Flavin-cyclodextrin conjugates: Effect of the structure on the catalytic activity in enantioselective sulfoxidations

Article abstract of DOI:10.1016/j.tetasy.2012.10.017

A series of flavin-cyclodextrin conjugates has been prepared and tested in the enantioselective oxidations of prochiral aromatic and aliphatic sulfides with hydrogen peroxide. The newly prepared conjugates contain isoalloxazinium or alloxazinium moieties attached to the primary rim of α- and β-cyclodextrins at the C-6 positions. In addition, flavinium units were attached to the secondary rim of the β-cyclodextrin macrocycle. The relationship between the structural features and the catalytic performance of the conjugates, including those recently reported by us, was analyzed. The rate and enantioselectivity of the sulfoxidations catalyzed by flavin-cyclodextrin conjugates are influenced mainly by the size of the cyclodextrin cavity, the type of flavin unit (alloxazine or isoalloxazine), and by the relative orientation of the flavin and cyclodextrin moieties.

Full text of DOI:10.1016/j.tetasy.2012.10.017

Tetrahedron: Asymmetry 23 (2012) 1571–1583
Contents lists available at SciVerse ScienceDirect
Tetrahedron: Asymmetry
journal homepage: www.elsevier.com/locate/tetasy
Flavin–cyclodextrin conjugates: effect of the structure on the catalytic
activity in enantioselective sulfoxidations
b
c
a,
⇑
Tomáš Hartman a, Vladimír Herzig a,b, Miloš Budešínsky , Jindrich Jindrich , Radek Cibulka
,
ˇ
´
ˇ
ˇ
Tomáš Kraus b,
⇑
a Department of Organic Chemistry, Institute of Chemical Technology, Prague, Technická 5, Prague 6, Czech Republic
b Institute of Organic Chemistry and Biochemistry AS CR, v.v.i., Flemingovo nám. 2, 166 10 Prague 6, Czech Republic
c Department of Organic Chemistry, Charles University in Prague, Faculty of Science, Hlavova 2030, 128 40 Prague 2, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 18 September 2012
Accepted 24 October 2012
A series of flavin–cyclodextrin conjugates has been prepared and tested in the enantioselective oxida-
tions of prochiral aromatic and aliphatic sulfides with hydrogen peroxide. The newly prepared conjugates
contain isoalloxazinium or alloxazinium moieties attached to the primary rim of a- and b-cyclodextrins
at the C-6 positions. In addition, flavinium units were attached to the secondary rim of the b-cyclodextrin
macrocycle. The relationship between the structural features and the catalytic performance of the conju-
gates, including those recently reported by us, was analyzed. The rate and enantioselectivity of the sul-
foxidations catalyzed by flavin–cyclodextrin conjugates are influenced mainly by the size of the
cyclodextrin cavity, the type of flavin unit (alloxazine or isoalloxazine), and by the relative orientation
of the flavin and cyclodextrin moieties.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
We have recently reported that conjugates of a- and b-cyclo-
dextrins with alloxazinium and isoalloxazinium moieties, respec-
tively, 1–5 (Fig. 1) displayed remarkable catalytic activities in
enantioselective oxidations of sulfides with hydrogen peroxide.7d,e
In particular, conjugates 1–4 bearing an alloxazinium moiety of the
structural type A (Fig. 2) attached onto the primary rim of the
cyclodextrins at the C-6 position via spacers with variable lengths
were very efficient and provided high conversions and, in some
cases, very high enantioselectivities (up to 91% ee). Moreover, oxi-
dation reactions proceeded in buffered aqueous solutions under
low catalyst loadings (0.2–5 mol %). Presumably, the cyclodextrin
macrocycle binds to a sulfide, which is then oxidized in the cavity
by the attached flavin hydroperoxide formed reversibly from the
flavinium unit of the catalyst upon reaction with hydrogen perox-
ide (for the structure of flavin–hydroperoxides, see Fig. 2). Variable
degrees of catalytic abilities exerted by conjugates 1–5 with re-
spect to the structure of the substrate convincingly showed that
the nature of the oxidizing moiety and its orientation to the cyclo-
dextrin cavity, along with the size of this cavity, have a strong im-
pact on the performance of the whole catalytic system. Herein, we
explore the catalytic activities of seven new flavin–cyclodextrin
In the last decade, great attention has been paid to environmen-
tally friendly enantioselective oxidations, particularly those that
use organocatalysts and hydrogen peroxide as terminal oxidants.1,2
Among oxidation reactions, the enantioselective sulfoxidation of
prochiral sulfides has been of high interest due to the application
of chiral sulfoxides as active substances in the pharmaceutical
industry or as chiral auxilliaries in organic synthesis.3 Metal com-
plexes with chiral ligands are considered as privileged sulfoxida-
tion catalysts that give high enantioselectivities (ee >90%).2b,3–5
Several highly enantioselective biological sulfoxidations have also
been reported.4a,b On the other hand, new developments in organ-
ocatalytic sulfoxidations have shown that metal-free chemo-cata-
lytic procedures can be viable alternatives.2a,6 The design of
catalysts containing chiral flavinium salts (both isoalloxazinium
and alloxazinium),7 inspired by the function of flavin cofactors in
monooxygenases,8 and chiral phosphoric acid derivatives possess-
ing a BINOL moiety9 has led to efficient organocatalytic systems for
the enantioselective oxidations of sulfides to sulfoxides with
hydrogen peroxide. Very recently, List et al. reported sulfoxidations
catalyzed by a chiral imidodiphosphoric acid allowing enantiose-
lectivity comparable with the best metal-based systems.10
conjugates 6–12. In addition to conjugate 6, which consists of
a-
cyclodextrin and the previously explored isoalloxazinium unit of
type B, we have designed catalysts 7–10, which bear new isoallox-
azinium and alloxazinium moieties of structures C and D, respec-
⇑
Corresponding authors. Fax: +420 220444288 (R.C.); fax: +420 220183560
(T.K.).
tively, which are attached to the primary rims of both
a- and b-
cyclodextrins. Finally, we were interested in the performance of
E-mail addresses: cibulkar@vscht.cz (R. Cibulka), kraus@uochb.cas.cz (T. Kraus).
0957-4166/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetasy.2012.10.017

1572
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
O
N
O
N
CH3
N
CH3
O
CH3
N
N
N
N
N
O
O
N
N
N
O
N
CH2
C
-
(CH2)2
-
ClO4
-
(CH2)n
ClO4
ClO4
O
C
O
C
O
HN
HN
(C-6)
NH
(C-6)
(C-6)
CD
CD
CD
1 (α-CD, n = 1)
2 (β-CD, n = 1)
3 (α-CD, n = 2)
4 (β-CD, n = 2)
5 (β-CD)
6 (α-CD)
7
8
(α-CD)
(β-CD)
O
CH3
N
CH3
N
CH3
N
N
O
O
N
N
N
O
N
N
O
N
N
N
-
CH2
ClO4
CH2
CH2
C
-
-
ClO4
ClO4
O
NH
NH
O
O
O
HN
(C-6)
O
O
(O-3)
(O-3)
CD
CD
CD
12
11
(β-CD)
9
α-CD)
(β-CD)
(
10 β-CD)
(
Figure 1. Overview of alloxazine- and isoalloxazine conjugates discussed herein: compounds 1–5 were reported in our previous work;7d,e compounds 6–12 were prepared
herein.
CH3
CH3
N
N
O
N
N
O
N
N
O
N
N
O
1
1
10
10
1
10
10
1
3 N
5
N
3 N
3 N
3 N
5
5
5
N
N
N
CH3
CH3
O
O
O
O
A
B
C
D
H2O2
-H+
CH3
N
CH3
N
N
N
N
O
O
N
O
O
N
N
N
O
O
N
N
O
N
N
N
N
N
CH3
CH3
O
O
O
O
O
OH
OH
OH
OH
Figure 2. Structural types of flavinium units (both alloxazinium and isoalloxazinium) differing in the position of the cyclodextrin attachment. The structures of the
corresponding flavin hydroperoxides are shown.

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1573
conjugates in which the flavinium moieties were attached to the
secondary rim of the macrocycle. Thus, conjugates 11 and 12, hav-
ing alloxazinium and isoalloxazinium units of types A and B,
respectively, attached to O-3 position of b-cyclodextrin via amido-
ethylene linker were designed. Herein we report the synthesis of
conjugates 6–12 and their ability to catalyze oxidation reactions
of aromatic and aliphatic sulfides by hydrogen peroxide. We also
draw conclusions on the structure–reactivity relationship across
the whole series of conjugates that we have explored so far.
pure alloxazine 20. Its alkylation with ethyl bromoacetate followed
by hydrolysis of the ester function in 21 afforded acid 14. (10-
Methylisoalloxazin-3-yl)acetic acid 15 and 3-methylalloxazin-1-
ylacetic acid 16 were prepared as previously described.7d,e
In order to attach the flavin moieties onto the secondary (wider)
rim of cyclodextrins, we developed the synthesis of 3-O-amino-
ethyl-b-cyclodextrin 24 (Scheme 2). The synthetic method takes
advantage of a known selective etherification (cinnamylation) at
the O-3 position.16 The 3-O-cinnamyl-b-cyclodextrin 22 was trans-
formed into aldehyde 23 via ozonolysis followed by decomposition
of the intermediary ozonide with dimethyl sulfide. The resulting
aldehyde 23 could only be unequivocally characterized by TLC
and ESI-MS; its NMR spectra appeared to be composed of more
than one species, probably due to the propensity of 23 to form hy-
drates and/or intramolecular hemiacetals. Aldehyde 23 was then
treated with ammonium acetate and sodium cyanoborohydride
to give amine 24.
The conjugates of cyclodextrins and flavins 27–33 were pre-
pared from the corresponding flavin carboxylic acids 13–16 and
cyclodextrin amines via amide coupling using a PyBOP reagent as
described in the literature (Schemes 3 and 4).7d,e These compounds
are stable in the absence of light. Their conversion into quaternary
5-ethyl ammonium salts was carried out by reductive amination
using acetaldehyde and hydrogen under the catalysis of palladium
on charcoal under acidic conditions, followed by exposure to oxy-
gen. Solutions of the resulting flavinium conjugates 6–12, which
exhibited limited stability, were used immediately in catalytic
experiments without isolation of the compounds.
2. Results and discussion
2.1. Synthesis of the conjugates
We have used an amide bond as the bridging function since its
formation by the reaction of flavin carboxylic acid and the amine-
containing cyclodextrin unit proved to be a reliable ligation meth-
od in our previous work.7d,e Isoalloxazine and alloxazine carboxylic
acids 13–16 (Schemes 1, 3 and 4) were used for the introduction of
the flavin moiety into the conjugates. 3-(3-Methylisoalloxazin-10-
yl)propanoic acid11,12 13 was prepared by starting from 1-fluoro-2-
nitrobenzene and ethyl 3-aminopropanoate (Scheme 1). After
hydrogenation of nitroaniline 17, the obtained N-(ethoxycarbonyl-
ethyl)benzene-1,2-diamine, due to its high reactivity toward poly-
merization, was immediately condensed with a 3-methylalloxan
hydrate to form isoalloxazine 18 possessing an ester function,
which was subsequently hydrolyzed to acid 13.13 The synthesis
of (1-methylalloxazin-3-yl)acetic acid 14 started from 6-chloro-
1-methyluracil14 to afford (N-phenylamino)uracil 19 by reaction
with aniline. Nitrosative cyclization of aminouracil 19 under condi-
tions originally described for isoalloxazine derivatives15 gave a
mixture of 1-methylalloxazine-5-oxide and 1-methylalloxazine
20 (5:1 according to 1H NMR), which was hydrogenated to give
2.2. Catalysis
All of the flavinium salts were tested as catalysts of oxidations
of alkyl methyl sulfides or aryl methyl sulfides by hydrogen perox-
1. H2, Pd(C)
COOR
2.
O
N
O
HO
N
H
CH3
F
N
N
N
N
O
O
HO
NH2CH2CH2COOEt
dioxane
COOEt
O
N
NO2
AcOH
NO2
CH3
17
18: R = Et
13: R = H
H2O, HCl
CH3
N
CH3
CH3
N
1. NaNO2
H
N
N
O
N
O
Cl
O
2. H2, Pd(C)
PhNH2
NH
O
NH
NH
AcOH
N
O
O
19
20
CH3
N
N
N
O
BrCH2COOEt
K2CO3, DMF
N
COOR
O
21: R = Et
14: R = H
H2O, HCl
Scheme 1. Synthesis of isoalloxazine 13 and alloxazine 14 carboxylic acids.

1574
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
OH
O
HO
OH
O
OH
O
HO
HO
1. NH4OAc
1. O3
O
O
O
2. NaBH3CN
2. Me2S
(
(
(
O
O
O
HO
O
HO
HO
O
MeOH
O
MeOH
OH
OH
OH
OH
OH
OH
O )
O
O )
)
6
6
6
H2N
O
24
23
22
Scheme 2. Synthesis of 3-O-aminoethyl-b-cyclodextrin 24.
CH3
N
N
O
O
CH3
N
N
N
O
O
N
CH2COOH
O
N
H
N
15
ii)
N
C
6
HO
(
HO
H2
O
O
27
(n=5)
HO
O
OH
OH
O )
n
i)
O
CH3
O
N
O
N
N
CH3
O
N
N
N
NH2
HO
N
N
(CH2)2COOH
O
O
(
O
ii)
HO
13
H
HO
7
8
(n=5)
(n=6)
O
OH
OH
H2C
N
HO
O )
2
n
i)
O
O
OH
(
25
(n=5)
HO
HO
O
26
OH
(n=6)
O )
n
28 (n=5)
29 (n=6)
CH3
N
N
N
O
CH3
N
N
i)
N
N
O
CH2COOH
O
O
N
14
ii)
H
N
9 (n=5)
C
HO
10 (n=6)
H2
O
O
O
(
30 (n=5)
31 (n=6)
HO
HO
O
OH
OH
O )
n
Scheme 3. Synthesis of conjugates 6–10. Reagents and conditions: (i) PyBOP, DIPEA, DMF; (ii) (1) H2/Pd, CH3CHO, EtOH–H2O, HClO4, (2) O2.
ide in buffered aqueous media. In line with our previous findings,
reactions catalyzed with isoalloxazinium and alloxazinium salts
were performed at pH 2.9 and 7.5, respectively.7d,e,17
sulfide undergo non-catalyzed oxidations to a significant extent
within 60 min of reaction time (Table 8). Similarly, benzyl- and
phenyl methyl sulfides also provide good conversions but negligi-
ble enantioselectivity.
Conjugate 6 was expected to catalyze the oxidations of alkyl
methyl sulfides due to the known affinity of the
a
-cyclodextrin
Isoalloxazinium conjugates 7 and 8 (structural type C, Fig. 2)
show comparable catalytic activities toward n-alkyl methyl sul-
fides (Tables 2 and 3): n-butyl- and n-hexyl methyl sulfides are oxi-
dized nearly quantitatively within 60 min of reaction time with
moderate enantioselectivities (up to 53% ee, Table 3, entry 4). Con-
versions are significantly lower for n-octyl methyl sulfides but be-
come higher for n-decyl and n-dodecyl methyl sulfides in reactions
catalyzed by conjugate 7 although conversions of these substrates
remained below 50% within 60 min when oxidations proceeded
under the catalysis of conjugate 8. Enantioselectivities tended to
decrease with an increase of the chain length. Oxidation of bulkier
cavity toward n-alkyl chains. However, conversion within 60 min
was satisfactory only in the case of n-butyl methyl sulfide (98%, Ta-
ble 1, entry 1). Increasing the chain length caused the conversion to
decrease dramatically to 11% and 15% for n-decyl methyl sulfide
and n-dodecyl methyl sulfide, respectively (Table 1, entries 4 and
5). The enantioselectivity in these reactions increased from n-bu-
tyl- toward n-octyl methyl sulfide and then decreased again. t-Bu-
tyl- and c-hexyl methyl sulfides furnished sulfoxide products in
quantitative yield, yet with negligible enantioselectivity. It should
be noted, however, that c-hexyl- and, in particular, t-butyl methyl

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1575
O
N
CH3
O
N
O
N
N
N
CH3
O
N
N
OH HO
N
CH2COOH
O
O
(
HO
ii)
16
O
O
11
OH
OH
)
O
CH2
6
i)
O
N
H
32
OH HO
O
O
(
HO
O
O
OH
24
OH
)
O
6
H2N
CH3
N
N
O
O
CH3
N
N
i)
N
15
O
O
CH2COOH
N
O
O
ii)
OH
)
O
12
O
6
N
N
NH
C
H2
O
33
Scheme 4. Synthesis of conjugates 11 and 12. Conditions: i) PyBOP, DIPEA, DMF; ii) (1)H2/Pd, CH3CHO, EtOH-H2O, HClO4; (2) O2.
Table 1
H2O2-sulfoxidations catalyzed by conjugate 6a
Table 2
H2O2-sulfoxidations catalyzed by conjugate 7a
Entry
Sulfide
Time (min)
Conv.b (%)
eec (%)
Opt. rot.d
Entry
Sulfide
Time (min)
10
Conv.b (%)
eec (%)
Opt. rot.d
1
2
3
4
5
6
7
8
9
n-C4H9SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
BnSCH3
60
60
60
60
60
60
60
60
120
98
68
28
11
15
100
100
83
4 (0)
19
51 (45)
30
10 (8)
5
4 (5)
0
1
2
3
4
5
6
7
8
n-C4H9SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C10H21SCH3
n-C12H25SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
c-C6H11SCH3
BnSCH3
90
22
94
97
8
21
42
30
21
27
30
—
10
—
13
0
0
0
—
0
—
0
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(+)
(+)
(+)
(+)
(ꢀ)
(ꢀ)
10
10 (5 mol %)
60
10
60
10
60
10
60
10
10
60
10
60
10
60
10
60
38
14
98
12
96
93
53
99
34
96
11
62
56
57
(ꢀ)
(ꢀ)
PhSCH3
99
4
9
10
11
12
13
14
15
16
17
18
19
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 2.9, rt,
catalyst loading 1 mol % (related to the substrate) if not stated otherwise; vigorous
shaking.
Conversion determined by 1H NMR.
b
Enantiomeric excess determined by 1H NMR using chiral shift agent and/or by
c
BnSCH3
PhSCH3
PhSCH3
p-MePhSCH3
p-MePhSCH3
HPLC on a chiral stationary phase (values in brackets).
Sense of optical rotations derived from a Chiralyser (426 nm) detector over the
course of analysis in a heptane–isopropyl alcohol mixture.
d
—
0
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 2.9, rt,
catalyst loading 1 mol % (related to the substrate) if not stated otherwise; vigorous
shaking.
aliphatic substrates (t-butyl, c-hexyl) catalyzed by 7 proceeded
rapidly without enantioselectivity, while with 8 possessing a b-
cyclodextrin cavity these substrates are oxidized with enantiose-
lectivity of up to 59% ee (Table 3, entries 12 and 13). Catalyst 8 is
also better than 7 for the oxidation of benzyl- and phenyl methyl
sulfides; in these cases enantioselectivities of up to 51% ee were
observed (Table 3, entries 14–19).
b
Conversion determined by 1H NMR.
c
Enantiomeric excess determined by HPLC on a chiral stationary phase.
Sense of optical rotations derived from a Chiralyser (426 nm) detector over the
d
course of analysis in a heptane–isopropyl alcohol mixture.
Conjugates 9 and 10, consisting of
a
- and b-cyclodextrins,
rim via an amidoethylene bridge linking the O3 positions with
the heterocyclic moiety. We expected that this structural change
may bring about changes in reactivity and, in particular, enantiose-
lectivity. Conjugate 11 with an alloxazinium unit provided excel-
lent rate enhancements for all of the aliphatic n-alkyl substrates
(Table 6, entries 1–5); conversions in the range of 92–97% in the
oxidation of n-alkyl methyl sulfides were the highest values that
we observed in the whole series of flavin–cyclodextrin conjugates
studied so far. Additionally, t-butyl and c-hexyl, benzyl- and phe-
nyl methyl sulfides were oxidized in quantitative yields, and only
p-tolyl methyl sulfide (Table 6, entry 10) was oxidized with a lower
conversion than with the most efficient catalyst 2. Enantioselectiv-
ities in this series of reactions were, however, modest; the highest
ee (47%) was encountered in the oxidation of the n-butyl methyl
respectively, and an alloxazinium moiety of type D, showed the
lowest catalytic activity in terms of conversions and, in particular,
enantioselectivities, which were either absent (conjugate 9, Ta-
ble 4) or negligible (conjugate 10, Table 5, values for pH 7.5). Nev-
ertheless, a comparison with control experiments carried out
without any catalyst (Table 8) showed that there is a rate enhance-
ment in the oxidation of several substrates, for example, n-butyl-
and c-hexyl. As a result, the catalysis of oxidation provided by
the flavinium moiety is likely to proceed without the inclusion of
the substrate into the cyclodextrin cavity.
Conjugates 11 and 12 differ from all of the conjugates prepared
so far in the orientation of the cyclodextrin cavity: the flavinium
moieties of type A and B are attached onto the secondary (wider)

1576
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
Table 3
Table 5
H2O2-sulfoxidations catalyzed by conjugate 8a
H2O2-sulfoxidations catalyzed by conjugate 10a
d
d
Entry
Sulfide
Time (min)
10
10 (5 mol %)
10 (5 mol %)
Conv.b (%)
eec (%)
Opt. rot.
Entry
Sulfide
Time (min)
Conv.b (%)
eec (%)
Opt. rot.
1
2
3
4
5
6
7
8
n-C4H9SCH3
n-C4H9SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C10H21SCH3
n-C12H25SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
BnSCH3
92
94
53
97
93
11
46
11
40
11
45
98
99
81
97
69
97
58
76
33
35
48
53
45
40
42
27
17
—
23
50
59
—
37
—
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
1
2
3
4
5
6
7
8
n-C4H9SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C10H21SCH3
n-C12H25SCH3
n-C12H25SCH3
t-C4H9SCH3
t-C4H9SCH3
c-C6H11SCH3
c-C6H11SCH3
BnSCH3
10
10
60
10
60
10
60
10
60
10
60
10
60
10
60
10
60
10
60
92
31
69
8
15
8
7
—
9
—
0
—
0
—
0
—
8
—
11
—
8
—
4
—
0
(ꢀ)
—
(ꢀ)
10 (5 mol %)
—
60
10
60
10
60
10
60
10
10
10
60
10
60
10
60
—
—
—
—
—
—
7
11
18
61
97
78
99
52
95
17
31
15
25
9
9
10
11
12
13
14
15
16
17
18
19
10
11
12
13
14
15
16
17
18
19
(ꢀ)
(ꢀ)
(ꢀ)
(ꢀ)
—
(ꢀ)
—
(+)
—
(+)
—
—
BnSCH3
PhSCH3
PhSCH3
p-MePhSCH3
p-MePhSCH3
(+)
(+)
(+)
BnSCH3
PhSCH3
PhSCH3
p-MePhSCH3
p-MePhSCH3
35
—
51
a
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 2.9, rt,
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 7.5, rt,
catalyst loading 1 mol % (related to the substrate) if not stated otherwise; vigorous
shaking.
catalyst loading 1 mol % (related to the substrate); vigorous shaking.
b
Conversion determined by 1H NMR.
b
c
Conversion determined by 1H NMR.
Enantiomeric excess determined by HPLC on a chiral stationary phase.
Sense of the optical rotations derived from a Chiralyser (426 nm) detector over
c
d
Enantiomeric excess determined by HPLC on a chiral stationary phase.
d
Sense of the optical rotations derived from a Chiralyser (426 nm) detector in the
the course of analysis in a heptane–isopropyl alcohol mixture.
course of analysis in a heptane–isopropyl alcohol mixture.
Table 6
H2O2-sulfoxidations catalyzed by conjugate 11a
Table 4
H2O2-sulfoxidations catalyzed by conjugate 9a
d
Entry
Sulfide
Time (min)
Conv.b (%)
eec (%)
Opt. rot.
Entry
Sulfide
Time (min)
Conv.b (%)
eec (%)
Opt. rot.
1
2
3
4
5
6
7
8
9
10
n-C4H9SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
BnSCH3
60
60
60
60
60
60
60
60
60
60
96
97
97
96
92
100
100
99
98
82
44(47)
11
(ꢀ)
(ꢀ)
1
2
3
4
5
6
7
8
n-C4H9SCH3
n-C6H13SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C10H21SCH3
n-C12H25SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
c-C6H11SCH3
BnSCH3
10
10
60
10
60
10
60
10
60
10
10
60
10
60
10
60
10
60
95
24
50
9
16
9
0
—
0
—
0
—
0
—
0
0
—
0
—
0
—
0
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
—
1
1
0
24(30)
28 (32)
30(22)
31(36)
26(31)
(ꢀ)
(ꢀ)
(+)
(+)
(+)
9
PhSCH3
p-MePhSCH3
14
49
95
64
96
39
68
12
20
12
17
9
10
11
12
13
14
15
16
17
18
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 7.5, rt,
catalyst loading 1 mol % (related to the substrate); vigorous shaking.
b
Conversion determined by 1H NMR.
c
Enantiomeric excess determined by 1H NMR using chiral shift agent and/or by
BnSCH3
PhSCH3
PhSCH3
p-MePhSCH3
p-MePhSCH3
HPLC on a chiral stationary phase (values in brackets).
d
Sense of optical rotations derived from a Chiralyser (426 nm) detector over the
course of analysis in a heptane–isopropyl alcohol mixture.
—
0
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 7.5, rt,
catalyst loading 1 mol % (related to the substrate); vigorous shaking.
b
Conversion determined by 1H NMR.
Enantiomeric excess determined by HPLC on a chiral stationary phase.
Table 7
c
H2O2-sulfoxidations catalyzed by conjugate 12a
d
Entry
Sulfide
Time (min)
Conv.b (%)
eec (%)
Opt. rot.
1
2
3
4
5
6
7
8
9
10
n-C4H9SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
BnSCH3
60
60
60
60
60
60
60
60
60
60
84
88
45
82
76
100
98
99
77
50
1
5
1
4
4
1
(ꢀ)
sulfide (Table 6, entry 1) but other n-alkyl methyl sulfides pro-
ceeded with low or negligible enantioselectivities.
Bulkier aliphatic sulfides and aromatic sulfides were oxidized
with ee values in the range of 30–36% (Table 6, entries 6–10). Con-
jugate 12 with an isoalloxazinium unit allowed significant rate
enhancements (Table 7) in the oxidation of aliphatic sulfides in
comparison to blank experiments, but enantioselectivities were
negligible or small for all substrates examined.
Lastly, a complete set of control experiments was carried out in
addition to that known from with the literature;7d,e the results are
summarized in Table 8. It can be seen that t-butyl- and n-butyl
methyl sulfides rapidly undergo non-catalyzed oxidations within
the timescale investigated, especially at pH 7.5. Thus, the low
enantioselectivities encountered in the oxidations of these sub-
17(17)
(ꢀ)
0
1
5
PhSCH3
p-MePhSCH3
(+)
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 2.9, rt,
catalyst loading 1 mol % (related to the substrate); vigorous shaking.
Conversion determined by 1H NMR.
b
Enantiomeric excess determined by 1H NMR using chiral shift agent and/or by
c
HPLC on a chiral stationary phase (values in brackets).
Sense of optical rotations derived from a Chiralyser (426 nm) detector over the
course of analysis in a heptane–isopropyl alcohol mixture.
d

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1577
Table 8
Control experiments: H2O2 sulfoxidations without a catalysta
Entry
Sulfide
Time (min)
Conv.b (%)
pH 2.9
Conv.b (%)
pH 7.5
1
2
3
4
5
6
7
8
9
10
n-C4H9SCH3
n-C6H13SCH3
n-C8H17SCH3
n-C10H21SCH3
n-C12H25SCH3
t-C4H9SCH3
c-C6H11SCH3
BnSCH3
60
60
60
60
60
60
60
60
60
60
90
67
3
0
2
28
44
36
5
97
10
2
0
0
98
39
20
2
PhSCH3
p-MePhSCH3
3
4
a
Conditions: substrate (0.1 mmol), H2O2 (2.3 equiv), phosphate buffer pH 2.9 or
7.5, rt; vigorous shaking;
b
Conversion determined by 1H NMR.
strates may be caused by the relatively low rates of the catalyzed
reactions compared to the non-catalyzed ones.
2.3. Structure–reactivity relationship
Figure 3. Comparison of the enantioselectivities in H2O2-sulfoxidations catalyzed
by flavin–cyclodextrin conjugates 1–12. Catalysts are ordered according to the
Next, we decided to study the effect of the structures of all of
the catalysts 1–12 and the substrates on the final outcome of the
oxidation reaction, especially with regard to enantioselectivity, in
a more systematic manner. For this purpose, we considered four
structural parameters of the catalyst: (i) the size (diameter) of
the cyclodextrin macrocycle; (ii) the position of the attachment
on the macrocycle (primary or secondary rim of cyclodextrin);
(iii) the position of the attachment on the flavin unit (upper
site—flavinium units A and C in Fig. 2, lower site units B and D);
and iv) the length of the linker. On the side of the substrate, we di-
vided the structures into three groups: (i) aliphatic n-alkyl methyl
sulfides; (ii) bulkier aliphatic sulfides (t-butyl-, c-hexyl-, benzyl
methyl sulfides); and (iii) aromatic sulfides.
structural type (see text and Fig. 2). Results for conjugates with
a-cyclodextrin
(blue), b-cyclodextrin substituted on primary rim (red) and secondary rim
(magenta) are distinguished by colours. Data for catalysts 1–5 were taken from
previous studies.7d,e
exclusively gave enantiomers with positive values of specific rota-
tions, the longer spacer allowed enantiomers with the opposite
configuration. Thus, the configuration of the enantiomers can be
tuned, in principle, by factors other than just the configuration of
the chiral moiety. On the other hand, the configuration of the aryl
methyl sulfoxides was not influenced by the structure of the cata-
lyst; positive values for the specific rotations were observed in all
cases. (ꢀ)-Enantiomers of sulfoxides were predominantly obtained
from the oxidations of c-hexyl and t-butyl methyl sulfides with the
only exception of c-hexyl methyl sulfoxide obtained from oxida-
tion catalyzed by 1.7e
Analyses of the achieved enantioselectivities (Fig. 3) published
herein as well as in the literature7d,e clearly show that
a-cyclodex-
trin macrocycles are only suitable for the oxidation of n-alkyl
methyl sulfides, while b-cyclodextrins are more versatile and that
some of these catalysts could be used for all types of substrates.
In general, alloxazine units turned out to be more efficient than
the isoalloxazine moieties, provided they were attached to the
cyclodextrin at the upper site of the alloxazine unit (structure A
Fig. 2, conjugates 1–4). Since the reactive center is located on the
lower site of the flavin unit (opposite to the side of the linker
attachment, see Fig. 2), the heterocycle had to adjust its position
so that the sulfide group of the complexed substrate could come
into contact with the oxygen of the flavin hydroperoxide moiety
(for the structure of flavin hydroperoxides see Fig. 2). Alloxazine
units with a linker attached at the lower site (catalysts 9 and 10
of type D) gave negligible enantioselectivities. In the isoalloxazine
series, specifically in the case of b-cyclodextrin derivatives (cata-
lysts 5 and 8), the position of the cyclodextrin attachment plays
a similar role. Catalyst 8 of type C with the cyclodextrin unit bound
to the upper site of the isoalloxazine allowed higher enantioselec-
tivities than catalyst 5 of type B. Nevertheless, this effect is less
remarkable than in the alloxazine series. Thus, the newly prepared
conjugate 8 seems to be the best among isoalloxazine catalysts. In
terms of the cyclodextrin macrocycle, the attachment of the flavin
unit to the primary rim of the cyclodextrin led to a better perfor-
mance of the catalysts as compared with those being substituted
at the secondary rim.
3. Conclusions
Cyclodextrin–flavin conjugates proved to be efficient catalysts
for sulfoxidations by hydrogen peroxide allowing good enantiose-
lectivities for most of the resulting sulfoxides, both aliphatic and
aromatic, used herein and in previous studies.7d,e The reactions
proceed in aqueous media under low loadings (typically 1 mol %)
of the catalysts. The structure of the catalysts can be tuned to im-
prove their performance with respect to particular substrates.
One drawback of the current system is the relatively low stabil-
ity of flavinium salts which, in common with the relatively high
rate of the non-catalyzed oxidation in the case of some substrates,
can lower the enantioselectivity of the transformation. This effect
could in principle be avoided by using oxygen as the oxidation re-
agent under Murahashi conditions.18 Such studies are currently
underway.
4. Experimental
4.1. General
NMR spectra were acquired with spectrometers Bruker AVANCE
500 (1H at 500.1 MHz and 13C at 125.8 MHz), AVANCE 600 (1H at
600.1 MHz and 13C at 150.9 MHz, equipped with a cryoprobe)
and Varian Mercury Plus (1H at 299.97 MHz and 13C at
The length of the linker connecting both components of the cat-
alyst turned out to have an effect on the configuration of the result-
ing n-alkyl methyl sulfoxides (Fig. 4): while the shorter spacer

1578
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
Figure 4. Enantioselectivities of sulfoxidations (ee) with the sign corresponding to sense of the optical rotations of n-alkyl methyl sulfoxides resulting from oxidations of the
corresponding sulfides catalyzed by flavin–cyclodextrin conjugates. Only catalysts achieving significant enantioselectivity are included; columns represent results for n-butyl
(a), n-hexyl (b), n-octyl (c), n-decyl (d), and n-dodecyl (e) methyl sulfoxides.
75.44 MHz) in CDCl3, d6-DMSO or D2O at 300 K. Homonuclear 2D-
NMR spectra (H,H-COSY) and heteronuclear 2D-NMR spectra (H,C-
HSQC and H,C-HMBC) were used for the structural assignment of
proton and carbon signals of compounds 27–33. Mass spectra were
measured using ES ionization either in a positive or negative mode
(Waters micromass ZQ). High resolution mass spectra were mea-
sured using nanoESI ionization on LTQ Orbitrap XL (Thermo Fisher
Scientific). Elemental analyses were carried out on Perkin Elmer
2400 II. UV spectra were measured on Varian Cary 50 spectropho-
tometer. Preparative reversed-phase chromatography (RP) was
carried out using medium pressure columns containing C-18 mod-
1.27 (t, 3H, J = 7.0 Hz, CH3); 2.70 (t, 2H, J = 6.6 Hz, CO–CH2–CH2–
N); 3.64 (t, 2H, J = 6.6 Hz, CO–CH2–CH2–N); 4.18 (q, 2H, J = 7.0 Hz,
CH2–CH3); 6.65 (ddd, 1H, J = 8.5, 6.8 and 1.2 Hz, Ar-H); 6.87 (dd,
1H, J = 8.5 and 1.2 Hz, Ar-H); 7.44 (ddd, 1H, J = 8.5, 6.8 and
1.2 Hz, Ar-H); 8.17 (dd, 1H, J = 8.5 and 1.2 Hz, Ar-H); 13C NMR
(DMSO-d6, 25 °C, 125.8 MHz), d: 14.36, 33.92, 38.51, 61.01,
113.39, 115.52, 127.04, 132.30, 136.19, 144.92, 171.25; MS (ESI+),
calcd for C11H14N2O4 m/z [M+Na]+ = 261.1, found = 261.1. Anal
Calcd for C11H14N2O4: C, 55.46; H, 5.92; N, 11.76. Found: C,
55.42; H, 6.00; N, 11.57.
ified silica (Phenomenex Luna, 15
l
m). Thin-layer (TLC) and re-
4.2.2. Ethyl 3-(3-methylisoalloxazin-10-yl)propanoate 18
versed-phase thin-layer (RPTLC) chromatographies were
performed with precoated Silica Gel 60F and RP-18F plates
(E. Merck). HPLC analyses were recorded on a Knauer Smartline
with a UV detector and a detector of optical rotatory power
CHIRALYSER.
Nitroaniline 17 (4.1 g; 17.2 mmol) was dissolved in acetic acid
(70 mL) and then palladium on carbon (10%) was added
(150 mg). The reaction mixture was stirred under hydrogen
(0.4 MPa) for 18 h. After filtration through Celite TLC, solution of
the obtained N-(ethoxycarbonylethyl)benzene-1,2-diamine was
used for the synthesis of 18. A solution of N-(ethoxycarbonyleth-
yl)benzene-1,2-diamine was added dropwise (over 1 h) to a solu-
tion of 3-methylalloxan hydrate (3.48 g, 20 mmol) in acetic acid
(70 mL) heated to 80 °C under a nitrogen atmosphere. After cooling
the solution, the solvents were evaporated in vacuo. A crude solid
product was recrystallized from ethanol, washed with diethyl
ether, and dried in vacuo. An orange powder (1.6 g, 28%, two steps)
was obtained. Mp 229–231 °C (Ref. 13 201–204 °C). 1H NMR
(DMSO-d6, 25 °C, 500.1 MHz), d: 1.19 (t, 3H, J = 7.1 Hz, CH3); 2.82
(t, 2H, J = 7.7 Hz, CO–CH2–CH2–N); 3.30 (s, 3H, N–CH3), 4.11 (q,
2H, J = 7.1 Hz, CH2–CH3); 4.85 (t, 2H, J = 7.7 Hz, CO–CH2–CH2–N);
7.68 (ddd, 1H, J = 8.5, 6.8 and 1.2 Hz, Ar-H); 7.97 (ddd, 1H, J = 8.5,
6.8 and 1.2 Hz, Ar-H); 8.08 (dd, 1H, J = 8.5 and 1.2 Hz, Ar-H); 8.19
(dd, 1H, J = 8.5 and 1.2 Hz, Ar-H); 13C NMR (DMSO-d6, 25 °C,
125.8 MHz), d: 14.44, 28.49, 31.08, 40.02, 60.90, 116.77, 126.60,
132.35, 132.73, 135.54, 135.61, 138.13, 149.40, 155.56, 159.89,
(Methylisoalloxazin-3-yl)acetic acid 15,7d,7e (3-methylalloxa-
zin-1-yl)acetic acid 16, 7e 6-chloro-1-methyluracil,14 3-methylallo-
xan hydrate,17 mono-6-deoxy-6-amino-b-cyclodextrin 26,19
mono-6-deoxy-6-amino-a
-cyclodextrin 25,7e and mono-3-O-cin-
namyl-b-cyclodextrin 2216 were prepared according to the de-
scribed procedures. All other chemicals used were commercially
available. Correct elemental analysis for compounds 27–33 could
not be obtained unless variable numbers of water molecules were
taken into account. Thus, calculations based on the accurate
weights of these compounds (molarity, optical rotation, and yields)
were corrected with respect to the water content as deduced from
the elemental analysis.
4.2. Synthesis of carboxylic acids 13 and 14
4.2.1. N-[2-(Ethoxycarbonyl)ethyl]-2-nitroaniline 17
Ethyl 3-aminopropanoate (3.0 g; 25.6 mmol) was dissolved in
dried dioxane (25 mL), after which 1-fluoro-2-nitrobenzene
(3.5 g; 24.8 mmol) and dried potassium carbonate (9 g; 65 mmol)
were added and the mixture was stirred at 80 °C for 1 h (TLC mon-
itoring with ethylacetate). Potassium carbonate was filtered off
and the solvents were evaporated in vacuo. The crude product
was purified by column chromatography (hexane:ethylacetate,
1:1). Orange crystals (4.1 g; 69%) were obtained. Mp 66–67 °C
(Ref. 13 66–68.5 °C). 1H NMR (DMSO-d6, 25 °C, 500.1 MHz), d:
170.72; HR-MS (ESI), calcd for
351.10538, found = 351.10690.
C
16H16N4O4 m/z [M+Na]+ =
4.2.3. 3-(3-Methylisoalloxazin-10-yl)propionic acid 13
Ester 18 (1.8 g; 5.5 mmol) was dissolved in hydrochloric acid
(35%; 8 mL) and water (16 mL) was added. The mixture was stirred
at 80 °C for 2 h, then diluted with cold water (200 mL) and cooled
to 4 °C in a refrigerator. An orange precipitate was then collected,
washed with water and diethyl ether, and then dried in vacuo.

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1579
An orange powder (1.4 g; 85%) was obtained. Mp 334–336 °C. 1H
NMR. (DMSO-d6, 25 °C, 500.1 MHz), d: 2.75 (t, 2H, J = 7.7 Hz, N–
CH2–CH2–COOH); 3.29 (s, 3H, N–CH3); 4.81 (t, 2H, J = 7.7 Hz, N–
CH2–CH2–COOH), 7.68 (ddd, 1H, J = 8.4, 7.2 and 1.2 Hz, Ar-H),
7.97 (ddd, 1H, J = 8.6, 7.2 and 1.2 Hz, Ar-H); 8.08 (dd, 1H, J = 8.6
and 1.2 Hz, Ar-H); 8.19 (dd, 1H, J = 8.4 and 1.2 Hz Ar-H); 13C
NMR (DMSO-d6, 25 °C, 125.8 MHz), d: 28.50, 31.13, 39.99, 116.70,
126.56, 132.31, 132.74, 135.55, 135.61, 138.13, 149.39, 155.61,
159.92, 172.26; HR-MS (ESI), calcd for C14H12O4N4 m/z [M+H]+ =
301.09313, found = 301.09333. Anal. Calcd for C14H12N4O4: C,
56.00; H, 4.03; N, 18.66. Found C, 55.85; H, 4.13; N, 18.10.
(ddd, 1H, J = 8.0, 6.6 and 1.2 Hz, Ar-H); 8.02 (dd, 1H, J = 8.0 and
1.2 Hz, Ar-H); 8.22 (dd, 1H, J = 8.2 and 1.2 Hz, Ar-H); 13C NMR
(DMSO-d6, 25 °C, 125.8 MHz), d: 29.30, 42.84, 127.46, 129.13,
130.37, 130.94, 133.94 138.90, 142.18, 145.56, 149.99, 158.65,
168.83; HR-MS (ESI), calcd for
287.0774, found = 287.0773. Anal. Calcd for C13H10N4O4: C, 54.55;
H, 3.52; N, 19.57. Found C, 54.89; H, 3.58; N, 19.25.
C
13H10N4O4 m/z [M+H]+ =
4.3. Synthesis of cyclodextrin 24
4.3.1. 2-(b-Cyclodextrin-3I-O-yl)acetaldehyde 23
3-O-Cinnamyl-b-cyclodextrin (0.924 g; 0.74 mmol) was dis-
solved in methanol (110 mL) under sonication. Gaseous ozone
was then introduced into the solution with stirring. The progress
of the reaction was monitored using TLC (n-propanol:H2O:NH3
(25%):EtOAc, 6:3:1:1). After 10 min, no starting material could be
detected in the reaction mixture. The introduction of ozone was
discontinued and the reaction was quenched by the addition of
dimethylsulfide (6 mL). The solvents were then evaporated under
reduced pressure. The solid residue was suspended in acetone
and sonicated in an ultrasonic bath. The solid material was further
washed with acetone (2 ꢁ 20 mL) on a glass frit and then dried un-
der vacuum at 85 °C for 5 h. A white powder (0.927 mg; 94%, calcu-
lated for nonahydrate) was obtained. MS (ESI) calcd for C44H72O36
[M+H]+ 1177, found 1177; calcd [M+Na]+ = 1199, found 1199. Anal.
Calcd for C44H72O36. 9H2O: C, 39.46; H, 6.77. Found C, 39.43; H,
6.46.
4.2.4. 6-(N-Phenylamino)-1-methyluracil 19
6-Chloro-1-methyluracil (250 mg; 1.56 mmol) was dissolved in
aniline (500 mg; 5.4 mmol) and stirred at 150 °C for 90 min. The
solid mixture obtained was washed with ether and methanol and
dried. White crystals (304 mg; 90%) were obtained. Mp 308–
310 °C. 1H NMR (DMSO-d6, 25 °C, 500.1 MHz), d: 3.39 (s, 3H, N–
CH3); 4.51 (s, 1H, CH), 7.20–7.27 (m, 3H, Ar-H); 7.36–7.45 (m,
2H, Ar-H); 8.55 (s, 1H, NH);13C NMR (DMSO-d6, 25 °C, 125.8
MHz), d: 25.93, 77.91, 123.06, 125.53, 129.79, 138.92, 151.72,
155.35, 162.95; HR-MS (ESI), calcd for
C11H11N3O2 m/z
[M+Na]+ = 240.07435, found = 240.07468.
4.2.5. 1-Methylalloxazine 20
Uracil 19 (3.8 g; 17.5 mmol) was dissolved in acetic acid
(40 mL) and while stirring, sodium nitrite (5.6 g; 81 mmol) was
added. The reaction mixture was then stirred at ambient tempera-
ture for 2 h. Next, water (100 mL) was added and the mixture was
cooled to 4 °C in a refrigerator. An orange precipitate was collected,
washed with water and diethyl ether, and then dried in vacuo. Or-
ange crystals (2.54 g) containing a mixture of 1-methylalloxazine-
5-oxide and 1-methylalloxazine (5:1) were dissolved in the mix-
ture of DMF and AcOH (1:1), after which palladium on carbon
(150 mg, 10%) was added and the reaction mixture was stirred in
an autoclave under hydrogen (0.6 MPa) at room temperature for
20 h. The palladium on carbon was filtered off and washed with
hot acetic acid and DMF. The solvents were evaporated, washed
with diethyl ether, and dried in vacuo. Orange crystals (1.72 g;
43%) were obtained. Mp >350 °C. 1H NMR (DMSO-d6, 25 °C,
500.1 MHz), d: 3.57 (s, 3H, N–CH3); 7.82 (ddd, 1H, J = 8.2, 6.6 and
1.2 Hz, Ar-H); 7.97 (ddd, 1H, J = 8.0, 6.6 and 1.2 Hz, Ar-H); 8.03
(dd, 1H, J = 8.0 and 1.2 Hz, Ar-H); 8.21 (dd, 1H, J = 8.2 and 1.2 Hz,
Ar-H); 12.06 (s, 1H, NH); 13C NMR (DMSO-d6, 25 °C, 125.8 MHz),
d: 28.84, 127.84, 129.13, 130.41, 132.57, 133.90, 138.89, 142.44,
147.47, 150.87, 160.03; HR-MS (ESI), calcd for C11H8N4O2 m/z
[M+Na]+ = 251.0545, found = 251.0540.
4.3.2. 2-(b-Cyclodextrin-3-O-yl)ethane-1-amine 24
Aldehyde 23 (107 mg; 0.080 mmol, calculated for the nonahy-
drate) and ammonium acetate (105 mg; 1.67 mmol) were dis-
solved in dry methanol (500 mL) under an argon atmosphere.
The mixture was allowed to react for 10 min and then sodium cya-
noborohydride (6.9 mg; 0.11 mmol) was added under stirring. The
progress of the reaction was monitored using TLC (n-propanol/
H2O/NH3 (25%)/EtOAc 6:3:1:1). The reaction was quenched with
water (1 mL) after 63 h of the reaction time. The solvents were re-
moved on a rotatory evaporator and the product was isolated by
chromatography (reversed phase, gradient elution MeOH:H2O:TFA,
from 4:95.9:0.1 to 30:59.9:0.1). The product (as a salt with the TFA
anion) was isolated by freeze drying of partly concentrated
fractions: white amorphous material (37 mg; 31%, calcd for deca-
hydrate).
½
a 2D0
ꢂ
¼ þ141 (c 0.3, H2O); 1H NMR (D2O, 27 °C,
600.1 MHz), d: 2.94 (m, 2H, N–CH2–CH2–O); 3.56–3.60(3H), 3.685
and 3.80–3.85(3H) (m, 7H, H-4); 3.596(2H), 3.64–3.66(4H) and
3.743 (m, 7H, H-2); 3.82–3.90 (12H) and 3.973(2H) (m, 14H, H-
6); 3.82–3.87(6H) and 3.987 (7H, H-5); 3.92–3.97 (m, 7H, H-3);
3.965 (m, 2H, N–CH2–CH2–O); 5.045–5.09 (m, 7H, H-1); 13C NMR
(D2O, 27 °C, 150.9 MHz), d: 42.85 (N–CH2–CH2–CO); 62.90(2C),
62.91(2C), 62.92(2C) and 62.98 (7 ꢁ C-6); 74.24, 74.30, 74.64,
74.69, 74.73(2C) and 75.18 (7 ꢁ C-2); 74.40, 74.43, 74.46,
74.48(2C), 74.51 and 74.96 (7 ꢁ C-5); 75.11 (N–CH2–CH2–O);
75.62, 75.73(4C), 75.77 and 75.79 (7 ꢁ C-3); 81.55, 83.57, 83.64,
83.70, 83.73, 83.74 and 83.83 (7 ꢁ C-4); 103.76, 104.47(3C),
104.50 and 104.53(2C) (7 ꢁ C-1); HR-MS (ESI), calcd for
4.2.6. 2-(1-Methylalloxazine-3-yl)acetic acid 14
1-Methylalloxazine 20 (1.7 g; 7.5 mmol) and potassium carbon-
ate (2 g; 14 mmol) were dissolved/suspended in dry DMF (20 mL).
Ethyl bromacetate (6.4 g; 38 mmol) was then added and the mix-
ture was stirred at 50–60 °C overnight. Potassium carbonate was
then filtered off and the solvents were evaporated in vacuo. The
crude product was purified by column chromatography (Ethylace-
tate:MeOH:Hexane, 50:5:1). The obtained ester 21 was used for
the synthesis of acid 14.
C
44H75NO35 m/z [M+H]+ = 1178.41924, found = 1178.41865; calcd
[M+Na]+ 1200.40118, found 1200.40198. Anal. Calcd for
Ester 21 (2.0 g; 6.36 mmol) was dissolved in hydrochloric acid
(45 mL; 35%) and the mixture was stirred at 80 °C for 90 min. Cold
water (200 mL) was then added, and the mixture was cooled to
4 °C in a refrigerator. The precipitate was collected, washed with
water and diethyl ether, and dried in vacuo. The crude product
was recrystallized from 4 M acetic acid. An orange powder
(460 mg; 22%, two steps) was obtained. Mp 278–280 °C. 1H NMR
(DMSO-d6, 25 °C, 500.1 MHz), d: 3.65 (s, 3H, N–CH3); 4.69 (s, 2H,
N–CH2–CO); 7.83 (ddd, 1H, J = 8.2, 6.6 and 1.2 Hz, Ar-H); 7.99
C
44H75NO35.CF3COOHꢃ10H2O: C, 37.53; H, 6.57; N, 0.95. Found C,
37.21; H, 6.61, N, 1.11.
4.4. Synthesis of conjugate 27
4.4.1. N-(6-Deoxy-a-cyclodextrin-6-yl)-2-(10-methylisoalloxa-
zin-3-yl)acetamide 27
Acid 15 (67 mg, 0.234 mmol) and mono-6-deoxy-6-amino-a-
cyclodextrin 25 (204 mg; 0.189 mmol, calculated for hexahydrate)

1580
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
were dissolved in DMF (5 mL), after which DIPEA (70
lL;
4.5.2. N-(6-Deoxy-b-cyclodextrin-6-yl)-3-(3-methylisoalloxazin-
0.41 mmol) and PyBOP (116 mg; 0.193 mmol) were added. The
mixture was stirred for 1 h under a nitrogen atmosphere at ambi-
ent temperature, after which DMF was evaporated off. The solid
material was dried under vacuum and purified by reverse phase
chromatography (MeOH:Water, 3:7). The collected main fractions
were evaporated in vacuo. Conjugate 27 was isolated as a yellow
amorphous material (184 mg; 72%, calculated for hexahydrate).
1H NMR (CD3OD+D2O, 27 °C, 600.1 MHz), d: 3.48–3.62 (m, 12H,
H-4 and H-2); 3.62–3.88 (m, 6H, H-5); 3.88–3.92 (m, 6H, H-3);
3.42, 3.66–4.02(10H) and 3.93 (12 ꢁ m, 12H, H-6); 4.20 (s, 3H,
N–CH3); 4.72, 4.86 (2 ꢁ d, 2H, J = 16.2 Hz, N–CH2–CO); 4.934,
4.955, 4.985, 4.997, 5.016, 5.022 (6 ꢁ d, 6H, J = 3.4 Hz, H-1); 7.80
(ddd, 1H, J = 8.2, 7.0 and 1.4 Hz, fl-H-8); 8.035 (dd, 1H, J = 8.6 and
1.4 Hz, fl-H-6); 8.10 (ddd, 1H, J = 8.6, 7.0 and 1.4 Hz, fl-H-7); 8.28
(dd, 1H, J = 8.2 and 1.4 Hz, fl-H-9); 13C NMR (CD3OD+D2O, 27 °C,
150.9 MHz), d: 33.46 (N–CH3); 44.81 (N–CH2–CO); 41.53, 61.14,
61.27, 61.38, 61.41 and 61.55 (6 ꢁ C-6); 71.70 and 71.70–
73.28(11C) (6 ꢁ C-2 and 6 ꢁ C-5); 74.66 and 74.90(5C) (6 ꢁ C-3);
82.41, 82.46, 82.55, 82.59, 82.75 and 84.76 (6 ꢁ C-4); 103.05(4C),
103.13 and 103.25 (6 ꢁ C-1); 117.71 (fl-C-6); 128.64 (fl-C-8);
133.16 (fl-C-9); 134.90 (fl-C-5a); 137.06 (fl-C-4a); 137.62 (fl-C-
9a); 137.85 (fl-C-7); 150.90 (fl-C-10a); 157.80 (fl-C-2); 161.86 (fl-
C-4); 170.14 (NH-CO-CH2). MS (ESI+) calcd for C49H69N5O32 m/z
[M+Na]+ 1262.4, found 1263.0; [M+K]+ calcd 1278.4, found
1279.0. Anal Calcd for C49H69N5O32 6H2O: C, 43.65; H, 6.06; N,
5.19. Found: C, 43.46; H 6.17; N 5.17.
10-yl)propanamide 29
This compound was prepared from acid 13 (50 mg; 167
mono-6-deoxy-6-amino-b-cyclodextrin 26 (251 mg; 200
calculated for hexahydrate), PyBOP (120 mg; 193 mol), and DIPEA
(83 L; 0.49 mmol) by the same procedure as described for 28.
Conjugate 29 was isolated as yellow amorphous material
l
l
mol),
mol,
l
l
a
(182 mg; 77%, calculated for hexahydrate). 1H NMR (D2O, 37 °C,
600.1 MHz), d: 2.89 (m, 2H, CH2–CO); 3.04, 3.61–3.90(12H) and
3.94 (14 ꢁ m, 14H, H-6); 3.332 and 3.52–3.64(6H) (7 ꢁ m, 7H, H-
4); 3.44 (s, 3H, N–CH3); 3.60–3.66 (m, 7H, H-2); 3.62–3.94 (m,
7H, H-5); 3.87–3.95 (m, 7H, H-3); 4.96 and 5.12 (2 ꢁ m, 2H, N–
CH2); 4.974, 5.014, 5.037, 5.046, 5.060 and 5.080 (7 ꢁ d, 7H,
J = 3.8 Hz, H-1); 7.83 (ddd, 1H, J = 8.3, 7.2 and 1.1 Hz, fl-H-7);
7.96 (dd, 1H, J = 8.8 and 1.1 Hz, fl-H-9); 8.07 (ddd, 1H, J = 8.8, 7.2
and 1.5 Hz, fl-H-8); 8.24 (dd, 1H, J = 8.3 and 1.5 Hz, fl-H-6); 13C
NMR (D2O, 37 °C, 150.9 MHz), d: 31.13 (N–CH3); 35.78 (CH2–CO);
42.79, 62.77(2C), 62.88(2C), 62.93, 62.96 (7 ꢁ C-6); 44.78 (N–
CH2–CH2–CO); 73.63, 74.46, 74.49, 74.52(2C) and 74.59(2C)
(7 ꢁ C-5); 74.63(2C), 74.66, 74.67, 74.72, 74.74 and 74.77 (7 ꢁ C-
2); 75.48, 75.67, 75.79 and 75.83(4C) (7 ꢁ C-3); 83.36, 83.42,
83.60, 83.62, 83.66, 83.67 and 85.91 (7 ꢁ C-4); 104.20, 104.24,
104.39, 104.46 and 104.51(3C) (7 ꢁ C-1); 118.81 (fl-C-9); 130.31
(fl-C-7); 130.74 (fl-C-6); 135.38 (fl-C-9a); 138.60 (fl-C-5a); 138.89
(fl-C-4a); 139.34 (fl-C-8); 151.77 (fl-C-10a); 160.19 (fl-C-2);
163.90 (fl-C-4); 174.79 (N–CO–CH2). HR-MS (ESI+) calcd for
C
56H81N5O37 m/z [M+Na]+ 1415.460, found 1415.450. UV–VIS
(phosphate buffer pH 2.0): 350 nm, 473 nm. Anal. Calcd for
C56H81N5O37 7H2O: C, 43.61; H, 6.21; N, 4.54. Found: C, 43.63; H,
4.5. Synthesis of conjugates 28–31
6.42; N, 4.90.
4.5.1. N-(6-Deoxy-a-cyclodextrin-6-yl)-3-(3-methylisoalloxazin-
10-yl)propanamide 28
4.5.3. N-(6-Deoxy-a-cyclodextrin-6-yl)-2-(1-methylalloxazin-3-
Acid 13 (50 mg; 167
lmol), mono-6-deoxy-6-amino-a-cyclo-
yl)acetamide 30
dextrin 25 (216 mg; 200
and PyBOP (120 mg; 193
l
l
mol, calculated for the hexahydrate)
mol) were dissolved in DMF (5 mL).
This compound was prepared from acid 14 (50 mg; 175
mono-6-deoxy-6-amino- -cyclodextrin 25 (233 mg; 216
calculated for hexahydrate), PyBOP (136 mg; 219 mol), and DIPEA
(80 L; 0.47 mmol) by the same procedure as described for 28.
Conjugate 30 was isolated as yellow amorphous material
l
mol),
a
lmol,
Next, DIPEA (80
lL; 0.47 mmol) was added. The mixture was stir-
l
red overnight under a nitrogen atmosphere at ambient tempera-
ture. Acetone (50 mL) was then added, and the precipitate was
decanted (centrifuge). The solid was dried in vacuo and purified
by reverse phase chromatography (MeOH:Water, 3:7). The col-
lected main fractions were evaporated in vacuo. Conjugate 28
was isolated as a yellow amorphous material (145 mg; 69%, calcu-
lated for hexahydrate). 1H NMR (D2O, 37 °C, 600.1 MHz), d: 2.90
(m, 2H, CH2–CO); 3.24, 3.592, 3.703, 3.71 and 3.76–3.96(8H)
(12 ꢁ m, 12H, H-6); 3.323 and 3.53–3.59(5H) (6 ꢁ m, 6H, H-4);
3.24, 3.592, 3.703, 3.71 and 3.76–3.96(8H) (12 ꢁ m, 12H, H-6);
3.43 (s, 3H, N–CH3); 3.57–3.62 (6 ꢁ m, 6H, H-2); 3.60–3.95
(6 ꢁ m, 6H, H-5); 3.89–3.97 (6 ꢁ m, 6H, H-3); 4.949, 4.975,
5.020(2H), 5.025 and 5.035 (6 ꢁ d, 6H, J = 3.5 Hz, H-1); 4.99 and
5.07 (2 ꢁ m, 2H, N–CH2–CH2–CO); 7.81 (ddd, 1H, J = 8.3, 7.1 and
1.1 Hz, fl-H-7); 8.01 (dd, 1H, J = 8.8 and 1.1 Hz, fl-H-9); 8.10 (ddd,
1H, J = 8.8, 7.1 and 1.5 Hz, fl-H-8); 8.23 (dd, 1H, J = 8.3 and 1.5 Hz,
fl-H-6); 13C NMR (D2O, 37 °C, 150.9 MHz), d: 31.30 (N–CH3);
35.72 (CH2–CO); 43.03, 62.79, 62.89, 63.02, 63.11 and 63.21
(6 ꢁ C-6); 44.78 (N–CH2–CH2–CO); 72.90, 74.30, 74.35, 74.38(2C)
and 74.40 (6 ꢁ C-5); 74.42, 74.59(2C), 74.65, 74.66 and 74.68
(6 ꢁ C-2); 75.74, 75.95, 75.98(2C), 76.00 and 76.04 (6 ꢁ C-3);
83.60, 83.81(2C), 83.83, 83.89 and 85.93 (6 ꢁ C-4); 103.94,
104.01, 104.03, 104.08 and 104.13(2C) (6 ꢁ C-1); 118.97 (fl-C-9);
130.50 (fl-C-7); 134.82 (fl-C-6); 135.46 (fl-C-9a); 138.69 (fl-C-5a);
138.75 (fl-C-4a); 139.71 (fl-C-8); 151.62 (fl-C-10a); 160.66 (fl-C-
2); 164.19 (fl-C-4); 174.93 (N-CO-CH2); HR-MS (ESI+) calcd for
l
a
(117 mg; 33%, calculated for pentahydrate). 1H NMR (D2O, 37 °C,
600.1 MHz), d: 3.445 and 3.49–3.65(5H) (m, 6H, H-4); 3.49,
3.584, 3.616, 3.69, 3.772, 3.855, 3.90, 3.905, 3.91, 3.911, 3.961
and 4.002 (12 ꢁ m, 12H, H-6); 3.562, 3.568, 3.601, 3.617, 3.639
and 3.672 (6 ꢁ m, 6H, H-2); 3.67–3.95 (m, 6H, H-5); 3.71 (s, 3H,
N–CH3); 3.89–3.97 (m, 6H, H-3); 4.77 and 4.99 (2 ꢁ d, 2H,
J = 16.2 Hz, N–CH2–CO); 4.953, 4.984, 5.015, 5.053, 5.072 and
5.076 (6 ꢁ d, 6H, J = 3.7 Hz, H-1); 7.87 (ddd, 1H, J = 8.5, 6.7 and
1.6 Hz, fl-H-7); 8.01 (ddd, 1H, J = 8.6, 6.7 and 1.4 Hz, fl-H-8); 8.04
(ddd, 1H, J = 8.6, 1.6 and 0.5 Hz, fl-H-9); 8.15 (ddd, 1H, J = 8.5, 1.4
and 0.5 Hz, fl-H-6); 13C NMR (D2O, 37 °C, 150.9 MHz), d: 32.26
(N-CH3); 43.43, 62.67, 62.96(3C) and 63.10 (6 ꢁ C-6); 46.97 (N–
CH2–CO); 72.95, 74.34(2C), 74.37 and 74.40(2C) (6 ꢁ C-5); 74.43,
74.59(2C) and 74.62(3C) (6 ꢁ C-2); 75.75, 75.98, 75.99, 76.01(2C)
and 76.09 (6 ꢁ C-3); 83.72(2C), 83.79(2C), 83.83 and 86.00
(6 ꢁ C-4); 104.03, 104.05, 104.07, 104.14 and 104.21(2C) (6 ꢁ C-
1); 130.10 (fl-C-9); 131.78 (fl-C-4a); 132.19 (fl-C-6); 132.91 (fl-C-
7); 137.62 (fl-C-8); 141.68 (fl-C-5a); 145.82 (fl-C-9a); 147.91 (fl-
C-10a); 153.75 (fl-C-2); 163.26 (fl-C-4); 171.87 (N–CO–CH2); HR-
MS (ESI+) calcd for C49H69N5O32 [M+Na]+ m/z 1239.390; found
1239.382. UV–VIS (phosphate buffer pH 2.0) 333 nm, 383 nm.
Anal. Calcd for C49H69N5O32 5H2O: C, 44.24; H, 5.99; N 5.27. Found:
C, 44.21; H, 5.95; 5.16.
C
50H71N5O32 m/z [M+Na]+ 1253.410, found 1253.399. UV–VIS
(phosphate buffer pH 2.0): 352 nm, 468 nm. Anal. Calcd for
50H71N5O32 7H2O: C, 43.51; H, 6.21; N, 5.07. Found: C, 43.70; H
6.15; N 4.91.
4.5.4. N-(6-Deoxy-b-cyclodextrin-6-yl)-2-(1-methylalloxazin-3-
yl)acetamide 31
C
This compound was prepared from acid 14 (50 mg; 175
lmol),
mono-6-deoxy-6-amino-b-cyclodextrin 26 (270 mg; 215
l
mol,

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1581
calculated for hexahydrate), PyBOP (136 mg; 219
(83 L; 0.49 mmol) by the same procedure as described for 28.
Conjugate 31 was isolated as yellow amorphous material
l
mol), and DIPEA
145.81 (fl-C-5a); 147.40 (fl-C-10a); 154.32 (fl-C-2); 164.05 (fl-C-
4); 172.92 (NH–CO–CH2); MS (ESI) calcd for [M+Na]+ m/z 1468.5;
found 1468.8. UV–VIS (phosphate buffer pH 2.0): 322 nm,
376 nm. Anal. Calcd for C57H83N5O38.8H2O: C, 43.05; H, 6.27; N,
4.40. Found C, 43.18; H, 6.28; N, 4.69.
l
a
(195 mg; 52%, calculated for pentahydrate). 1H NMR (D2O, 37 °C,
600.1 MHz), d: 3.28, 3.54–3.85(5H) and 4.07 (7 ꢁ m, 1H, H-5);
3.435, 3.50, 3.53, 3.67–3.75(4H), 3.82–3.90(6H) and 3.96 (m, 14H,
H-6); 3.482, 3.504, 3.535-3.65(3H), 3.542 and 3.573 (7 ꢁ m, 7H,
H-4); 3.538, 3.607, 3.611, 3.613, 3.622, 3.664 and 3.714 (7 ꢁ m,
7H, H-2); 3.706, 3.794–3.847(4H), 3.959 and 4.00 (7 ꢁ m, 7H, H-
3); 3.75 (s, 3H, N-CH3); 4.84 and 4.96 (2 ꢁ d, 2H, J = 16.0 Hz, N–
CH2–CO); 4.944, 4.972, 5.013, 5.043, 5.062, 5.082 and 5.088
(7 ꢁ d, 7H, J = 3.8 Hz, H-1); 7.97 (ddd, 1H, J = 8.5, 6.8 and 1.4 Hz,
fl-H-7); 8.11 (dd, 1H, J = 8.6, 6.8 and 1.4 Hz, fl-H-8); 8.16 (dd, 1H,
J = 8.6 and 1.4 Hz, fl-H-9); 8.28 (dd, 1H, J = 8.5 and 1.4 Hz, fl-H-6);
13C NMR (D2O, 37 °C, 150.9 MHz), d: 32.21 (N–CH3); 43.23, 62.62,
62.70, 62.76, 62.77, 62.80 and 62.82 (7 ꢁ C-6); 47.30 (N–CH2–
CO); 73.32, 74.39, 74.41(2C), 74.46 and 74.53(2C) (7 ꢁ C-5);
74.55, 74.67(2C), 74.68, 74.72, 74.81 and 74.82 (7 ꢁ C-2);
75.70(2C), 75.77, 75.80(2C), 75.83 and 75.86 (7 ꢁ C-3); 82.93,
83.37, 83.60(2C), 83.65, 83.89 and 86.00 (7 ꢁ C-4); 104.10,
104.31, 104.54, 104.55, 104.61, 104.64 and 104.66 (7 ꢁ C-1);
130.09 (fl-C-9); 132.21 (fl-C-4a); 132.32 (fl-C-6); 132.88 (fl-C-7);
137.61 (fl-C-8); 141.95 (fl-C-5a); 145.80 (fl-C-9a); 148.20 (fl-C-
10a); 153.86 (fl-C-2); 163.49 (fl-C-4); 171.81 (N-CO-CH2). HR-MS
(ESI+) calcd for C55H79N5O37 [M+H+] m/z 1401.45; found 1401.45.
UV–VIS (phosphate buffer pH 2.0) 331 nm, 383 nm. Anal. Calcd
for C55H79N5O37 5H2O: C, 44.27; H, 6.01; N, 4.69. Found: C,
44.02; H, 5.99; N, 4.56.
4.6.2. N-[2-(b-Cyclodextrin-3-O-yl)ethyl]-2-(10-methylisoallo-
xazine-3-yl)acetamide 33
This compound was prepared from the hydrochloride salt of
amine 24 (18.9 mg; 0.014 mmol), PyBOP (9.75 mg; 0.019 mmol),
DIPEA (14 lL; 0.08 mmol), and 2-(10-methylisoalloxazin-3-yl)ace-
tic acid (4.11 mg; 0.014 mmol) by the same procedure as described
for 32. Conjugate 33 was isolated as a yellow amorphous material
(18 mg; 77%). 1H NMR (D2O, 27 °C, 600.1 MHz, 40 °C), d: 3.13, 3.55
and 3.59–3.64(5H) (7 ꢁ m, 7H, H-2); 3.52–3.58(5H), 3.62 and 3.74
(7 ꢁ m, 7H, H-4); 3.55–3.94 (14 ꢁ m, 14H, H-3 and H-5); 3.74–3.87
(14 ꢁ m, 14H, H-6); 4.63, 4.95, 5.044, 5.049, 5.051 and 5.055(2H)
(7 ꢁ d, 7H, J = 3.8 Hz, H-1); 3.37 and 3.51 (2 ꢁ m, 2H, CH2–CH2–
NH); 3.92 and 4.11 (2 ꢁ m, 2H, O–CH2–CH2); 4.17 (s, 3H, N–CH3);
4.73 and 4.83 (2 ꢁ d, 2H, J = 16.5 Hz, N–CH2–CO); 7.83 (ddd, 1H,
J = 8.3, 7.1 and 1.2 Hz, fl-H-8); 8.03 (dd, 1H, J = 8.8 and 1.2 Hz, fl-
H-6); 8.11 (ddd, 1H, J = 8.8, 7.1 and 1.6 Hz, fl-H-7); 8.21 (dd, 1H,
J = 8.3 and 1.6 Hz, fl-H-9). 13C NMR (D2O, 27 °C, 150.9 MHz,
40 °C), d: 35.59 (N–CH3); 43.25 (CH2–CH2–N); 47.07 (CO–CH2–N);
62.54, 62.82(5C) and 62.94 (7 ꢁ C-6); 72.76 (O–CH2–CH2); 74.15,
74.20, 74.24, 74.34, 74.36(2C) and 82.96 (7 ꢁ C-5); 74.38, 74.47,
74.56, 74.62(2C), 74.74 and 74.80 (7 ꢁ C-3); 74.98, 75.57, 75.62,
75.65, 75.68(2C) and 75.71 (7 ꢁ C-2); 81.16, 83.56, 83.57, 83.60,
83.68(2C), and 84.20 (7 ꢁ C-4); 103.85, 104.05, 104.30, 104.34,
104.39(2C), and 104.43 (7 ꢁ C-1); 119.41 (fl-C-6); 130.87 (fl-C-8);
134.39 (fl-C-9); 136.26 (fl-C-9a); 138.45 (fl-C-5a); 138.90 (fl-C-
4a); 139.94 (fl-C-7); 152.17 (fl-C-10a); 159.43 (fl-C-2); 163.74 (fl-
C-4); 172.48 (N–CO–CH2). MS (ESI) calcd for C57H83N5O38
[M+Na]+ m/z 1468.5, found 1468.8. UV–VIS (phosphate buffer pH
4.6. Synthesis of conjugates 32 and 33
4.6.1. N-[2-(b-Cyclodextrin-3I-O-yl)ethyl]-2-(3-methylalloxazine-
1-yl)acetamide 32
The TFA salt of amine 24 (24 mg; 0.016 mmol, calculated for the
decahydrate) was transformed into the HCl salt by passing its solu-
tion through Dowex 18 (0.1 mL, in the Clꢀ cycle). The residue
(20.5 mg) was dried in vacuo and then dissolved in dry DMF
(0.5 mL). To this solution, PyBOP (9.9 mg; 0019 mmol), DIPEA
(0.15 mL; 0.086 mmol), and 2-(3-methylalloxazin-1-yl)ethanoic
acid 16 were added subsequently with stirring and under an argon
atmosphere. The mixture was allowed to react for 20 h. The solvent
was then evaporated under reduced pressure and the product was
isolated by chromatography (reversed phase, MeOH–H2O, gradient
from 5:95 to 30:70). White amorphous solid (18 mg, 71%, calcd for
octahydrate). 1H NMR (D2O, 40 °C, 600.1 MHz), d: 2.22, 3.468,
3.762, 3.967, 4.077, 4.45 and 4.463 (7 ꢁ m, 7H, H-3); 2.803,
3.452, 3.465, 3.86, 3.905, 3.908 and 4.064 (7 ꢁ m, 7H, H-5); 3.15,
3.49, 3.531, 3.626, 3.721, 3.907 and 4.064 (7 ꢁ m, 7H, H-4);
3.434, 3.546, 3.559, 3.601, 3.724, 3.755 and 3.832 (7 ꢁ m, 7H, H-
2); 3.550, 3.598, 3.686, 3.806, 3.875(2H), 3.90, 3.903, 3.95, 3.976,
3.986, 4.036, 4.109 and 4.175 (14 ꢁ m, 14H, H-6); 4.811, 4.920,
5.002, 5.078, 5.123, 5.154 and 5.188 (7 ꢁ d, 7H, J = 3.5 Hz, H-1);
3.32 and 3.66 (2 ꢁ m, 2H, CH2–CH2–NH); 3.59 (s, 3H, N–CH3);
3.76 and 4.18 (2 ꢁ m, 2H, O–CH2–CH2); 5.11 and 5.45 (2 ꢁ d,
J = 16.4 Hz, N–CH2–CO); 7.87 (ddd, 1H, J = 8.6, 6.9 and 1.3 Hz, fl-
H-8); 8.04 (ddd, 1H, J = 8.5, 6.9 and 1.4 Hz, fl-H-7); 8.27 (dd, 1H,
J = 8.5 and 1.3 Hz, fl-H-6); 8.36 (dd, 1H, J = 8.6 and 1.4 Hz, fl-H-9);
13C NMR (D2O, 40 °C, 150.9 MHz), d: 31.97 (N–CH3); 43.42 (CH2–
CH2–N); 47.97 (CO–CH2–N); 62.67, 62.89, 62.91, 62,99, 63,04,
63,10 and 63,13 (7 ꢁ C-6); 74.11, 74.58, 74.63(2C), 74.82, 74.89
and 76.19 (7 ꢁ C-2); 74.14, 75.14, 75.28, 75.74, 75.83, 75.90 and
82.13 (7 ꢁ C-5); 74.63, 74.86, 75.50(2C), 75.57, 76.00 and 76.29
(7 ꢁ C-3); 76.38 (O–CH2–CH2); 83.29, 83.70, 83.72, 83.80, 83.92,
84.45 and 85.00 (7 ꢁ C-4); 104.27, 104.49, 104.56, 104.54,
104.66(2C) and 105.27 (7 ꢁ C-1); 130.27 (fl-C-6); 132.26 (fl-C-8);
132.36 (fl-C-9); 132.57 (fl-C-4a); 136.82 (fl-C-7); 142.46 (fl-C-9a);
2.0): 347 nm, 432 nm. Anal. Calcd for
43.05; H, 6.27; N, 4.40. Found C, 42.67; H, 5.92; N, 4.40.
C57H83N5O38.8H2O: C,
4.7. Synthesis of flavinium catalysts 6–12
4.7.1. 3-{[N-(6-Deoxy-
5-ethyl-10-methylisoalloxazin-5-ium perchlorate 6
Conjugate 27 (4.78 mg; 3.8 mol) and acetaldehyde (60
1 mmol) were dissolved in a mixture of ethanol (320 L), water
(250 L), and 0.1 M HClO4 (320 L). The mixture was degassed
a-cyclodextrin-6-yl)carbamoyl]methyl}-
l
lL;
l
l
l
by a vacuum–argon cycle, after which palladium on charcoal
(10% w/w, 0.51 mg) was added. The reaction mixture was placed
into an autoclave equipped with a magnetic stirring bar and hydro-
gen was introduced (6 bar) for 24 h. The palladium was then fil-
tered off, after which ethanol and the residual acetaldehyde were
evaporated off under reduced pressure. The resulting solution of
6 was diluted to 2 mL. The concentration of the catalyst in the stock
solution was recalculated with respect to the concentration of the
conjugates prior to the alkylation and the actual volume of the
sample. MS (ESI) calcd for C51H74N5O32 [M+Na]+ m/z 1291.4; found
1292.0. UV–VIS (1 M HCl): 272 nm, 301 nm, 351 nm, 460 nm; after
addition of excess of H2O2: 395 nm, 533 nm.
4.7.2. 10-{3-[N-(6-Deoxy-
5-ethyl-3-methylisoalloxazin-5-ium perchlorate 7
Acetaldehyde (170 L, 3.03 mmol) and palladium on carbon
(10%, 2.5 mg, 2.52 mol) were added to solution of 28
(14.55 mg, 10.5 mol, calculated for the heptahydrate) in ethanol
L), perchloric acid (0.1 M in water, 800 L), and water
L). The resulting mixture was stirred for 21 h in the auto-
a-cyclodextrin-6-yl)carbamoyl]ethyl}-
l
l
a
l
(800
(600
l
l
l
clave under a hydrogen atmosphere (0.7 MPa) at room tempera-
ture. The palladium was then filtered off, and ethanol and the

1582
T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
residual acetaldehyde were evaporated off under reduced pressure.
The resulting solution of 7 was diluted to 2 mL. This stock solution
was frozen and stored at ꢀ78 °C. The concentration of the catalyst
in the stock solution was recalculated with respect to the concen-
tration of the conjugates prior to the alkylation and the actual vol-
ume of the sample. UV–VIS (phosphate buffer pH 2.0): 351 nm,
473 nm, 567 nm. HR-MS (ESI+) calcd for C52H76O32N5 m/z: [M]+
1282.4458, found 1282.4486.
vacuum–argon cycle, and then palladium on charcoal (10% w/w,
1.9 mg) was added. The reaction mixture was placed into an auto-
clave equipped with a magnetic stirrer bar and hydrogen was
introduced (6 bar) for 24 h. The hydrogen was then released and
the catalyst was filtered off under argon. The remaining solution
was directly used for catalytic sulfoxidations. UV–VIS (phosphate
buffer pH 2.0): 347 nm, 432 nm. HR-MS (ESI+) calcd for
C
59H88N5O38 [M]+ m/z 1474.51018; found 1474.51063.
4.7.3. 10-{3-[N-(6-Deoxy-b-cyclodextrin-6-yl)carbamoyl]ethyl}-
4.7.7. 3-{N-[2-(b-Cyclodextrin-3-O-yl)ethyl]carbamoylmethyl}-
5-ethyl-3-methylisoalloxazin-5-ium perchlorate 8
5-ethyl-10-methylisoalloxazin-5-ium-perchlorate 12
Acetaldehyde (170
(10%, 3 mg, 2.79 mol) were added to a solution of 29 (14.77 mg,
9.6 mol, calculated for the heptahydrate) in ethanol (800 L), per-
chloric acid (0.1 M in water, 800 L), and water (600 L). The
l
L, 3.03 mmol) and palladium on carbon
Conjugate 33 (13.05 mg; 0.0082 mmol), acetaldehyde (170 lL;
l
3 mmol), and Pd/C (10% w/w; 1.66 mg; 0.0015 mmol) were reacted
as described above for compound 11. UV–VIS (phosphate buffer pH
2.0): 354 nm, 480 nm, 562 nm. HR-MS (ESI+) calcd for C59H88N5O38
[M]+m/z 1474.51018; found 1474.51045.
l
l
l
l
resulting mixture was stirred for 21 h in the autoclave under a
hydrogen atmosphere (0.7 MPa) at room temperature. The palla-
dium was then filtered off, and ethanol and the residual acetalde-
hyde were evaporated off under reduced pressure. The resulting
solution of 8 was diluted to 2 mL. This stock solution was frozen
and stored at ꢀ78 °C. The concentration of the catalyst in the stock
solution was recalculated with respect to the concentration of the
conjugates prior to the alkylation and the actual volume of the
sample. UV–VIS (phosphate buffer pH 2.0): 352 nm, 470 nm,
567 nm. HR-MS (ESI+) calcd for C58H86O37N5 [M]+ m/z 1444.4996,
found 1444.4997.
4.8. Catalytic procedures
All catalytic oxidations of alkyl or aryl methyl sulfides were per-
formed in 1 mL thick-walled screw-capped vials (Supelco). The
reaction mixtures were prepared by the addition of a sodium–
phosphate buffer (pH 2.9 for isoalloxazinium catalysts 5–8 and
12, pH 7.5 for alloxazinium catalysts 1–4 and 9–11; 0.05 M,
300
l
L), a liquid substrate (3–4 ꢁ 10ꢀ5 mol), an appropriate vol-
ume of catalyst solution (1–5 mol %), and then the hydrogen perox-
ide (2.3 mol equiv with respect to substrate). The vial was capped
and the reaction mixture was shaken with a Wrist Action Shaker
for a given period of time. The reaction was then quenched by
the addition of a solution of sodium dithionite in water (1.4 M,
0.17 mL). The product and the remaining substrate were extracted
with carbon tetrachloride (2 ꢁ 0.5 mL) and then with CDCl3
(0.50 mL). The combined extracts were mixed and dried over so-
dium sulfate. One third of the extracts was used for the NMR anal-
ysis of the conversion. The remaining solution was used for the
analysis of the enantiomeric excess of the isolated sulfoxides by
HPLC on chiral phases or/and by NMR employing chiral shift re-
agent (R)-(ꢀ)-3,5-dinitro-N-(1-phenylethyl)benzamide.20,21
4.7.4. 3-{2-[N-(6-Deoxy-
5-ethyl-1-methylalloxazin-5-ium perchlorate 9
Acetaldehyde (170 L, 3.03 mmol) and palladium on carbon
(10%, 2.6 mg, 2.62 mol) were added to a solution of 30 (14.50
mg, 10.9 mol, calculated for pentahydrate) in ethanol (800 L),
perchloric acid (0.1 M in water, 800 L), and water (600 L). The
a-cyclodextrin-6-yl)carbamoyl]methyl}-
l
l
l
l
l
l
resulting mixture was stirred for 21 h in the autoclave under a
hydrogen atmosphere (0.7 MPa) at room temperature. The palla-
dium was then filtered off, and ethanol and the residual acetalde-
hyde were evaporated under reduced pressure. The resulting
solution of 9 was diluted to 2 mL. This stock solution was frozen
and stored at ꢀ78 °C. The concentration of the catalyst in the stock
solution was recalculated with respect to the concentration of the
conjugates prior to the alkylation and the actual volume of the
sample. UV–VIS (phosphate buffer pH 2.0): 385 nm, 450 nm. HR-
MS (ESI+) calcd for C57H84N5O37 [M]+ m/z 1430.48, found 1430.48.
Acknowledgements
This work was financially supported by the Czech Science Foun-
dation (Grants Nos. P207/12/0447, 203/09/1919), IOCB (RVO:
61388963) and Ministry of Education, Youth, and Sports of the
Czech Republic (MSM0021620857).
4.7.5. 3-{2-[N-(6-Deoxy-b-cyclodextrin-6-yl)carbamoyl]-
methyl}-5-ethyl-1-methylalloxazin-5-ium perchlorate 10
Acetaldehyde (170
(10%, 2.3 mg, 2.14 mol) were added to a solution of 31 (15 mg,
10.1 mol, calculated for the pentahydrate) in ethanol (800 L),
perchloric acid (0.1 M in water, 800 L), and water (600 L). The
lL, 3.03 mmol) and palladium on carbon
l
References
l
l
1. Noyori, R.; Aoki, M.; Sato, K. Chem. Commun. 2003, 1977–1986.
2. (a) Russo, A.; De Fusco, C.; Lattanzi, A. ChemCatChem 2012, 4, 901–916; (b)
Bäckvall, E. Modern Oxidation Methods; Wiley-VCh: Weinheim, 2010.
l
l
resulting mixture was stirred for 21 h in the autoclave under a
hydrogen atmosphere (0.7 MPa) at room temperature. The palla-
dium was then filtered off, and ethanol and the residual acetalde-
hyde were evaporated off under reduced pressure. The resulting
solution of 10 was diluted to 2 mL. This stock solution was frozen
and stored at ꢀ78 °C. The concentration of the catalyst in the stock
solution was recalculated with respect to the concentration of the
conjugates prior to the alkylation and the actual volume of the
sample. UV–VIS (phosphate buffer pH 2.0): 385 nm, 449 nm. HR-
MS (ESI+) calcd for C51H75N5O32 [M]+ m/z 1268.43, found 1268.43.
3. (a) Carreno, M. C.; Hernandez-Torres, G.; Ribagorda, M.; Urbano, A. Chem.
Commun. 2009, 46, 6129–6144; (b) Legros, J.; Dehli, J. R.; Bolm, C. Adv. Synth.
Catal. 2005, 347, 19–31; (c) Caron, S.; Dugger, R. W.; Ruggeri, S. G.; Ragan, J. A.;
Ripin, D. H. B. Chem. Rev. 2006, 106, 2943–2989.
4. (a) Wojaczynska, E.; Wojaczynski, J. Chem. Rev. 2010, 110, 4303–4356; (b)
Fernandez, I.; Khiar, N. Chem. Rev. 2003, 103, 3651–3705; (c) Bryliakov, K. P.;
Talsi, E. P. Curr. Org. Chem. 2008, 12, 386–404; (d) Kagan, H. B. Organosulfur
Chemistry in Asymmetric Synthesis; Wiley-VCH: Weinheim, 2008.
5. (a) Fujisaki, J.; Matsumoto, K.; Katsuki, T. J. Am. Chem. Soc. 2011, 133, 56–61; (b)
Wu, Y.; Liu, J.; Li, X.; Chan, A. S. C. Eur. J. Org. Chem. 2009, 2607–2610; (c) Drago,
C.; Caggiano, L.; Jackson, R. F. W. Angew. Chem. 2005, 44, 7221–7223; (d) Zeng,
Q.; Wang, H.; Wang, T.; Cai, Y.; Weng, W.; Zhao, Y. Adv. Synth. Catal. 2005, 347,
1933–1936; (e) Matsumoto, K.; Yamaguchi, T.; Katsuki, T. Chem. Commun.
2008, 1704–1706; (f) Yamaguchi, T.; Matsumoto, K.; Saito, B.; Katsuki, T.
Angew. Chem., Int. Ed. 2007, 46, 4729–4731; (g) Legros, J.; Bolm, C. Angew. Chem.
2003, 42, 5487–5489; (h) Legros, J.; Bolm, C. Angew. Chem. 2004, 43, 4225–
4228; (i) Legros, J.; Bolm, C. Chemistry 2005, 11, 1086–1092; (j) Egami, H.;
Katsuki, T. Synlett 2008, 1543–1546; (k) Scarso, A.; Strukul, G. Adv. Synth. Catal.
2005, 347, 1227–1234; (l) Massa, A.; Mazza, V.; Scettri, A. Tetrahedron:
4.7.6. 1-{N-[2-(b-Cyclodextrin-3-O-yl)ethyl]carbamoylmethyl}-
5-ethyl-3-methylalloxazine-5-ium-perchlorate 11
Conjugate 32 (12.53 mg; 7.9 lmol) and acetaldehyde (170 lL;
3 mmol) were dissolved in a mixture of ethanol (1 mL), water
(0.1 mL), and 0.1 M HClO4 (1 mL). The mixture was degassed by a

T. Hartman et al. / Tetrahedron: Asymmetry 23 (2012) 1571–1583
1583
Asymmetry 2005, 16, 2271–2275; (m) Egami, H.; Katsuki, T. J. Am. Chem. Soc.
2007, 129, 8940–8941; (n) Matsumoto, K.; Yamaguchi, T.; Fujisaki, J.; Saito, B.;
Katsuki, T. Chem. Asian J. 2008, 3, 351–358; (o) Tanaka, H.; Nishikawa, H.;
Uchida, T.; Katsuki, T. J. Am. Chem. Soc. 2010, 132, 12034–12041; (p) Sun, J.;
Zhu, C.; Dai, Z.; Yang, M.; Pan, Y.; Hu, H. J. Org. Chem. 2004, 69, 8500–8503.
6. (a) Stingl, K. A.; Tsogoeva, S. B. Tetrahedron: Asymmetry 2010, 21, 1055–1074;
(b) Armstrong, A. Enantioselective Organocatalysis; Wiley-VCH: Weinheim,
2007.
11. The analogous (3-methylisoalloxazin-10-yl)acetic acid has not been used for
the synthesis of conjugates due to its extremely low photostability; see the
literature.12
12. Kino, K.; Kobayashi, T.; Arima, E.; Komori, R.; Miyazawa, H. Bioorg. Med. Chem.
Lett. 2009, 19, 2070–2074.
13. Takeda, J.; Ohta, S.; Hirobe, M. J. Am. Chem. Soc. 1987, 109, 7677–7688.
14. Youssef, S.; Pfleiderer, W. J. Heterocycl. Chem. 1998, 35, 949–954.
15. Yoneda, F.; Sakuma, Y.; Ichiba, M.; Shinomura, K. J. Am. Chem. Soc. 1976, 98,
830–835.
7. (a) Shinkai, S.; Yamaguchi, T.; Manabe, O.; Toda, F. J. Chem. Soc., Chem. Commun.
1988, 1399–1401; (b) Murahashi, S. I. Angew. Chem., Int. Ed. 1995, 34, 2443–
ˇ
16. Jindrich, J.; Tišlerová, I. J. Org. Chem. 2005, 70, 9054–9055.
ˇ
ˇ
ˇ
2465; (c) Jurok, R.; Cibulka, R.; Dvoráková, H.; Hampl, F.; Hodacová, J. Eur. J. Org.
17. Cibulka, R.; Baxová, L.; Dvoráková, H.; Hampl, F.; Ménová, P.; Mojr, V.; Plancq,
ˇ
´
Chem. 2010, 5217–5224; (d) Mojr, V.; Herzig, V.; Budešínsky, M.; Cibulka, R.;
B.; Sayin, S. Collect. Czech. Chem. Commun. 2009, 74, 973–993.
18. Imada, Y.; Iida, H.; Ono, S.; Murahashi, S.-I. J. Am. Chem. Soc. 2003, 125, 2868–
2869.
ˇ
´
Kraus, T. Chem. Commun. 2010, 46, 7599–7601; (e) Mojr, V.; Budešínsky, M.;
Cibulka, R.; Kraus, T. Org. Biomol. Chem. 2011, 9, 7318–7326.
8. (a) Gelalcha, F. G. Chem. Rev. 2007, 107, 3338–3361; (b) Ghisla, S.; Massey, V.
Eur. J. Biochem. 1989, 181, 1–17; (c) Moonen, M. J. H.; Fraaije, M. W.; Rietjens, I.;
Laane, C.; van Berkel, W. J. H. Adv. Synth. Catal. 2002, 344, 1023–1035; (d)
Ménová, P.; Cibulka, R. J. Mol. Catal. A: Chem. 2012, 363–364, 362–370.
9. Liu, Z.-M.; Zhao, H.; Li, M.-Q.; Lan, Y.-B.; Yao, Q.-B.; Tao, J.-C.; Wang, X.-W. Adv.
Synth. Catal. 2012, 354, 1012–1022.
19. Petter, R. C.; Salek, J. S.; Sikorski, C. T.; Kumaravel, G.; Lin, F. T. J. Am. Chem. Soc.
1990, 112, 3860–3868.
20. Deshmukh, M.; Dunach, E.; Juge, S.; Kagan, H. B. Tetrahedron Lett. 1984, 25,
3467–3470.
21. The addition of carbon tetrachloride to a solution of complexes with a shift
reagent significantly improved the separation of the peaks of the
diastereomers compared to neat chloroform solutions and allowed more
precise integration.
ˇ
´
10. Liao, S.; Coric, I.; Wang, Q.; List, B. J. Am. Chem. Soc. 2012, 134, 10765–
10768.