Synthesis of organoselenium-modified β-cyclodextrins possessing a 1,2-benzisoselenazol-3(2H)-one moiety and their enzyme-mimic study

Article abstract of DOI:10.1002/1522-2675(200201)85:1<9::AID-HLCA9>3.0.CO;2-H

Six novel H₂O-soluble β-cyclodextrin derivatives containing a 1,2-benzisoselenazol-3(2H)-one moiety were synthesized by a convenient method in 25-60% yield and characterized by MS, elemental analysis, IR, ¹H-NMR, and UV/VIS spectrosc

Full text of DOI:10.1002/1522-2675(200201)85:1<9::AID-HLCA9>3.0.CO;2-H

Helvetica Chimica Acta Vol. 85 (2002)
9
Synthesis of Organoselenium-Modified b-Cyclodextrins Possessing a
1,2-Benzisoselenazol-3(2H)-one Moiety and Their Enzyme-Mimic Study
by Yu Liu*, Bin Li, Li Li, and Heng-Yi Zhang
Department of Chemistry, Nankai University, 300071 Tianjin, China
Six novelH 2O-soluble b-cyclodextrin derivatives containing a 1,2-benzisoselenazol-3(2H)-one moiety were
synthesized by a convenient method in 25 60% yield and characterized by MS, elemental analysis, IR,
1H-NMR, and UV/VIS spectroscopy. The conformations of these b-cyclodextrin derivatives 1 6 were analyzed
by circular dichroism and fluorescence-lifetime experiments. The superoxide dismutase (SOD) activities of 1
6
were determined by auto-oxidation of pyrogallol at 25.08 in buffer solution (pH 8.2), giving relatively high SOD
activities of up to 121 330 U/mg. Also, the glutathione peroxidase (GPX) activities of hosts 1 6, determined
by the method of Wilson at 378 in buffer solution (pH 7.0), show good GPX activities in the range of 0.34
0.86 U/mmol. The mimicking results of the bifunctional artificial enzyme models 1 6 were globally compared
with regard to their structuraland conformationaldifference.
Introduction. Imitation of the highly effective catalysis by natural enzymes is
currently a significant topic in chemistry and biochemistry [1][2]. A good artificial
enzyme model usually has two prerequisites: one is the catalytically active center, the
other is a substrate-binding-site group. Cyclodextrins possessing well-defined hydro-
phobic cavities can selectively bind various guests to form host-guest complexes or
supramolecular species, which have been employed successfully as excellent models to
mimic various enzymes [3 6]. A generalmethod to design an effective enzyme model
based on a cyclodextrin is the introduction of a suitably modified moiety at the primary
or secondary rim of the cyclodextrin cavity, which can act as the catalytic functional
group in an artificialmimic, acceel rating the reactions of substrates accommodated in
the cyclodextrin cavity. Recently, the biomimetic reactions catalyzed by cyclodextrins
and their derivatives have been reviewed comprehensively by Breslow and Dong [6].
However, artificialimitation of superoxide dismutase and glutathione peroxidase with
cyclodextrin derivatives has rarely been reported, except for the recent studies by Ni
and co-workers [7] and Shen and co-workers [8].
Superoxide dismutases (SODs, EC 1.15.1.1) are metalloenzymes that can protect
.
cells from the highly reactive O2À radicalby acceelrating the disproportionation
.
‡
reaction 2 O2À ‡ 2 H ! H2O2 ‡ O2. Glutathione peroxidase (GPX, EC 1.11.1.9) is a
selenoenzyme, which can prevent biomembranes from oxidative damage by catalyzing
the reaction 2 GSH ‡ ROOH ! ROH ‡ GSSG (GSH ˆ glutathione). Both enzymes
belong to the antioxidative enzyme family that can cooperatively compose the
antioxidative defence system to dispose of free radicals and reactive oxygen species,
which may be the main causes of aging, cardiovascular, tumorous, and some endemic
diseases [9]. To obtain novelmedicines, a great dealof effort has been devoted to the
artificial imitation of GPX [10 14]. Some small molecular mimics like ebselen (PZ51)
have been used as successfulcilnicalmedicines. So far, the most effective mimics are

10
Helvetica Chimica Acta Vol. 85 (2002)
Wilson×s models, the GPX activities of which are approximately 10-fold that of PZ51
[12]. Recently, Shen and co-workers [8] reported that the GPX activity of a selenium-
bridged bis(b-cyclodextrin) mimic is also up to 4.3 times that of PZ51. The common
feature of these GPX mimics is that there exists a catalytic cycle of selenium in the
catalysis procedure [12]. However, the artificial imitations of SOD have mainly focused
on metalcomplexes of the macrocyclic compounds [15]. Extensive studies toward SOD
imitation demonstrated that there is a catalytic cycle of copper in the disproportio-
nation reaction catalyzed by natural Cu, ZnÀSOD [16]. No report has appeared about
imitating SOD by organoselenium-modified cyclodextrin derivatives.
Although ebselen possessing a high GPX activity has been employed as an excellent
selenoenzyme model and an effective drug, its low solubility in H2O limits its utility. In
the present study, we introduced the 1,2-benzisoselenazol-3(2H)-one moiety at the
primary rim of b-cyclodextrin, thus obtaining the bifunctional enzyme models 1 6
which can simultaneously imitate SOD and GPX. As expected, the synthesized mimics
1 6 display excellent solubility in H2O, with the exception of 5. The SOD and GPX
activities of these modified b-cyclodextrins were determined by autooxidation of
pyrogallol at 25.08 in buffer solution (pH 8.2) and as described by Wilson et al. [12] at
378 in buffer solution (pH 7.0), respectively. The mimicking results of the bifunctional

Helvetica Chimica Acta Vol. 85 (2002)
11
artificialenzyme modesl were discussed with regard to their structuraland conforma-
tionaldifferences.
Results and Discussion. Synthesis. The modified b-cyclodextrins 1 4 and 6 were
synthesized in satisfactory yields from 6-O-monotosyl-b-cyclodextrin, while the
modified b-cyclodextrin derivative 5 was obtained by esterification of the 6-hydroxy
group of b-cyclodextrin (see Scheme).
Conformational Analysis by Means of CD Spectra and Fluorescence Lifetimes. As
can be seen from the Figure, the induced circular dichroism (ICD) spectra of modified
Scheme

12
Helvetica Chimica Acta Vol. 85 (2002)
Figure. Circular dichroism of mimics 1 6 (0.1 mm) in phosphate buffer solution (pH 7.2) at 258
b-cyclodextrins 1 4 and 6 in buffer solution show two positive Cotton effects at
around 232 246 nm (De ˆ 0.51 25.56) and 259 264 nm (De ˆ 0.50 10.21). Accord-
ing to the results of Harata and Uedaira [17] and Kodaka [18], these positive Cotton
effects indicate that the 1,2-benzisoselenazol-3(2H)-one moiety lies outside the chiral
cavity of b-cyclodextrin, and both of the corresponding transition moments may be
perpendicular to the axis of b-cyclodextrin cavity. In contrast, the ICD spectrum of b-
cyclodextrin derivative 5 displays one negative Cotton effect at 229 nm (De ˆ À1.59)
and a positive Cotton effect at 248 nm (De ˆ 2.70). It may be inferred that the
modifying moiety is located inside the chiral cavity of b-cyclodextrin, thus the transition
moments at 229 nm may be perpendicular to the axis of the b-cyclodextrin cavity, and
the transition moments at 248 nm may be in parallel alignment [19].
To further analyze and confirm the conformations of the modifying moiety of 1 6,
time-resolved fluorescence decay of 8-anilinonaphthalene-1-sulfonate (ANS) was
investigated in the presence and absence of the b-cyclodextrin derivatives. Since the
rates of complexation/decomplexation are much lower than that of the fluorescence
decay, the decay profile of fluorescence intensity (F(t)) can be described by Eqn. 1 as a
sum of unimolecular decays for all fluorescing species present in the solution, where Ai
and ti represent the initialabundance and ilfetime of the ith fluorescing species.
n
X
F(t) ˆ
Ai exp (Àt/ti) (n ˆ 1, 2, etc.)
(1)
iˆ1

Helvetica Chimica Acta Vol. 85 (2002)
13
In the absence of the host, the observed decay profile of ANS fluorescence is
absolutely single-exponential in the aqueous phosphate buffer, giving a short lifetime of
0.4 ns for free ANS in the bulk solution [20]. In contrast, the decay profile in the
presence of native b-cyclodextrin or modified b-cyclodextrins 1 4 and 6 is successfully
analyzed only by a linear combination of two exponential functions, affording a short
(0.5 1.5 ns) and a long (3.1 8.6 ns) lifetime assigned to free and included ANS,
respectively. However, the decay curve obtained with added 5 is well-fitted to a single-
exponential function, giving short lifetimes (0.6 ns). The fluorescence lifetimes and
relative quantum yields thus obtained are summarized in Table 1. The longer lifetime of
ANS (3.1 8.6 ns) in the presence of native b-cyclodextrin or mimics 1 4 and 6 is
reasonably accounted for in terms of a more hydrophobic environment around the
ANS molecule included in the cavity. Since the hydrophobic moieties of these modified
b-cyclodextrins may enhance the microenvironmental hydrophobicity of host com-
pounds, the elongated lifetimes (7.2 8.6 ns) of ANS in the presence of modified
cyclodextrins are much longer than that in the presence of native b-cyclodextrin (3.1
3.2 ns). Interestingly, the addition of an excess amount of 5 does not appreciably alter
the original lifetime (0.4 ns) obtained in the bulk solution. This somewhat unexpected
result clearly indicates that the modified group self-included in the cavity of 5 prevents
the inclusion of ANS.
Table 1. Short and/or Long Fluorescence Lifetimes (tS, tL ) and Relative Quantum Yields (FS, FL ) for 8-
Anilinonaphthalene-1-sulfonate (ANS) in the Presence and Absence of Host Compounds 1 6 in Aqueous
Phosphate Buffer Solution (pH 7.20; 0.1m) at 25.08
ANS conc./mm
Host
Host equiv.
tS/ns
FS/%
tL/ns
FL/%
c2
10
500
10
250
10
10
10
10
10
none
none
b-CD
b-CD
1
2
3
4
5
6
0.4
0.4
0.5
1.5
0.8
1.5
0.7
1.1
0.6
0.8
100
100
1.42
1.46
1.00
1.24
1.38
1.51
1.48
1.32
1.04
1.17
40
10
21
21
21
22
22
22
96.5
67.6
87.0
76.8
80.5
83.6
100
3.1
3.2
8.6
8.1
7.2
7.9
3.5
32.4
13.0
23.2
19.5
16.4
10
79.9
8.3
20.1
SODActivities of Mimics 1 6. The superoxide dismutase reaction can be expressed
as by Eqn. 2. The SOD activity can be calculated according to Eqn. 3, where Vtotal
/
Vdefined ˆ 1:1, v and v' are the auto-oxidation speed of pyrogallol in the absence and
presence of mimics (or naturalSOD enzyme), respectiveyl, Tdilution indicates the times
of enzyme-mimic dilution, i.e. Tdilution ˆ Vtotal/Venzyme, and Venzyme is the volume of the
natural-enzyme or mimic solution. The results obtained are listed in Table 2. When
repeated measurements were made, the SOD activity value was reproducible within an
error of Æ 5%.
SOD
ƒƒ!
.
.
‡
O2À ‡ O2À ‡ 2 H
H O ‡ O
(2)
2
2
2
v À v'
Vtotal Tdilution
v
SOD activity [U/ml] ˆ
¥
¥
(3)
50% vdefined Venzyme

14
Helvetica Chimica Acta Vol. 85 (2002)
Table 2. SOD Activities of Mimics 1 6, of a Natural SOD Enzyme Sample, as well as of Native and Oligoamino-
Modified b-Cyclodextrins
Activity/U/mg
Activity/U/mmol
b-CD
b-CDen
0
0
0
0
b-CDdien
0
0
b-CDtrien
0
0
b-CDprn
0
0
1
2
3
4
5
6
179
330
221
206
121
162
269
510
343
301
184
262
naturalSOD enzyme sampel
3400
As can be seen from Table 2, native and oligoamino-modified b-cyclodextrins
(including b-CDen, b-CDdien, b-CDtrien, b-CDprn; see Scheme) do not show SOD
activities at all, which clearly indicates that the b-cyclodextrin cavity acts only as a
substrate-binding site. All of the mimics 1 6 with the organoselenium moiety at the
primary rim of b-cyclodextrin display good SOD activities in the range of 121 330 U/
mg. The SOD activity of the best mimic 2 is one tenth of that of naturalSOD from
bovine erythrocytes. These results indicate that the organoselenium moiety is the
catalytic group of these mimics. As was reported in the previous study [14], we may
deduce that the underlying principle of the SOD activities of these mimics is the
similarity in catalytic cycles of selenium and copper in the mechanism.
It is very interesting to compare the SOD activities of mimics possessing similar
structures. For mimics 1 4, the unique difference is the oligoamino chain linking the b-
cyclodextrin cavity and the 1,2-benzisoselenazol-3(2H)-one moiety. As shown in
Table 2, SOD activities of mimics 1 4 are in the following sequence: 2 > 3 > 4 > 1,
which obviously can be roughly correlated to the length of the oligoamino chain: 3 >
2 > 4 > 1. Due to the difference of length and flexibility of the oligoamino chain, mimics
1 4 possess different conformations, as analyzed by the circular-dichroism and
fluorescence-lifetime experiments. One possible explanation for the highest SOD
activity of mimic 2 is that there may exist the best cooperation between the catalytic
group (the 1,2-benzisoselenazol-3(2H)-one moiety) and the substrate-binding site (the
b-cyclodextrin cavity), due to the suitable length/flexibility of the oligoamino chain of
2. Mimics possessing either a longer (3) or a shorter (1, 4) oligoamino chain show lower
SOD activities than mimic 2, because they cannot adopt the excellent cooperative
relationship of 2 between the organoselenium catalytic group and the b-cyclodextrin
cavity.
Both mimics 5 and 6 display relatively low SOD activities (see Table 2). Indeed, the
self-included moiety in 5 prevents the inclusion of the substrates, which explains why 5
shows the lowest SOD activity of all of the mimics. However, it is somewhat unexpected
that mimic 6 with the catalytic group located outside the b-cyclodextrin cavity also
exhibits a similarly low SOD activity. This may be attributed to the rigid benzene-ring

Helvetica Chimica Acta Vol. 85 (2002)
15
linker between the 1,2-benzisoselenazol-3(2H)-one and oligoamino moiety, which
might prevent the catalytic group from approaching the b-cyclodextrin cavity.
GPX Activities of Mimics 1 6. The GPX activities of 1 6, ebselen, and native and
oligoamino-modified b-cyclodextrins were determined by the decrease of the NADPH
absorption at 366 nm (see Table 3). When repeated measurements were made, the
GPX activity value was reproducible within an error of Æ 5%.
Table 3. GPX Activities of Mimics 1 6, of Ebselen, as well as of Native and Oligoamino-Modified b-
Cyclodextrins
Activity/U/mg
Activity/U/mmolRef.
Ebselen (PZ51)
b-CD
b-CDen
b-CDdien
b-CDtrien
b-CDprn
1
2
3
4
5
6
3.61
0
0
0
0
0.99
0
0
0
0
[12]
a
)
)
)
)
)
)
)
)
a
a
a
a
0
0
a
0.25
0.28
0.31
0.26
0.22
0.53
0.37
0.43
0.48
0.38
0.34
0.86
a
a
a
)
)
a
a
)
a) This work.
Ebselen (PZ51) is one of the best small molecular mimics of GPX, but its low
solubility in H2O limits its use (see above). On the contrary, our b-cyclodextrins 1 6,
carrying a 1,2-benzisoselenazol-3(2H)-one moiety, are GPX mimics with excellent
solubility in H2O. Native and oligoamino-modified b-cyclodextrins (including b-CDen,
b-CDdien, b-CDtrien, and b-CDprn) do not possess GPX activities, indicating that the
b-cyclodextrin cavity acts only as a substrate-binding site. On the contrary, all of the
organoselenium mimics 1 6 display moderate to good GPX activities in the range of
0.34 0.86 U/mmol, these values being somewhat lower than that of ebselen. Therefore,
we can deduce that the organoselenium moiety plays the role of the catalytic center. It
is interesting to investigate the relationship between the GPX activity and the structure
of these mimics. In mimics 1 4, only the 1,2-benzisoselenazol-3(2H)-one moiety, the
most important part of ebselen, is located at the primary face of b-cyclodextrin.
Nevertheless, the GPX activities of 1 4 are lower than that of ebselen, only 0.37
0.48 U/mmol. It can be inferred that the 2-phenyl group in ebselen might play an
important role in its GPX activity. Thus, we synthesized mimic 5 possessing the whole
structure of ebselen. Unfortunately, the GPX activity of mimic 5 is very weak, only
0.34 U/mmol. One possible explanation for this low activity is the self-inclusion of the
modifying moiety, which is suggested by the lower solubility in H2O and the results of
the ICD and fluorescence-lifetime experiments. To obtain a well H2O-soluble
compound possessing the ebselen moiety that is well-soluble in H2O, we designed
and synthesized mimic 6 for which the results of the ICD and fluorescence-lifetime
experiments established the absence of self-inclusion of the modifying moiety. As

16
Helvetica Chimica Acta Vol. 85 (2002)
expected, mimic 6 is easily dissolved in H2O and shows the highest GPX activity of our
mimics, up to 0.86 U/mmol, which is close to the value of ebselen (0.99 U/mmol).
Previous studies have demonstrated that there is a catalytic cycle involving
organoselenium compounds as artificial GPX mimics [12][14]. Therefore, we can
reasonably propose a similar cycle in the catalytic mechanism of mimics 1 6, which
may be the original cause of their GPX activities. Particularly, that 6 exhibits the highest
GPX activity suggests that the 2-phenylgroup in ebselen is very important for the high
GPX activity of the latter.
Conclusion. The above results indicate that the artificial enzyme models 1 6 can
perform the bifunctionalimitation of SOD and GPX, giving good mimicking activities.
Since these enzyme models possess excellent solubilities in H2O, they can be used as a
novel medicine in place of ebselen.
Experimental Part
General. b-Cyclodextrin of reagent grade (Suzhou Monosodium Glutamate Works) was recrystallized twice
from H2O and dried in vacuo for 12 h at 1008. N,N-Dimethylformamide (DMF) was dried over CaH2 for 2 days
and then distilled under vacuum prior to use. Ethane-1,2-diamine, propane-1,3-diamine, diethylenetriamine
(ˆ N-(2-aminoethyl)ethane-1,2-diamine), and triethylene-tetramine (ˆ N,N'-bis(2-aminoethyl)ethane-1,2-dia-
mine) were dried over K2CO3 and then distilled under vacuum prior to use. 2-(Chloroseleno)benzoyl chloride
was prepared according to [21]. Superoxide dismutase (from bovine erythrocytes, lyophilized powder) was
obtained from Sigma Chemical Company. Commercially available Tris ¥ HCl( Beijing Yiji Chemical Company),
dimethylarsinic acid sodium salt (Shanghai Chemical Regents Plant), diethylenetriamine-pentaacetic acid
(ˆ N,N-bis{2-[bis(carboxymethyl)amino]ethyl}glycine; Beijing Chemical Plant), and pyrogallol (ˆ benzene-
1,2,3-triol; Zhejiang WenzhouOuhai Chem. Plant ) were used without further purification. Glutathione (GSH),
NADPH, and glutathione reductase (type III) were purchased from Sigma. The superoxide dismutase (SOD)
activities and glutathione peroxidase (GPX) activities were measured with a ShimadzuUV-2401 spectrometer.
FT-IR Spectra: Nicolet FT-IR-5DX; in cmÀ1. Circular dichroism (CD) spectra: Jasco J-720W spectropolarim-
eter. 1H-NMR Spectra: at 200 MHz in (D)6DMSO; Bruker AM200 spectrometer; d in ppm. Elemental analyses:
Perkin-Elmer 240 instrument.
4-(2,3-Dihydro-3-oxo-1,2-benzisoselenazol-2-yl)benzoic Acid. To a stirred soln. of 4-aminobenzoic acid
(5 mmol) in Et2O (50 ml) in an ice bath, 2-(chloroseleno)benzoyl chloride (5 mmol) in Et2O (25 ml) and Et3N
(10 mmol) in Et2O (25 ml) were added dropwise simultaneously within 0.5 h. Then the mixture was heated to r.t.
with stirring for an additional3 h. The precipitate was filtered off and washed with Et 2O, 1m HCl, and H2O. The
crude product was dissolved in 0.5m aq. Na2CO3 (10 ml), the soln. then filtered, the filtrate acidified with HCl
soln. to pH < 1, and the precipitate filtered off, washed with H2O, and dried in vacuo: white solid (85% yield).
M.p. > 3008 ([22] > 3008).
6-Deoxy-6-{[2-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)ethyl]amino}-b-cyclodextrin (1). The 6-O-[(4-
methylphenyl)sulfonyl]-b-cyclodextrin (6-O-Ts-b-CD) was prepared from b-cyclodextrin with TsCl in dry
pyridine [23]. The 6-[(2-aminoethyl)amino]-6-deoxy-b-cyclodextrin (b-CDen) was prepared from 6-O-Ts-b-
CD and ethane-1,2-diamine [24]. Then, Et3N (0.2 ml) was added to the soln. of b-CDen (0.7 g, 0.6 mmol) in dry
DMF (50 ml) in an ice bath, with stirring and under N2. Then, the soln. of 2-(chloroseleno)benzoyl chloride [21]
(0.15 g, 0.6 mmol) in dry DMF (25 ml) was added dropwise, and the soln. was stirred for another 5 h at 5 8. After
evaporation, the yellow solid was dissolved in a minimum amount of hot H2O, and then the soln. was poured into
acetone (100 ml). The precipitate formed was filtered to give a yellow powder. This crude product was purified
by column chromatography (Sephadex G-25, distilled, deionized H2O): pure 1 (0.5 g, 60%). IR (KBr): 3399.5,
2914.0, 1702.6, 1635.2, 1595.6, 1541.2, 1520.9, 1444.0, 1418.3, 1381.0, 1343.9, 1298.8, 1272.4, 1238.9, 1154.4, 1074.5,
1022.6, 999.3, 940.1, 887.2, 839.6, 750.0, 701.1, 646.9, 604.7. 1H-NMR: 2.9 4.1 (m, 44 H); 4.2 4.6 (m, 2 H); 4.8
‡
‡
5.1 (m, 7 H); 5.6 6.0 (m, 14 H); 7.3 8.0 (m, 4 arom. H). FAB-MS (NaI): 1358 ([M ‡ H] ), 1380 ([M ‡ Na] ).
Anal. calc. for C51H78N2O35Se ¥ 8 H2O: C 40.78, H 6.31, N 1.86; found: C 40.67, H 5.88, N 2.04.
6-Deoxy-6-{{2-{[2-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)ethyl]amino}ethyl}amino}-b-cyclodextrin
(2). The 6-{{2-[(2-aminoethyl)amino]ethyl}amino}-6-deoxy-b-cyclodextrin (b-CDdien) was prepared by the

Helvetica Chimica Acta Vol. 85 (2002)
17
reaction of 6-O-Ts-b-CD with diethylenetriamine [24]. Then 2 was synthesized from b-CDdien and 2-
(chloroseleno)benzoyl chloride as described for 1: 55% of 2. IR (KBr): 3367.5, 2903.5, 1750.9, 1736.2, 1653.4,
1629.2, 1587.9, 1538.9, 1504.2, 1398.7, 1331.8, 1307.9, 1272.0, 1252.7, 1226.1, 1147.3, 1071.8, 1021.3, 940.1, 886.9,
835.4, 747.2, 691.7, 600.2. 1H-NMR: 2.8 4.1 (m, 48 H); 4.2 4.6 (m, 2 H); 4.8 5.1 (m, 7 H); 5.6 6.0 (m, 14 H);
‡
‡
7.3 8.0 (m, 4 arom. H). FAB-MS (NaI): 1401 ([M ‡ H] ), 1423 ([M ‡ Na] ). Anal. calc. for C53H83N3O35Se ¥
8 H2O: C 41.19, H 6.41, N 2.72; found: C 41.07, H 6.49, N 2.54.
6-Deoxy-6-{{2-{{2-{[2-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)ethyl]amino}ethyl}amino}ethyl}ami-
no}-b-cyclodextrin (3). The 6-{{2-{{2-[(2-aminoethyl)amino]ethyl}amino}ethyl}amino}-6-deoxy-b-cyclodextrin
(b-CDtrien) was prepared by the reaction of 6-O-Ts-b-CD with triethylenetetramine [24]. Then 3 was
synthesized from b-CDtrien and 2-(chloroseleno)benzoyl chloride as described for 1: 60% of 3. IR (KBr):
3370.0, 2909.5, 1754.1, 1720.1, 1652.6, 1586.0, 1541.5, 1458.2, 1381.5, 1335.3, 1304.0, 1254.7, 1231.6, 1147.8, 1072.2,
1020.5, 940.4, 883.7, 832.3, 747.6, 697.5, 601.2, 573.4. 1H-NMR: 2.8 4.1 (m, 52 H); 4.2 4.6 (m, 2 H); 4.8 5.1
‡
(m, 7 H); 5.6 6.0 (m, 14 H); 7.3 8.0 (m, 4 arom. H). FAB-MS (NaI): 1442 ([M ‡ H] ). Anal. calc. for
C55H88N4O35Se ¥ 11 H2O: C 40.19, H 6.70, N 3.41; found: C 40.28, H 6.36, N 2.98.
6-Deoxy-6-{[3-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)propyl]amino}-b-cyclodextrin (4). The 6-[(3-
aminopropyl)amino]-6-deoxy-b-cyclodextrin (b-CDprn) was prepared from 6-O-Ts-b-CD and propylenedi-
amine [24]. Then 4 was synthesized from b-CDprn and 2-(chloroseleno)benzoyl chloride as described for 1: 58%
of 4. IR (KBr): 3367.5, 2903.5, 1750.9, 1736.2, 1653.4, 1629.2, 1587.9, 1538.9, 1504.2, 1398.7, 1331.8, 1307.9, 1272.0,
1252.7, 1226.1, 1147.3, 1071.8, 1021.3, 940.1, 886.9, 835.4, 747.2, 691.7, 600.2. 1H-NMR: 2.8 4.1 (m, 46 H); 4.2 4.6
‡
(m, 2 H); 4.8 5.1 (m, 7 H); 5.6 6.0 (m, 14 H); 7.3 8.0 (m, 4 arom. H). FAB-MS (NaI): 1376 ([M ‡ H] ).
Anal. calc. for C53H83N3O35Se ¥ 8 H2O: C 41.19, H 6.41, N 2.72; found: C 41.07, H 6.49, N 2.54.
6-O-[4-(2,3-Dihydro-3-oxo-1,2-benzisoselenazol-2-yl)benzoyl]-b-cyclodextrin (5). To a soln. of DMF
(100 ml) containing 4-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)benzoic acid (1.55 g) and dicyclohexylcar-
bodiimide (DCC; 1.65 g), b-cyclodextrin (60 g) and dry pyridine (25 ml) as well as some 4-ä molecular sieves
were added. The mixture was stirred for 12 h in an ice bath and then at r.t. for another 18 h. The mixture was left
for 2 or 3 days to complete precipitation. The precipitate was filtered off, the filtrate evaporated, the residue
dissolved in a minimal amount of hot H2O, and the soln. poured into acetone (150 ml). The precipitate was
filtered off and the white powder washed with a large amount of acetone. The crude product was recrystallized
from EtOH/H2O 1:1 and then recrystallized twice from H2O: pure 5 (1.5 g, 25%). Light yellow powder. IR
(KBr): 3371.5, 2911.0, 1701.3, 1627.4, 1595.5, 1561.2, 1513.4, 1401.0, 1323.5, 1270.7, 1145.0, 1072.0, 1019.7, 938.2,
843.7, 794.2, 747.2, 697.6, 663.4, 573.4. 1H-NMR: 3.1 3.9 (m, 40 H); 4.1 4.6 (m, 2 H); 4.8 5.2 (s, 7 H); 7.0 8.0
‡
(m, 8 H). FAB-MS (NaI): 1461 ([M ‡ Na] ). Anal. calc. for C56H77NO37Se ¥ 5 H2O: C 44.09, H 5.71, N 0.92;
found: C 44.03, H 5.51, N 0.97.
6-Deoxy-6-{{2-{[4-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)benzoyl]amino}ethyl}amino}-b-cyclodex-
trin (6). As described for 5, from b-CDen [24] and 4-(2,3-dihydro-3-oxo-1,2-benzisoselenazol-2-yl)benzoic acid:
35% of 6. IR (KBr): 3391.2, 2927.6, 2150.8, 1634.2, 1591.0, 1546.4, 1450.5, 1399.3, 1387.7, 1332.1, 1262.4, 1202.3,
1
1155.6, 1081.0, 1029.2, 943.8, 853.0, 750.7, 704.8, 646.3, 578.3. H-NMR: 2.8 4.0 (m, 44 H); 4.1 4.6 (m, 2 H);
‡
4.8 5.2 (s, 7 H); 7.0 8.0 (m, 8 H). FAB-MS (NaI): 1481 ([M ‡ H] ). Anal. calc. for C58H83N3O35Se ¥ 8 H2O:
C 43.39, H 6.17, N 2.62; found: C 43.68, H 6.19, N 2.94.
Determination of SOD Activity. Superoxide dismutase (SOD) activities of the modified b-cyclodextrins 1
6 were determined by a modification of the method of Marklund and Marklund [25] using autooxidation of
pyrogallol at pH 8.2 and 258. The absorption at 319 nm of the assaying mixture (1 ml), containing 54 mm Tris ¥
HClbuffer (pH 8.2), 54 m m dimethylarsinic acid sodium salt, 1.07 mm diethylenetriaminepentaacetic acid,
84 mm pyrogallol, and an appropriate amount of mimic (usually 30 50 mm) or 1 mg/mlof the naturalSOD
sample, was recorded at 258. One activity unit corresponds to the amount of enzyme which inhibits the rate of
autooxidation of pyrogallol by 50% at pH 8.2 and 258.
Determination of GPX Activity. The GPX activity was determined by the method of Wilson et al. [12] with
hydrogen peroxide as the substrate in the presence of GSH. The reactions (see Eqns. 4 6) were performed at
378 in 700 mlof assay soln. containing 50 m m potassium phosphate buffer (pH 7.0), 1 mm EDTA, 1 mm sodium
azide, 1 mm GSH, 0.25 mm NADPH, 1 unit of GSSG reductase, and an appropriate amount of mimic (usually
10 50 mm finalconcentration). Gultathione reductase was used to reduce the oxidized GSH (GSSG) with
NADPH as a cofactor (Eqn. 4). The decrease in NADPH absorption at 366 nm is a measure of GPX activity.
The reaction was initiated by addition of 0.5 mm H2O2. Appropriate controls were run without enzyme or mimic
and were subtracted. The GPX activity unit of enzyme is defined as the amount of enzyme that utilizes 1 mmolof
NADPH per min. The activity is expressed in U/mg or U/mmolof enzyme.

18
Helvetica Chimica Acta Vol. 85 (2002)
GSH peroxidase
ƒƒƒƒƒƒƒƒ!
2 GSH ‡ H2O2
GSSG ‡ 2H2O
(4)
(5)
(6)
GSH reductase
ƒƒƒƒƒƒƒ!
‡
‡
GSSG ‡ NADPH ‡ H
2 GSH ‡ NADP
‡
‡
H ‡ NADPH ‡ H2O2
!
NADP ‡ 2H2O
This work was supported by the National Outstanding Youth Fund (Grant No. 29625203), Natural Science
Foundation of China (Grant No. 29992590-8 and 29972029), and Natural Science Fund of Tianjin Municipality
(Grant No. 013613511), which are gratefully acknowledged.
REFERENCES
[1] Y. Murakami, J. Kikuchi, Y. Hisaeda, O. Hayashida, Chem. Rev. 1996, 96, 721.
[2] R. B. Silverman, −The Organic Chemistry of Enzyme-Catalyzed Reaction×, Academic Press, San Diego,
2000.
[3] M. L. Bender, M. Komiyama, −Cyclodextrin Chemistry×, Springer Verlag, Berlin, 1978.
[4] R. Breslow, Acc. Chem. Res. 1995, 28, 146.
[5] M. Komiyama, −Cyclodextrins as Enzyme Models in Comprehensive Supramolecular Chemistry×, Eds. J. L.
Atwood, J. E. Davies, D. D. MacNicol, and F. Vˆgtle, Pergamon, Oxford, U.K., 1996, Vol. 3, p. 401.
[6] R. Breslow, S. D. Dong, Chem. Rev. 1998, 98, 1997.
[7] X.-Y. Ma, Y.-H. Wu, L. Ding, D.-Q. Zhao, J.-Z. Ni, Y. Liu, Chem. J. Chinese Univ. 1999, 20, 1163.
[8] J.-Q. Liu, S.-J. Gao, G.-M. Luo, G.-L. Yan, J.-C. Shen, Biochem. Biophys. Res. Commun. 1998, 237, 397.
[9] K. M. Anderson, T. Seed, D. Ou, J. E. Harris, Med. Hypoth. 1999, 52, 451.
[10] L. Engmann, A. Hallberg, J. Org. Chem. 1989, 54, 2964.
[11] N. J. Reich, C. P. Jasperse, J. Am. Chem. Soc. 1987, 109, 5549.
[12] S. R. Wilson, P. A. Sucker, R. C. Hung, A. Spector, J. Am. Chem. Soc. 1989, 111, 5936, and refs. cited
therein.
[13] T. G. Back, B. P. Dyck, J. Am. Chem. Soc. 1997, 119, 2079.
[14] A. M¸ller, E. Cadenas, P. Graf, H. Sies, Biochem. Pharmacol. 1984, 33, 3235.
[15] D. P. Riley, Chem. Rev. 1999, 99, 2573.
[16] M. E. Stroppolo, S. Nuzzo, A. Pesce, C. Rosano, A. Battistoni, M. Bolognesi, S. Mobilio, A. Desideri,
Biochem. Biophys. Res. Commun. 1998, 249, 579.
[17] K. Harata, U. Uedaira, Bull. Chem. Soc. Jpn. 1975, 48, 375.
[18] M. Kodaka, J. Am. Chem. Soc. 1993, 115, 3702.
[19] Y. Liu, C.-C. You, T. Wada, Y. Inoue, J. Org. Chem. 1999, 64, 3630; Y. Liu, B. Li, T. Wada, Y. Inoue,
Supramol. Chem. 1999, 10, 173; Y. Liu, B. Li, B.-H. Han, T. Wada, Y. Inoue, J. Chem. Soc., Perkin Trans. 2
1999, 563.
[20] Y. Liu, C.-C. You, Y. Chen, T. Wada, Y. Inoue, J. Org. Chem. 1999, 64, 7781.
[21] N. Kamigata, H. Iizuka, A. Izuoka, M. Kobayashi, Bull. Chem. Soc. Jpn. 1986, 59, 2179.
[22] Y.-X. Xiao, X.-F. Liu, H.-S. Xu, S.-Y. Sun, B. Xu, Chem. J. Chinese Univ. 1996, 107, 914.
[23] Y. Matsui, T. Yokoi, K. Mochida, Chem. Lett. 1976, 1037.
[24] B. L. May, S. D. Kean, C. J. Easton, S. F. Lincoln, J. Chem. Soc., Perkin Trans. 1 1997, 3157.
[25] S. Marklund, G. Marklund, Eur. J. Biochem. 1974, 47, 469.
Received May 17, 2001