Solid-phase Friedel-Crafts acylation on polystyrene resins-synthesis of antiepiletic 1-aryl-3,5-dihydro-4H-2,3-benzodiazepin-4-ones

Article abstract of DOI:10.1016/S0040-4039(01)01602-1

Friedel-Crafts acylation with various acyl chlorides of resin-bound 3,4-dimethoxyphenylacetate afforded resin bound ketones which, following treatment with hydrazine, were converted into the corresponding 2,3-benzodiazepines in good yields and purities.

Full text of DOI:10.1016/S0040-4039(01)01602-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 7683–7685
Solid-phase Friedel–Crafts acylation on polystyrene
resins-synthesis of antiepiletic
1-aryl-3,5-dihydro-4H-2,3-benzodiazepin-4-ones
Francesca Bevacqua,a Andrea Basso,b Rosaria Gitto,a Mark Bradleyb,* and Alba Chimirria,*
aDipartimento Farmaco-Chimico, Universita` di Messina, viale Annunziata 96168, Messina, Italy
bDepartment of Chemistry, University of Southampton, Highfield SO17 1BJ, Southampton, UK
Received 11 July 2001; accepted 24 August 2001
Abstract—Friedel–Crafts acylation with various acyl chlorides of resin-bound 3,4-dimethoxyphenylacetate afforded resin bound
ketones which, following treatment with hydrazine, were converted into the corresponding 2,3-benzodiazepines in good yields and
purities. © 2001 Published by Elsevier Science Ltd.
The preparation and screening of combinatorial
libraries has in recent years become an attractive
method for the discovery of pharmaceutical lead com-
pounds, with much of the work concentrated on explor-
ing and exploiting the synthesis of heterocycles.1 We
report here the solid-phase synthesis of 1-aryl-3,5-dihy-
dro-4H-2,3-benzodiazepin-4-ones, which are potentially
useful for the treatment of epilepsy. The key synthetic
step was a Friedel–Crafts acylation reaction on a
polystyrene supported substrate, an unusual solid-phase
reaction due to the nature of the solid support.
logical studies have shown that a series of 2,3-benzo-
diazepine derivatives acting as non-competitive AMPA
receptor antagonists are potentially useful for the treat-
ment of epilepsy.4 Some of these compounds are more
potent, less toxic and longer lasting anticonvulsant
agents than their analogue GYKI 52466.5
In this paper we report the first results of our study
towards the solid-phase synthesis of 1-aryl-3,5-dihydro-
4H-2,3-benzodiazepin-4-ones for the potential treat-
ment of epilepsy (Fig. 1).
Epilepsy is a pathological state of the central nervous
system (CNS). It affects 1% of the world’s population
and at the present time clinically available drugs pro-
duce satisfactory seizure control in only 60% of
O
O
O
O
O
N
N
N
R1
N
patients. L-Glutamate, the major excitatory neurotrans-
mitter in the CNS, plays an important role in the
convulsant phenomena and Glutamate receptors
(GluRs) are involved in fundamental processes such as
neuronal development, learning and memory. However,
excessive stimulation of GluRs can cause damage of
neuronal tissue during ischemic episodes, epileptic
seizures and other neurological diseases. Antagonists of
one subset of GluRs, (AMPA [2-amino-3-(-hydroxy-5-
methylisoxazol-4-yl)propionic acid] type receptors2)
have shown promise in the prevention and treatment of
a broad range of neurological diseases3 and pharmaco-
R2
H2N
GYKI 52466
2,3-Benzodiazepines
Figure 1. AMPA receptor antagonists.
The synthetic pattern followed is outlined in Scheme 1.
The key steps were a Friedel–Crafts acylation of a
resin-bound 3,4-dimethoxyphenylacetate and a con-
comitant hydrazine-mediated cleavage and ring closure.
Given the reactivity of the polystyrene backbone, the
Friedel–Crafts acylation reaction is not a routine solid-
phase transformation and only few examples have been
reported in the literature, most of which have been
directed at modifications of the polystyrene backbone
Keywords: solid-phase synthesis; Friedel–Crafts acylation; 2,3-benzo-
diazepines.
* Corresponding authors. Fax +39 090 355613; e-mail: chimirri@
pharma.unime.it
0040-4039/01/$ - see front matter © 2001 Published by Elsevier Science Ltd.
PII: S0040-4039(01)01602-1

7684
F. Be6acqua et al. / Tetrahedron Letters 42 (2001) 7683–7685
Scheme 1. (i) 4-(Hydroxymethyl)benzoic acid (2 equiv.), DIC (2 equiv.), HOBt (2 equiv.), DCM, rt, 2 h ; then 1 M NaOH
(aq)/dioxane 1:1, rt, 4 h; (ii) 3,4-dimethoxyphenylacetic acid (2 equiv.), DMAP (0.5 equiv.), DIC (2 equiv.), DCM, rt, 3 h; (iii)
acyl chloride (1 equiv.), SnCl4 or FeCl3 (5 equiv.), DCM, rt, 6–12 h; (iv) 1 M NaOH (aq)/dioxane 1:1, rt, 4 h; (v) hydrazine (10
equiv.), EtOH, reflux, 24 h.
and not at the attached compound. For example
Ajayaghosh and Pillai6 reacted 2-nitrobenzoylchloride
with polystyrene in the presence of AlCl3 in
dichloromethane (DCM) for the preparation of the
photosensitive o-nitrobenzhydrylamino (NBHA) linker,
Ferderigos7 published an alternative method for the
synthesis of high capacity aminomethyl polystyrene
Lewis acids FeCl3 or SnCl4. No difference in reactivity,
selectivity or purity of the resulting compounds 4 was
observed upon varying the Lewis acid. Cleavage of
compounds 3a–e from the resin with hydrazine
afforded the final benzodiazepines 5a–e in good purity
and isolated yields (Table 1), which were fully charac-
terized.9 In attempt to obtain the analogous alkyl
derivatives, propionyl chloride was coupled onto resin
3, but the resulting ketone 4 (obtained with 88% HPLC
purity) did not give the desired benzodiazepine upon
treatment with hydrazine. Fig. 2 shows the synthetic
sequence in the synthesis of 5a, and the high purity of
the crude material.
resin involving
a
Friedel–Crafts alkylation of
polystyrene with N-(chloromethyl)phthalimide in the
presence of FeCl3, while in 1997 Park and Lee8 tested a
wide variety of Lewis acids for the alkylation of
polystyrene with propylene oxide with SnCl4 being the
most effective.
In conclusion we have demonstrated how 2,3-benzo-
diazepines 5 can be straightforwardly synthesized on
the solid-phase, using the Friedel–Crafts acylation on
polystyrene supports. The cleavage/cyclization step
using substituted hydrazines to generate a second
source of diversity is now under investigation.
Initially, hydroxymethyl polystyrene was used to
directly attach 3,4-dimethoxyphenylacetic acid; how-
ever, difficulties were encountered, due to proximity of
the resin backbone. Thus, as shown in Scheme 1,
commercially available aminomethyl polystyrene resin
(1% DVB, 1.2 mmol/g) was coupled with 4-(hydroxy-
methyl)benzoic acid using standard peptide coupling
conditions, DIC/HOBt in DCM. This was then treated
with 1 M NaOH (aq)/dioxane 1:1 to remove any linker
oligomers formed via ester bond formation between the
hydroxymethyl group and the benzoic acid, to give 1.
Table 1.
4a
5a,b
HPLC purity (%)
HPLC purity (%)
Yieldc (%)
a
b
c
d
e
91
91
77
86
40
81
67
27
57
41
62
36
32
29
27
3,4-Dimethoxyphenylacetic acid was esterified to the
linker via the resin-bound hydroxymethyl group giving
compound 2. This was then quantitatively and selec-
tively acylated with five different commercially avail-
able acyl chlorides: benzoyl chloride, 4-chlorobenzoyl
chloride, 3-chlorobenzoyl chloride, 4-fluorobenzoyl
chloride and 3-fluorobenzoylchloride. Acylation was
performed in DCM at room temperature using the
a HPLC purity at 220 nm.
b Products were identified by 1H NMR and MS spectroscopy.
c After semi-preparative HPLC purification.

F. Be6acqua et al. / Tetrahedron Letters 42 (2001) 7683–7685
7685
Figure 2. Crude HPLC traces of intermediates (3a, 4a) and benzodiazepine 5a (u=220 nm).
Acknowledgements
5. Chimirri, A.; De Sarro, G.; De Sarro, A.; Gitto, R.;
Quartarone, S.; Zappala`, M.; Costanti, A.; Libri, V. J.
Med. Chem. 1998, 41, 3409–3416.
Financial support for this research by Fondi di Ateneo
di Ricerca (1998, Messina, Italy) is gratefully acknow-
ledged. We also thank Sunil Rana for valuable proof
reading.
6. Ajayaghosh, A.; Pillai, V. N. R. Tetrahedron Lett. 1995,
36, 777–780.
7. Zikon, C. C.; Ferderigos, N. G. Tetrahedron Lett. 1995,
36, 3741–3744.
8. Park, B.-D.; Lee, H.-I.; Ryoo, S.-J.; Lee, Y.-L. Tetra-
hedron Lett. 1997, 38, 591–594.
1
References
9. Data for 5b: H NMR (400 MHz, CDCl3): l (ppm)=3.42
(s, 2H, CH2), 3.65 and 3.89 (s, 3H, OCH3-7 and OCH3-8),
6.57 (s, 1H, H-6), 6.78 (s, 1H, H-9), 7.27–7.60 (m, 4H, Ar),
8.32 (brs, 1H, NH); ES-MS (ES+): 331 (100%, [M+H]+).
1. (a) Fru¨chtel, J. S.; Jung, G. Angew. Chem., Int. Ed. Engl.
1996, 35, 17–42; (b) Booth, S.; Hermenks, P. H. H.;
Ottenheijm, H. C. J.; Rees, D. C. Tetrahedron 1998, 54,
15385–15443.
2. Dingledine, R.; Borges, K.; Bowie, D.; Traynelis, S. R.
Pharmacol. Rev. 1999, 51, 7–61.
1
Data for 5c: H NMR (400 MHz, CDCl3): l (ppm)=3.46
(s, 2H, CH2), 3.67 and 3.90 (s, 3H, OCH3-7 and OCH3-8),
6.59 (s, 1H, H-6), 6.78 (s, 1H, H-9), 7.29–7.51 (m, 4H, Ar),
8.36 (brs, 1H, NH); ES-MS (ES+): 315 (100%, [M+H]+).
1
3. Chimirri, A.; Gitto, R.; Zappala`, M. Exp. Opin. Ther.
Data for 5e: H NMR (400 MHz, CDCl3): l (ppm)=3.46
Patents. 1999, 9, 557–570.
(s, 2H, CH2), 3.68 and 3.92 (s, 3H, OCH3-7 and OCH3-8),
6.58 (s, 1H, H-6), 6.79 (s, 1H, H-9), 7.12–7.63 (m, 4H, Ar),
8.48 (brs, 1H, NH); ES-MS (ES+): 315 (100%, [M+H]+).
Data for 5a and 5d are consistent with the published
literature.4
4. Chimirri, A.; De Sarro, G.; De Sarro, A.; Gitto, R.;
Grasso, S.; Quartarone, S.; Zappala`, M.; Giusti, P.; Libri,
V.; Costanti, A.; Chapman, A. J. Med. Chem. 1997, 40,
1258–1269.