13.3
f
0.0 (0)
0.0 (0)
-0.9 (-7)
-2.3 (-16)
-2.4 (-15)
-3.2 (-18)
-0.2 (-2)
-0.5 (-4)
-0.6 (-4)
f
f
f
f
f
f
365
365
391
418
428
346
408
438
436
f
-13
-13
-7
0
49
53
53
53
55
52
51
47
51
f
11.7
12.9
12.8
12.8
13.2
12.0
11.8
10.9
12.1
f
-0.5 (-5)
-0.6 (-5)
-1.9 (-13)
-3.2 (-20)
-3.9 (-23)
-0.3 (-2)
-0.5 (-4)
0.6 (5)
0.6 (5)
f
f
-0.5 (-4)
f
2
-18
-3
4
4
f
f
-8
f
f
f
53
f
f
388
f
12.5
f
a
b
d
Underlined ) encapsulation observed. GRASP. c GRASP, probe radius 1.4 Å. (VC - 420)/420. e (Lf - L0)/L0. f The guest was found
outside the capsule in all obtained structures.
were used without further purification. Compound 1 was
prepared following literature procedures, after changing reac-
tion conditions from 80 °C, 5 h to room temperature, 10 h.9
1
En ca p su la tion Stu d ies. H NMR experiments were car-
ried out using a 600 MHz spectrometer at 296 K. Deuterated
mesitylene was used as purchased from Aldrich. In encapsula-
tion experiments, the concentration of 1 and individual guest
was 2.0 and 5.0 mM, respectively. In a typical competition
experiment, a solution of 1 (2.0 mM), 14 (5.0 mM), and 15 (5.0
mM) in mesitylene-d12 was placed in an NMR tube and allowed
to stand at room temperature at least for 1 day and then
analyzed by 1H NMR spectroscopy. The relative binding
affinities of the guests to the capsule were determined by direct
integration of the corresponding peaks of the encapsulated
guests. NOE experiment was carried out by using a GOESY
sequence, with d1 )2.5 s and d8 ) 400 ms.
Gen er a l P r oced u r e for th e Syn th esis of N-r-P r otected
F IGURE 7. Comparison of experimental (filled circles) and
estimated (empty circles) relative binding affinities ln(Ki/K14
)
Am in o Acid Ester s. P r oced u r e A. N,N′-Dicyclohexylcarbo-
diimide (DCC) (1.1 equiv) and 1-hydroxybenzotriazole (HOBt)
were added to a solution of N-R-tert-butoxycarbonylamino acid
in dry dichloromethane at 0 °C, and the mixture was allowed
to stand at the same temperature while being stirred for 20
min. Corresponding alcohol (1.0 equiv) was added to the
mixture, and the resulting mixture was stirred for 4 h at 0 °C
and then overnight at room temperature. Precipitates that
formed (N,N′-dicyclohexylurea) were removed by filtration, the
solvent was eliminated under reduced pressure, and the
residue was dissolved in ethyl acetate. The solution was then
washed with 10% aqueous citric acid, saturated aqueous
sodium chloride, and 4% aqueous sodium hydrogen carbonate
in this sequence. After being dried (MgSO4), the solution was
evaporated to dryness under reduced pressure. The residue
was chromatographed on a column of silica gel with ethyl
acetate-hexane as eluant. The product fraction was collected
and dried in vacuo. The crude product was purified by gel
filtration chromatography on a column of Sephadex LH-20
with methanol-chloroform (1:1 v/v) as eluant. The product
fraction was collected and dried in vacuo.
for the N-Boc-L-Ala (left) and N-Boc-â-Ala (right) ester series.
In conclusion, the size and shape selectivity for mo-
lecular encapsulation of N-R-protected amino acid esters
in capsule 1‚1 was revealed by NMR spectroscopy. The
most strongly bound guest to the capsule was guest 14.
The capsule also accommodated guests that were slightly
longer than the internal dimensions of the cavity, and
these guests adopted compacted conformations. The
conformational changes of the alkyl chains of these guests
were also supported by the computer-aided molecular
modeling analysis of the complexes. A good correlation
was found between the experimental and calculated
relative binding affinities in a homologous series of Boc-
L-alanine and Boc-â-alanine n-alkyl esters. The compu-
tational analysis of the host/guest complexes may be
useful in predicting relative binding affinities in other
cases of molecules within molecules.
P r oced u r e B. Amino acid ester hydrochloride (p-tosylate)
and equivalent of triethylamine were dissolved in chloroform.
One equivalent of di-tert-butyl dicarbonate [(Boc)2O] in chlo-
roform was added dropwise to the solution at room tempera-
Exp er im en ta l Section
Gen er a l. 1H NMR spectra were recorded on a 600 MHz
spectrometer at 296 K. All commercially available chemicals
8296 J . Org. Chem., Vol. 67, No. 24, 2002

Self-Assembled Cylindrical Capsule
ture, and the mixture was stirred for overnight at room
temperature. The resulting mixture was evaporated to dryness
under reduced pressure. The residue was dissolved in ethyl
acetate. After removal of an amount of insoluble materials by
filtration, the filtrate was then washed with 10% aqueous citric
acid, saturated aqueous sodium chloride, and 4% aqueous
sodium hydrogen carbonate in this sequence. After being dried
(MgSO4), the solution was evaporated to dryness under
reduced pressure. The residue was chromatographed on a
column of silica gel with ethyl acetate-hexane as eluant. The
product fraction was collected and dried in vacuo. The crude
product was purified by gel filtration chromatography on a
column of Sephadex LH-20 with methanol-chloroform (1:1 v/v)
as eluant. The product fraction was collected and dried in
vacuo.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e m eth yl ester (2):
prepared following the general procedure B on a 144 mM scale
of L-alanine methyl ester hydrochoride. 1H NMR (600 MHz,
CDCl3, 296 K) δH 1.32 (3H, t, J 7.2 Hz, CHCH3), 1.38 [9H, s,
(CH3)3CO], 3.68 (3H, s, OCH3), 4.26(1H, m, CHCH3), 5.12 (1H,
m, NH). Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v)
0.74.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e eth yl ester (3): pre-
pared following the general procedure A on a 106 mM scale of
N-R-tert-butoxycarbonyl-L-alanine and ethanol. 1H NMR (600
MHz, CDCl3, 296 K) δH 1.26 (3H, t, J 7.1 Hz, CH2CH3), 1.37
(3H, d, J 7.2 Hz, CHCH3), 1.43 [9H, s, (CH3)3CO], 4.18 (2H, q,
J 7.1 Hz, CH2CH3), 4.27 (1H, m, CHCH3), 5.17 (1H, m, NH).
Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.74.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e n -p r op yl ester (4):
prepared following the general procedure A on a 106 mM scale
of N-R-tert-butoxycarbonyl-L-alanine and 1-propanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 0.88 (3H, t, J 7.1 Hz, CH2CH3),
1.32 (3H, d, J 7.2 Hz, CHCH3), 1.37 [9H, s, (CH3)3CO], 1.60
(2H, q, J 7.1 Hz, CH2CH2CH3), 4.04 (2H, m, CH2CH2CH3), 4.23
(1H, m, CHCH3), 5.15 (1H, m, NH). Rf (K6F Silica Gel; CHCl3:
CH3OH:AcOH ) 95:5:3 v/v) 0.74.
CH2CH3), 1.40 [9H, s, (CH3)3CO], 1.4-1.8 (8H, m, cyclic-CH2),
4.22 (1H, m, OCH), 5.08 (1H, m, CHCH3), 5.18 (1H, m, NH).
Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.69.
N-r-ter t-Bu t oxyca r b on yl-L-a la n in e cycloh exyl est er
(10): prepared following the general procedure A on a 106 mM
scale of N-R-tert-butoxycarbonyl-L-alanine and cyclohexanol.
1
1H NMR (600 MHz, CDCl3, 296 K) H 1.36 (3H, t, J 7.2 Hz,
CH2CH3), 1.44 [9H, s, (CH3)3CO], 1.3-1.8 (10H, m, cyclic-CH2),
4.27 (1H, m, OCH), 4.80 (1H, m, CHCH3), 5.08 (1H, m, NH).
Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.69.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e cycloh ep tyl ester
(11): prepared following the general procedure A on a 106 mM
scale of N-R-tert-butoxycarbonyl-L-alanine and cycloheptanol.
1H NMR (600 MHz, CDCl3, 296 K) δH 1.35 (3H, t, J 7.2 Hz,
CH2CH3), 1.43 [9H, s, (CH3)3CO], 1.3-1.8 (12H, m, cyclic-CH2),
4.24 (1H, m, OCH), 4.96 (1H, m, CHCH3), 5.08 (1H, m, NH).
Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.72.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e ben zyl ester (12):
prepared following the general procedure B on a 114 mM scale
of L-alanine benzyl ester p-tosylate. 1H NMR (600 MHz, CDCl3,
296 K) δH 1.33 (3H, d, J 7.0 Hz, CHCH3), 1.48 [9H, s, (CH3)3-
COCONH], 4.4 (1H, m, CHCH2), 5.1 (1H, m, NH), 5.2 (2H, m,
CH2), 7.39 (5H, m, Ar-H). Rf (K6F Silica Gel; CHCl3:CH3OH:
AcOH ) 95:5:3 v/v) 0.82.
N-r-ter t-Bu toxyca r bon yl-â-a la n in e m eth yl ester (13):
prepared following the general procedure A on a 106 mM scale
1
of N-R-tert-butoxycarbonyl-â-alanine and methanol. H NMR
(600 MHz, CDCl3, 296 K) δH 1.42 [9H, s, (CH3)3CO], 2.52 (2H,
t, J 6.0 Hz, NHCH2CH2), 3.39 (2H, m, NHCH2CH2), 3.69 (3H,
s, OCH3), 5.04 (1H, m, NH). Rf (K6F Silica Gel; CHCl3:CH3-
OH:AcOH ) 95:5:3 v/v) 0.68.
N-r-ter t-Bu t oxyca r b on yl-â-a la n in e et h yl est er (14):
prepared following the general procedure A on a 106 mM scale
of N-R-tert-butoxycarbonyl-â-alanine and ethanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 1.26 (3H, t, J 7.2 Hz, CH2CH3),
1.42 [9H, s, (CH3)3CO], 2.51 (2H, t, J 6.0 Hz, NHCH2CH2), 3.39
(2H, m, NHCH2CH2), 4.15 (2H, q, J 7.2 Hz, OCH2CH3), 5.03
(1H, m, NH). Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:
5:3 v/v) 0.68.
N-r-ter t-Bu toxyca r bon yl-â-a la n in e n -p r op yl ester (15):
prepared following the general procedure A on a 106 mM scale
of N-R-tert-butoxycarbonyl-â-alanine and 1-propanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 0.88 (3H, t, J 7.3 Hz, CH2CH3),
1.37 [9H, s, (CH3)3CO], 1.66 (2H, m, OCH2CH2CH3), 2.47 (2H,
t, J 6.4 Hz, NHCH2CH2), 3.33 (2H, m, NHCH2CH2), 4.00 (2H,
q, J 6.4 Hz, OCH2CH2CH3), 5.16 (1H, m, NH). Rf (K6F Silica
Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.70.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e n -bu tyl ester (5):
prepared following the general procedure A on a 106 mM scale
1
of N-R-tert-butoxycarbonyl-L-alanine and 1-butanol. H NMR
(600 MHz, CDCl3, 296 K) δH 0.94 (3H, t, J 7.1 Hz, CH2CH3),
1.39 (3H, d, J 7.2 Hz, CHCH3), 1.4 [2H, m, O(CH2)2CH2CH3],
1.45 [9H, s, (CH3)3CO], 1.64 (2H, m, OCH2CH2CH2CH3), 4.16
(2H, m, OCH2CH2), 4.3 (1H, m, CHCH3), 5.06 (1H, m, NH). Rf
(K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.74.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e n -p en tyl ester (6):
prepared following the general procedure A on a 106 mM scale
of N-R-tert-butoxycarbonyl-L-alanine and 1-pentanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 0.88 (3H, t, J 7.1 Hz, CH2CH3),
1.29-1.31 [7H, O(CH2)2(CH2)2CH3, CHCH3], 1.45 [9H, s,
(CH3)3CO], 1.62 [2H, m, OCH2CH2(CH2)2CH3], 4.11 [2H, m,
OCH2(CH2)3CH3], 4.27 (1H, m, CHCH3), 5.09 (1H, m, NH). Rf
(K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.74.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e n -h exyl ester (7):
prepared following the general procedure A on a 106 mM scale
N-r-ter t-Bu toxyca r bon yl-â-a la n in e n -bu tyl ester (16):
prepared following the general procedure A on a 106 mM scale
1
of N-R-tert-butoxycarbonyl-â-alanine and 1-butanol. H NMR
(600 MHz, CDCl3, 296 K) δH 0.88 (3H, t, J 7.3 Hz, CH2CH3),
1.30 [9H, s, (CH3)3CO], 1.2-1.5 [4H, m, OCH2(CH2)2CH3], 2.40
(2H, t, J 6.4 Hz, NHCH2CH2), 3.25 (2H, m, NHCH2CH2), 3.97
[2H, q, J 6.4 Hz, OCH2(CH2)2CH3], 5.21 (1H, m, NH). Rf (K6F
Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.71.
1
of N-R-tert-butoxycarbonyl-L-alanine and 1-hexanol. H NMR
N-r-ter t-Bu toxyca r bon yl-â-a la n in e n -p en tyl ester (17):
prepared following the general procedure A on a 106 mM scale
of N-R-tert-butoxycarbonyl-â-alanine and 1-pentanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 0.92 (3H, t, J 7.2 Hz, CH2CH3),
1.45 [9H, s, (CH3)3CO], 1.3-1.7 [6H, m, OCH2(CH2)3CH3], 2.53
(2H, t, J 6.4 Hz, NHCH2CH2), 3.41 (2H, m, NHCH2CH2), 4.10
[2H, q, J 6.4 Hz, OCH2(CH2)3CH3], 5.09 (1H, m, NH). Rf (K6F
Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.71.
N-r-ter t-Bu toxyca r bon yl-glycin e eth yl ester (18): pre-
pared following the general procedure A on a 114 mM scale
N-R-tert-butoxycarbonyl-glycine and ethanol. 1H NMR (600
MHz, CDCl3, 296 K) δH 1.22 (3H, t, J 7.1 Hz, CH2CH3), 1.39
[9H, s, (CH3)3CO], 3.84 (2H, d, J 5.4 Hz, CH2), 4.15 (2H, q, J
7.1 Hz, CH2CH3), 5.12 (1H, m, NH). Rf (K6F Silica Gel; CHCl3:
CH3OH:AcOH ) 95:5:3 v/v) 0.71.
(600 MHz, CDCl3, 296 K) δH 0.84 (3H, t, J 7.1 Hz, CH2CH3),
1.2-1.6 (11H, OCH2(CH2)4CH3, CHCH3), 1.39 [9H, s, (CH3)3-
CO], 1.59 (2H, m, OCH2CH2), 4.08 [2H, m, OCH2(CH2)4CH3],
4.24 (1H, m, CHCH3), 5.12 (1H, m, NH). Rf (K6F Silica Gel;
CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.71.
N-r-ter t-Bu t oxyca r b on yl-L-a la n in e ter t-Bu t yl est er
(8): prepared following the general procedure B on a 110 mM
scale of L-alanine tertert-butyl ester hydrochoride. 1H NMR
(600 MHz, CDCl3, 296 K) δH 1.30 (3H, d, J 7.0 Hz, CHCH3),
1.40 [9H, s, (CH3)3COCOCH], 1.42 [9H, s, (CH3)3COCONH],
4.14 (1H, m, CH), 5.1 (1H, m, NH). Rf (K6F Silica Gel; CHCl3:
CH3OH:AcOH ) 95:5:3 v/v) 0.77.
N-r-ter t-Bu toxyca r bon yl-L-a la n in e cyclop en tyl ester
(9): prepared following the general procedure A on a 106 mM
scale of N-R-tert-butoxycarbonyl-L-alanine and cyclopentanol.
1H NMR (600 MHz, CDCl3, 296 K) δH 1.33 (3H, t, J 7.2 Hz,
N-r-ter t-Bu toxyca r bon yl-glycin e ter t-Bu tyl ester (19):
prepared following the general procedure B on a 120 mM scale
J . Org. Chem, Vol. 67, No. 24, 2002 8297

Hayashida et al.
of glycine tert-butyl ester hydrochoride. 1H NMR (600 MHz,
CDCl3, 296 K) δH 1.45 [18H, s, (CH3)3COCOCH, (CH3)3COC-
ONH], 3.80 (2H, m, NHCH2), 5.00 (1H, m, NH). Rf (K6F Silica
Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.65.
N-r-ter t-Bu toxyca r bon yl-L-va lin e m eth yl ester (20):
prepared following the general procedure B on a 120 mM scale
of L-valine methyl ester hydrochoride. 1H NMR (600 MHz,
CDCl3, 296 K) δH 0.84 [nonequivalent, 3H, d, J 6.9 Hz, CH-
(CH3)2], 0.90 [nonequivalent, 3H, d, J 6.9 Hz, CH(CH3)2], 1.39
[9H, s, (CH3)3CO], 2.07 [1H, m, CH(CH3)2], 3.68 (3H, s, OCH3),
4.17 (1H, m, NHCH), 5.03 (1H, m, NH). Rf (K6F Silica Gel;
CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.78.
N-r-ter t-Bu toxyca r bon yl-L-va lin e eth yl ester (21): pre-
pared following the general procedure A on a 92 mM scale of
N-R-tert-butoxycarbonyl-L-valine and ethanol. 1H NMR (600
MHz, CDCl3, 296 K) δH 0.87 [nonequivalent, 3H, d, J 6.8 Hz,
CH(CH3)2], 0.94 [nonequivalent, 3H, d, J 6.8 Hz, CH(CH3)2],
1.26 (3H, d, J 7.1 Hz, CHCH3), 1.42 [9H, s, (CH3)3CO], 2.11
[1H, m, CH(CH3)2], 4.17 (2H, m, CH2CH3), 4.17 (1H, m,
CHCH3), 5.02 (1H, m, NH). Rf (K6F Silica Gel; CHCl3:CH3-
OH:AcOH ) 95:5:3 v/v) 0.78.
N-r-ter t-Bu toxyca r bon yl-L-p h en yla la n in e m eth yl es-
ter (22): prepared following the general procedure B on a 76
mM scale of L-phenylalanine methyl ester hydrochoride. 1H
NMR (600 MHz, CDCl3, 296 K) δH 1.48 [9H, s, (CH3)3CO], 3.11
(2H, m, CH2Ph), 3.73 (3H, s, OCH3), 4.62 (1H, m, CHCH2),
5.11 (1H, m, NH), 7.2 (5H, m, Ar-H). Rf (K6F Silica Gel;
CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.80.
N-r-9-flu or en ylm eth oxyca r bon yl-L-a la n in e m eth yl es-
ter (25): prepared following the general procedure A on a 90
mM scale of N-R-9-fluorenylmethoxycarbonyl-L-alanine and
methanol. 1H NMR (600 MHz, CDCl3, 296 K) δH 1.38 (3H, t, J
6.9 Hz, CHCH3), 3.74 (3H, s, OCH3), 4.25 (1H, m CHCH3), 4.42
(3H, m, CH2Ar, CO2CH2CH), 5.39 (1H, m, NH), 7.3-7.8 (8H,
m, Ar-H). Rf (K6F Silica Gel; CHCl3:CH3OH:AcOH ) 95:5:3
v/v) 0.58.
Molecu la r Mod elin g. For simplification, the C11 chains in
1 were replaced by Me groups. Initial structures of molecular
complexes were generated by a conjugate gradient minimiza-
tion with the AMBER* force-field and the BatchMin program
implemented within Maestro/MacroModel 7.0.11 These struc-
tures were further refined by a 3000-4000 step Monte Carlo
multiple minimum simulation in a vacuum using a user
defined cutoffs: van der Waals 3.5 Å, Electrostatic 6.0 Å,
H-Bond 4.0 Å. All conformations within 20 kJ mol-1 of the
computed global minimum were stored, and the representative
lowest-energy structure was analyzed.
The volume of cavities of the capsules and the volume of
the guests were estimated using GRASP (probe radius 1.4 Å).14
The length of the capsule was measured as a distance between
the centroids of methine carbons of the capsule.
For energy estimation all structures were further minimized
using the GB/ SA solvation model for chloroform (Amber*,
Macromodel/Maestro 7.0, extended cutoff). The guest binding
affinities were estimated as a difference between the total
energy of the complex and the energies of the free guest and
the free capsule. The relative binding affinities were estimated
from these guest binding affinities using guest 14 as a
standard.
N-r-ter t-Bu toxyca r bon yl-L-p h en yla la n in e eth yl ester
(23): prepared following the general procedure A on a 76 mM
scale of N-R-tert-butoxycarbonyl-L-phenylalanine and ethanol.
1H NMR (600 MHz, CDCl3, 296 K) δH 1.24 (3H, d, J 7.1 Hz,
CHCH3), 1.45 [9H, s, (CH3)3CO], 3.11 (2H, m, CH2Ph), 4.18
(2H, q, J 7.1 Hz, CH2CH3), 4.59 (1H, m, CHCH2), 5.12 (1H, m,
NH), 7.2 (5H, m, Ar-H). Rf (K6F Silica Gel; CHCl3:CH3OH:
AcOH ) 95:5:3 v/v) 0.80.
Ack n ow led gm en t. We thank the Skaggs Institute
for support and for a fellowship to L.S. O.H. thanks The
Ministry of Education, Culture, Sports, Science and
Technology of J apan.
N-r-ben zyloxyca r bon yl-L-a la n in e m eth yl ester (24):
prepared following the general procedure A on a 90 mM scale
of N-R-benzyloxycarbonyl-L-alanine and methanol. 1H NMR
(600 MHz, CDCl3, 296 K) δH 1.26 (3H, t, J 7.3 Hz, CHCH3),
3.76 (3H, s, OCH3), 4.40 (1H, m CHCH3), 5.11 (2H, m, CH2-
Ph), 5.35 (1H, m, NH), 7.3-7.4 (5H, m, Ph-H). Rf (K6F Silica
Gel; CHCl3:CH3OH:AcOH ) 95:5:3 v/v) 0.58.
Su p p or tin g In for m a tion Ava ila ble: NOE data and Table
with the comparison of experimental and estimated relative
binding affinities calculated using AMBER* force-field. This
material is available free of charge via the Internet at
http://pubs.acs.org.
J O0201171
8298 J . Org. Chem., Vol. 67, No. 24, 2002