The synthesis of S 18986, a chiral AMPA receptor modulator, via catalytic asymmetric hydrogenation

Article abstract of DOI:10.1016/S0957-4166(03)00547-0

The asymmetric hydrogenation of a dihydropyrrolo-benzothiadiazine dioxide 1 with a diphosphine ruthenium diamine catalyst has been used to synthesize the AMPA receptor positive modulator tetrahydropyrrolo-benzothiadiazine dioxide 2 (S 18986) in high enantiomeric excess. The use of factorial experimental design led to significant improvements in the process parameters and improved enantioselectivity.

Full text of DOI:10.1016/S0957-4166(03)00547-0

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 14 (2003) 3431–3433
The synthesis of S 18986, a chiral AMPA receptor modulator,
via catalytic asymmetric hydrogenation
a
b
b
a,
Christopher J. Cobley, Elsa Foucher, Jean-Pierre Lecouve, Ian C. Lennon, *
a,
b
James A. Ramsden * and Gilles Thominot
a
Dowpharma, Chirotech Technology Ltd., A Subsidiary of The Dow Chemical Company, Unit 321 Cambridge Science Park,
Milton Road, Cambridge CB4 0WG, UK
b
Oril Industrie, 13 rue Auguste Desgen e´ tais, BP17-76210, Bolbec, France
Received 7 May 2003; accepted 23 May 2003
Abstract—The asymmetric hydrogenation of a dihydropyrrolo-benzothiadiazine dioxide 1 with a diphosphine ruthenium diamine
catalyst has been used to synthesize the AMPA receptor positive modulator tetrahydropyrrolo-benzothiadiazine dioxide 2 (S
1
8986) in high enantiomeric excess. The use of factorial experimental design led to significant improvements in the process
parameters and improved enantioselectivity.
2003 Elsevier Ltd. All rights reserved.
©
Various benzothiadiazines have been shown to be
potential drugs for memory and learning disorders,
CNS trauma and neurodegenerative disease. In particu-
achieved in this reaction was reasonably high (576%
ee), however this route presented significant disadvan-
tages. The method required a stoichiometric amount of
the chiral modifier and the reaction was incomplete,
with 1–2% of the starting material present. The unre-
acted dihydropyrrolo-benzothiadiazine dioxide 1 had to
be removed from the product before the enantiomeric
purity of the enriched product could be increased via
lar,
racemic
2,3,3a,4-tetrahydro-1H-pyrrolo[2,1-
c][1,2,4]benzothiadiazine-5,5-dioxide 2 (Fig. 1) was
shown to be a selective AMPA receptor positive
1
modulator.
1
two crystallisations from acetonitrile. In order to fur-
nish a more efficient route to (S)-tetrahydropyrrolo-
benzothiadiazine dioxide 2 (S 18986) the use of
asymmetric hydrogenation of dihydropyrrolo-benzoth-
2
iadiazine dioxide 1 was examined.
In general, imine hydrogenation has proved to be a
challenging area for metal-catalyzed asymmetric hydro-
genation with only a few examples being reported
where moderately high selectivity has been demon-
Figure 1. Hydrogenation substrate and racemic product.
The enantiomers of racemic tetrahydropyrrolo-benzoth-
iadiazine dioxide 2 were separated by preparative chiral
HPLC and it was found that, in common with similar
compounds within this therapeutic class, all activity
resided in a single stereoisomer. X-Ray crystallography
3
strated. The structurally related benzisothiazole diox-
ides have been successfully converted to their
corresponding benzoisothiazoline dioxides via either
4
catalytic asymmetric transfer hydrogenation or hydro-
1
5
identified the (S)-enantiomer as the active component.
genation. Preliminary screening experiments demon-
In earlier studies the enantioselective synthesis of 2
relied on the use of lithium aluminium hydride in the
presence of a chiral modifier. The enantiomeric excess
strated that none of these approaches were suitable for
the asymmetric reduction of 1. A wide range of
rhodium, ruthenium and iridium catalysts were
screened for both pressure and transfer hydrogenation
of 1, typically little or no reactivity was observed (<10%
conversion) and the products that were obtained had
very low enantiomeric excesses.
*
Corresponding authors. Tel.: +44-(0)1223-728037; fax: +44-(0)1223-
06701; e-mail: ilennon@dow.com
5
0
957-4166/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0957-4166(03)00547-0

3
432
C. J. Cobley et al. / Tetrahedron: Asymmetry 14 (2003) 3431–3433
Ruthenium precatalysts of the type [(diphos-
liminary experiments, although this also lead to lower
solubility of the substrate and a slight reduction in
conversion. Lower pressure led to slightly improved
enantioselectivity as did lower temperature. Further
analysis of this data identified conditions whereby com-
plete conversion could be achieved without significant
impact on the selectivity of the process. Suitable process
phine)RuCl (1,2-diamine)] have been shown to be
2
extremely effective in the asymmetric hydrogenation of
6
7
ketones and more recently imines. In excess of 40
precatalysts were screened, containing a variety of
diphosphine/1,2-diamine combinations, for the asym-
metric hydrogenation of 1. In the initial screen a multi-
well vessel was used, with 220 mg of substrate in 5 ml
conditions identified were 40°C, 60 psi H and a 75/25
2
8
mixture of toluene/isopropanol.
of solvent. Only biaryl ligands such as BINAP, Tol-
8
9
BINAP and Cl-MeO-BIPHEP (Fig. 2) proved to be
useful for the asymmetric hydrogenation of 1. In addi-
tion, only two 1,2-diamines gave suitable results,
diphenylethylenediamine (DPEN) and diaminocyclo-
Most studies up to this point had been carried out at a
molar substrate to catalyst ratio (S/C) of 500. A much
lower catalyst loading could be achieved (S/C=2,500),
but at the expense of reaction time. However, it had
already been demonstrated that a higher temperature
significantly increased the overall rate, while having
only a marginally deleterious effect on selectivity. There-
fore the reaction was re-examined at a higher tempera-
hexane (DACH). Unexpectedly, the non-C symmetric
2
diamine DIAPEN in combination with BINAP gave a
completely unreactive catalyst. For ketone reduction
DIAPEN often provides the most selective and active
catalyst in combination with diphosphines based on
1
2
1
0
ture and a lower catalyst loading (higher S/C).
BINAP.
Performing the reaction at 60°C and 60 psi H , the
2
product was isolated in 97% yield and 87% ee (Scheme
1
). A further recrystallisation from acetonitrile provided
1
enantiomerically pure product 2 in >99% ee.
Figure 2. Selection of ligands used during initial screening.
Scheme 1. Asymmetric hydrogenation of dihydropyrrolo-ben-
zothiadiazine 1.
In the initial screen high conversions (93–99%) and good
enantiomeric excesses (67–75%) were obtained using
catalysts based on the ligands in Figure 2. The best
catalysts were examined in greater detail using single
well vessels with improved stirring capability. Based on
these results and the ready availability of the BINAP
ligand, we chose to characterize the reaction using the
We have demonstrated that the precatalyst [(R)-BINAP
RuCl (R,R)-DPEN] 3 can be used for the asymmetric
2
hydrogenation of a novel N-sulfonylimine substrate, the
dihydropyrolo-benzothiadiazine dioxide 1. This could
provide an efficient process for the manufacture of the
chiral AMPA receptor positive modulator 2 (S 18986).
[
(R)-BINAP RuCl (R,R)-DPEN] 3 precatalyst.
2
Further applications of [(diphosphine) RuCl (1,2-
2
It was established that the amount of base used in the
reaction was important to attaining high conversion.
When utilized in the hydrogenation of ketones, precata-
lysts of type 3 require activation with a sub-stoichiomet-
ric amount of base (with respect to ketone). In the
hydrogenation of dihydropyrrolo-benzothiadiazine
dioxide 1 between 0.8 and 1.5 equiv. of base (with
respect to 1) was required to effect full conversion.
diamine)] precatalysts are being investigated and will be
reported in due course.
Acknowledgements
We are grateful to Catherine Ripp e´ and David Baldwin
of Chirotech and Eric Bourdais of Oril Industrie for the
development of analytical methods.
A more thorough examination of the reaction parame-
ters were undertaken via a series of factorially designed
1
1
experiments. The process parameters examined were
temperature, pressure and solvent composition. It was
of no surprise that higher temperatures led to higher
conversion, but it was interesting that hydrogen pres-
sure had a negligible effect on conversion. The use of a
mixture of toluene and isopropanol had already been
identified as leading to higher enantioselectivity in pre-
References
1. Desos, P.; Serkiz, B.; Morain, P.; Lepagnol, J.; Cordi, A.
Bioorg. Med. Chem. Lett. 1996, 6, 3003–3008.
2. Cordi, A.; Spedding, M.; Serkiz, B.; Lepagnol, J.; Desos,
P.; Morain, P. US Patent 5,536,719, 1996; Chem. Abstr.
1996, 124, 261085.

C. J. Cobley et al. / Tetrahedron: Asymmetry 14 (2003) 3431–3433
3433
3
. (a) Kobayashi, S.; Ishitani, H. Chem. Rev. 1999, 99,
069–1094; (b) Blaser, H. U.; Buser, H. P.; H a¨ usel, R.;
Jalett, H. P.; Spindler, F. J. Organomet. Chem. 2001, 621,
4–38; (c) Sablong, R.; Osborn, J. A. Tetrahedron: Asym-
metry 1996, 7, 3059–3062; (d) Willoughby, C. A.; Buch-
wald, S. L. J. Am. Chem. Soc. 1994, 116, 8952–8965; (e)
Morimoto, T.; Nakajima, N.; Achiwa, K. Synlett 1995,
10. Doucet, H.; Ohkuma, T.; Murata, K.; Yokozawa, T.;
Kozawa, M.; Katayama, E.; England, A. F.; Ikariya, T.;
Noyori, R. Angew. Chem., Int. Ed. 1998, 37, 1703–1707.
11. MODDE 5.0, Umetrics AB (www.umetrics.com).
12. A pressure vessel was charged with 2,3-dihydro-1H-
pyrrolo[2,1-c][1,2,4]benzothiadiazine-5,5-dioxide 1 (100 g,
450 mmol) and toluene (750 ml). A nitrogen atmosphere
was established. A solution of potassium butoxide (56.3
g, 502 mmol) in propan-2-ol (250 ml) was added followed
1
3
7
48–750; (f) Bianchini, C.; Barbaro, P.; Scapacci, G.;
Farnetti, E.; Graziani, M. Organometallics 1998, 17,
308–3310.
3
by [(R)-BINAP RuCl (R,R)-DPEN] (186 mg, 0.185
2
4
5
6
7
. Ahn, K. H.; Ham, C.; Kim, S. K.; Cho, C. W. J. Org.
Chem. 1997, 62, 7047–7048.
. Oppolzer, W.; Wills, M.; Starkemann, C.; Bernardinelli,
G. Tetrahedron Lett. 1990, 31, 4117–4120.
. Noyori, R.; Ohkuma, T. Angew. Chem., Int. Ed. 2001, 40,
mmol). The mixture was heated to 60°C and the vessel
was charged with 60 psi hydrogen. The reaction mixture
was stirred for 6.5 h, maintaining the hydrogen pressure
between 50–65 psi. Once cooled and acidified, the
product was collected by filtration and dried to give 98 g
(97% yield) of (S)-2,3,3a,4-tetrahydro-1H-pyrrolo[2,1-c]-
[1,2,4]benzothiadiazine-5,5-dioxide 2. HPLC analysis
showed 99.4% chemical purity 87% ee. Chiral HPLC:
Daicel Chiralpak AS column, UV 212 nm, 1.0 ml/min,
30% heptane, 70% ethanol for 10 min, ramp to 100%
ethanol in 1 min, hold for 12 min. Retention time (R)=
5.0 min, (S)=10.6 min. Two recrystallisations from
4
0–73.
. (a) Cobley, C. J.; Henschke, J. P. Adv. Synth. Cat. 2003,
45, 195–201; (b) Abdur-Rashid, K.; Lough, A. J.; Mor-
3
ris, R. H. Organometallics 2001, 20, 1047–1049.
. Kumobayashi, H.; Miura, T.; Sayo, N.; Saito, T.; Zhang,
X. Synlett 2001, 1055–1064.
8
9
. Laue, C.; Schr o¨ der, G.; Arlt, D. US Patent 5,710,339,
1
1
998; Chem. Abstr. 1997, 126, 104246.
acetonitrile afforded 2 in >99% ee.