Molecular Recognition Study on Supramolecular System. 14.1 Synthesis of Modified Cyclodextrins and Their Inclusion Complexation Thermodynamics with L-Tryptophan and Some Naphthalene Derivatives

Article abstract of DOI:10.1021/jo971466b

A series of β-cyclodextrin derivatives, carrying pyridinio (4-6), phosphonyl (7, 8), seleno (9-11), m- and p-picolinyl (12, 13), o-chloroanilino (16), 8-quinolyl (17), furfuryl (18), and 9-fluorenyl (19) moieties in the side chain, were newly synthesized, and their complexation behavior was assessed and discussed thermodynamically, using L-tryptophan and a few naphthalene derivatives as representative guests. Calorimetric titrations have been performed at 25.0°C in buffered aqueous solution (pH 7.20) to give the complex stability constants and thermodynamic parameters for the 1:1 inclusion complexation of these guests with the native and modified α-, β-, and/or γ-cyclodextrins (1-20). All of the chemical modifications to the primary side of cyclodextrins examined led to significant changes in complex stability and thermodynamic parameters, which are elucidated in terms of the conformational, electrostatic, hydrogen-bonding, and hydration effects. Thermodynamically, the inclusion complexation is mainly enthalpy-driven with a negative or minor positive entropic contribution, which in some cases determines the complex stability. The induced circular dichroism spectral analyses of these cyclodextrin derivatives indicated that the aromatic moiety in modified β-cyclodextrins (4-6, 9-19) only shallowly penetrates into the hydrophobic cavity of β-cyclodextrin, while the phenyl phosphate and fluorenyl moieties in 7 and 20 are embedded into the hydrophobic cavity of β-cyclodextrin because of the longer linking chain. Using all the thermodynamic data for a wide variety of cyclodextrin derivatives obtained in this and previous studies, the entropy changes (TΔS) were plotted against the enthalpy changes (ΔH) to give an excellent linear relationship. The slope (α) of 1.02 and an intercept (TΔS₀) of 4.3 of the regression line indicate substantial conformational changes and extensive desolvation caused upon complexation, respectively.

Full text of DOI:10.1021/jo971466b

1444
J . Org. Chem. 1998, 63, 1444-1454
Molecu la r Recogn ition Stu d y on Su p r a m olecu la r System . 14.1
Syn th esis of Mod ified Cyclod extr in s a n d Th eir In clu sion
Com p lexa tion Th er m od yn a m ics w ith L-Tr yp top h a n a n d Som e
Na p h th a len e Der iva tives
Yu Liu,*,† Bao-Hang Han,† Bin Li,† Yi-Min Zhang,† Peng Zhao,† Yun-Ti Chen,†
Takehiko Wada,‡ and Yoshihisa Inoue*,‡
Department of Chemistry, Nankai University, Tianjin, 300071, China, Department of Molecular
Chemistry, Osaka University, Yamadaoka, Suita 565, J apan, and Inoue Photochirogenesis Project,
ERATO, J ST, 4-6-3 kamishinden, Toyonaka 565, J apan
Received August 6, 1997
A series of â-cyclodextrin derivatives, carrying pyridinio (4-6), phosphonyl (7, 8), seleno (9-11),
m- and p-picolinyl (12, 13), o-chloroanilino (16), 8-quinolyl (17), furfuryl (18), and 9-fluorenyl (19)
moieties in the side chain, were newly synthesized, and their complexation behavior was assessed
and discussed thermodynamically, using L-tryptophan and a few naphthalene derivatives as
representative guests. Calorimetric titrations have been performed at 25.0 °C in buffered aqueous
solution (pH 7.20) to give the complex stability constants and thermodynamic parameters for the
1:1 inclusion complexation of these guests with the native and modified R-, â-, and/or γ-cyclodextrins
(1-20). All of the chemical modifications to the primary side of cyclodextrins examined led to
significant changes in complex stability and thermodynamic parameters, which are elucidated in
terms of the conformational, electrostatic, hydrogen-bonding, and hydration effects. Thermody-
namically, the inclusion complexation is mainly enthalpy-driven with a negative or minor positive
entropic contribution, which in some cases determines the complex stability. The induced circular
dichroism spectral analyses of these cyclodextrin derivatives indicated that the aromatic moiety in
modified â-cyclodextrins (4-6, 9-19) only shallowly penetrates into the hydrophobic cavity of
â-cyclodextrin, while the phenyl phosphate and fluorenyl moieties in 7 and 20 are embedded into
the hydrophobic cavity of â-cyclodextrin because of the longer linking chain. Using all the
thermodynamic data for a wide variety of cyclodextrin derivatives obtained in this and previous
studies, the entropy changes (T∆S) were plotted against the enthalpy changes (∆H) to give an
excellent linear relationship. The slope (R) of 1.02 and an intercept (T∆S0) of 4.3 of the regression
line indicate substantial conformational changes and extensive desolvation caused upon complex-
ation, respectively.
In tr od u ction
of separation science and technology.7-18 Consequently,
a good deal of effort has been devoted to the synthesis of
various cyclodextrin derivatives in order to examine their
molecular recognition behavior and has recently been
focused on the studies of the models of biological sub-
strate-receptor interactions.19-24 Unfortunately, the
A wide variety of modified cyclodextrins with nucleo-
philic or electrophilic substituents attached to the pri-
mary side of cyclodextrin have been designed and syn-
thesized to enhance the original binding ability and
selectivity of intact cyclodextrins through further ligation
or stereochemical complement of the functional side arm
to the anionic or molecular guests accommodated in the
cyclodextrin cavity.1-6 Indeed, some modified cyclodex-
trins, which feature the bifunctionalized structures ca-
pable of multipoints recognition and induced-fit interac-
tion with the guest accommodated in the cyclodextrin
cavity, have been employed successfully in several areas
(7) Szejtli, J . Cyclodextrin Technology. Dordrecht: Kluwer-Academic:
1988.
(8) Saenger, W. Angew. Chem., Int. Ed. Engl. 1980, 19, 344.
(9) Liu, Y.; Li, B.; Zhang, Y.-M.; Bu, X.-H.; Li, Y.-M.; Chen, R.-T.
Chin. Sci. Bull. 1996, 41, 117.
(10) Kawabata, Y.; Matsumoto, M.; Nakamura, T.; Tanaka, M.;
Manda, E.; Takahashi, H.; Tamura, S.; Tagaki, W.; Nakahara, H.;
Fukuda, K. Thin Solid Films 1988, 159, 353.
(11) Li, G.; McGown, L. B. Science 1994, 264, 249.
(12) Manka, J . S.; Lawrence, D. S. J . Am. Chem. Soc. 1990, 112,
2440.
(13) Rao, T. V. S.; Lawrence, D. S. J . Am. Chem. Soc. 1990, 112,
3614.
(14) Nelles, G.; Weisser, M.; Back, R.; Wohlfart, P.; Wenz, G.;
Mittler-Neher, S. J . Am. Chem. Soc. 1996, 118, 5039.
(15) Rojas, M. T.; Koniger, R.; Stoddart, J . F.; Kaifer, A. E. J . Am.
Chem. Soc. 1995, 117, 336.
(16) J ullien, L.; Canceill, J .; Valeur, B.; Bardez, E.; Lehn, J .-M.
Angew. Chem., Int. Ed. Engl. 1994, 33, 2438.
(17) J ullien, L.; Canceill, J .; Valeur, B.; Bardez, E.; Lefe`vre, J .-P.;
Lehn, J .-M.; Marchi-Artzner, V.; Pansu, R. J . Am. Chem. Soc. 1996,
118, 5432.
† Nankai University.
‡ ERATO, Osaka University.
(1) (a) Part 13: Liu, Y.; Li, B.; Han, B.-H.; Li, Y.-M.; Chen, R.-T. J .
Chem. Soc., Perkin Trans. 2 1997, 1275. (b) Part 12: Liu, Y.; You, C.-
C.; Han, B.-H.; Li, Y.-M.; Li, B.; Chen, R.-T. Chem. J . Chin. Univ. 1997,
18, 1316. (c) Part 10: Liu, Y.; Zhang, Y,-M.; Qi, A.-D.; Chen, R.-T.;
Yamamoto, K.; Wada, T.; Inoue, Y. J . Org. Chem. 1997, 62, 1826.
(2) Martin, K. A.; Mortellaro, M. A.; Sweger, R. W.; Fikes, L. E.;
Winn, D. T.; Clary, S.; J ohnson, M. P.; Czarnik, A. W. J . Am. Chem.
Soc. 1995, 117, 10443.
(3) Ueno, A.; Kuwabara, T.; Nakamura, A.; Toda, F. Nature 1992,
356, 136.
(4) Ueno, A.; Suzuki, I.; Osa, T. J . Am. Chem. Soc. 1989, 111, 6391.
(5) Ikeda, H.; Nakamura, M.; Ise, N.; Oguma, N.; Nakamura, A.;
Ikeda, T.; Toda, F.; Ueno, A. J . Am. Chem. Soc. 1996, 118, 10980.
(6) Tarkka, R. M.; Buncel, E. J . Am. Chem. Soc. 1995, 117, 1503.
(18) Harada, A. Coord. Chem. Rev. 1996, 148, 115.
(19) Croft, A. P.; Bartsch, R. A. Tetrahedron 1983, 39, 1417.
(20) Maletic, M.; Wennemers, H.; McDonald, D. Q. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 1490.
S0022-3263(97)01466-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/14/1998

Molecular Recognition Study on Supramolecular System
J . Org. Chem., Vol. 63, No. 5, 1998 1445
work on the molecular recognition thermodynamics for
model substances such as phenethylamine and ephe-
drines,25 hydrocarbons,26 aliphatic alcohols,27 aliphatic
diols,28 phenols,29 amino acids,9,29 cyclohexane deriva-
tives,30 naphthalene derivatives,31 and other aromatic
compounds29 has been concentrated mainly on natural
R-, â-, and γ-cyclodextrins. So far, less attention has been
paid to the inclusion complexation thermodynamics of
chemically modified R-, â-, and γ-cyclodextrins.32
alkylamino moiety (18-20), as shown in Chart 1. The
inclusion complexation behavior of these novel hosts with
some representative guest molecules, shown in Chart 2,
was investigated by using calorimetric titration to reveal
the consequences of the modifications upon complexation
from the thermodynamic point of view. The induced
circular dichroism (ICD) spectral study with modified
â-cyclodextrins enables us to elucidate the conformation
of the aromatic moiety in the hosts (4-7, 9-20) and to
get significant insights into conjecturing inclusion com-
plexation mechanism.
The thermodynamic parameters for the complexation
with cyclodextrin derivatives (1-20), together with those
reported,32 will serve our further understanding of this
recently developing, but thermodynamically less inves-
tigated, area of chemically modified cyclodextrins. An-
other aspect of interest of this study is to test and discuss
in further depth the general validity and the significance
of the enthalpy-entropy compensation effect in the
complexation of modified cyclodextrins, using a wider
variety of host-guest combinations.
Recently, we have demonstrated that all of the deriva-
tizations of â-cyclodextrin diminish the complex stability
with 2-naphthalenesulfonate, which are mostly attribut-
able to the highly negative entropy changes (T∆S) that
exceed the increased enthalpic gains (-∆H) arising from
the enhanced hydrophobic interaction with lipophilic side
chain(s) in the modified cyclodextrins. Using the com-
piled thermodynamic parameters reported for most avail-
able host-guest combinations which interact with each
other through weak forces, it was further demonstrated
that the enthalpy (∆H) and entropy (∆S) changes of
complexation are compensatory to each other, displaying
good linear relationship between ∆H and T∆S. The slope
and intercept of the ∆H - T∆S plot are used as quantita-
tive measures of the conformational changes and the
extent of desolvation caused upon complexation, res-
pectively.32a These results prompted us to investigate the
complexation thermodynamics of much wider variety of
modified cyclodextrins with several types of guest mol-
ecules in order to obtain deeper thermodynamic insights
into the factors that govern the complexation phenomena
between molecular receptor (host) and model substrate
(guest).
Exp er im en ta l Section
Gen er a l P r oced u r es. Combustion analyses were per-
formed on a Perkin-Elmer-240 instrument. 1H NMR spectra
were recorded at 200 or 400 MHz in [2H6]dimethyl sulfoxide
(DMSO-d6) on a Bruker AM200 or AM400 spectrometer. FT-
IR and UV spectra were obtained on a Nicolet FT-IR 5DX and
Shimadzu UV-2401PC spectrometer, respectively. Circular
dichroism (CD) spectra were recorded on a J ASCO J -720
spectropolarimeter. A TRONAC model 458 isoperibol titration
calorimeter was used for all of the thermodynamic measure-
ments.
In the present study, we synthesized several series of
R-, â-, and γ-cyclodextrin derivatives, which include novel
R-, â-, and γ-cyclodextrins carrying a positively charged
pyridinio group (4-6, 12, 13), â-cyclodextrin 6-O-mono-
phosphates (7, 8), mono[6-(phenylseleno)-6-deoxy]-â-cy-
clodextrins (9, 10), mono[6-(benzylseleno)-6-deoxy]-â-
cyclodextrin (11), â-cyclodextrin derivatives bearing an
anilino moiety (14-16), mono[6-O-(8-quinolyl)]-â-cyclo-
dextrin (17), and â-cyclodextrin derivatives bearing a
(furfurylideneamino)alkylamino or (9-fluorenylamino)-
Ma ter ia ls. Commercially available L-tryptophan (Tianjin
Chemical Reagent Factory), sodium 3-hydroxy-2-naphthoate
(Aldrich), sodium salts of 2-naphthalenesulfonate and 2,7-
naphthalenedisulfonate (Nakarai) were dried in vacuo prior
to use. R- and γ-cyclodextrins purchased from Nakarai were
used without further purification. â-Cyclodextrin of reagent
grade (Suzhou Monosodium Glutamate Works) was recrystal-
lized twice from water and dried for 12 h in vacuo at 100 °C.
N,N-Dimethylformamide (DMF) was dried over calcium hy-
dride for 2 days and then distilled under a reduced pressure
prior to use. 9-Bromofluorene and diphenyl phosphorochlo-
ridate (DojinDo) and 8-hydroxyquinoline of analytical reagent
grade were used without further purification. Mono[6-(1-
pyridinio)-6-deoxy]-R-, -â-, and -γ-cyclodextrins (4-6) were
synthesized according to the procedures reported by Matsui
et al.33 Mono(6-O-diphenylphosphoro)- (7) and mono(6-O-
ethoxyhydroxyphosphoro)-â-cyclodextrin (8) were synthesized
as reported recently.1a Mono[6-(m-toluidino)-6-deoxy]-â-cyclo-
dextrin (15) and mono[6-[[(9-fluorenylamino)ethyl]amino]-6-
deoxy]-â-cyclodextrin (19) were prepared by the reported
procedures.1c Disodium hydrogen phosphate and sodium
dihydrogen phosphate were dissolved in distilled, deionized
water to make a 0.10 M phosphate buffer solution of pH 7.20
for calorimetric titration.
(21) Ikeda, T.; Yoshida, K.; Schneider, H.-J . J . Am. Chem. Soc. 1995,
117, 1453.
(22) Lehn, J .-M. Angew. Chem., Int. Ed. Engl. 1990, 29, 1304.
(23) Breslow, R. Science 1982, 218, 532.
(24) Dugas, H. Bioorganic Chemistry, 2nd ed.; Springer: New York,
1989.
(25) Rekharsky, M. V.; Goldberg, R. N.; Schwarz, F. P.; Tewari, Y.
B.; Ross, P. D.; Yamashoji, Y.; Inoue, Y. J . Am. Chem. Soc. 1995, 117,
8830.
(26) Tucker, E. E.; Christian, S. D. J . Am. Chem. Soc. 1984, 106,
1942.
(27) (a) Rekharsky, M. V.; Schwarz, F. P.; Tewari, Y. B.; Goldberg,
R. N.; J . Phys. Chem. 1994, 98, 10282. (b) Andini, S.; Castronuovo, G.;
Elia, V.; Gallotta, E. Carbohydr. Res. 1991, 49, 281.
(28) Bastos, M.; Briggner, L.-E.; Shehatta, I.; Wadso¨, I. J . Chem.
Thermodyn. 1990, 22, 1181.
(29) (a) Lewis, F. A.; Hansen, L. D. J . Chem. Soc., Perkin Trans. 2
1973, 2081. (b) Eftink, M. R.; Harrison, J . C. Bioorg. Chem. 1988, 10,
388.
(30) Rekharsky, M. V.; Schwarz, F. P.; Tewari, Y. B.; Goldberg, R.
N.; Tanaka, M.; Yamashoji, Y. J . Phys. Chem. 1994, 98, 4098.
(31) Inoue, Y.; Hakushi, T.; Liu, Y.; Tong, L.-H.; Shen, B.-J .; J in,
D.-S. J . Am. Chem. Soc. 1993, 115, 475.
Syn th esis of Mon o(6-p h en ylselen o-6-d eoxy)-â-cyclo-
d extr in (9). Mono[6-O-(p-toluenesulfonyl)]-â-cyclodextrin was
prepared by a reaction of â-cyclodextrin with p-toluenesulfonyl
chloride in dry pyridine.33 The compound (9) was prepared
by the following procedures. Sodium borohydride (NaBH4) was
added to a yellow solution of diphenyl diselenide34a in anhy-
drous ethanol with stirring under nitrogen at room tempera-
(32) (a) Inoue, Y.; Liu, Y.; Tong, L.-H.; Shen, B.-J .; J in, D.-S. J . Am.
Chem. Soc. 1993, 115, 10637. (b) Liu, Y.; Zhang, Y.-M.; Sun, S.-X.; Li,
Y.-M.; Chen, Y.-T. J . Chem. Soc., Perkin Trans. 2 1997, 1609. (c)
Harata, K.; Tsuda, K.; Uekama, K.; Otagiri, M.; Hirayama, F. J .
Inclusion Phenom. 1988, 6, 135. (d) Liu, Y.; Zhang, Y.-M.; Sun, S.-X.;
Zhang, Z.-H.; Chen, Y.-T. Acta Chim. Sin. (Huaxue Xuebao) 1997, 55,
779.
(33) (a) Matsui, Y.; Okimoto, A. Bull. Chem. Soc. J pn. 1978, 51,
3030. (b) Matsui, Y.; Ogawa, K.; Mikami, S.; Yoshimoto, M.; Mochida,
K. Bull. Chem. Soc. J pn. 1987, 60, 1219. (c) Matsui, Y.; Fujie, M.;
Hanaoka, K. Bull. Chem. Soc. J pn. 1989, 62, 1451
(34) (a) Reich, H. J .; Cohen, M. L.; Clark, P. S. Org. Synth. 1979,
59, 141. (b) Gu¨nther, W. H. H. J . Org. Chem. 1967, 32, 3929.

1446 J . Org. Chem., Vol. 63, No. 5, 1998
Ch a r t 1. Cyclod extr in Der iva tives (Mostly â-Cyclod extr in , u n less Noted Oth er w ise)
Liu et al.
ture. After the solution turned colorless, a solution of mono[6-
O-(p-toluenesulfonyl)]-â-cyclodextrin in DMF was added
dropwise into the solution, heated to 60 °C for 2 h with stirring.
The resulting solution was evaporated under a reduced pres-
sure to give light-yellow powder, which was dissolved in a
minimal amount of hot water and then poured into acetone
(100 mL). The participate was collected by filtration to obtain
white powder (yield 45%): 1H NMR (DMSO-d6, TMS) δ 2.8-
3.9 (m, 40H), 4.0-4.6 (m, 8H), 4.8-5.2 (m, 7H), 5.3-5.8 (m,
14H), 7.1-7.9 (m, ArH); FT-IR (KBr) ν 3386.5, 2914.5, 1731.7,
1698.5, 1629.2, 1600.7, 1558.2, 1537.6, 1513.6, 1397.7, 1363.5,
1302.6, 1273.5, 1232.8, 1170.3, 1149.3, 1073.2, 1023.2, 939.7,
883.5, 830.0, 774.6, 751.5, 726.9, 699.1, 661.6 cm-1
. Anal.
Calcd for C48H74O34Se‚2H2O: C, 43.48; H, 5.93. Found: C,
43.57; H, 5.55.
Syn th esis of Mon o[6-[(p-tolyl)selen o]-6-d eoxy]-â-cy-
clod extr in (10). Compound 10 was prepared from mono[6-
O-(p-toluenesulfonyl)]-â-cyclodextrin and di(p-tolyl) diselenide34a
according to procedures similar to those employed in the
synthesis of 9 (yield 47%). 10: 1H NMR (DMSO-d6, TMS) δ

Molecular Recognition Study on Supramolecular System
J . Org. Chem., Vol. 63, No. 5, 1998 1447
941.4, 859.4, 752.7, 703.5, 572.3 cm-1
Anal. Calcd for
.
Ch a r t 2
C
48H75O34N‚3H2O: C, 45.61; H, 6.4; N, 1.11. Found: C, 45.9;
H, 6.51; N, 1.14.
Syn th esis of Mon o[6-(o-ch lor oa n ilin o)-6-d eoxy]-â-cy-
clod extr in (16). Compound 16 was prepared from mono[6-
O-(p-toluenesulfonyl)]-â-cyclodextrin and o-chloroaniline as
above.1c 16: FAB-MS m/z 1244; 1H NMR (DMSO-d6, TMS) δ
3.2-4.3 (m, 42H), 6.7-7.5 (m, ArH); FT-IR (KBr) ν 3385.9,
2926.6, 1638.7, 1605.7, 1507.4, 1417.2, 1368, 1335.2, 1302.3,
1253.1, 1237, 1154.7, 1080.9, 1031.6, 941.4, 859.4, 752.7, 703.5,
572.3 cm-1. Anal. Calcd for C48H74O34Cl‚3H2O: C, 44.41; H,
6.17; N, 1.08. Found: C, 44.22; H, 6.35; N, 1.02.
Syn th esis of Mon o[6-O-(8-qu in olyl)]-â-cyclodextr in (17).
This compound was prepared by the reaction of mono[6-O-(p-
toluenesulfonyl)]-â-cyclodextrin (2 g) with 8-hydroxyquinoline
(1 g) in the presence of potassium carbonate (0.3 g) in N,N-
dimethylformamide (30 mL) with stirring at 90 °C for 20 h
under a nitrogen atmosphere. A workup procedure similar to
that above gave yellow precipitate. After drying, the precipi-
tate was purified by column chromatography using Sephadex
G-25; 0.93 g (yield 41.4%) of light yellow solid was obtained:
1H NMR (DMSO-d6, TMS) δ 3.2-4.3 (m, â-cyclodextrin pro-
tons); 7.1-9.2 (m, 8-quinolyl protons); FT-IR (KBr) ν 3320,
2910, 2160, 1652, 1625, 1503, 1407, 1370, 1328, 1258, 1200,
2.1 (s, 3H), 3.1-3.8 (m, 40H), 4.0-4.6 (m, 8H), 4.8-5.2 (7H),
5.3-5.8 (m, 14H), 7.1 (t, 4H); FT-IR (KBr) ν 3383.0, 2914.0,
1731.8, 1698.0, 1632.0, 1556.3, 1512.7, 1397.8, 1364.9, 1297.3,
1267.4, 1233.8, 1149.3, 1072.9, 1023.3, 939.6, 890.2, 830.9,
799.4, 772.9, 751.5, 699.7, 660.2 cm-1
.
Anal. Calcd for
49H76O34Se‚2H2O: C, 44.40; H, 6.09. Found: C, 44.32; H,
6.01.
C
1153, 1078, 1025, 940, 860, 820, 788, 752, 710 cm-1
. Anal.
Calcd for C51H75O35N‚6H2O: C, 44.70; H, 6.30; N, 1.02.
Found: C, 44.44; H, 5.96; N, 1.17.
Syn th esis of Mon o(6-ben zylselen o-6-d eoxy)-â-cyclo-
d extr in (11). Compound 11 was prepared from mono[6-O-
(p-toluenesulfonyl)]-â-cyclodextrin and dibenzyl diselenide34b
according to procedures similar to those employed in the
synthesis of 9 (yield 50%). 11: 1H NMR (DMSO-d6, TMS) δ
3.1-3.9 (m, 40H), 4.1-4.6 (m, 8H), 4.8-5.2 (9H), 5.3-5.8 (m,
14H), 7.3 (t, 5H). FT-IR (KBr) ν 3369.0, 2912.5, 1730.9, 1701.5,
1638.6, 1615.4, 1574.6, 1536.7, 1514.6, 1398.6, 1365.0, 1336.5,
1302.9, 1273.9, 1235.1, 1149.4, 1127.1, 1073.8, 1022.3, 938.4,
Syn th esis of Mon o[6-[[(fu r fu r ylid en ea m in o)eth yl]a m i-
n o]-6-d eoxy]-â-cyclod extr in (18). To an ethanol solution
(4 mL) of furfural (0.2 mL) was added an aqueous solution
(10 mL) of mono[6-(ethylenediamino)-6-deoxy]-â-cyclodextrin
(0.5 g)35 with stirring at 0 °C. The mixture was stirred for
additional 1 h and filtrated to give a precipitate, which was
washed twice with ethanol and then subjected to column
chromatography on a Sephadex G-25 to yield the product, 0.35
885.6, 789.7, 769.3, 692.7, 658.7 cm-1
.
Anal. Calcd for
1
g (yield 66%): FAB-MS m/z 1255; H NMR (DMSO-d6, TMS)
C49H76O34Se‚6H2O: C, 42.15; H, 6.35. Found: C, 42.12; H, 6.10.
Syn th esis of Mon o[6-(m -p icolin yl)-6-d eoxy]-â-cyclo-
d extr in (12).33 Mono[6-O-(p-toluenesulfonyl)]-â-cyclodextrin
(2 g) was added to m-picoline (30 mL), and the solution was
refluxed for 12 h under nitrogen atmosphere. The resulting
solution was evaporated under a reduced pressure to give
yellow residue, which was subjected to column chromatogra-
phy on Sephadex G-25 with the elution of aqueous ammonium
hydrogen carbonate solutions (yield: 91%). 12: FAB-MS m/z
δ 7.1-7.4, 7.5-7.8, 8.1-8.2 (furyl); FT-IR (KBr) ν 3385.9,
2926.6, 1638.7, 1556.6, 1409.6, 1354, 1335.2, 1244.9, 1154.7,
1080.9, 1031.6, 941.41, 752.7, 703.5, 580.5 cm-1. Anal. Calcd
for C49H79O35N2‚4H2O: C, 44.18; H, 6.41; N, 1.97. Found: C,
44.18; H, 6.54; N, 1.97.
Syn th esis of Mon o[6-[[[[(9-flu or en yla m in o)eth yl]a m i-
n o]eth yl]a m in o]-6-d eoxy]-â-cyclod extr in (20). A solution
of mono(6-alkylamino-6-deoxy)-â-cyclodextrin35 and 9-bromo-
fluorene in dry DMF (30 mL) was stirred for 10 min under
nitrogen at 0 °C. A solution of triethylamine (10 mL) was
added dropwise into the clear solution in 1 h with stirring
under nitrogen. The solution was allowed to warm and stirred
for additional 6 h at room temperature, and then the resultant
mixture was evaporated under reduced pressure. The residue
was dissolved in water and then poured into acetone to give a
yellow precipitate, which was washed with acetone. The crude
product was purified by the column chromatography on
Sephadex G-25 with the elution of distilled, deionized water:
FAB-MS m/z 1384; 1H NMR (DMSO-d6, TMS) δ 3.58-4.7 (m,
42H), 7.30-7.80 (m, ArH); FT-IR (KBr) ν 3377.7, 2926.6,
1704.3, 1638.7, 1425.4, 1368, 1302.3, 1296.7, 1154.7, 1080.9,
1
1210; H NMR (DMSO-d6, TMS) δ 1.07 (t, CH3), 3.3-3.6 (m,
42H), 8.0 (t, 1H), 8.4 (d, 1H), 8.8 (d, 2H); FT-IR (KBr) ν 3372.5,
2910.0, 1629.5, 1479.2, 1403.4, 1377.9, 1328.9, 1293.4, 1147.6,
1115.9, 1073.8, 1023.0, 937.5, 832.2, 750.6, 699.4, 679.3, 601.2,
567.1 cm-1. Anal. Calcd for C49H77O37N‚3H2O: C, 44.38; H,
6.31; N, 1.06. Found: C, 44.30; H, 6.06; N, 1.35.
Syn th esis of Mon o[6-(p-p icolin yl)-6-d eoxy]-â-cyclod ex-
tr in (13). Compound 13 was synthesized from mono[6-O-(p-
toluenesulfonyl)]-â-cyclodextrin and p-picoline according to the
procedures described above (yield: 89%). 13: FAB-MS m/z
1210; 1H NMR (DMSO-d6, TMS) δ 1.08 (m, CH3), 3.1-3.7 (m,
42H), 7.9 (d, 2H), 8.8 (d, 2H); FT-IR (KBr) ν 3376.0, 2908.0,
1640.0, 1580.2, 1555.8, 1513.1, 1405.8, 1380.3, 1346.8, 1295.9,
1140.1, 1116.6, 1074.2, 1024.6, 935.2, 835.4, 809.9, 748.4,
699.1, 678.3, 604.1, 566.1 cm-1. Anal. Calcd for C49H77O37N:
C, 46.26; H, 6.10; N, 1.10. Found: C, 46.06; H, 5.88; N, 1.50.
Syn th esis of Mon o(6-a n ilin o-6-d eoxy)-â-cyclod extr in
(14). The 6-anilino-â-cyclodextrin was prepared by the reac-
tion of mono[6-O-(p-toluenesulfonyl)]-â-cyclodextrin (2 g) with
aniline (10 mL) in N,N-dimethylformamide (20 mL) at 85 °C
with stirring for 12 h under a nitrogen atmosphere. The
reaction mixture was evaporated in vacuo at 40 °C to dryness.
The residue was dissolved in water, and the resulting mixture
was poured into acetone to give gray precipitate. After drying,
the precipitate was purified by recrystallization from water
twice; 0.85 g (yield 45%) of light yellow solid was obtained:
1H NMR (DMSO-d6, TMS) δ 0.85, 1.23 (m, 1H, N-H), 2.65 (d,
2H, CH2), 3.2-4.4 (m, C-H, O-H), 6.7-7.5 (m, 4H, Ar-H);
FT-IR (KBr) ν 3385.9, 2926.6, 1638.7, 1605.9, 1507.4, 1417.2,
1368.0, 1335.2, 1302.3, 1253.1, 1237.0, 1154.7, 1080.9, 1031.6,
1031.6, 941.4, 851.1, 752.7, 703.5, 580.6, 539.4 cm-1
. Anal.
Calcd for C59H90O34N4‚3H2O: C, 48.76; H, 6.60; N, 3.86.
Found: C, 48.52; H, 6.88; N, 3.51.
Sp ectr a l Mea su r em en ts. Circular dichroism spectra were
measured in a buffer solution (pH 7.20) at room temperature
in order to elucidate the inclusion of the host’s aromatic moiety
into the cyclodextrin cavity.
Ca lor im etr ic Mea su r em en ts. Calorimetric titrations in
an aqueous buffer solution were performed at pH 7.20 in a
temperature-controlled water bath maintained at 25 °C, by
using a TRONAC model 458 isoperibol titration calorimeter
connected to a personal computer for automated titration and
data processing.36 Typically, a solution of 2-naphthalene-
(35) Schneider, H.-J .; Xiao, F. J . Chem. Soc., Perkin Trans. 2 1992,
387.
(36) Liu, Y.; Han, B.-H.; Li, Y.-M.; Chen, Y.-T.; Ouchi, M.; Inoue, Y.
J . Phys. Chem. 1996, 100, 17361.

1448 J . Org. Chem., Vol. 63, No. 5, 1998
Liu et al.
Sch em e 1
sulfonate was continuously introduced at a rate of 0.3321 mL/
min into a solution of cyclodextrin derivatives (4-20) (2-5
mM) placed in the calorimeter. To obtain the net heat of
complexation (Qnet), the total apparent heat observed (Qobs) was
corrected for the dilution of titrant (QD), the nonchemical
contributions (QHL), including agitation, heat flow between the
vessel and its surrounding, and resistance heating by the
thermistor used, and the temperature difference between
titrant and titrate (QTC) in each run: Qnet ) Qobs - QD - QHL
- QTC
.
A titration curve was obtained by plotting the
temperature change (measured by voltage) against the amount
of the aromatic guest solution added, from which the complex
stability constant (K) and the enthalpy change (∆H) were
calculated. Reliability of the whole system and the calculation
procedures were doubly checked as previous37 by comparison
of the obtained thermodynamic data with the reported val-
ues,38,39 and satisfactory results were obtained.
Resu lts a n d Discu ssion
F igu r e 1. Circular dichroism spectrum of â-cyclodextrin
derivative (17) (0.094 mM) in aqueous buffer solution (pH 7.20)
at room temperature.
Syn th esis. As shown in Scheme 1, the modified
cyclodextrins (4-6, 9-20) were synthesized in satisfac-
tory yields starting from 6-O-monotosylcyclodextrin,
while the modified â-cyclodextrins (7, 8) were synthesized
by the esterification of the 6-hydroxyl group of cyclodex-
trin.
CD Sp ectr a . Possessing a chiral cavity, cyclodextrins
can accommodate the chromophoric substituent group
attached to its primary side, which leads to the induced
circular dichroism. As can be judged from the relatively
weak induced circular dichroism (ICD) spectra exempli-
fied in Figures 1 and 2, the aromatic moieties of most
â-cyclodextrin derivatives (5, 9-19) with lipophilic sub-
stituents are deduced to penetrate only shallowly into
(37) Liu, Y.; Tong, L.-H.; Huang, S.; Tian, B.-Z.; Inoue, Y.; Hakushi,
T. J . Phys. Chem. 1990, 94, 2666.
(38) Hansen, L. D.; Lewis, E. R. J . Chem. Thermodyn. 1971, 3, 35.
(39) Inoue, Y.; Hakushi, T.; Liu, Y.; Tong, L.-H.; Hu, J .; Zhao, G.-
D.; Huang, S.; Tian, B.-Z. J . Phys. Chem. 1988, 92, 2371.

Molecular Recognition Study on Supramolecular System
J . Org. Chem., Vol. 63, No. 5, 1998 1449
The stability constant K and the enthalpy change ∆H°
for the complexation of model substances with molecular
receptors (1-20) were calculated by computer simulation
by continuously changing K, i.e., Nt, to minimize the U
value. For each host-guest combination, the measure-
ment was repeated more than three times and the U
value obtained was minimized satisfactorily to give the
optimized set of K and ∆H° with standard deviations. No
serious deviation was found in the fitting process, verify-
ing the 1:1 stoichiometry of complexation as assumed
above. The complex stability constants and thermody-
namic parameters obtained are listed in Table 1. For
comparison purpose, the thermodynamic quantities re-
ported for the complexation with â-cyclodextrin deriva-
tives (21-35) are also included in Table 1.
Bin d in g Con sta n ts a n d Th er m od yn a m ic P a r a m -
eter s. Thermodynamic studies of molecular recognition
by modified cyclodextrins have shown that an important
feature of the inclusion complexation is simultaneous
operation of several weak forces working between guest
molecule and host cyclodextrin, which play important
roles in determining how the guest molecule fits into the
host cavity, according to its size, shape, dipole, charge,
and functional group. Therefore, the attachment of
nucleophiles or electrophiles to cyclodextrins is expected
to affect the inclusion behavior of these modified cyclo-
dextrins. From the ICD spectra, it is deduced that some
functional groups, attached to the edge of cyclodextrin
cavity, are embedded into the cavity of cyclodextrin,
preventing the guest inclusion at least in part. On the
other hand, the hydrophilic functional groups situated
just above the cavity may increase the host hydration
and therefore decrease the microenvironmental hydro-
phobicity to a great extent.32a As can be seen from Table
1, the complex stability constants, relative selectivity, and
thermodynamic parameters for the inclusion complex-
ation of the guest molecules (Chart 2) with modified
cyclodextrins (1-20) are influenced by several factors:
relative size between the cyclodextrin’s cavity to the guest
molecule, induced dipole of functional side arm attached
to the edge of cyclodextrin cavity, spatial conformation,
microenvironmental hydrophobicity, van der Waals, hy-
drogen-bonding interactions, and so on. To visualized the
molecule-binding behavior of cyclodextrins (1-20) from
the thermodynamic point of view, the free energy (-∆G),
enthalpy (-∆H), and entropy changes (T∆S) for the
inclusion complexation of L-tryptophan and the naph-
thalene derivatives are plotted for the host cyclodextrins
(1-20) in Figures 3a-d.
F igu r e 2. Circular dichroism spectrum of â-cyclodextrin
derivative (20) (0.110 mM) in aqueous buffer solution (pH 7.20)
at room temperature.
the hydrophobic cavity. In contrast, more hydrophilic
and/or bulky substituents such as phosphate or 9-fluo-
renyl in 7 or 20 do not appear to be accommodated in
the cavity. Typically, the circular dichroism spectra of
mono[6-O-(8-quinolyl)]-â-cyclodextrin (17) (0.094 mM) in
aqueous solution (Figure 1) shows a negative Cotton
effect peak at a wavelength corresponding to the 1La band
(λext 245.8 nm, ∆ꢀ ) -0.909) and a positive Cotton effect
peak for the 1Lb band (λext 295 nm, ∆ꢀ ) 0.708). The
modified â-cyclodextrin (20) with a fluorenyl group shows
1
two weaker positive Cotton effect peaks for the La band
1
at 305 nm (∆ꢀ ) 0.301) and for the Lb band at 216.6 nm
(∆ꢀ ) 0.242) (Figure 2). According to the sector rule
proposed by Kajtar et al.,40 the Cotton effects observed
for the 1La and 1Lb bands indicate that the 8-quinolyl
moiety of 17 penetrates only shallowly into the hydro-
phobic cavity of cyclodextrin. Similarly the very weak
ICD peaks observed for 20 indicates that the (9-fluore-
nyl)alkylamino group is perching on the edge of cyclo-
dextrin cavity.
Ca lor im etr y. The treatment of the calorimetric data
obtained has been described previously.41,42 Assuming
a 1:1 stoichiometry for the inclusion complexation of some
selected guest molecules (G) with native and modified
cyclodextrins (1-20) (CD), the complex stability constant
(K) and the enthalpy change (∆H°) were determined
simultaneously by using the least-squares method to
minimize the error square sum (U):43,44
As can be recognized more easily from Figures 3a-d,
the ∆H values for the inclusion complexation of cyclo-
dextrins (1-20) with the guests are all negative and
stabilizing, whereas the T∆S values are either negative
or slightly positive. This means that these inclusion
reactions are primarily enthalpy-driven processes, al-
though some host-guest combinations, including 2, 5,
8, and 13 as hosts, are evidently assisted by the entropy
term to a great extent. For the inclusion complexation
of L-tryptophan, 3-hydroxy-2-naphthoate, 2-naphthale-
nesulfonate, and 2,7-naphthalenedisulfonate with cyclo-
dextrins, the enthalpic (-∆H) and entropic (T∆S) con-
tributions to the complex stability (-∆G) are discussed
below.
G + CD y
\
Kz G‚CD
U(K,∆H°) ) (Q - ∆H°× N )2
(1)
(2)
m
∑
t
t
t)1
where Qt refers to the net heat of complexation measured
at time t in minutes and Nt denotes the amount in moles
of the complex formed at time t and is directly related to
the complex stability constant K.
(40) Kajtar, M.; Harvoth-Toro, C.; Kuthi, E.; Szejtli, J . Acta Chin.
Acad. Sci. Hung. 1982, 110, 327.
(41) Liu, Y.; Wang, Y.; Guo, Z.-Q.; Yang, S.-Y.; J in, D.-S. Acta Chim.
Sin. (Huaxue Xuebao) 1986, 44, 22.
(42) De J ong, F.; Reinhoudt, D. N. Stability and Reactivity of Crown
Ether Complexes; Academic Press: New York, 1981.
Com p lexa tion of L-Tr yp top h a n . As can be seen
from Table 1 and Figure 3a, the size-fitted combination
of L-tryptophan with â-cyclodextrin (2) gives the strongest

1450 J . Org. Chem., Vol. 63, No. 5, 1998
Liu et al.
Ta ble 1. Com p lex Sta bility Con sta n t (K) a n d Th er m od yn a m ic P a r a m eter s (in k ca l/m ol) for 1:1 Host-Gu est
Com p lexa tion of Som e Ar om a tic Gu est Molecu les w ith Cyclod extr in s a n d Mod ified Cyclod extr in s (1-20) in Aqu eou s
Solu tion (T ) 298 K)
host
guest
log K
-∆G°
-∆H°
T∆S°
ref
1
2
3
4
5
6
7
8
9
10
11
2
5
7
1
2
3
4
L-tryptophan
1.79 ( 0.03
3.26 ( 0.04
1.95 ( 0.07
2.39 ( 0.05
3.71 ( 0.03
2.22 ( 0.01
1.98 ( 0.08
3.36 ( 0.06
2.07 ( 0.03
2.24 ( 0.04
2.02 ( 0.02
2.15 ( 0.02
2.88 ( 0.08
2.34 ( 0.02
0.98 ( 0.06
2.44 ( 0.02
2.58 ( 0.02
2.56 ( 0.07
2.01 ( 0.04
1.92 ( 0.06
3.68 ( 0.03
3.86 ( 0.06
1.92 ( 0.05
2.42 ( 0.11
2.54 ( 0.05
2.64 ( 0.04
2.69 ( 0.08
5.37 ( 0.07
3.04 ( 0.03
2.78 ( 0.01
2.84 ( 0.01
2.03 ( 0.05
2.56 ( 0.01
3.51 ( 0.08
2.58 ( 0.06
2.14 ( 0.04
2.53 ( 0.07
3.61 ( 0.08
2.94 ( 0.03
3.13 ( 0.02
2.50 ( 0.03
3.13 ( 0.10
3.42 ( 0.02
3.21 ( 0.03
3.81 ( 0.06
3.93 ( 0.05
2.95 ( 0.02
3.58 ( 0.05
3.62 ( 0.02
1.85 ( 0.04
3.94 ( 0.02
2.74 ( 0.06
2.74
2.44
4.45
2.66
3.26
5.06
3.02
2.70
4.58
2.82
3.06
2.76
2.93
3.92
3.19
1.34
3.33
3.52
3.49
2.74
2.69
5.02
5.27
2.62
3.30
3.47
3.60
3.67
7.33
4.15
3.79
3.88
2.77
3.49
4.79
3.52
2.92
3.45
4.93
4.01
4.27
3.41
4.27
4.67
4.38
5.20
5.36
4.02
4.88
4.94
2.52
5.37
3.74
3.73
4.17
4.07
3.85
3.91
4.16
4.74
4.60
4.35
4.86
4.54
5.08
4.55
4.30
3.98
4.08
4.97
4.91
4.57
4.28
4.17
4.99
2.32 ( 0.07
1.11 ( 0.06
5.18 ( 0.11
2.64 ( 0.12
1.06 ( 0.05
5.71 ( 0.05
7.61 ( 0.07
1.47 ( 0.09
8.20 ( 0.13
2.68 ( 0.11
2.40 ( 0.05
10.45 ( 0.12
10.90 ( 0.13
7.46 ( 0.06
5.99 ( 0.05
6.75 ( 0.08
0.86 ( 0.01
1.87 ( 0.08
9.23 ( 0.05
4.42 ( 0.04
2.62 ( 0.05
1.98 ( 0.06
7.54 ( 0.02
6.89 ( 0.09
6.35 ( 0.07
6.34 ( 0.12
4.47 ( 0.09
7.01 ( 0.06
8.20 ( 0.11
9.61 ( 0.08
9.07 ( 0.07
6.31 ( 0.09
4.52 ( 0.13
4.07 ( 0.09
3.08 ( 0.09
10.05 ( 0.06
4.60 ( 0.13
3.46 ( 0.05
2.89 ( 0.08
13.75 ( 0.09
10.30 ( 0.10
7.19 ( 0.08
4.00 ( 0.09
8.01 ( 0.08
3.29 ( 0.04
6.63 ( 0.04
6.75 ( 0.06
10.90 ( 0.06
10.72 ( 0.06
20.39 ( 0.25
2.99 ( 0.07
3.88 ( 0.08
16.4
0.12
3.34
-2.52
0.62
a
a
a
a
a
a
a
a
a
a
a
a
a
a
b
b
b
a
a
a
a
a
a
a
a
a
a
b
a
a
a
a
a
a
a
a
a
a
c
c
c
c
c
c
c
c
c
c
c
c
c
c
d
d
d
d
d
d,e
d
f
4.00
-2.69
-4.91
3.11
-5.38
0.38
0.36
sodium 3-hydroxy-2-naphthoate
-7.52
-6.98
-4.27
-4.65
-3.42
2.66
disodium 2,7-naphthalenedisulfonate
1.62
5
6
-6.49
-1.73
2.40
12
13
14
15
17
19
20
2
3.29
-4.92
-3.59
-2.88
-2.74
-0.80
0.32
-4.05
-5.82
-5.19
-3.54
-1.03
0.72
sodium 2-naphthalenesulfonate
5
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
0.44
-7.13
-1.15
1.47
1.12
-9.48
-6.89
-2.92
0.67
-3.63
1.91
-1.27
-2.73
-6.02
-5.78
-17.87
2.38
-0.14
-12.8
-13.1
-10.1
-6.3
benzoic acid
m-hydroxybenzoic acid
p-hydroxybenzoic acid
m-nitrophenol
p-nitrophenol
m-nitroaniline
p-nitroaniline
L-alanine
D-alanine
L-valine
D-valine
L-isoleucine
L-proline
L-serine
D-serine
L-cysteine
L-aspartic acid
L-leucine
D-leucine
L-alanine
D-alanine
L-valine
3.06
2.98
2.82
2.87
3.05
3.48
3.37
3.19
3.56
3.33
3.72
3.34
3.15
2.92
2.99
3.65
17.3
14.4
10.2
14.9
11.5
14.4
5.88
3.92
-4.23
2.42
17.52
-3.07
6.08
6.62
7.28
35
4
-11.0
-8.1
-9.8
-1.30
0.44
f
f
f
f
f
f
f
f
f
f
f
g
g
g
9.09
2.12
-12.40
7.62
-1.76
-2.63
-3.19
6.23
-1.25
1.28
8.85
8.49
4.89
0.92
3.60
3.35
3.14
3.06
3.63
-4.26
-4.21
-0.72
4.07
5
3.65

Molecular Recognition Study on Supramolecular System
J . Org. Chem., Vol. 63, No. 5, 1998 1451
Ta ble 1 (Con tin u ed )
host
guest
D-valine
L-isoleucine
L-proline
L-serine
D-serine
L-cysteine
L-aspartic acid
L-leucine
D-leucine
L-lysine
L-methionine
L-glutamic acid
L-alanine
D-alanine
L-valine
log K
-∆G°
-∆H°
T∆S°
ref
3.34
3.90
3.08
3.07
2.82
2.43
3.61
3.69
3.46
2.31
2.52
2.88
3.03
2.96
3.46
3.65
3.06
3.00
2.77
2.69
3.58
3.50
3.22
3.62
3.52
3.82
3.79
3.64
3.49
3.43
2.65
4.17
2.57
4.04
3.70
3.67
4.56
5.33
4.19
4.18
3.85
3.31
4.92
5.03
4.72
3.15
3.80
3.92
4.13
4.03
4.72
4.98
4.18
4.09
3.78
3.67
4.89
4.77
4.40
4.93
4.80
5.21
5.17
4.96
4.74
4.67
3.61
5.69
3.49
5.47
5.05
5.01
-10.19
1.52
-1.15
9.38
14.76
3.73
5.34
-5.17
-8.34
5.61
7.73
3.36
9.68
8.03
13.20
7.10
-6.22
0.58
0.42
8.72
11.18
-1.50
-1.10
8.19
g
g
g
g
g
g
g
g
g
g
g
g
f
f
f
f
f
f
f
f
f
12.74
-2.29
-2.21
1.67
-5.02
-4.88
-9.50
-3.18
10.49
3.46
6
4.25
L-isoleucine
L-proline
L-serine
-3.74
-6.70
5.60
D-serine
4.88
L-cysteine
L-aspartic acid
L-leucine
D-leucine
L-alanine
D-alanine
L-valine
L-isoleucine
L-proline
L-serine
-4.52
-2.64
1.84
3.84
-8.02
7.27
9.59
11.38
4.02
-6.28
-5.14
15.75
17.06
-7.17
7.21
7.53
2.93
0.57
f
f
14
12.98
-2.48
-4.38
-6.17
0.95
11.02
10.03
-12.08
-11.31
10.66
-1.70
12.21
1.83
g
g
g
g
g
g
g
g
g
g
g
g
g
D-serine
L-cysteine
L-glutamic acid
L-lysine
L-aspartic acid
L-leucine
-7.16
3.18
D-leucine
a
b
This work; calorimetric titrations performed at 25.0 °C in a buffered aqueous solution at pH 7.20. Reference 31. c Reference 32a.
Reference 32c, thermodynamic parameters determined by variable-temperature circular dichroism spectrometer at 20-60 °C. e Data
d
not used in the plot, since ∆G * ∆H - T∆S within an error limit of 0.2 kcal/mol. f Reference 32d (VT UV-vis). Reference 32b (VT
g
UV-vis).
inclusion complex (log K ) 3.26) among the native R- to
γ-cyclodextrins and forms a typical entropy-driven com-
plex. Thus, the complex is inferred to be stabilized by
the entropic gain arising from the desolvation of the host
and guest molecules upon complexation, as well as the
release of water molecules trapped originally in the
cavity. Both the smaller R-cyclodextrin and larger γ-cy-
clodextrin give the poor inclusion abilities forming en-
thalpy-driven complexes with accompanying negligible
or negative entropic contribution. The decreases in the
complex stability constants for hosts 1 and 3 (log K )
1.79 and 1.95) are mainly attributable to the reduced or
negative entropic contribution, probably due to the less
extensive desolvation. Although R- and γ-cyclodextrins
incidentally afford comparable K values, their thermo-
dynamic profiles are completely different.
As compared with the thermodynamic quantities ob-
tained with parent R-, â-, and γ-cyclodextrins (1-3), the
modified R-, â-, and γ-cyclodextrins (4-6), bearing a
positively charged pyridinio group, are expected to exhibit
increased complex stability through the electrostatic
interaction by ion-pairing between N+ and the carboxyl-
ate of tryptophan accommodated in the cyclodextrin
cavity. Indeed, the complex stability constants K for 4-6
are larger by roughly 0.3-0.6 order of magnitude than
those for the parent cyclodextrins 1-3 as a consequence
of the increased enthalpic and/or entropic gain.
phosphates 7 and 8 gave contrasting results. Although
the thermodynamic parameters obtained for the hydro-
gen phosphate 8 are almost identical with those for the
parent 2, the introduction of a diphenyl phosphate group
leads to much weaker complexation between tryptophan
and 7, which is evidently enthalpy-driven but the en-
thalpic gain (-∆H° 4.61) is exceedingly canceled out by
the accompanying entropic loss (T∆S° -4.91). This large
entropic loss may be attributable to the drastic confor-
mational changes upon extrusion of one of the phos-
phate’s phenyl groups originally perching just above the
cavity, which is suggested by the ICD studies.
A series of structurally related arylseleno derivatives
9-11 showed quite interesting and contrasting complex-
ation behavior and affords some insights into the struc-
tural requirements for self-inclusion and its thermody-
namic consequences. As can be seen from Table 1, all
the seleno derivatives give almost comparable complex
stability constants around 2.0-2.2, but the thermody-
namic profiles differ strikingly between the phenylseleno
derivative 9 and the p-tolyl- or benzylseleno derivative
(10 or 11). Beyond the apparent differences in structure,
the thermodynamic behavior of 9 is similar to that of 7,
i.e., the large enthalpic gain accompanied by the similarly
large entropic loss. This may be attributable to the fact
that both modified cyclodextrins possess a phenyl group
which is connected to the cyclodextrin through a short
linkage and is perching just above the cavity. On the
other hand, the tolyl and benzyl groups in 10 and 11 in
Possessing the same host moiety and a similar phos-
phate group at the peripheral position, â-cyclodextrin

1452 J . Org. Chem., Vol. 63, No. 5, 1998
Liu et al.
Com p lexa tion of 2,7-Na p h th a len ed isu lfon a te. As
shown in Table 1, the effects of cyclodextrin’s cavity size
on the thermodynamic parameters are best illustrated
by the complexation of 2,7-naphthalenedisulfonate with
the parent R-, â-, and γ-cyclodextrins.31 In sharp contrast
to this seeming reasonable result, the pyridinio cyclo-
dextrins 4-6 show an opposite profile of the complex
stability sequence which decreases monotonically with
increasing cavity size.
Although the pyridinio â-cyclodextrin 5 gives the
largest enthalpic gain (-∆H°) in the series as is the case
with the parent 2, very drastic changes in the thermo-
dynamic parameters are induced by appending a pyri-
dinio group to R- and γ-cyclodextrins yielding 4 and 6.
The enthalpic gain for 4 (-∆H° 1.87) is much smaller
than that for 1 (5.99), but the increased entropic gain
(T∆S°) from -4.65 to 1.62 exceeds the loss in enthalpy,
affording a higher complex stability for 4. The small
negative ∆H° and positive T∆S° values for 4 jointly
suggest that the inclusion interaction is very weak but
the desolvation is rather extensive. This kind of situation
may be realized in the case of simple ion pairing without
intimate inclusion of hydrophilic disulfonate with the
positively charged guest like 4. By contrast, γ-cyclodex-
trin derivative 6 affords more negative ∆H° and T∆S°
values than the parent 3. This corresponds to stronger
inclusion and structural freezing, probably arising from
the through-cavity complexation with two sulfonate
moieties interacting with the peripheral hydroxyl and
positively charged pyridinio groups.
The introduction of a methyl group to the pyridinio
moiety of 5 does enhance the original log K of 2.01 for 5
up to 3.7-3.9 for 12 and 13, for which the dramatic
increases of the T∆S° value are responsible. Although
the origin of the positive entropy changes are not neces-
sarily clear, one possible rationalization for the enhanced
stability for 12 and 13 is the intervention of CH-π
interaction between the electron-deficient methyl group
on the positively charged pyridinio group and the guest’s
naphthyl group accommodated in the cavity.45 It is also
totally unexpected that, irrespective of the diversity of
the substituents introduced, the more sophisticated â-cy-
clodextrin derivatives 14, 15, 17, and 20 show quite
comparable thermodynamic quantities, which are inci-
dentally fairly similar to those for the parent â-cyclodex-
trin 2. Thus, thermodynamically these modifications do
not very much affect the complexation behavior of â-cy-
clodextrin.
Com p lexa tion of 2-Na p h th a len esu lfon a te. As has
been demonstrated previously for the inclusion complex-
ation of 2-naphthalenesulfonate with the modified â-cy-
clodextrins 21-34,32a all the present modifications ap-
plied to â-cyclodextrin lead to substantially reduced
complex stabilities for 5 and 12-20. Even pyridinio (5,
12, 13), quinolyl (17), and fluorenyl cyclodextrins (19, 20),
which give enhanced binding of L-tryptophan, 3-hydroxy-
2-naphthoate, and 2,7-naphthalenedisulfonate, never
promote the binding of 2-naphthalenesulfonate, but
rather reduce the binding constant by 2-3 orders of
magnitude, as compared with the parent cyclodextrin (2).
F igu r e 3. Free energy (-∆G), enthalpy (-∆H), and entropy
changes (T∆S) for the inclusion complexation of L-tryptophan
(a), 3-hydroxy-2-naphthoate (b), 2-naphthalenesulfonate (c),
and 2,7-naphthalenedisulfonate (d) with cyclodextrins (1-20)
in a buffered aqueous solution (pH 7.20) at 25 °C.
conjunction with the much longer C-Se bonds would lead
to their tighter self-inclusion, which discourages the
replacing inclusion of the guest tryptophan.
Com p lexa tion of 3-Hyd r oxy-2-n a p h th oa te. Pos-
sessing a more rigid planar structure as compared with
that of L-tryptophan, 3-hydroxy-2-naphthoate as a guest
gave stronger hydrophobic and hydrogen-bonding inter-
actions upon complexation with 2, 5, and 7, forming
typical enthalpy-driven complexes with the large negative
entropy changes. However, as can be seen from Table 1
and Figure 3b, the larger enthalpic gains (-∆H°) do not
immediately mean high complex stabilities and are
canceled out by the large entropic loss arising from the
structural freezing. It is noted that the modified â-cy-
clodextrins 5 and 7 show higher complex stabilities (log
K ) 2.88 and 2.34) than the parent â-cyclodextrin (log K
) 2.15), probably through the extra electrostatic and/or
hydrogen-bonding interactions.
(43) Christensen, J . J .; Ruchman, J .; Eatough, D. J .; Izatt, R. M.
Thermochim. Acta 1972, 3, 203.
(44) Eatough, D. J .; Christensen, J . J .; Izatt, R. M. Thermochim.
Acta 1972, 3, 219.
(45) Nishio, M.; Umezawa, Y.; Hirata, M.; Takeuchi, Y. Tetrahedron
1995, 51, 8665.

Molecular Recognition Study on Supramolecular System
J . Org. Chem., Vol. 63, No. 5, 1998 1453
Close examinations of the thermodynamic quantities,
however, reveal some interesting features of the inclusion
complexation of modified â-cyclodextrins with 2-naph-
thalenesulfonate as a guest molecule. Possessing a
positively charged group linked to the edge of cyclodextrin
cavity through a short chain, the host compounds 5, 12,
and 13 appear to develop the hydrophobic and electro-
static interactions between cyclodextrin and 2-naphtha-
lenesulfonate to give much higher enthalpic gains (-∆H°),
which are however entirely canceled out by the entropic
loss arising from the more rigid structural fixation in the
inclusion complex produced, resulting in the much lower
stability constants (log K ) 2.78-3.04) than that for 2
(5.37). On the other hand, the anilino or quinolyl moiety
in host cyclodextrins 14-17, as revealed by the ICD
study, is only shallowly accommodated in the cavity of
free host and the thermodynamic parameters for these
hosts resemble to those for the methylated cyclodextrin
21 and the bridged cyclodextrins 33 and 34 reported
previously.31 The mutually resembling thermodynamic
parameters for these apparently very diverse types of
modified cyclodextrins suggest some common behavior
occurring upon their inclusion complexation. Linked
with a short chain to cyclodextrin, the pendent or
bridging group can exclude only partly the bound water
molecules residing originally in the cavity and therefore
the inclusion of guest by these hosts may become similar.
En th a lp y-En tr op y Com p en sa tion . We have pre-
viously demonstrated that the thermodynamic param-
eters for the inclusion complexation of various neutral
and charged organic guests with native R- to γ-cyclodex-
trins exhibit a compensatory relationship between ∆H
and T∆S.31 The slope (R) and intercept (T∆S0) of the
regression line of the ∆H - T∆S plot, as measures of the
conformational changes and the extent of desolvation
caused by inclusion,46 are 0.90 and 3.1, respectively,
which indicate fairly large conformational changes caused
by the rearrangement of the hydrogen-bond network
around the rim and the moderate dehydration mostly
from the peripheral hydroxyl groups. More recently,32a
we have studied the complexation thermodynamics of
2-naphthalenesulfonate with some cyclodextrin deriva-
tives (21-35), the results of which afford an excellent
∆H - T∆S compensation plot of larger slope (R ) 1.07)
and intercept (T∆S0 5.0) than those for the native
cyclodextrins.
F igu r e 4. Enthalpy-entropy compensation plot for inclusion
complexation with modified R-, â- and γ-cyclodextrins (4-35);
see Table 1 for the original data.
Ta ble 2. Slop e (r) a n d In ter cep t (T∆S0) of th e ∆H - T∆S
P lots for 1:1 Host-Gu est Com p lexa tion in Hom ogen eou s
Solu tion
host
glyme/podand
crown ether
cryptand
bis(crown ether)
ionophore antibiotics
cyclodextrin
guest
a
T∆S0
ref
cation
cation
cation
cation
cation
molecule
0.86
0.76
0.51
1.03
0.95
0.90
1.02
2.3
2.4
4.0
4.6
5.6
3.1
4.3
a
a
a
a
a
b
c
modified cyclodextrin (4-35) molecule
a
b
Reference 37. Reference 31. c This work.
representative cation binders such as glyme/podand,
crown ether, bis(crown ether), cryptand, ionophore an-
tibiotics, and the parent cyclodextrins.
For the series of acyclic, cyclic, and bicyclic cation
binders, the slope (R) decreases with increasing host’s
dimensionality or structural rigidity from 0.86 (glyme/
podand) to 0.76 (crown ether) and then to 0.51 (cryptand),
while the intercept (T∆S0) increases in this order from
2.3 to 4.0 with increasing coverage of the cation surface
by ligand’s donor. The bis(crown ethers) and ionophore
antibiotics give much larger slope near unity and inter-
cept around 5, indicating substantial conformatinal
changes and extensive desolvation upon complexation.
In this context, the slope of unity and the intercept of
4.3, which are incidentally very similar to those for bis-
(crown ethers) irrespective of the weak interaction in-
volved, reveal that the inclusion complexation with
modified cyclodextrins accompanies substantial changes
in conformation and solvation. This seems reasonable,
since the appended side arm, either self-included or
excluded in the cavity, is considered to change its
conformation upon guest inclusion.
The present study enables us to examine more pre-
cisely and comprehensively the general validity of the
enthalpy-entropy compensation effect in the complex-
ation thermodynamics with a wider variety of both guests
and modified cyclodextrins. Using the compiled thermo-
dynamic quantities listed in Table 1 (number of data set
(n) ) 101), the entropy change (T∆S) was plotted against
the enthalpy change (∆H) to give an excellent regression
line (correlation coefficient (r) ) 0.993) of a large slope
(R ) 1.02) and intercept (T∆S0 4.3), Figure 4. The
obtained extrathermodynamic parameters, R and T∆S0,
are in good agreement with the parameters calculated
previously with a smaller sample size32a but are much
larger than those for native cyclodextrins.31 For com-
parison purpose, the slopes and intercepts are sum-
marized in Table 2, along with those obtained for the
On the other hand, the larger intercept (T∆S0) indi-
cates that the guest molecule accommodated in the cavity
suffers extensive desolvation, which is characteristic of
the three-dimensional complexation by cryptand, anti-
biotic, bis(crown ether),37 and modified cyclodextrin,
although the species and the weak interactions involved
are completely different. Quantitatively, the unit slope
(R) and the large intrinsic entropic gain (T∆S0) indicate
that the enthalpic gain from guest inclusion is completely
canceled out by the entropic loss arising from the modi-
(46) Inoue, Y.; Hakushi, T.; Liu, Y. In Cation Binding by Macro-
cycles; Inoue, Y., Gokel, G. W., Eds.; Marcel Dekker: New York, 1990;
Chapter 1.

1454 J . Org. Chem., Vol. 63, No. 5, 1998
Liu et al.
fied cyclodextrin’s conformational change required upon
complexation, but the inclusion complexation is still
favored even in the absence of enthalpic stabilization
exclusively owing to the entropic gain originating from
the extensive desolvation of both guest and host. En-
deavors to measure and compile the thermodynamic
quantities in the homogeneous phase for complexation
of various guest molecules with hosts of other category
with different topologies are currently in progress.
Ack n ow led gm en t. This work was supported by
National Outstanding Youth Fund (Grant 29625203)
and Natural Science Foundation (Grant 29672021) of
China, Trans-Century Training Program Foundation for
the Talents of the State Education Commission of
China, and Natural Science Fund of Tianjin Municipal-
ity (Grants 953604811 and 973602211), which are
gratefully acknowledged.
J O971466B