Synthesis and anticonvulsant activity of new N-1′,N-3′-disubstituted-2′H,3H,5′H-spiro-(2-benzofuran-1,4′-imidazolidine)-2′,3,5′-triones

Article abstract of DOI:10.1016/j.bmcl.2006.05.102

Thirteen new N-1′,N-3′-disubstituted-2′H,3H,5′H-spiro-(2-benzofuran-1,4′-imidazolidine)-2′,3,5′-triones were synthesized and their pharmacological activity determined with the objective to better understand their SAR for anticonvulsant activity. The anticonvulsant effects of these compounds were evaluated by standard pentylenetetrazol (scPTZ test) and maximum electroshock seizure (MES test) models in mice. Most of the compounds showed ability to protect against the pentylenetetrazol-induced convulsions. Compound 3o (the N-1′-p-nitrophenyl, N-3′-ethyl derivative) in the N-1′-aryl, N-3′-alkyl disubstituted series exhibited maximum activity with ED₅₀ of 41.8 mg/kg in scPTZ convulsion model.

Full text of DOI:10.1016/j.bmcl.2006.05.102

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4644–4647
Synthesis and anticonvulsant activity of new N-10,N-30-
disubstituted-20H,3H,50H-spiro-(2-benzofuran-1,40-
imidazolidine)-20,3,50-triones
Hardik J. Patel,a Joe Sarra,a Francesco Caruso,c Miriam Rossi,d
Utkarsh Doshia and Ralph A. Stephania,b,
*
aDepartment of Pharmaceutical Sciences, St. John’s University, 8000 Utopia Parkway, Jamaica, NY 11439, USA
bDepartment of Chemistry, St. John’s University, 8000 Utopia Parkway, Jamaica, NY 11439, USA
cIstituto di Chimica Biomolecolare, CNR, Rome, Italy
dVassar College Department of Chemistry, Poughkeepsie, NY 12604-0484, USA
Received 9 May 2006; revised 31 May 2006; accepted 31 May 2006
Available online 21 June 2006
Abstract—Thirteen new N-10,N-30-disubstituted-20H,3H,50H-spiro-(2-benzofuran-1,40-imidazolidine)-20,3,50-triones were synthe-
sized and their pharmacological activity determined with the objective to better understand their SAR for anticonvulsant activity.
The anticonvulsant effects of these compounds were evaluated by standard pentylenetetrazol (scPTZ test) and maximum electro-
shock seizure (MES test) models in mice. Most of the compounds showed ability to protect against the pentylenetetrazol-induced
convulsions. Compound 3o (the N-10-p-nitrophenyl, N-30-ethyl derivative) in the N-10-aryl, N-30-alkyl disubstituted series exhibited
maximum activity with ED50 of 41.8 mg/kg in scPTZ convulsion model.
Ó 2006 Elsevier Ltd. All rights reserved.
Epilepsy, characterized by the periodic and unpredict-
able occurrence of seizures, is the most prevalent neuro-
logical disorder, affecting approximately 50 million
people worldwide.1,2 Even though significant advances
have been made in epilepsy research, convulsions in
25% of epileptics are inadequately controlled by stan-
dard drug therapy.3 In recent times several new drugs,
for example, felbamate, lamotrigine, gabapentin, and
topiramate, have been approved to treat epilepsy.4
Although these drugs have been shown to be effective
in epileptic syndromes in a number of patients, their effi-
cacy does not appear to be superior to that of the estab-
lished antiepileptic drugs. Therefore, there is need for
new antiepileptic drugs with greater efficacy and fewer
side effects. Most of the compounds with anticonvulsant
activity contain a cyclic ureide ring system.5,6 Phenytoin,
a hydantoin, represents one of the important anticon-
vulsant drugs in the treatment of the generalized convul-
sion and psychomotor seizures since its discovery over
six decades ago.5,7 Since then many hydantoin analogs
have been synthesized and shown to have anticonvul-
sant activity. Spirohydantoins, 5,5-(cyclic hydrocarbon)
hydantoins were also synthesized and reported for their
antiepileptic activity,5 Figure 1. Compounds of the type
1 were found to be similar or more effective than diphe-
nylhydantoin and methylphenylhydantoin to a stimulus
test,8 while compound 2 has sedative, hypnotic, and
anticonvulsant properties.9
In a previous paper,10 from our laboratory it was shown
that oxidation of N-1,N-3-disubstituted urea–ninhydrin
cycloadducts with sodium periodate leads to products
R1
CH3
O
O
R2
O
HN NH
N
O
H
N
R3
O
Keywords: 20H,3H,50H-spiro-(2-benzofuran-1,40-imidazolidine)-20,3,50-
triones anticonvulsants; Spirohydantoins; Ninhydrin; N,N0-substituted
ureas; Sodium periodate.
1
2
R1, R2 = H, halogen
3 = H, alkyl
R
*
Corresponding author. Tel: +1 718 990 5215; fax: +1 718 990 1872;
e-mail: stephanr@stjohns.edu
Figure 1. Spirohydantoins with antiepileptic activity.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.102

H. J. Patel et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4644–4647
4645
which were assigned benzodiazocine structure. Howev-
er, crystallographic studies11 have revealed that they
are spirohydantoins of the type 3, Scheme 1. In the
homologous series of these compounds bearing dimethyl
(3a, R1, R2 = CH3), diethyl (3b, R1, R2 = C2H5), and
diallyl (3c, R1, R2 = CH2–CH@CH2), the diethyl (3b)
was the most active with ED50 of 190 mg/kg in scPTZ
test. In this research, a series of new N-10,N-30-disubsti-
tuted, 20H,3H,50H-spiro-(2-benzofuran-1,40-imidazoli-
dine)-20,3,50-triones (3d–p) Scheme 1, were synthesized
to better understand the SAR of these compounds. Sub-
stituents were selected in order to determine the opti-
mum chain length of the nitrogen substituents and to
increase the lipophilicity of the spirohydantoins to make
them more permeable to the blood–brain barrier (BBB).
The anticonvulsant effects of these compounds have
been evaluated through subcutaneous pentylenetetrazol
(scPTZ)-induced convulsions and maximal electroshock
(MES) tests. It emerges that the p-nitrophenyl deriva-
tive, compound 3o in the N-10-aryl, N-30-alkyl series, is
the most active in scPTZ animal model.
the corresponding disubstituted ureas. Commercially
unavailable ureas 6h–p were prepared by reaction of
the corresponding alkylamine with the appropriately
substituted phenylisocyanate or phenylisothiocyanate.
The indeno-[1,2-d]-imidazole intermediates (7d–p) were
readily prepared by refluxing the disubstituted ureas
(6d–p) with ninhydrin in benzene for 1 h. These were
then oxidized to spirohydantoins 3d–p in good yields
with sodium periodate by previously reported meth-
ods.10 The structures were assigned on the basis of infra-
red, NMR, mass spectra, and X-ray crystallography.12
The anticonvulsant activities of the synthesized com-
pounds were determined13 through the evaluation of
the ability of the compounds to protect mice against
convulsions induced by PTZ14,15 (scPTZ test) and elec-
troshock14,16 (MES test) as two routine models. Pheno-
barbital and phenytoin were taken as the reference
compounds in the scPTZ and MES models, respectively.
As shown in Table 1, compound 3e with a three-carbon
chain at N-30 position is approximately two and half
times more active than parent compound 3b (the diethyl
analog) in scPTZ model and also showed protection in
MES model. But, increasing the chain length to four
carbons (3f) resulted in decreased activity in both scPTZ
and MES tests. Also, decreasing the length of the alkyl
group by one-carbon at N-10 (3d) results in substantial
loss in activity in both scPTZ and MES models.
Replacement of the straight chain propyl group with
the branched isopropyl (3g) group or bulky cyclohexane
ring (3h) abolishes the activity in both the anticonvul-
sant models. The activity was regained by replacement
of the alkyl group at the N-10 position with a phenyl ring
(3i). These results suggest that the length of the alkyl
chain is critical at the N-30 position of the spirohydan-
toin and there is tolerance up to a three-carbon alkyl
group only. Bulk tolerance is not very critical at N-10
position of the ring as compound 3i which has a phenyl
ring, shows substantial activity in scPTZ test.
The designed compounds were synthesized according to
Scheme 2. All the compounds, 3d–p, were prepared from
O
O
O
N3'
1N'
R2
R1
X
3a-p
Compd
3a*
3b*
3c*
3d
R1
R2
X
CH3
C2H5
CH3
O
O
O
O
O
O
O
O
O
S
C2H5
allyl
allyl
CH3
C2H5
3e
n-C3H7
n-C4H9
i-C3H7
n-C3H7
n-C4H9
i-C3H7
cyclo-C6H11
CH3
3f
3g
3h
cyclo-C6H11
C6H5
3i
The oxygen of the C@O group at C-20 is thought to be
involved in H-bonding, since there is decrease in activity
when the carbonyl group in compound 3i is replaced by
a thiocarbonyl (3j), which is in agreement with the
reported SAR for hydantoins.17 Substitution of a small
lipophilic group like fluorine (3k) at the para position of
the N-10-phenyl ring of 3i resulted in increased activity
in both scPTZ and MES. However, substitution with
CF3 (3l) at the para position resulted in substantial de-
3j
C6H5
CH3
O
O
O
O
O
O
3k
p-F-C6H4
p-CF3-C6H4
C6H5
CH3
3l
CH3
3m
3n
C2H5
p-F-C6H4
p-NO2-C6H4
p-OCH3-C6H4
C2H5
3o
C2H5
3p
C2H5
Scheme 1. The structure of designed compounds. *Previously synthe-
sized and reported by Sarra et al.10
Scheme 2. The synthesis of 3d–p. Reagents and conditions: (i) dry THF, 0 °C; (ii) ninhydrin, C6H6, reflux, 1 h; (iii) NaIO4, EtOH, 12–16 h.

4646
H. J. Patel et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4644–4647
Table 1. Anticonvulsant test results (ED50) in mice of the synthesized compounds
O
O
O
N3'
1N'
R2
R1
X
3b and 3d-p
R1
R2
X
milogPf
ED50
a,b
Compound
scPTZ test
MES test
3b
3d
C2H5
CH3
C2H5
C2H5
O
O
O
O
O
O
O
S
1.457
1.081
2.462
3.581
2.051
4.516
2.403
2.943
2.566
3.298
2.779
2.942
2.738
2.835
190c
ꢀ
+
78.5 (63.7–96.7)
ꢀ
3e
n-C3H7
n-C4H9
i-C3H7
n-C3H7
n-C4H9
i-C3H7
cyclo-C6H11
CH3
>100
+
3f
3g
>100
+
ꢀ
3h
3i
cyclo-C6H11
C6H5
C6H5
ꢀ
ꢀ
96.8 (70.1–133.9)
+
+
3j
3k
CH3
CH3
ꢀ
p-F-C6H4
p-CF3-C6H4
C6H5
O
O
O
O
O
O
87.1 (65.6–123.8)
>100
>100
>100
3l
3m
CH3
C2H5
54.8 (19.0–162.4)
85.8 (40.7–181.0)
41.8 (39.2-44.5)
82.5 (61.8–110.0)
20d (16)e
81 (13–530)
>100
3n
3o
p-F-C6H4
p-NO2-C6H4
p-OCH3-C6H4
C2H5
C2H5
C2H5
ꢀ
3p
Phenobarbital
Phenytoin
+
ꢀ
ꢀ
8.1d (9)e
a ED50 values are in mg/kg of test drug delivered intraperitoneally and measured at 0.5 h. The time of peak effect was not determined.
b n = 3, numbers in parentheses are 95% confidence intervals.
c Previously synthesized and published data (Sarra et al.10). ‘ꢀ’ means not active up to 300 mg/kg. ‘+’ denotes activity at 300 mg/kg but ED50 could
not be determined.
d Experimental ED50 value.
e Literature ED50 value.
f Partition coefficient values calculated by MolinspirationÒ interactive logP calculator.
crease in anticonvulsant activity, possibly due to its in-
creased bulk. Increasing the chain length to two carbons
at N-30 position (3m) resulted in increased activity, when
compared to the N-30 methyl compound (3i), in agree-
ment with previous SAR of linear chain substituents.9
Introducing the p-nitrophenyl group as the R1-substitu-
ent resulted in compound 3o having the best activity in
scPTZ test with ED50 of 41.8 mg/kg, five times more po-
tent than the parent compound 3b, but abolished its
activity in MES test.
significant anticonvulsant activity against the scPTZ test
with most ED50 values in the narrow range of 78–97 mg/
kg. In the dialkyl series, based on the SAR, an un-
branched, unsaturated alkyl chain up to three carbons
is essential for protection, with di-n-propyl derivative
(3e) exhibiting strongest activity. Compound 3o18 (the
N0-30-ethyl-N-10-p-nitrophenyl derivative) in the N-10-
aryl-N-30-alkyl disubstituted series exhibited maximum
activity with an ED50 of 41.8 mg/kg which is five times
more potent than the parent compound (3b) in the
scPTZ test. It should be noted all of these compounds
are prepared as racemic mixtures and no attempt has
been made to resolve the enantiomers. Most compounds
were not significantly active in the MES assay.
Electron-donating groups like methoxy (3p) at the 4 po-
sition of the phenyl ring cause a slight decrease in activ-
ity compared to 3m but was not significant.
Unfortunately, further increasing the chain length to
three carbons at N-30 engendered very high toxicity
(data not shown). Based on the calculated logP values
shown in Table 1 the compounds showing significant
activity in scPTZ model have partition coefficient values
in the range of 2.4–3.0 compared to that of the lead
compound 3b (logP = 1.457). Although the drug levels
in cerebrospinal fluid (CSF) or brain concentrations
were not measured, higher logP values for the new series
suggest that there is correlation between activity and
lipophilicity.
Acknowledgment
´
The authors thank Dr. Istvan Lengyel for interpreting
the spectroscopic data.
References and notes
1. Chang, B. S.; Lowenstein, D. H. N. Eng. J. Med. 2003,
349, 1257.
2. McNamara, J. O. In Goodman and Gilman’s The Pharma-
cological Basis of Therapeutics; Hardman, J. G., Limbird,
L. E., Eds., 10th ed.; McGraw Hill: New York, 2001; pp
521–547.
In summary, several new N-10, N-30-disubstituted
phthalidyl spirohydantoins were synthesized and shown
to have appreciable anticonvulsant activity. Among the
compounds tested 3e, 3i, 3k, 3m, 3n, and 3o displayed

H. J. Patel et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4644–4647
4647
3. Upton, N. Trends Pharmacol. Sci. 1994, 15, 452.
4. Pastalos, P. N. Curr. Opin. CPNS Invest. Drugs 1999, 1,
549.
5. Vida, J. A. In Anticonvulsants in Medicinal Chemistry;
Academic Press, 1977; Vol. 15, pp 58, 59, 28, 29, 73, 74,
152, 155.
6. Brodie, M. J. Epilepsy Res. 2001, 45, 3.
7. Kung, C.; Wurpel, J.; Kwon, C. Drug Dev. Res. 1999, 47,
17.
extensor component were recorded in scPTZ and MES,
respectively. Experiments were done in triplicates and the
ED50 values were determined by non-linear regression
analysis and reported as the mean value with 95%
confidence intervals.
14. Swinyard, E. A.; Brown, W. C.; Goodman, L. S. J.
Pharmacol. Exp. Ther. 1952, 102, 319–330.
15. Sarra, J. D.; Stephani, R. A. Res. Commun. Biol. Psychol.
Psychiat. 1998, 23, 73.
8. Arnold, H.; Kuehas, E.; Brock, N. DE 1,135,915, Chem.
Abstr. 1962, 58, 20770.
9. Brimelow, H. C.; Vasey, C. H. GB 807,676 and 807,678,
Chem. Abstr. 1959, 53, 67757.
10. Sarra, J. D.; Stephani, R. A. Med. Chem. Res. 2000, 10, 81.
11. Rossi, M.; Phillips, C.; Conant, C.; Stephani, R. A.;
Caruso, F. American Crystallographic Association Annu-
al Meeting, May 28–June 2, Orlando, FL, 2005; 04.01.03
(W0188).
12. X-ray coordinates of compound 3i have been deposited
with the Cambridge Crystallographic Data Centre and the
deposition number is CCDC 608866.
13. Compounds 3d–p, positive controls or vehicle (DMSO)
were administered intraperitoneally 30 min before the
subcutaneous injection of PTZ, 80 mg/kg, or administra-
tion of electroshock (50 mA for 0.2 s). Occurrences of
clonic seizure for more than 5 s and hind limb tonic
16. Vogel, G. H.; Vogel, W. H. In Drug Discovery and
Evaluation; Pharmacological Assays; Springer Publication:
1997; pp 227, 220, 221.
17. Poupaert, J. H.; Vandervost, D.; Moustafa, M. M.;
Dumont, P. J. Med. Chem. 1984, 27, 76.
18. Spectroscopic data for 3o. IR (KBr) m: 1799, 1783, 1738,
1520 cmꢀ1. 1H NMR (400 MHz; DMSO-d6) d: 1.039 (3H,
t, J = 7.2 Hz), 3.160 (1H, sextet, J = 7.2 Hz), 3.397 (1H,
sextet, J = 7.3 Hz), 7.880 (1H, t, J = 7.5 Hz), 7.907 (2H, d,
J = 8.7 Hz), 7.995 (1H, t, J = 7.6 Hz), 8.097 (1H, d,
J = 7.6 Hz), 8.221 (1H, d, J = 7.7 Hz), 8.417 (2H, d,
J = 8.9 Hz). 13C NMR (400 MHz; DMSO-d6) d: 14.06,
35.23, 92.20, 124.15, 124.66, 125.74, 126.74, 132.55,
135.80, 136.67, 140.52, 146.47, 152.79, 165.5, 166.42. m/
z: 367 (MÅ+), 323, 175, 160, 131, 104, 90, 76, 50. Anal.
Calcd for C18H13N3O6: C, 58.87; H, 3.56; N, 11.44.
Found: C, 58.63; H, 3.56; N, 11.45.