Synthesis and anticonvulsant activities of 3,3-dialkyl- and 3-alkyl-3- benzyl-2-piperidinones (δ-valerolactams) and hexahydro-2H-azepin-2-ones (ε- caprolactams)

Article abstract of DOI:10.1021/jm960561u

A series of 3-substituted 2-piperidinone (δ-valerolactam) and hexahydro-2H-azepin-2-one (ε-caprolactam) derivatives were prepared and evaluated as anticonvulsants in mice. In the 2-piperidinone series, 3,3- diethyl compound 7b is the most effective antico

Full text of DOI:10.1021/jm960561u

44
J . Med. Chem. 1997, 40, 44-49
Syn th esis a n d An ticon vu lsa n t Activities of 3,3-Dia lk yl- a n d
3-Alk yl-3-ben zyl-2-p ip er id in on es (δ-Va ler ola cta m s) a n d
Hexa h yd r o-2H-a zep in -2-on es (E-Ca p r ola cta m s)
P. Amruta Reddy,† Karen E. Woodward,‡ Sarah M. McIlheran,‡ Bonnie C. H. Hsiang,† Tammy N. Latifi,†
Matthew W. Hill,§ Steven M. Rothman,§,| J ames A. Ferrendelli,‡ and Douglas F. Covey*,†
Departments of Molecular Biology and Pharmacology, Anatomy and Neurobiology, and Neurology and Neurosurgery,
Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, Missouri 63110, and Department of
Neurology, University of Texas Health Science Center, 6431 Fannin Street, Houston, Texas 77030
Received J uly 30, 1996X
A series of 3-substituted 2-piperidinone (δ-valerolactam) and hexahydro-2H-azepin-2-one (ꢀ-
caprolactam) derivatives were prepared and evaluated as anticonvulsants in mice. In the
2-piperidinone series, 3,3-diethyl compound 7b is the most effective anticonvulsant against
pentylenetetrazole-induced seizures (ED50, 37 mg/kg; PI (TD50/ED50), 4.46), and 3-benzyl
compound 4c (ED50, 41 mg/kg; PI, 7.05) is the most effective anticonvulsant against seizures
induced by maximal electroshock. By contrast, none of the ꢀ-caprolactams tested had
anticonvulsant effects below doses causing rotorod toxicity. log P values were correlated with
neurotoxicity and [35S]TBPS displacement, but not with anticonvulsant activity. Electrophysi-
ological evaluations of selected compounds from each series indicated that both the δ-valero-
lactams and ꢀ-caprolactams potentiated GABA-mediated chloride currents in rat hippocampal
neurons.
Sch em e 1a
A recent study from this laboratory described the
anticonvulsant activities of a series of 2-pyrrolidinones
(γ-butyrolactams) containing alkyl and/or phenylmethyl
(benzyl) substituents at C-3.1 Overall, it was found that
3,3-disubstituted 2-pyrrolidinones had anticonvulsant
activities against seizures induced either by pentylene-
tetrazole (PTZ) or by maximal electroshock (MES) which
were superior to those reported for the previously
studied congeneric R,R-dialkyl-substituted γ-butyrolac-
tones (GBLs) and γ-thiobutyrolactones (TBLs).2 Within
this context, to further define the structure-activity
relationships of anticonvulsant lactams, we have inves-
tigated the effect of ring size on pharmacological activ-
ity. A series of 3,3-disubstituted 2-piperidinone (δ-
valerolactam) and hexahydro-2H-azepin-2-one (ꢀ-capro-
lactam) derivatives has been prepared, and their anti-
convusant activities against PTZ- and MES-induced
seizures has been studied. The effect of the compounds
on [35S]-tert-butylbicyclophosphorothionate ([35S]TBPS)
binding to the picrotoxin site on GABAA receptors was
examined, and the effects of selected compounds on
GABAA receptor function were measured using electro-
physiological methods.
a
(a) R1X, NaOEt; (b) (i) NaOH, (ii) heat; (c) PhCH2Br, KOH,
Bu4NHSO4; (d) R2X, LDA; (e) Li, NH3; (f) (i) Me3SiCl, Et3N, (ii)
EtI, LDA.
Ch em istr y
Monosubstituted 2-piperidinones 4a -c were prepared
from 3-(ethoxycarbonyl)-2-piperidinone (2) by reported
procedures3 as summarized in Scheme 1. Attempted
C-alkylation of lactams 4a and 4b with benzyl bromide
in the presence of 2 equiv of n-butyllithium in THF at
0 °C as described in the preparation of 3,3-dialkyl-2-
piperidinones7 led only to small amounts of the corre-
sponding products 7c and 7d , respectively. Hence, an
alternative approach, in which N-benzyl lactams 5a ,b
(prepared from 4a ,b under phase-transfer conditions8)
were initially alkylated9,10 with appropriate alkyl ha-
lides in the presence of lithium diisopropylamide (LDA)
in THF-HMPA at -78 °C, followed by removal of the
N-benzyl group from resulting 6a -c with lithium in
liquid ammonia,9,11 was used to prepare lactams 7a , 7c,
and 7d (Scheme 1). The diethyl lactam 7b12 was
prepared in a slightly different manner. Thus, lactam
4b was reacted with Me3SiCl in the presence of excess
Et3N in benzene to afford the corresponding N-silyl
lactam, which was treated with LDA at -23 °C in THF
to form the enolate and then alkylated with ethyl iodide
at -78 °C to provide lactam 7b after aqueous desilyla-
† Department of Molecular Biology and Pharmacology.
‡ Department of Neurology.
§ Department of Anatomy and Neurobiology.
| Department of Neurology and Neurosurgery.
X Abstract published in Advance ACS Abstracts, December 15, 1996.
S0022-2623(96)00561-4 CCC: $14.00 © 1997 American Chemical Society

δ-Valerolactams and ꢀ-Caprolactams
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 45
Sch em e 2a
a
(a) (i) Me3SiCl, Et3N, (ii) EtI, LDA; (b) (i) Me3SiCl, Et3N, (ii)
RX, LDA.
tion.13 As summarized in Scheme 2, ꢀ-caprolactams 9,
10a , and 10b were prepared from ꢀ-caprolactam (8) by
a similar silylation, alkylation, and desilylation route.
Resu lts a n d Discu ssion
The anticonvulsant potencies (ED50) against PTZ- and
MES-induced seizures in mice, neurotoxicity (TD50) as
evaluated by a rotorod test, and protective indices (PI
) TD50/ED50) for the 2-piperidinone and ꢀ-caprolactam
derivatives are summarized in Table 1. For comparison,
earlier results1,14,15 from studies which utilized identical
methods to obtain these values for 3-(phenylmethyl)-2-
pyrrolidinone (11a ), 3,3-diethyl-2-pyrrolidinone (11b),
and antiepileptic drugs ethosuximide, phenobarbital,
and valproic acid are also included in Table 1.
Among the 2-piperidinone derivatives, the unsubsti-
tuted lactam 1, which was reported previously to
produce mild sedation in mice at a dose of 750 mg/kg,16
has neither anticonvulsant activity nor neurotoxicity at
doses up to 600 mg/kg. The 3-methyl lactam 4a is a
very weak anticonvulsant, and the corresponding 3-eth-
yl compound 4b displays both greater anticonvulsant
activity and neurotoxicity. The 3-benzyl lactam 4c has
an anticonvulsant activity which is significantly higher
than that found for compounds 4a and 4b. The dialkyl
lactams 7a and 7b have better anticonvulsant activity
than monoalkyl-substituted lactams 4a and 4b, with
diethyl compound 7b being a significantly more potent
anticonvulsant than compound 7a in the two seizure
models examined. An additional 3-alkyl substituent
along with the 3-benzyl group (compare compound 4c
with compounds 7c and 7d ) causes an increase in
rotorod toxicity, but does not significantly alter anti-
convulsant activity. For comparison, anticonvulsant
activity also has been reported for 3-methoxy-3-phenyl-
2-piperidinone (ED50 ) 50 mg/kg against subcutaneous
F igu r e 1. Plot of log 1/C against Hlog P, where C is TD50
expressed in moles per kilogram mouse: γ-butyrolactams
(circles), δ-valerolactams (squares), ꢀ-caprolactams (triangles).
The substituents on the 3-position in each lactam are indicated
next to the data points.
In contrast to the lack of behavioral effects produced
by unsubstituted 2-piperidinone 1, unsubstituted ꢀ-ca-
prolactam 8 caused a nonsedating toxicity in five out of
six mice at a dose of 300 mg/kg in the PTZ screen. At
a dose of 500 mg/kg, compound 8 produced clonic, but
not tonic seizures in four out of six mice. These results
are in general agreement with results from an earlier
study in which lactam 8 was found to cause slight
sedation at 300 mg/kg and Straub tail and clonic and
tonic convulsions at 500 mg/kg when injected intra-
peritonially into mice.16
The favorable effects (enhancement of anticonvulsant
activity relative to neurotoxicity) of C-3 substitution for
compounds in the 2-piperidinone series were not ob-
served in the ꢀ-caprolactam series. Compounds 9 and
10a had anticonvulsant activity only at doses which
caused rotorod neurotoxicity. Compound 10b, which
was very insoluble, had no neuroactivity at the highest
dose that could be evaluated. Hence, whereas the
structure-activity relationships for C-3 alkyl and benzyl
substituents and anticonvulsant activity are similar in
the 2-pyrrolidinone and 2-piperidinone series, these
relationships are clearly different in the ꢀ-caprolactam
series. Most notably, the high anticonvulsant activities
found for compounds 7b and 11b are not found for
compound 10a .
Metrazol and 121 mg/kg against MES; rotorod TD50
)
239 mg/kg).17
Overall, the structure-activity relationships of the
2-piperidinone derivatives investigated are very similar
to those reported earlier for the corresponding 2-pyrro-
lidinone derivatives.1 In each series, the compounds
with the best anticonvulsant profiles have either the
3-benzyl (compounds 4c and 11a ) or 3,3-diethyl (com-
pounds 7b and 11b) substituents. Compounds 7b and
11b are equally potent and superior to all but phe-
nobarbital in anti-PTZ activity. Compounds 4c and 11a
exhibit similar anticonvulsant potency in the MES test,
but compound 11a is about twice as neurotoxic as 4c.
The lower neurotoxicity of 4c gives a PI value of 7.05
against MES, which is greater than the PI value of
phenobarbital. With regard to the PI value for anti-
PTZ activity, 7b (PI ) 4.46) ranked ahead of phenobar-
bital and second only to the five-membered ring ana-
logue 11b.
Other investigators have also failed to find any
substitution patterns which produce useful anticonvul-
sant ꢀ-caprolactams. The convulsant activity of lactam
8 was found to be enhanced significantly by introduction
of alkyl groups at C-4 and C-6 of the lactam ring. By
contrast, ꢀ-caprolactams substituted at C-5 were inac-
tive, while some with bulky substituents at C-7 were
reported to be depressants.18,19
The interactions of the lactams with the picrotoxin
binding site of the GABAA receptors found in rat brain
membranes were assessed by determining IC50 values
for [35S]TBPS displacement and are provided in Table
1. The best anticonvulsant lactams studied display
weak interactions with the picrotoxin site on the GABAA
receptor, and there is clearly no correlation between
anticonvulsant activity and IC50 values for TBPS dis-

46 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Reddy et al.
Ta ble 1. Anticonvulsant Activity, Neurotoxicity, [35S]TBPS Binding Data, and Protective Index Values for 2-Piperidinones (1-7d ),
Hexahydro-2H-azepin-2-ones (8-10b), 2-Pyrrolidinones (11a ,b), Ethosuximide, Phenobarbital, and Valproic Acid
anticonvulsant potency
protective index
(PI) TD50/ED50
[
35S]TBPS
a
C-3 substituents
R1 R2
ED50 (mg/kg)
rotorod toxicity
TD50 (mg/kg)
displacement
b
compd
PTZ
MES
PTZ
MES
IC50c (mM)
1
H
H
>600
>600
>600
197 ( 4.5
27.7 ( 0.8
(0/6)d
(0/6)
(0/6)
4a
Me
Et
H
>600
(1/6)
>600
(0/6)
550
[462-654]
41
[31-55]
199
[175-227]
54
[33-87]
50
[41-60]
70
[28-175]
>300
(0/6)
249
[186-333]
178
[99-320]
>100
(0/6)
41
[20-63]
174
[133-229]
>562
>600
(2/6)
4b
H
297
>325
>1.09
3.52
2.25
4.46
2.23
1.69
<1
>0.59
7.05
1.90
3.06
2.50
1.47
<1
8.78 ( 0.24
1.44 ( 0.04
4.41 ( 0.11
1.31 ( 0.06
0.73 ( 0.01
0.27 ( 0.00
17.4 ( 0.24
3.20 ( 0.07
0.38 ( 0.02
0.09 ( 0.01
2.44 ( 0.17
5.81 ( 0.12
[196-449]e
82
{325,450}f
289
4c
Bn
Me
Et
H
[61-109]
168
[143-197]
37
[25-56]
56
[36-89]
61
[38-99]
>300g
(0/6)
210
[172-257]
226
[175-478]
378
[339-422]
165
[89-307]
125
[100-155]
103
7a
Et
Et
7b
7c
Me
Et
Bn
Bn
H
7d
[70-152]
8
H
<300
(9/12)
9
Et
H
145
[106-198]
131
[82-208]
>100
(0/6)
144
[125-168]
260
[219-320]
378
[320-436]
88
0.69
0.58
0.58
0.74
10a
10b
11a h
11bh
Et
Et
Bn
H
[180-285]
>100
(0/6)
Et
Bn
Et
71
2.03
5.65
2.35
4.00
2.12
3.51
1.49
[43-101]
46
[30-63]
161
[137-189]
22
[17-28]
133
Et
ethosuximidei
phenobarbitalj
valproic acidi
<0.67
5.18
17
[14-23]
206
[71-112]
282
1.37
[109-154]
[176-247]
[258-315]
a
Dose at which 50% of the mice were protected from clonic seizures induced by pentylenetetrazole (85 mg/kg) or tonic hindlimb seizures
induced by maximal electroshock. A minimum of four doses of each compound dissolved in 30% polyethylene glycol was administered
b
through intraperitonial injections (0.01 mL/g of body weight). A group of six animals was used at each dose tested. Dose at which 50%
of the mice failed the rotorod toxicity test. c Binding data are presented as the mean and range from duplicate experiments performed in
d
triplicate. Values reported for 8, 9, 10a , and 10b are the mean ( SEM of three experiments performed in triplicate. Fractions in
parentheses indicate number of mice protected/number of mice in group or number of mice toxic/number of mice in group at the dose
reported. e Numbers in brackets are the 95% fiducial limits. f Numbers in braces are the highest dose tested at which all mice were
nontoxic and the lowest dose tested at which all mice were toxic, respectively. These values are given in place of 95% fiducial limits which
g
could not be calculated from the data available. Caused toxicity in five out of six mice at a dose of 300 mg/kg, but the animals were not
h
notably sedated. At a dose of 500 mg/kg, only clonic but not tonic seizures were observed in four out of six mice. Data is from ref 1.i Data
j
is from ref 14. Data is from ref 15.
placement in the binding experiments (e.g., compare the
results shown for compounds 7b, 7d , and 10a in Table
1). Additionally, no correlation between the calculated
log P values, Hlog P,20 and the anticonvulsant activity
(r2 ) 0.208 for PTZ) of substituted 2-pyrrolidinones,
2-piperidinones, and ꢀ-caprolactams described in this
and our previous paper1 was observed. By contrast, a
plot of log 1/C (where C is the TD50 value in moles per
kilogram) against Hlog P of the same lactams (Figure
1) shows that there is a reasonably high degree of
correlation between the neurotoxicity and lipophilic
character of the lactams (r2 ) 0.759). Similar correla-
tions have been observed previously for other congeneric
series of CNS active compounds.21 Lastly, a correlation
(r2 ) 0.698) was observed between lipophilicity and the
ability of these lactams to displace TBPS from the
picrotoxin site on the GABAA receptor.
R-isopropyl-R-methyl-γ-butyrolactone (Table 2). The
potencies are comparable to those reported previously
for the corresponding 2-pyrrolidinones (for example, 1
mM of compound 11b gives a response of 137 ( 6%
relative to 3 µM GABA control ) 100%), indicating that
increasing ring size by one or two carbons has little
effect on this response. The results obtained with
ꢀ-caprolactam 10a in this assay would suggest that this
compound should be an effective anticonvulsant, but
this was not found to be the case. Previously, the
modulatory actions of another ꢀ-caprolactam, hexahy-
dro-3,3-diallyl-6,6-dimethyl-2H-azepin-2,4-dione, at
GABAA receptors also was found not to correlate with
the compound’s in vivo activity. This compound, which
at a concentration of 1 mM enhances GABA-mediated
chloride current (114% of control) in mouse spinal cord
neurons,23 was found to be a potent convulsant in mice
(CD50 ) 40 mg/kg).24
The effects of compounds 7b, 7d , 10a , and 10b on
GABA receptor function were determined using elec-
trophysiological methods. All four compounds enhanced
3 µM GABA-induced chloride currents in cultured rat
hippocampal neurons (Figure 2), and the actions of the
compounds were blocked by the picrotoxin antagonist22
In summary, the structure-activity relationships for
anticonvulsant lactams has been extended to 3-alkyl-
substituted 2-piperidinones and ꢀ-caprolactams. The
anticonvulsant profiles of the 2-piperidinones and ear-
lier studied 2-pyrrolidinones are similar, whereas the

δ-Valerolactams and ꢀ-Caprolactams
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 47
Ta ble 2. Antagonism of 2-Piperidinone and
Hexahydro-2H-azepin-2-one Derivative Induced Potentiation of
3 µM GABA-Mediated Current by the Picrotoxin Antagonist
R-Isopropyl-R-methyl-γ-butyrolactone (R-IMGBL)
compd potentiation
% response relative
to current produced
compd
by GABAa
Nb
7b (1 mM)
7b (1 mM) + R-IMGBL (3 mM)
7d (300 µM)
7d (300 µM) + R-IMGBL (3 mM)
10a (200 µM)
10a (200 µM) + R-IMGBL (3 mM)
10b (600 µM)
10b (600 µM) + R-IMGBL (3 mM)
142 ( 10c
102 ( 6*
138 ( 8
6
6
6
5
6
5
9
14
93 ( 4*
117 ( 4
105 ( 2**
128 ( 8
107 ( 3**
a
To calculate the percentage response, the magnitude of the
peak current produced by 3 µM GABA plus compound(s) was
normalized with respect to the peak current produced by 3 µM
GABA alone on the same cell. A percentage response of 100%
reflects no change in the current compared to 3 µM GABA alone.
b
N ) Number of cells examined. c Values are the mean ( SEM.
*p < 0.01 by Student’s t test. **p < 0.05 by Student’s t test.
Statistical significance calculated for test compound alone vs test
compound and R-IMGBL.
mmol) in THF (200 mL) was treated with powdered 85% KOH
(2.9 g, 44 mmol) at 0 °C in a N2 atmosphere. The reaction
mixture was stirred at room temperature for 24 h, poured into
saturated NH4Cl (60 mL), and extracted with ether (3 × 50
mL), and the combined organic extract was washed with brine
(50 mL). Solvent removal gave a yellow oil, which was flash
chromatographed over silica gel (hexanes-CHCl3-EtOAc, 7:2:
1) to give 7.33 g (90%) of N-benzyl lactam 5a as a colorless
oil, with spectroscopic characteristics identical to those re-
ported earlier.26
F igu r e 2. Concentration-response curves for the potentiation
of 3 µM GABA-mediated chloride currents in cultured hippo-
campal neurons by compounds 7b and 7d (panel A) and 10a
and 10b (panel B). Data points are the mean ( SEM for at
least four cells at each concentration tested.
5b was similarly prepared.
3-Eth yl-1-(p h en ylm eth yl)-2-p ip er id in on e (5b). Lactam
5b27 (2.03 g, 89%) was prepared from 4b (1.33 g, 10.5 mmol)
as a colorless oil: 13C NMR δ 172.79, 137.48, 128.52, 127.93,
127.22, 50.29, 47.44, 42.90, 25.74, 24.91, 21.59, 11.47. Anal.
(C14H19NO) C, H, N.
ꢀ-caprolactams are much weaker anticonvulsants. In
particular, compounds 4c and 7b exhibit an anticon-
vulsant profile resembling that of phenobarbital. Ad-
ditional studies to elucidate the interactions of these
compounds with GABAA receptors and to identify the
role of other receptors in the anticonvulsant actions of
these compounds are in progress.
3-Eth yl-3-m eth yl-1-(ph en ylm eth yl)-2-piper idin on e (6a).
A solution of lactam 5a (3.25 g, 16.0 mmol) in THF (5 mL)
was added slowly (ca. 5 min) to a solution of LDA [prepared
from diisopropylamine (2.42 g, 24.0 mmol) and n-butyllithium
in hexanes (2.5 M, 9.2 mL, 23 mmol)] in THF (30 mL) at 0 °C
in a N2 atmosphere, and the mixture was stirred for 45 min.
The temperature was then reduced to -78 °C, and a solution
of iodoethane (3.92 g, 25.1 mmol) in THF (5 mL) and HMPA
(2.4 mL, 13.8 mmol) was added over a period of 10 min.
Stirring was continued for 2 h at -78 °C, and the reaction
was quenched by addition of saturated NH4Cl (50 mL). The
layers were separated, the aqueous phase was further ex-
tracted with ether (3 × 50 mL), and the combined organic
extract was washed with brine (50 mL). Solvent removal gave
5.90 g of a pale yellow liquid, which upon column chromatog-
raphy over silica gel (hexanes-EtOAc, 9:1) followed by bulb-
to-bulb distillation [pot temperature 110-115 °C (1 mmHg)]
afforded lactam 6a (2.52 g, 68%) as a colorless viscous liquid:
IR 1635 (CdO), 1595 (CdC) cm-1; 1H NMR δ 7.34-7.22 (m, 5,
PhH), 4.66 (d, 1, J ) 14.6 Hz, diastereotopic H of CH2Ph), 4.49
(d, 1, J ) 14.6 Hz, diastereotopic H of CH2Ph), 3.24-3.14 (m,
2, H-6), 1.89-1.68 (m, 4, H-4, H-5), 1.59-1.48 (m, 2, CH3CH2),
1.23 (s, 3, CH3), 0.88 (t, 3, J ) 7.5 Hz, CH3CH2); 13C NMR δ
175.43, 137.56, 128.33, 127.73, 126.98, 50.18, 47.66, 41.72,
32.42, 31.84, 25.65, 19.30, 8.47. Anal. (C15H21NO) C, H, N.
Similarly prepared were the following.
Exp er im en ta l Section
Gen er a l Meth od s. 2-Piperidinone (1), 3-(ethoxycarbonyl)-
2-piperidinone (2), and hexahydro-2H-azepin-2-one (8) were
obtained from Aldrich Chemical Co. (Milwaukee, WI). Dichlo-
romethane (CH2Cl2) was distilled over calcium hydride. Hex-
amethylphosphoramide (HMPA) was dried over activated 13X
molecular sieves. Tetrahydrofuran (THF) was distilled over
sodium-benzophenone. Solvents used for extractions were
dried over MgSO4, filtered, and removed on a rotary evapora-
tor. Elemental analyses were performed at M-H-W Labora-
tories in Phoenix, AZ. NMR spectra, IR spectra, melting
points, and chromatographic separations were obtained by
methods described previously.1
3-Meth yl-2-p ip er id in on e (4a ): mp 55-56 °C (lit.4 mp 46
°C; lit.25 mp 53-55 °C).
3-Eth yl-2-p ip er id in on e (4b): mp 67-68 °C (lit.5 mp 66-
68 °C); IR 3279-3072 (NH), 1657 (CdO) cm-1; 1H NMR δ 7.36
(br, 1, NH), 3.36-3.22 (m, 2, H-6), 2.26-2.15 (m, 1, H-3), 2.00-
1.46 (m, 6, CH3CH2, H-4, H-5), 0.95 (t, 3, J ) 7.5 Hz, CH3-
CH2); 13C NMR δ 175.33, 42.05, 42.02, 25.23, 24.23, 21.08,
11.15.
3-Meth yl-1,3-bis(p h en ylm eth yl)-2-p ip er id in on e (6b).
Lactam 6b (3.72 g, 71%) was prepared from 5a (3.65 g, 18.0
3-(P h en ylm eth yl)-2-p ip er id in on e (4c): mp 119-121 °C
(lit.6 mp 115-116 °C); IR 3251-3026 (NH), 1661 (CdO), 1605
mmol) as a colorless oil: IR 1635 (CdO), 1605 (CdC) cm-1
;
(CdC) cm-1
;
13C NMR δ 174.30, 139.85, 129.25, 128.36, 126.14,
1H NMR δ 7.30-7.12 (m, 10, PhH), 4.74 (d, 1, J ) 14.6 Hz,
diastereotopic H of NCH2), 4.39 (d, 1, J ) 14.6 Hz, diaste-
reotopic H of NCH2), 3.33 (d, 1, J ) 13.2 Hz, diastereotopic H
of CH2Ph), 3.11-2.95 (m, 2, H-6), 2.65 (d, 1, J ) 13.2 Hz,
diastereotopic H of CH2Ph), 1.84-1.44 (m, 4, H-4, H-5), 1.31
42.87, 42.45, 37.43, 25.42, 21.24. Anal. (C12H15NO) C, H, N.
3-Meth yl-1-(p h en ylm eth yl)-2-p ip er id in on e (5a ): A well-
stirred mixture of the lactam 4a (4.52 g, 40.0 mmol), benzyl
bromide (8.21 g, 48.0 mmol), and Bu4NHSO4 (2.72 g, 8.00

48 J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1
Reddy et al.
(s, 3, CH3); 13C NMR δ 174.64, 138.06, 137.20, 130.69, 128.31,
127.78, 127.75, 126.99, 126.09, 50.43, 47.70, 45.49, 43.04,
31.91, 26.86, 19.20. Anal. (C20H23NO) C, H, N.
-78 °C with stirring in an atmosphere of N2. The resulting
solution was kept at -23 °C for 1 h and recooled to -78 °C,
and iodoethane (1.76 g, 11.3 mmol) was added to it in one
portion. The reaction mixture was allowed to warm to room
temperature (ca. 3 h), and stirring was continued for another
2 h period. Water (25 mL) was added, the layers were
separated, the aqueous phase was extracted with ether (3 ×
30 mL), and the combined organic extract was washed with
brine (30 mL). Removal of the solvent gave 1.39 g of viscous
oil, which upon flash chromatography over silica gel (hexanes-
acetone-EtOAc, 2:1:1) afforded the pure lactam 7b (0.93 g,
80%) as a colorless solid: mp 62-63 °C (from hexanes at -5
°C) (lit.12 mp 61-64 °C); IR 3284-3090 (NH), 1654 (CdO)
cm-1; 1H NMR δ 5.90 (br, 1, NH), 3.30-3.25 (m, 2, H-6), 1.86-
1.66 (m, 6, H-4, H-5, 2 × diastereotopic H of CH3CH2,), 1.56-
1.44 (m, 2, 2 × diastereotopic H of CH3CH2,), 0.88 (t, 6, J )
7.4 Hz, 2 × CH3CH2); 13C NMR δ 177.51, 44.78, 42.37, 30.75,
28.25, 19.81, 8.58. Anal. (C9H17NO) C, H, N.
3-Eth yl-1,3-bis(ph en ylm eth yl)-2-piper idin on e (6c). Lac-
tam 6c (3.10 g, 84%) was prepared from 5b (2.60 g, 12.0 mmol)-
as a colorless solid: mp 88-90 °C (from CH2Cl2-hexane); IR
1
1625 (CdO), 1595 (CdC) cm-1; H NMR δ 7.32-7.17 (m, 10,
PhH), 4.67 (d, 1, J ) 14.5 Hz, diastereotopic H of NCH2), 4.50
(d, 1, J ) 14.5 Hz, diastereotopic H of NCH2), 3.34 (d, 1, J )
13.0 Hz, diastereotopic H of CH2Ph), 3.11-3.03 (m, 1, diaste-
reotopic H of H-6), 2.96-2.89 (m, 1, diastereotopic H of H-6),
2.59 (d, 1, J ) 13.0 Hz, diastereotopic H of CH2Ph), 1.99-
1.90 (m, 1, diastereotopic H of CH3CH2), 1.71-1.26 (m, 5, H-4,
H-5, diastereotopic H of CH3CH2), 0.92 (t, 3, J ) 7.4 Hz, CH3-
CH2); 13C NMR δ 174.08, 138.43, 137.37, 130.59, 128.34,
127.96, 127.86, 127.05, 126.15, 50.54, 47.57, 46.88, 44.63,
32.70, 28.24, 19.74, 8.79. Anal. (C21H25NO) C, H, N.
3-Eth yl-3-m eth yl-2-p ip er id in on e (7a ). To a well-stirred
solution of lactam 6a (2.08 g, 9.00 mmol) in THF (27 mL) and
liquid NH3 (180 mL) was added Li metal (0.63 g, 90 mmol) at
-78 °C slowly over a period of 5 min. The cooling bath was
removed and the reaction mixture stirred for 1 h at refluxing
NH3. Then it was heated for 10-15 min at 60 °C to evaporate
NH3, the white residue was treated carefully with water (50
mL) at 0 °C and extracted with CH2Cl2 (3 × 50 mL), and the
combined organic extract was washed with brine (50 mL).
Removal of the solvent gave 1.81 g of a colorless viscous
residue, which upon flash chromatography over silica gel (1%
MeOH in CHCl3-EtOAc, 1:1) afforded the lactam 7a (1.10 g,
87%) as a colorless solid: mp 67-69 °C (from CH2Cl2-hexane);
IR 3285-3064 (NH), 1646 (CdO) cm-1; 1H NMR δ 6.20 (br, 1,
NH), 3.34-3.22 (m, 2, H-6), 1.87-1.68 (m, 4, H-4, diaste-
reotopic H of CH3CH2 and H-5), 1.59-1.43 (m, 2, diastereotopic
H of CH3CH2 and H-5), 1.19 (s, 3, CH3), 0.88 (t, 3, J ) 7.5 Hz,
CH3CH2); 13C NMR δ 178.43, 42.68, 41.33, 32.08, 31.86, 25.36,
19.42, 8.47. Anal. (C8H15NO) C, H, N.
Hexa h yd r o-3-eth yl-2H-a zep in -2-on e (9). A solution of
the known hexahydro-1-(trimethylsilyl)-2H-azepin-2-one (9.25
g, 50.0 mmol)28 in THF (25 mL) was added slowly (ca. 15 min)
to LDA [(prepared from diisopropylamine (5.05 g, 50.0 mmol)
and n-butyllithium 2.5 M, 20 mL, 50 mmol)] in THF (25 mL)
at -78 °C with stirring in an atmosphere of N2. After 45 min,
this was transferred (by cannula) over a 10 min period to a
-78 °C solution of iodoethane (7.02 g, 45.0 mmol) in THF (50
mL). The reaction mixture was allowed to warm to room
temperature (ca. 3 h), and stirring was continued for another
2 h period. Water (50 mL) was added, the layers were
separated, the aqueous phase was extracted with ether (3 ×
50 mL), and the combined organic extract was washed with
brine (50 mL). Removal of the solvent gave 6.51 g of colorless
solid, which upon recrystallization from CH2Cl2-hexanes
afforded the lactam 9 (5.54 g, 79%) as colorless needles: mp
97-99 °C; IR 3302-3078 (NH), 1656 (CdO) cm-1; 1H NMR δ
6.18 (br, 1, NH), 3.34-3.12 (m, 2, H-7), 2.33-2.25 (m, 1, H-3),
2.03-1.27 (m, 8, H-4, H-5, H-6, CH3CH2), 0.95 (t, 3, J ) 7.4
Hz, CH3CH2); 13C NMR δ 180.17, 45.36, 41.79, 29.47, 29.42,
24.34, 12.14. Anal. (C8H15NO) C, H, N.
Similarly prepared by the procedure described in 7b via the
intermediate hexahydro-3-ethyl-1-(trimethylsilyl)-2H-azepin-
2-one [colorless liquid: IR 1640 (CdO) cm-1; 1H NMR δ 3.33-
3.17 (m, 2, H-7), 2.47-2.38 (m, 1, H-3), 1.93-1.21 (m, 8, H-4,
H-5, H-6, CH3CH2), 0.91 (t, 3, J ) 7.3 Hz, CH3CH2), 0.24 (s, 9,
Si(CH3)3)] were the following.
Hexa h yd r o-3,3-d ieth yl-2H-a zep in -2-on e (10a ). Lactam
10a (0.85 g, 67%) was a colorless solid: mp 77-78 °C (from
hexanes at -5 °C); IR 3278-3069 (NH), 1645 (CdO) cm-1; 1H
NMR δ 5.89 (br, 1, NH), 3.24-3.19 (m, 2, H-7), 1.84-1.55 (m,
10, H-4, H-5, H-6, 2 × CH3CH2), 0.87 (t, 6, J ) 7.5 Hz, 2 ×
CH3CH2); 13C NMR δ 180.39, 47.99, 42.10, 31.62, 29.06, 27.85,
23.63, 8.39. Anal. (C10H19NO) C, H, N.
Similarly prepared were the following.
3-Meth yl-3-(p h en ylm eth yl)-2-p ip er id in on e (7c). Lac-
tam 7c (1.75 g, 86%) was prepared from 6b (2.93 g, 10.0 mmol)
as a colorless solid: mp 76-78 °C (from ether-hexane); IR
1
3210-3028 (NH), 1657 (CdO), 1600 (CdC) cm-1; H NMR δ
7.30-7.18 (m, 5, PhH), 5.69 (br, 1, NH), 3.23 (d, 1, J ) 13.3
Hz, diastereotopic H of CH2Ph), 3.16-3.08 (m, 2, H-6), 2.67
(d, 1, J ) 13.3 Hz, diastereotopic H of CH2Ph), 1.81-1.63 (m,
3, H-4, diastereotopic H of H-5), 1.52-1.41 (m, 1, diastereotopic
H of H-5), 1.26 (s, 3, CH3); 13C NMR δ 177.57, 137.85, 130.46,
127.83, 126.15, 44.77, 42.54, 42.30, 31.68, 26.07, 19.81. Anal.
(C13H17NO) C, H, N.
3-Eth yl-3-(p h en ylm eth yl)-2-p ip er id in on e (7d ). Lactam
7d (1.43 g, 73%) was prepared from 6c (2.76 g, 9.00 mmol) as
a colorless solid: mp 101-102 °C (from CH2Cl2-hexane); IR
1
3203-3028 (NH), 1656 (CdO), 1595 (CdC) cm-1; H NMR δ
7.29-7.18 (m, 5, PhH), 6.06 (br, 1, NH), 3.25 (d, 1, J ) 13.2
Hz, diastereotopic H of CH2Ph), 3.22-3.01 (m, 2, H-6), 2.59
(d, 1, J ) 13.2 Hz, diastereotopic H of CH2Ph), 1.95-1.84 (m,
1, diastereotopic H of CH3CH2), 1.69-1.59 (m, 3, H-4, diaste-
reotopic H of H-5), 1.51-1.36 (m, 2, diastereotopic H of
CH3CH2 and H-5), 0.93 (t, 3, J ) 7.4, CH3CH2); 13C NMR δ
176.70, 138.19, 130.41, 127.91, 126.18, 46.35, 44.20, 42.47,
32.07, 27.94, 19.74, 8.71. Anal. (C14H19NO) C, H, N.
H e xa h yd r o-3-e t h yl-3-(p h e n ylm e t h yl)-2H -a ze p in -2-
on e (10b). Lactam 10b (1.92 g, 69%) was a colorless solid:
mp 100-102 °C (from EtOAc-hexanes at -5 °C); IR 3287-
3062 (NH), 1647 (CdO) cm-1 1H NMR δ 7.28-7.16 (m, 5,
;
PhH), 6.02 (br, 1, NH), 3.27-3.04 (m, 2, H-7), 3.05 (d, 1, J )
13.5 Hz, diastereotopic H of CH2Ph), 2.87 (d, 1, J ) 13.5 Hz,
diastereotopic H of CH2Ph), 1.79-1.48 (m, 8, H-4, H-5, H-6,
CH3CH2), 0.99 (t, 3, J ) 7.5 Hz, CH3CH2); 13C NMR δ 179.75,
138.40, 130.74, 127.69, 125.96, 49.60, 42.14, 41.42, 31.18,
28.47, 28.40, 23.04, 8.71. Anal. (C15H21NO) C, H, N.
Neu r ologica l E va lu a t ion s, 35[S]TBP S Bin d in g, a n d
Electr op h ysiology. The methods used have been described
previously.1,29 Fiducial limits were analyzed by the method
of Litchfield and Wilcoxon30 using the computer program
reported by Tallarida and Murray.31
3,3-Dieth yl-2-p ip er id in on e (7b). To a solution of lactam
4b (1.27 g, 10.0 mmol) and triethylamine (10.1 g, 100 mmol)
in benzene (30 mL) was added chlorotrimethylsilane (4.36 g,
40.0 mmol) with stirring in an atmosphere of N2. The resulting
suspension was stirred at room temperature for 12 h, diluted
with benzene (30 mL), and filtered through a sintered glass
funnel. The filtrate was concentrated and the residue bulb-
to-bulb distilled [pot temperature 70 °C (0.7 mmHg)] to give
3-ethyl-1-(trimethylsilyl)-2-piperidinone (1.51 g, 76%) as a
Ack n ow led gm en t. We thank Ms. Nancy Lancaster
for the preparation and maintenance of the primary
hippocampal cell cultures and Dr. Rino Ragno for the
calculation of the log P values. The work was supported
by NIH Grant NS14834 and by Training Grant 5 T32
GM07805. Assistance was also provided by the Wash-
ington University High Resolution NMR Facility sup-
colorless liquid: IR 1657 (CdO) cm-1 1H NMR δ 3.24-3.11
;
(m, 2, H-6), 2.30-2.15 (m, 1, H-3), 2.01-1.39 (m, 6, H-4, H-5,
CH3CH2), 0.92 (t, 3, J ) 7.5 Hz, CH3CH2), 0.25 (s, 9, Si(CH3)3).
A solution of the above N-trimethylsilyl lactam (1.49 g, 7.50
mmol) in THF (10 mL) was added slowly (ca. 5 min) to LDA
[prepared from diisopropylamine (0.95 g, 9.4 mmol) and
n-butyllithium (2.5 M, 3.8 mL, 9.5 mmol)] in THF (10 mL) at

δ-Valerolactams and ꢀ-Caprolactams
J ournal of Medicinal Chemistry, 1997, Vol. 40, No. 1 49
(17) Brouillette, W. J .; Grunewald, G. L. Synthesis and Anticonvul-
sant Activity of Some Substituted Lactams and Amides. J . Med.
Chem. 1984, 27, 202-206.
ported in part by NIH Grant 1 S10 RR00204 and a gift
from Monsanto Company.
(18) Duong, T.; Prager, R. H.; Tippett, J . M.; Ward, A. D.; Kerr, D. I.
B. Central Nervous System Active Compounds. II The Synthesis
of Some 4-, 5-, 6-, and 7-Substituted Lactams. Aust. J . Chem.
1976, 29, 2667-2682.
(19) Hutchison, G. I.; Prager, R. H.; Ward, A. D. Central Nervous
System Active Compounds. III The Synthesis of Some 4- and
6-Substituted Caprolactam Derivatives by the Schmidt and
Beckmann Rearrangements. Aust. J . Chem. 1980, 33, 2477-
2486.
(20) Kellogg, G. E.; Abraham, D. J . HINT 1.1A, Department of
Medicinal Chemistry, Virginia Commonwealth University, Rich-
mond, VA.
(21) Hansch, C.; Bjorkroth, J . P.; Leo, A. Hydrophobicity and Central
Nervous System Agents: On the Principle of Minimum Hydro-
phobicity in Drug Design. J . Pharm. Sci. 1987, 76, 663-687 and
the references cited therein.
Refer en ces
(1) Reddy, P. A.; Hsiang, B. C. H.; Latifi, T. N.; Hill, M. W.;
Woodward, K. E.; Rothman, S. M.; Ferrendelli, J . A.; Covey, D.
F. 3,3-Dialkyl- and 3-Alkyl-3-Benzyl-Substituted 2-Pyrrolidi-
nones: A New Class of Anticonvulsant Agents. J . Med. Chem.
1996, 39, 1898-1906.
(2) Canney, D. J .; Holland, K. D.; Levine, J . A.; McKeon, A. C.;
Ferrendelli, J . A.; Covey, D. F. Synthesis and Structure-Activity
Studies of Alkyl-Substituted γ-Butyrolactones and γ-Thiobutyro-
lactones: Ligands for the Picrotoxin Receptor. J . Med. Chem.
1991, 34, 1460-1467.
(3) The lactams 4a 4 and 4b5 were prepared according to the
procedure utilized in the synthesis of 4c.6
(4) Kariyone, K. Studies on Insecticides. VI. Synthesis of Proline
Derivatives. Chem. Pharm. Bull. 1960, 8, 1110-1113.
(5) Koelsch, C. F. A Synthesis of 3-Alkylpiperidones. J . Am. Chem.
Soc. 1943, 65, 2458-2459.
(6) Rodriguez, M.; Heitz, A.; Martinez, J . “Carba” Peptide Bond
Surrogates: Different Approaches to Gly-ψ(CH2-CH2)-D,L-Xaa
Pseudo-Dipeptide Units. Int. J . Pept. Protein Res. 1992, 39, 273-
277.
(7) Potts, K. T.; Rochanapruk, T. Intramolecular 1,4-Dipolar Cy-
cloadditions of Cross-Conjugated Heterocyclic Betaines. Syn-
thesis of Hexahydrojulolidines and Related Peri-Fused Ring
Systems. J . Org. Chem. 1995, 60, 3795-3805.
(8) Knight, J .; Parsons, P. J . New Approaches to the Synthesis of
â-Lactam Antibiotics. J . Chem. Soc., Perkin Trans. 1 1987,
1237-1242.
(9) Hiemstra, H.; Klaver, W. J .; Speckamp, W. N. ω-Alkoxy Lactams
as Dipolar Synthons. Silicon-Assisted Synthesis of Azabicycles
and a γ-Amino Acid. J . Org. Chem. 1984, 49, 1149-1151.
(10) Meyers, A. I.; Kunnen, K. B.; Still, W. C. Conformational Effects
on the Regiochemical Metalation of C5-C13 N-Benzyllactams.
J . Am. Chem. Soc. 1987, 109, 4405-4407.
(11) Sugasawa, S.; Fuji, T. Debenzylation of N-Benzyl Acylamides.
Chem. Pharm. Bull. 1958, 6, 587-590.
(12) Baker, J . A.; Harper, J . F. The Synthesis of 3,3-Diethylpiperid-
2-one. J . Chem. Soc. C 1967, 2148.
(22) Holland, K. D.; Yoon, K.-W.; Ferrendelli, J . A.; Covey, D. F.;
Rothman, S. M. γ-Butyrolactone Antagonism of the Picrotoxin
Receptor: Comparison of a Pure Antagonist and
a Mixed
Antagonist/Inverse Agonist. Mol. Pharmacol. 1991, 39, 79-84.
(23) Skerrit, J . H.; J ohnston, G. A. R.; Chow, S. C.; Macdonald, R.
L.; Prager, R. H.; Ward, A. D. Differential Modulation of
γ-Aminobutyric Acid Receptors by Caprolactam Derivatives with
Central Nervous System Depressant or Convulsant Activity.
Brain Res. 1985, 331, 225-233.
(24) Mooney, B. A.; Prager, R. H.; Ward, D. A. Central Nervous
system Active compounds. V. Claisen Rearrangement Products
of Allyl Vinyl Ethers Obtained from Caprolactam Derivatives.
Aust. J . Chem. 1980, 33, 2717-2728.
(25) Overberger, C. G.; J on, Y. S. Synthesis and Solution Properties
of Poly(2-methylproline). J . Polym. Sci. 1977, 15, 1413-1421.
(26) Zhang, Z.; Ojima, I. Synthesis of Nitrogen Heterocycles by Means
of Amine-Directed Carbonylation and Hydrocarbonylation. J .
Organomet. Chem. 1993, 454, 281-289.
(27) Yamaguchi, M.; Hirao, I. A Direct Synthesis of [(tert-Butoxy-
carbonyl)methylidine]azacycloalkanes from N-Alkyl Lactams. J .
Org. Chem. 1985, 50, 1975-1977.
(28) Lane, T. H.; Frye, C. L. Structure-Reactivity Relationships of
N-Alkyl(trimethylsilyl)amides. J . Org. Chem. 1978, 43, 4890-
4891.
(13) Fisher, M. J .; Overman, L. E. Intramolecular N-(Acyloxy)-
iminium Ion-Alkyne Cycloadditions. A New Route to Bicyclic
R-Amino Ketones. J . Org. Chem. 1990, 55, 1447-1459.
(14) Ferrendelli, J . A.; Holland, K. D.; McKeon, A. C.; Covey, D. F.
Comparison of the Anticonvulsant Activities of Ethosuximide,
Valproate, and a New Anticonvulsant, Thiobutyrolactone. Epi-
lepsia 1989, 30, 617-622.
(15) Holland, K. D.; McKeon, A. C.; Canney, D. J .; Covey, D. F.;
Ferrendelli, J . A. Relative Anticonvulsant Effects of GABAmi-
metic and GABA Modulatory Agents. Epilepsia 1992, 33, 981-
986.
(29) Peterson, E. M.; Xu, K.; Holland, K. D.; McKeon, A. C.; Rothman,
S. M.; Ferrendelli, J . A.; Covey, D. F. R-Spirocyclopentyl- and
R-Spirocyclopropyl-γ-Butyrolactones: Conformationally Con-
strained Derivatives of Anticonvulsant and Convulsant R,R-
Disubstituted γ-Butyrolactones. J . Med. Chem. 1994, 37, 275-
286.
(30) Litchfield, J . R., J r.; Wilcoxon, R. A. Simplified Method of
Evaluating Dose-Effect Experiments. J . Pharmacol. 1949, 96,
99-113.
(31) Tallarida, R. J .; Murray, R. B. Manual of Pharmacologic
Calculations with Computer Programs; Springer-Verlag: New
York, 1987.
(16) Elison, C.; Lien, E. J .; Zinger, A. P.; Hussain, M.; Tong, G. L.;
Golden, M. CNS Activities of Lactam Derivatives. J . Pharm. Sci.
1971, 60, 1058-1062.
J M960561U