Synthesis of novel benzoxathiazepine-1,1-dioxides by means of a one-pot multicomponent reaction

Article abstract of DOI:10.1016/j.tetlet.2009.12.099

A simple one-pot synthesis of benzoxathiazepine-1,1-dioxides is described. Increased yields are afforded when suitable substituents are present in one of the starting materials. These additional substituents also provide a handle for further functionalization.

Full text of DOI:10.1016/j.tetlet.2009.12.099

Tetrahedron Letters 51 (2010) 1079–1082
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Synthesis of novel benzoxathiazepine-1,1-dioxides by means of a one-pot
multicomponent reaction
*
Ed Cleator , Carl A. Baxter, Michael O’Hagan, Timothy J. C. O’Riordan, Faye J. Sheen, Gavin W. Stewart
Department of Process Research, Merck Sharp and Dohme, Hertford Road, Hoddesdon, Hertfordshire EN11 9BU, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A simple one-pot synthesis of benzoxathiazepine-1,1-dioxides is described. Increased yields are afforded
when suitable substituents are present in one of the starting materials. These additional substituents also
provide a handle for further functionalization.
Received 27 October 2009
Revised 8 December 2009
Accepted 16 December 2009
Available online 24 December 2009
Ó 2009 Published by Elsevier Ltd.
Multicomponent reactions (MCRs) are convergent reactions in
which three or more starting materials react to form a product. A
MCR has distinct advantages over a stepwise process, including
the removal of the need to purify after each chemical transforma-
tion and the ability to produce a large number of derivatives
around a common scaffold by varying each of the components
involved in the reaction.1 However, it has the requirement that
the reagents must react in a defined and predictable way to yield
a major desired product, whilst minimizing side reactions which
would result in the formation of by-products. The outcome of a
MCR is dependent on a variety of reaction parameters, such as sol-
vent, temperature, catalyst, concentration, as well as the interac-
tion of the functional groups in the starting materials and the
products. Such considerations are of particular importance in con-
nection with the design and discovery of novel MCRs.
We envisaged that the application of this methodology to sub-
strates bearing an adequately positioned leaving group in the
intermediate 7 could lead to a rapid synthesis of benzoxathiaze-
pines 1 via a MCR, (Scheme 2). Herein we report the development
of a novel MCR in which an ortho-halo sulfonyl chloride 5, an
amine and an epoxide react in a sequential fashion to furnish
exclusively, a benzoxathiazepine derivative 1. This one-pot process
would be expected to give entry to privileged structures in high
yields, with the added benefits of not requiring purification of
any of the reaction intermediates, additionally it would also benefit
from using readily available starting materials. Furthermore, suit-
ably functionalized examples of 1, would allow the option for fur-
ther structural elaboration.
The proposed synthetic route to benzoxathiazepines 1 involves
the addition of a primary amine to an ortho-halo sulfonyl chloride
5 and the subsequent reaction of the resulting sulfonamide 6 with
an epoxide to give alcohol 7. Finally, the aromatic nucleophilic
substitution reaction (SNAr) of the aryl halide would result in the
formation of the desired cyclized compound 1.
Benzo-fused rings are prevalent in pharmacologically active
drugs. In particular, benzoxathiazepines such as 1 (Fig. 1) have gen-
erated considerable interest as potential treatments for hyperten-
sive disorders2 as well as diabetes.3
Although there are several syntheses of scaffolds analogous to 1
reported in the literature, they generally rely on the use of poten-
tially hazardous key steps and/or multistep synthetic sequences.4
As a result of their potential interest to the pharmaceutical indus-
try and due to the lack of satisfactory synthetic methods to prepare
them, an expedient synthesis of benzoxathiazepines 1 would be
desirable.
During recent studies within our research group, we have
shown that sulfonamides such as 2 efficiently ring-open epoxides
when heated in 1,4-dioxane in the presence of catalytic amounts
of tetra-n-butylammonium halides. Subsequent activation and
cyclization allowed us to prepare a range of N-alkylated-4-substi-
tuted isothiazolidine-1,1-dioxides, 4 (Scheme 1).5
On the basis of the previous work developed within our group,
the sulfonamide formation would be expected to occur at ambient
temperature whilst the epoxide ring-opening would require heat-
ing.5 The first step was expected to be very fast and quantitative; in
addition, the SNAr reaction could not proceed until the epoxide had
reacted with the sulfonamide 6 to form an alcohol 7. Bearing these
points in mind, we thought that there was an attractive opportu-
nity for the potential development of a MCR synthesis of com-
pounds with the general structure 1.
O
S
O
R'
N
R
O
R''
1
* Corresponding author. Tel.: +44 01992452179; fax: +44 01992 470437.
E-mail address: Edward_Cleator@merck.com (E. Cleator).
Figure 1. 3,4-Dihydro-2H-benzo[b][1,4,5]oxathiazepine-1,1-dioxides 1.
0040-4039/$ - see front matter Ó 2009 Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2009.12.099

1080
E. Cleator et al. / Tetrahedron Letters 51 (2010) 1079–1082
and 4, 10), and in both cases an increase in yield was observed
O O
S
O
S
a
b,c
R
O
when compared to the unsubstituted cases (Table 1, entries 1
and 7). The lower yield observed when the chlorine was at the
5-position could be rationalized on the basis of steric hindrance
(Table 1, entry 5). Compound 1c (Table 1, entry 3) was isolated
in a 91% yield, which is considerably higher than the 62% overall
yield of the stepwise process. This result illustrates the conve-
nience and efficiency of the MCR for the synthesis of compounds
of the general structure 1. When enantiopure epoxides were used
no loss of chirality was observed during the sequence (Scheme 4).8
In order to study the generality of this method, 2-chlorosulfonyl
chlorides were also utilized as starting materials. A wide range of
substituted 2-chlorosulfonyl chlorides are readily available. When
2-chlorosulfonyl chloride 5b (R = H) was subjected to the general
cyclization conditions with benzylamine and epoxide 9b, only
the corresponding alcohol was isolated from the reaction mixture.
The solvent was changed to N,N-dimethylformamide and the tem-
perature of the reaction mixture was increased to 120 °C, and using
these new conditions, a modest 25% yield of the benzoxathiazepine
1g was obtained (Scheme 5, Table 2, entry 1).
Once again the introduction of electron-withdrawing groups to
the aromatic ring resulted in increased reaction yields, as had
already been observed for the corresponding fluorides. With these
more active substrates the MCR could be performed at lower
temperatures with 1,4-dioxane as the preferred solvent. When
2-chloro-6-methylsulfonyl chloride was used as the starting
material, the intermediate alcohol 7b was the only product isolated
(Table 2, entry 4). This result could be explained by the steric hin-
drance imposed by the methyl group that forces the sulfonyl group
to twist and prevents the interaction between the alkoxide moiety
and the leaving group.
O O
S
R
N
N
R
O
N
HO
H
R'
R'
2
3
4
R'
Scheme 1. Synthesis of N-alkylated-4-substituted isothiazolidine-1,1-dioxides 4.
Reagents and conditions: (a) 1,4-dioxane, K2CO3, n-Bu4NCl, 100 °C; (b) PhSO2Cl
(1.2 equiv), pyridine, DMAP (cat.), 50 °C; (c) n-BuLi (2.2 equiv), THF, À78 °C ? rt,
2 h.
The synthesis depicted in Scheme 2 was initially attempted in a
stepwise fashion. Treatment of 2-fluorobenzenesulfonyl chloride
5a with benzylamine and potassium carbonate in 1,4-dioxane at
room temperature yielded the corresponding sulfonamide 6a in a
92% yield. A solution of 6a in 1,4-dioxane was then heated at
100 °C in the presence of tetra-n-butylammonium bromide
(n-Bu4NBr), K2CO3 and O-benzyl glycidyl ether 9a.6 We were
pleased to observe that, after 16 h, not only had the ring-opening
of the epoxide occurred to give 38% of alcohol 7a, but the SNAr
reaction had also occurred to some extent, generating the ben-
zoxathiazepine 1a in a 34% yield (Scheme 3).
This observation encouraged us to attempt the one-pot MCR
using the same experimental conditions. Thus, 2-fluorosulfonyl
chloride 5a (R = H) was treated with benzylamine and the com-
mercially available epoxides, benzyl glycidyl ether 9a or (R)-1,2-
epoxy-hex-5-ene 9b, in refluxing 1,4-dioxane, using K2CO3 as the
base (Scheme 4, Table 1, entries 1 and 7). After 72 h, the desired
benzoxathiazepines 1a and 1g were isolated in 76% and 58% yield,
which equate to a step average of 91% and 83%, respectively.7
The same experimental protocol was then successfully applied
to a variety of starting 2-fluorosulfonyl chlorides substituted with
additional halogen atoms on the benzene ring. The yields of iso-
lated products were similar or higher to those observed for the
unsubstituted systems (Table 1), which is the expected trend when
electron-withdrawing substituents are present on the aromatic
ring during SNAr reactions. The relative position of the electron-
withdrawing group with respect to the leaving group also played
a role in the observed yields. Thus, a chlorine atom causes a larger
increase in the yield when it is at the 3-position in the starting
material, than when it is at the 4-position (Table 1, entries 3, 9
The halogen substituents present on the aromatic ring of the
benzoxathiazepines 1 confer a drug-like profile and also provide
a synthetic handle for the introduction of further functionalization
by means of a palladium-catalyzed cross-coupling reaction. An
example of this type of modification is highlighted below
(Scheme 6).
A selection of the prepared benzoxathiazepines, 1 were sub-
jected to palladium-catalyzed Suzuki cross-couplings (Table 3),
using a variety of boronic acids 10.9 In general, the yields of these
cross-coupling products 11a–c were good, however, a lower yield
O O
S
O
S
O
O O
O O
R'
R'
amine
epoxide
R'
N
S
S
SNAr
N
Cl
addition
N
opening
H
R''
OH
X
O
X
X
R
R
R
R
R''
5
6
7
1
Scheme 2. Synthetic strategy for the MCR synthesis of benzoxathiazepines 1.
O O
S
N
Ph
OBn
F
O O
S
7a
1a
OH
SO2Cl
F
a
b
N
Ph
+
H
F
O
O
Ph
S
N
5a
6a
O
OBn
Scheme 3. Stepwise benzoxathiazepine 1 formation. Reagents and conditions: (a) benzylamine, K2CO3, 1,4-dioxane, 92%. (b) O-benzyl glycidyl ether 9a, K2CO3, n-Bu4NBr,
1,4-dioxane, 100 °C, 16 h, 7a (38%) and 1a (34%).

E. Cleator et al. / Tetrahedron Letters 51 (2010) 1079–1082
1081
O
S
O
S
O O
S
O
O
Ph
Ph
N
N
H
a
b
Cl
F
O
O
O
BnO
R
R
R
OBn
O
1a-f
5a
1g-j
9a
9b
Scheme 4. One-pot MCR of benzoxathiazepines 1. Reagents and conditions: (a) benzylamine 8, K2CO3, n-Bu4NBr, 1,4-dioxane, 100 °C, 72 h, 76% (when R = H); (b)
benzylamine 8, K2CO3, n-Bu4NBr, 1,4-dioxane, 100 °C, 72 h, 58% (when R = H).
Table 1
O
S
O O
S
O
Ph
Synthesis of benzoxathiazepines 1 from 2-fluorosulfonyl chlorides 5a
N
a or b
Cl
R
R
Entry
Product
Time (h)
Yield (%)
76
O
Cl
O
R
O
S
R
O
Ph
9
N
5b
1g,k,l
1
1a
1b
1c
1d
72
O
Scheme 5. Synthesis of benzoxathiazepines 1 from 2-chlorosulfonyl chlorides.
Reagents and conditions: (a) benzylamine 8, 9a or 9b, K2CO3, n-Bu4NBr, DMF,
120 °C, 72 h; (b) benzylamine 8, K2CO3, 1,4-dioxane, 100 °C, 72 h.
OBn
O
O
O
O
Ph
S
F
N
2
3
4
72
72
72
79
91
86
Br
O
OBn
Ph
Table 2
O
Synthesis of benzoxathiazepines 1 from 2-chlorosulfonyl chlorides 5b
S
Cl
N
N
Entry
Product
Temp (°C)
Yield (%)
25
O
S
O
O
Ph
OBn
Ph
N
O
1
1g
120
S
O
H
Cl
O
O
OBn
O
O
Ph
S
N
O
O
Ph
2
3
4
1k
1l
100
100
100
48
67
59
S
N
Br
Br
O
5
6
7
1e
1f
72
168
72
77
72
58
H
N
O
OBn
O
S
Cl
Ph
O
S
O
Ph
F
N
O
OBn
O
N
O O
S
OBn
O
S
N
Ph
O
Ph
7b
Cl
1g
OH
R''
O
H
O
O
O
O
Ph
Ph
Ph
S
F
N
OH
8
9
1h
1i
72
72
72
76
84
76
B
O
S
O
S
N
Br
Cl
O
OH
O
O
R
R
H
N
N
10
(R'' = Ar or Het)
Pd catalysis
O
S
Hal
R''
O
O
R'
R'
O
Scheme 6. Palladium-catalyzed diversification of halogenated benzoxathiazepines 1.
H
N
O
S
In summary, we have identified a cheap, general, high yielding
and rapid route to benzoxathiazepines, which can be completed in
one pot from the commercially available starting materials. Suit-
ably functionalized products have been shown to be useful sub-
strates for further elaboration via palladium-catalyzed cross-
couplings. This transformation broadens the applicability of this
methodology, and allows the rapid synthesis of complex novel
molecules of potential pharmacological interest. Further studies
directed towards the chemical elaboration of benzoxathiazepines
10
1j
Cl
O
H
and incomplete conversion was afforded when the substrate 1h
containing an alkene functionality was employed, affording 11d
in a 36% isolated yield (Table 3, entry 4).10,11

1082
E. Cleator et al. / Tetrahedron Letters 51 (2010) 1079–1082
K. G.; Waring, M. J. PCT Int. Appl., WO2006125972, 2006; Chem. Abstr. 2006,
Table 3
146, 27865.
Palladium-catalyzed cross-couplings
4. As an example, in Ref. 3a, the ring closure of an alcohol onto an aryl fluoride
using NaH in DMF has been used. See also: (a) White, E. H.; Lim, H. M. J. Org.
Chem. 1987, 52, 2162–2166; (b) Bliss, A. D.; Cline, W. K.; Hamilton, C. E.;
Sweeting, O. J. Org. Chem. 1963, 28, 3537–3541; (c) Penso, M.; Albanese, D.;
Landini, D.; Lupi, V.; Tagliabue, A. J. Org. Chem. 2008, 73, 6686–6690.
5. Cleator, E.; Sheen, F. J.; Bio, M. M.; Brands, K. M. J.; Davies, A. J.; Dolling, U.-H.
Tetrahedron Lett. 2006, 47, 4245–4248.
Entry
Product
Starting
material
Yield (%)
O
O
Ph
S
F
N
6. The ring-opening of epoxides with sulfonamides has been outlined in the
literature, see: Baker, B. R.; Kadish, A. F.; Querry, M. V. J. Org. Chem. 1950, 2,
400–401; The conditions outlined here were discovered in our previous work,
see Ref. 5, which were adapted from Albanese et al., see: Albanese, D.; Landini,
D.; Penso, M.; Petricci, S. Tetrahedron 1999, 20, 6387–6394.
1
11a
11b
1b
1b
88
O
OBn
N
N
O
S
7. General procedure for the multicomponent cyclization reaction: sulfonyl chloride
O
Ph
(1.0 mmol) was added to
a solution of K2CO3 (2.5 mmol) and tetra-n-
F
N
butylammonium bromide (0.1 mmol) in 1,4-dioxane (10 mL/g). Benzylamine
(1.0 mmol) and epoxide (1.0 mmol) were added and the reaction mixture was
heated to reflux and stirred for 72 h. The reaction mixture was allowed to cool
to room temperature and was diluted with water (10 mL/g). The mixture was
extracted with EtOAc (3 Â 15 mL/g) and the combined organic extracts were
dried over MgSO4, filtered and concentrated in vacuo. The crude products were
purified by silica gel flash chromatography. Characterization of
benzoxathiazepine 1d. Colourless oil. 1H NMR (400 MHz, CDCl3) dH 7.75 (1H,
d, J = 8.4, Ar–H), 7.30–7.13 (12H, m, Ar–H), 4.49 (2H, s, OCH2Ph), 4.45 (1H, d,
J = 14.4, NCHACHBPh), 4.26–4.20 (1H, m, OCH), 3.84 (1H, d, J = 14.4,
NCHACHBPh), 3.76 (1H, dd, J = 15.2 and 10.8, CHACHBOBn), 3.66–3.62 (1H, m,
NCHACHBCH), 3.48–3.44 (1H, m, NCHACHBCH), 3.20 (1H, dd, J = 15.2 and 1.6,
CHACHBOBn); 13C NMR (100 MHz, CDCl3) dC 155.8, 139.8, 137.6, 135.2, 133.0,
130.1, 129.0, 128.7, 128.4, 128.1, 127.4, 124.9, 124.0, 77.8, 73.8, 70.1, 51.9,
49.5; HRMS (ESI+) calcd for C23H22ClNNaO4S 466.0856 [M+Na]+, found
466.0859 [M+Na]+.
2
71
O
OBn
F3C
O
S
O
Ph
N
3
4
11c
11d
1d
1h
72
36
O
OBn
Ph
N
N
O
S
O
F
N
8. Determined by chiral HPLC. Conditions, Chiral AD; eluent 20% iso-propyl
alcohol in hexanes containing 0.1% TFA.
O
9. Billingsley, K.; Buchwald, S. L. J. Am. Chem. Soc. 2007, 129, 3358–3366.
10. General procedure 1 for the Suzuki coupling: the aryl halide (1.0 equiv), boronic
acid (1.05 equiv), PdCl2dppf–CH2Cl2 adduct (10 mol %) and potassium
phosphate, dibasic (2.0 equiv) were added to a sealable tube containing a
magnetic stir bar. Dioxane (10 mL/g) was added and the mixture was degassed
by evacuating the system and back-filling with nitrogen for three times. The
tube was sealed, heated to 100 °C and stirred for 19 h. The reaction was
quenched by the addition of water (10 mL/g), the product was extracted with
EtOAc (3 Â 10 mL/g) and the combined organics were concentrated in vacuo to
yield the crude product. The product was purified using silica gel flash
chromatography.
1 and the extension of this methodology towards the synthesis of
benzothiadiazepines are currently underway.
Acknowledgements
11. General procedure 2 for the Suzuki coupling: the benzoxathiazepine (1 equiv)
was dissolved in dioxane (10 mL/g) and the boronic acid (1.05 equiv) and
palladium (diphenylphosphino)dichloride were added (3 mol %). Sodium
carbonate (2 M, 1.2 equiv) was then added and the reaction mixture was
heated to reflux for 90 min. The reaction was quenched by the addition of
water (10 mL/g) and then extracted with EtOAc (3 Â 10 mL/g). The combined
organic layers were dried (MgSO4), filtered and then concentrated in vacuo
We thank Simon Hamilton and Sophie Strickfuss from the MSD
analytical research group at Hoddesdon for mass spectrometry
work and chiral HPLC analysis.
References and notes
to give
a yellow oil which was purified by column chromatography.
Characterization of 11a. Clear oil. 1H NMR (400 MHz, CDCl3) dH 7.84 (2H, m)
7.70–7.62 (4H, m) 7.58–7.46 (4H, m) 7.42–7.23 (4H, m), 4.63 (3H, m) 4.27–4.25
(1H, m) 4.0 (3H, s), 3.92–3.87 (2H, m), 3.78–3.74 (1H, m), 3.58–3.55 (1H, m)
3.28 (1H, d, J = 14 Hz); 13C NMR (60 MHz) dC 138.0, 137.6, 135.1, 132.2, 132.1,
131.9, 130.1, 130.0, 128.8, 128.5 (m), 128.4, 128.2, 127.9, 127.6, 121.4 (d),
116.6, 116.3, 77.2, 73.6, 70.1, 51.6, 49.5, 39.3; HRMS (ESI+) calcd for
C27H27FN3O4S 508.1706 [M+H]+, found: 508.1726 [M+H]+.
1. (a) Eckert, H. Heterocycles 2007, 73, 149–158; (b) Hulme, C.; Gore, V. Curr. Med.
Chem. 2003, 10, 51–80; (c) Weber, L. Curr. Med. Chem. 2002, 9, 2085–2093; (d)
Weber, L.; Illgen, K.; Almstetter, M. Synlett 1999, 366–374.
2. Santora, V. J.; Covel, J. A.; Ibarra, J. B.; Semple, G.; Smith, B.; Smith, J.;
Weinhouse, M. I.; Schultz, J. A. PCT Int. Appl., WO 2008005338, 2008; Chem.
Abstr. 2008, 148, 144502.
3. (a) Campbell, L.; Pike, K. G.; Suleman, A.; Waring, M. J. PCT Int. Appl., WO
2008050101, 2008; Chem. Abstr. 2008, 148, 495984.; (b) McKerrecher, D.; Pike,