The anticonvulsant activities of N-benzyl 3-methoxypropionamides

Article abstract of DOI:10.1016/S0968-0896(99)00186-8

We recently reported that the ED₅₀ value for (R,S)-2,3- dimethoxypropionamide (1) in the maximal electroshock (MES)-induced seizure test in mice was 30 mg/kg (Choi, D.; Stables, J.P., Kohn, H. Bioorg. Med. Chem. 1996, 4, 2105). This value is comparable to that observed for phenobarbital (ED₅₀ = 22 mg/kg). Compound 1 is structurally similar to a class of MES-selective anticonvulsant agents, termed functionalized amino acids (2), that were developed in our laboratory. The distinguishing feature of 2 is the differential activities observed for enantiomers. In this study, we asked whether comparable differences in activities were observed in the MES-induced seizure test for (R)- and (S)-1. We developed stereospecific syntheses for these enantiomers and showed that both compounds exhibit nearly equal anticonvulsant activity in mice (ip) (MES ED₅₀ = 79-111 mg/kg). The surprisingly high ED₅₀ values for (R)- and (S)-1 required our redetermining the ED₅₀ value for (R,S)-1. We revised this value to 79 mg/kg. A limited structure-activity relationship study for 1 was conducted. Special attention was given to the C(2) methoxy unit in 1. We found that replacement of this moiety led to only modest differences in the MES activities upon ip administration to mice. Significantly, we observed an enhancement in the anticonvulsant activity for (R,S)-N-benzyl 2-hydroxy-3-methoxypropionamide ((R,S)-6) upon oral administration to rats ((R,S)-6: mice (ip) ED₅₀ > 100, < 300 mg/kg; rat (oral) ED₅₀ = 62 mg/kg). The activities of 3- methoxypropionamides, functionalized amino acids, and related compounds are discussed. (C) 1999 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(99)00186-8

Bioorganic & Medicinal Chemistry 7 (1999) 2381±2389
The Anticonvulsant Activities of N-Benzyl
3-Methoxypropionamides
Shridhar V. Andurkar, a James P. Stables b and Harold Kohn a,*
aDepartment of Chemistry, University of Houston, Houston, TX 77204-5641, USA
bEpilepsy Branch, National Institute of Neurological Disorders and Stroke, National Institutes of Health,
Federal Building, Room 114, Bethesda, MD 20892-9020, USA
Received 3 December 1998; accepted 17 May 1999
AbstractÐWe recently reported that the ED50 value for (R,S)-2,3-dimethoxypropionamide (1) in the maximal electroshock (MES)-
induced seizure test in mice was 30 mg/kg (Choi, D.; Stables, J.P., Kohn, H. Bioorg. Med. Chem. 1996, 4, 2105). This value is
comparable to that observed for phenobarbital (ED50=22 mg/kg). Compound 1 is structurally similar to a class of MES-selective
anticonvulsant agents, termed functionalized amino acids (2), that were developed in our laboratory. The distinguishing feature of 2
is the di€erential activities observed for enantiomers. In this study, we asked whether comparable di€erences in activities were
observed in the MES-induced seizure test for (R)- and (S)-1. We developed stereospeci®c syntheses for these enantiomers and
showed that both compounds exhibit nearly equal anticonvulsant activity in mice (ip) (MES ED50=79±111 mg/kg). The surpris-
ingly high ED50 values for (R)- and (S)-1 required our redetermining the ED50 value for (R,S)-1. We revised this value to 79 mg/kg.
A limited structure±activity relationship study for 1 was conducted. Special attention was given to the C(2) methoxy unit in 1. We
found that replacement of this moiety led to only modest di€erences in the MES activities upon ip administration to mice. Sig-
ni®cantly, we observed an enhancement in the anticonvulsant activity for (R,S)-N-benzyl 2-hydroxy-3-methoxypropionamide
((R,S)-6) upon oral administration to rats ((R,S)-6: mice (ip) ED50 > 100, < 300 mg/kg; rat (oral) ED50=62 mg/kg). The activities
of 3-methoxypropionamides, functionalized amino acids, and related compounds are discussed. # 1999 Elsevier Science Ltd. All
rights reserved.
Introduction
(R)-stereoisomer exceeded the (S)-enantiomer by 10- to
22-fold.4,5,7,8 This di€erential activity is among the
highest ever reported for stereoisomers in the MES-
induced seizure test.
We reported that (R,S)-N-benzyl 2,3-dimethoxy-
propionamide (1) prevented maximal electroshock
(MES)-induced seizures in mice (ED50=30 mg/kg)
when administered intraperitoneally (ip)1 at levels com-
parable to phenobarbital (ED50=22 mg/kg).2 Com-
pound 1 is structurally similar to a recently discovered
class of anticonvulsant agents, termed functionalized
amino acids (2),3,4 which includes N-benzyl-2-acetami-
do-3-methoxypropionamide (3).5 A detailed structure±
activity relationship study for 2 revealed that placement
of a substituted heteroatom one atom removed from the
chiral C(2) atom led to enhanced anticonvulsant activ-
ity.5,6 Moreover, we showed that the principal activity
for 2 resided in the (R)-enantiomer. Comparison of the
anticonvulsant activities in the MES test for four di€er-
ent pairs of enantiomers revealed that the potency of the
The mode of action of 2 has not been determined. In the
absence of a pharmacological basis for its activity, we
asked whether the (R)- and (S)-enantiomers of 1 exhibited
di€erential anticonvulsant activity in the MES-induced
Key words: Anticonvulsants; amino acids and derivatives; peptoids;
substituents e€ects.
* Corresponding author. Tel.: +1-919-966-2680; fax: +1-919-966-
6919; e-mail: harold_kohn@unc.edu
0968-0896/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(99)00186-8

2382
S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
seizure test. In this paper, we report stereospeci®c
syntheses for (R)- and (S)-N-benzyl 2,3-dimethoxy-
propionamide (1), demonstrate that both stereoisomers
have comparable activity in the MES-induced seizure
test, revise the MES e€ective dose (ED50) value for 1,
and delineate the structural components in 1 necessary
for anticonvulsant activity. Our ®ndings suggest that
while 1 and 3 share common structural motifs their
mode of anticonvulsant activity may be di€erent.
and (S)-1. Accordingly, we coupled (R,S)-7 with (±)-
menthoxyacetic acid and (S)-phenylbutyric acid using
1,10-carbonyldiimidazole to give diastereomeric esters 9
and 10, respectively. Attempted separations by PTLC
using a variety of solvents and solvent mixtures were
unsuccessful.
Results and Discussion
Selection of test compounds
We wished to determine the anticonvulsant activities for
the two stereoisomers of N-benzyl 2,3-dimethoxy-
propionamide (1) and learn the structural elements in 1
necessary for anticonvulsant activity. Accordingly, an
abbreviated set of compounds was prepared and eval-
uated; they included (R)-1 and (S)-1. Added to our list
were 4 and 5. These compounds, like 1, contain sub-
stituted oxygen groups one atom removed from the
We then developed stereospeci®c syntheses of (R)- and
(S)-1 (Scheme 1). Commercially available (S)-(+)-2,2-
dimethyl-1,3-dioxolane-4-methanol ((S)-11) was oxi-
dized to (R)-129,10 with KMnO4, and then coupled with
benzylamine using 1,10-carbonyldiimidazole to provide
amide (R)-4. Deprotection of the acetonide group in
(R)-4 with aqueous acetic acid followed by methylation
with MeI and Ag2O gave (R)-1. An identical procedure
was utilized to prepare (S)-1 beginning with commer-
cially available (R)-11. The enantiopurity of (R)- and
(S)-1 was veri®ed by their optical rotation and the
demonstration that adding the NMR chemical resolving
agent11±13 (R)-mandelic acid to a CDCl3 solution con-
taining either (R)- or (S)-1 led to a single peak for the
terminal methoxy group ((R)-1: d 3.38; (S)-1: d 3.39). In
agreement with this ®nding, we observed that adding
(R)-mandelic acid to (R,S)-1 gave two distinct signals of
chiral C(2) atom. The pharmacological data for 1, 4 and
5 were compared with the monohydroxy-6 and dihy-
droxy-71 adducts. Finally, we added 8, where we
replaced the C(2)-methoxy group in 1 with a methyl
substituent. Our synthesis of 8 permitted us to prepare
both the (R)- and the (S)-enantiomers.
Synthesis
Our initial objective was to prepare the individual ste-
reoisomers of (R,S)-N-benzyl 2,3-dimethoxypropion-
amide ((R,S)-1) and to determine their anticonvulsant
activities. Accordingly, we attempted ®rst to resolve
(R,S)-7 by chemical derivatization with a chiral acid. It
served as the penultimate precursor of (R,S)-1 in our
original synthesis.1 We expected that chromatographic
separation of the diastereomeric mixture followed by
ester hydrolysis and methylation would provide (R)-
Scheme 1. Stereospeci®c synthesis of (R)-1.

S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
2383
equal height for the terminal methoxy group. Compound
(R,S)-4 was prepared using the procedure outlined in
Scheme 1, beginning with (R,S)-11. Syntheses of 5 and 6
were straight forward and proceeded from (R,S)-7. Addi-
tion of mesyl chloride (1 equiv) to a dichloromethane
solution of (R,S)-7 and triethylamine produced the C(3)-
monomesylate 13 and the C(2), C(3)-dimesylate 14,
respectively, in a ꢀ6:1 ratio (Scheme 2). The mesylates
were separated by chromatography, and 13 was con-
verted to epoxide (R,S)-5 with NaH. Methanolysis of 5
in acid led to a regiospeci®c opening of the epoxide ring
to give (R,S)-6. Preparation of (R)-8 proceeded from
commercially available (R)-15 (Scheme 3). Treatment of
(R)-15 with benzylamine gave amide (R)-16 without
racemization, and conversion of (R)-16 to ether (R)-8
was accomplished with methyl iodide and Ag2O in ace-
tonitrile. The same methodology was used to prepare
(S)-8, starting with (S)-15.
1 includes information concerning neurological impair-
ment (TD50) for these compounds in the rotorod test16
and, where appropriate, the protective index
(PI=TD50/ED50) for these adducts.
We ®rst determined the anticonvulsant activity for the
two enantiomers of (R,S)-1 in the MES-induced seizure
test. The compounds displayed comparable activity.
The ED50 value for (R)-1 was 79 mg/kg while (S)-1
exhibited slightly less activity, ED50=111 mg/kg. The
unexpectedly high ED50 values obtained for the stereo-
isomers of 1 were inconsistent with the original mea-
surement reported for the racemate (MES ED50=30
mg/kg).1 In subsequent experiments, we attempted to
reproduce the original ®nding for (R,S±1) utilizing
newly synthesized compound at the previously deter-
mined ED50 value of 30 mg/kg. Eight animals were
predosed with (R,S)-1 at the previously determined time
of peak e€ect of 0.25 h as well as at 0.5 h. All other
parameters were consistent with the original experi-
ment. No protection was a€orded against MES induced
seizures at either time point. Accordingly, we quantita-
tively re-evaluated (R,S)-1 determining a new MES
ED50 value of 79 mg/kg. The diminished protection
against MES-induced seizures con®rmed that the origi-
nal measurement could not be reproduced. The factors
that contributed to the inaccurate ED50 value for (R,S)-
11 could not fully be determined. The testing para-
meters17 for the pharmacological studies (e.g. time of
test, animal species, weight, sex) were kept constant so it
is unlikely that these contributed to the experimental
error. All the test animals had free access to food and
water except when used for experiments. Therefore,
pharmacokinetic interactions resulting from the animal
feed should have had little e€ect on the absorption rates
of (R,S)-1. Likewise, structural con®rmation of the test
compounds ((R)-1, (S)-1, (R,S)-1) were veri®ed by TLC
and optical measurements. We suspect the most likely
explanation for the variance in data is due to physical
characteristics of (R,S)-1 itself. This compound is an oil.
Oils are often dicult to get into either uniform sus-
pension or solution. This factor can lead to appreciable
di€erences in the actual concentration of compound
dilutions being prepared for administration. In the later
experiments special care was taken in preparation of all
suspensions. It is likely, administration of a more con-
centrated suspension was responsible for the originally
reported MES ED50 for (R,S)-1.1 Finally, we note that
the experimental slopes in the calculation of the ED50
value for (R,S)-1, (R)-1, and (S)-1 and the con®dence
limits are large, leading us to view the ED50 values in
Table 1 with additional caution. We concluded that (R)-
and (S)-1 are only moderate anticonvulsants with com-
parable activity. These experimental concerns do not
exist for functionalized amino acid 3 or for other com-
pounds similar to 1 in structure (e.g. 7, 17). Compounds
3 and 7 are solids, while 17 although an oil is a primary
amine. All these compounds readily dissolved in the
testing vehicle.
Pharmacological evaluation
Table 1 lists the anticonvulsant activities in the MES
(ip) and the subcutaneous (sc) Metrazol seizures tests
conducted in mice for compounds (R,S)-1, (R)-1, (S)-1,
(R,S)-4, (R,S)-5, (R,S)-6, (R)-8, and (S)-8 using the
procedure described by Stables and Kupferberg.14 The
corresponding values for (R)-3, (S)-3, (R,S)-3,5 (R,S)-71
and the clinically proven antiepileptic agents phenytoin,15
phenobarbital.2 and valproate15 are also provided. Table
Scheme 2. Synthesis of (R,S)-5 and (R,S)-6.
Several of the remaining compounds evaluated also dis-
played similar anticonvulsant activity in the MES-
induced seizure test. Although actual ED50 values were
Scheme 3. Stereospeci®c synthesis of (R)-8.

2384
S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
Table 1. Pharmacological data in micea
Compound
mpb (ꢁC)
Ðg
MESc ED50
scMetd ED50
Toxe TD50
P.I.f TD50/MES ED50
3.5
(R)-1
79 [0.5]
(73±88)
111 [1]
(86±129)
79 [0.25]
(67±114)
4.5 [0.5]
(3.7±5.5)
>100, <300
[0.5]
8.3 [0.5]
(7.9±9.8)
>30, <100
[0.25]
>100, <300
[0.5]
>100, <300
[0.5]
>100, <300
[0.5]
>30, <100
[0.25]
81 [0.25]
(60±96)
9.5 [2]
180 [0.5]
(125±241)
150 [0.25]
(114±189)
145 [0.25]
(113±187)
>25 [0.25]
279 [0.25]
(257±293)
308 [0.25]
(279±348)
280 [0.25]
(253±308)
27 [0.25]
(S)-1
Ðg
Ðg
2.8
3.5
6.0
Ð
(R,S)-1
(R)-3h
143±144
143±144
121±122
98±100
61±63
Ðg
(26±28)
>300 [0.5]
(S)-3h
>300 [0.5]
>300 [0.5]
(R,S)-3h
(R,S)-4
(R,S)-5
(R,S)-6
(R,S)-7i
(R)-8
43 [0.25]
(38±47)
>100, <300
[0.5]
>100, <300
[0.5]
>100, <300
[0.5]
>300
[0.5]
>300
[0.5]
221 [0.25]
(183±259)
66 [2]
(53±72)
69 [0.5]
(63±73)
426 [0.25]
(369±450)
5.2
Ð
>300
[0.25]
>300
[0.5]
>300
[0.5]
Ð
Ð
83±84
Ðg
>100, <300
[0.5]
Ð
>300
[0.5]
>300
[0.5]
ꢂNPk
Ð
(S)-8
Ðg
2.7
6.9
3.1
1.6
Phenytoinj
Phenobarbitall
Valproatej
(8.1±10.4)
22 [1]
(15±23)
272 [0.25]
(247±338)
13
(5.9±16)
149 [0.25]
(123±177)
a
The compounds were administered intraperitoneally unless otherwise indicated. ED50 and TD50 values are in mg/kg. Numbers in parentheses are
95% con®dence intervals. A dose±response curve was generated for all compounds that displayed sucient activity. The dose e€ect data for these
compounds was obtained at the ``time of peak e€ect'' (indicated in hours in the brackets). The compounds were tested through the auspices of the
National Institute of Neurological and Communicative Disorders and Stroke at the National Institute of Health unless otherwise indicated.
Melting points (ꢁC) are uncorrected.
b
c
MES=maximal electroshock seizure text.
scMet=subcutaneous pentylenetetrazole (metrazole) seizure test.
Neurologic toxicity determined using the rotorod test.
PI=protective index (TD50/MES ED50).
Oil.
Ref 5.
Ref 1.
Ref 15.
d
e
f
g
h
i
j
k
NP=no protection.
Data from Antiepileptic Drugs, 2nd ed.; Woodbury, D. M.; Penry, J. K.; Pippenger, C. E.; Eds.; Raven Press: New York, 1982; p 120.
l
not calculated for (R,S)-4-7, (R)-8 and (S)-8, qualitative
evaluations using two time points and three di€erent
doses (30, 100 and 300 mg/kg) provide us with some
estimates for the protective ranges of observed anti-
convulsant activity. Recognizing this limitation we
observed that acetonide (R,S)-4 exhibited a range of
activity comparable to (R,S)-1 and that substituting the
2-methoxy group in 1 with a methyl unit to provide 8
gave no appreciable losses in MES-seizure protection.
Signi®cantly, (R)- and (S)-8 possessed comparable
activities that paralleled our ®nding for (R)- and (S)-1.
Finally, epoxide (R,S)-5 and alcohol (R,S)-6 exhibited
decreased activity compared with (R,S)-1. The e€ect of
the C(2) substituent on the MES anticonvulsant activity
for N-benzyl 3-methoxypropionamides is highlighted in
Table 2. We have included values previously determined
for C(2) acetamido ((R,S)-3) and C(2) amino ((R)-1718)
3-methoxypropionamides. Also listed in Table 2 are the
MES ED50 values for select compounds after oral
administration to rats. In the mice (ip) test data we
observed that most of the C(2) substituted N-benzyl 3-
methoxypropionamides displayed fairly uniform activity
with only the 2-acetamido derivative (R,S)-35 a€ording
excellent protection against MES-induced seizures.
Comparable levels of activity were observed regardless
of the electronic nature of the C(2) substituent and the
ability of this moiety to undergo hydrogen bond donor/
acceptor interactions. Enhancement in the MES anti-
convulsant activity was observed for (R,S)-6 upon oral
administration to rats (ED50>100, <300 mg/kg, mice,
ip; ED50=62 mg/kg, rat, po). Furthermore, (R,S)-6
provided increased levels of protection against MES-
induced seizures with time over the 4 h test period.
These ®ndings parallel those observed for (R)-1718 and
indicate that (R,S)-6 is a moderately potent anti-
convulsant with an extended duration of action.
Table 1 includes the ED50 values obtained for the
test compounds in the Metrazol (sc) seizure assay.
We observed that both (R)- and (S)-1 exhibited

S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
2385
Table 2. Comparative pharmacological data for 2-substituted 3-methoxypropionamides in mice (ip) and rats (po)
mice (ip)a
rat (po)d
Compound
X
MESb ED50
Toxc TD50
MES ED50
>30 [1]
Toxc TD50
>30 [1]
(R,S)-1
OCH3
79 [0.25]
(67±114)
>100, <300
[0.5]
>30, <100
[0.5]
>30, <100
[0.25]
81 [0.25]
(60±96)
280 [0.25]
(253±308)
>100, <300
[0.5]
>100, <300
[0.5]
>300
(R,S)-7
(R)-17
(R)-8
OH
NH2
62 [4]
(43±88)
18
[4]
>30
[4]
>30
[4]
3.8 [2]
(2.9±5.5)
>200 [4]
500 [4]
Ðe
CH3
[0.5]
(S)-8
CH3
221 [0.25]
(183±259)
43 [0.25]
(38±47)
Ðe
(R,S)-3f
N(H)C(O)CH3
8.3 [0.5]
(7.4±9.8)
390 [1]
(320±510)
a
The compounds were administered intraperitoneally. ED50 and TD50 values are in mg/kg. Numbers in parentheses are 95% con®dence intervals.
A dose±response curve was generated for all compounds that displayed sucient activity. The dose e€ect data for these compounds was obtained at
the ``time of peak e€ect'' (indicated in hours in the brackets). The compounds were tested through the auspices of the National Institute of Neuro-
logical and Communicative Disorders and Stroke at the National Institute of Health unless otherwise indicated.
MES=maximal electroshock seizure test.
Tox=neurologic toxicity determined from rotorod test.
The compounds were administered orally.
Not determined.
Ref 5.
b
c
d
e
f
anticonvulsant activity (ED50=150±180 mg/kg) com-
parable to valproate (ED50=149 mg/kg). By compar-
ison, the structurally similar amino acid derivative
(R,S)-3 was devoid of activity in this assay at con-
centrations below 300 mg/kg.5 None of the remaining
compounds ((R,S)-4, (R,S)-5, (R,S)-6, (R)-8, (S)-8) in
the current study showed detectable anticonvulsant
activity in the Metrazol (sc) seizure assay at the highest
concentration employed (300 mg/kg).
Experimental
General methods
Melting points were determined with a Thomas±Hoover
melting point apparatus and are uncorrected. Infrared
spectra (IR) were run on an ATI Mattson Genesis Ser-
ies FTIRTM spectrometer. Absorption values are
expressed in wave-numbers (cm 1). Proton (1H NMR)
and carbon (13C NMR) nuclear magnetic resonance
spectra were taken on a General Electric QE-300 NMR
instrument. Chemical shifts (d) are in parts per million
(ppm) relative to tetramethylsilane, and coupling con-
stants (J values) are in Hertz. Low resolution mass
spectra (CI+) were obtained with a Varian MAT CH-5
spectrometer by Dr. M. Moini at the University of Texas±
Austin. The high-resolution chemical ionization mass
spectrum was performed on a Finnigan MAT TSQ-70 by
Dr. M. Moini at the University of Texas±Austin. Optical
rotations were recorded on Perkin±Elmer 241 MC and
Jasco P-1030 polarimeters. Microanalyses were provided
by Atlantic Microlab, Inc. (Norcross, GA). Thin-layer
chromatography was performed on precoated silica gel
GHLF microscope slides (2.5Â10 cm; Analtech No.
21521).
Conclusions
The overall pharmacological pro®le for 1 di€ers sig-
ni®cantly from functionalized amino acid 3 in two
important ways. First, no appreciable di€erences in
anticonvulsant activities were found for the (R)- and
(S)- enantiomers of 1. The ED50 values for (R)- and (S)-
1 were within a factor of two and were noticeably less
than the 22-fold di€erence in activity observed for the
enantiomers of 3.5 Second, both (R)- and (S)-1 exhibited
moderate levels of anticonvulsant activity in the Metra-
zol (sc) seizure test in mice (ED50=150±180 mg/kg)
while neither (R,S)- nor (S)-3 (ED50>300 mg/kg) did.5
The anticonvulsant activity for (R)-3 in the Metrazol
(sc) seizure test was not tested above 25 mg/kg;5 how-
ever, in corresponding tests conducted in rats the ED50
values in the Metrazol (sc) test for (R)-3 and (R,S)-3
exceeded 250 mg/kg (data not shown). These results
taken together suggest that 1 and 3 may express their
anticonvulsant activities by di€erent mechanisms and
that 3-methoxypropionamides may represent a phar-
macologically distinct class of anticonvulsants. Future
research will focus on understanding the oral activity of
(R,S)-6 and its amino counterpart (R)-17.
Synthesis of N-benzyl 2-hydroxy-3-(2-menthoxyacetylox-
y)propionamide (9). A THF solution (1 mL) of (±)-men-
thoxyacetic acid (0.06 g, 0.3 mmol) and 1,10-carbonyl-
diimidazole (0.04 g, 0.3 mmol) was heated at re¯ux (45
min), allowedto cool toroom temperature and thena THF
solution (0.5 mL) of (R,S)-7 (0.05 g, 0.3 mmol) was added
and the solution was heated at re¯ux (4 h). The solvent was
evaporated in vacuo and the residue puri®ed by pre-
parative TLC (SiO2, 10% MeOH±CHCl3) to a€ord the
oily ester (0.06 g, 54%) as an ꢀ1:1 mixture of diaster-

2386
S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
eomers: Rf 0.75 (10% MeOH±CHCl3);19 IR (liquid
®lm) 3365, 2954, 2869, 1760, 1651, 1532, 1455, 1122, 699
cm
with water (2Â25 mL), and the organic layer was dried
(Na2SO4) and evaporated in vacuo. The residue was pur-
i®ed by column chromatography (SiO2, 5% MeOH±
CHCl3) followed by recrystallization from ethyl ether-
petrolꢁeum ether to a€ord pure (R)-4 (1.22 g, 73%): mp
81±84 C; [a]2d3 +17.1ꢁ (c 0.1, MeOH); Rf 0.70 (5%
MeOH±CHCl3); IR (KBr) 3336, 1649, 1540, 1221, 1090,
735, 500 cm 1; 1H NMR (CDCl3) d 1.39 (s, CCH3), 1.45
(s, CCH3), 4.16 (dd, J=5.4, 9.0 Hz, OCHH0), 4.32 (dd,
J=7.4, 9.0 Hz, OCHH0), 4.49 (d, J=6.0 Hz, CH2Ph), 4.55
(dd, J=5.4, 7.4 Hz, OCH), 6.80±6.95 (m, NH), 7.23±7.41
(m, 5 PhH); 13C NMR (CDCl3) 25.1 (CCH3), 26.3
(CCH3), 43.0 (CH2Ph), 67.9 (CH2), 75.2 (CH), 111.0
(C(CH3)2), 127.7, 128.9, 137.9 (Ph), 171.3 (C(O)) ppm, the
remaining aromatic signal was not detected and is believed
to overlap with nearby peaks; MS (CI+) (rel. int.) 236
(M++1, 100), 208 (72), 178 (43); Mr (+ CI) 236.128 99
[M++1] (calcd for C13H18NO3 236.128 67); Anal. calcd
1
;
1H NMR (CDCl3) d 0.77 (d, J=7.2 Hz,
C(50)CH3, diastereomer a or b), 0.78 (d, J=6.9 Hz,
C(50)CH3 diastereomer a or b), 0.80±1.05 (m, 9 H),
1.20±1.40 (m, 2 H), 1.60±1.67 (m, 2 H), 1.95±2.05 (m, 1
H), 2.15±2.30 (m, 1 H), 3.05±3.20 (m, 1 H), 4.10 (ABq,
J=16.2 Hz, OCH2C(O)O diastereomer a or b), 4.11
(ABq, J=16.2 Hz, OCH2C(O)O diastereomer a or b),
4.35±4.65 (m, 6 H), 7.15±7.25 (m, 5 PhH and NH); 13C
NMR (CDCl3) 16.2, 20.9, 22.2, 23.2, 25.5, 31.4, 34.3,
39.8 (2-menthoxyacetyloxy, diastereomer a and b), 43.1
(CH2Ph), 47.9 (2-menthoxyacetyloxy, diastereomer a and
b), 65.7 (CHOCH2 diastereomer a or b), 65.8 (CH-OCH2
diastereomer a or b), 66.9 (C(O)OCH2), 71.0 (OCH2-
C(O)), 80.4 (CHOH diastereomer a or b), 80.5 (CHOH
diastereomer a or b), 127.5, 127.6, 128.7, 137.7 (Ph), 170.2
(C(O)O or C(O)NH), 171.7 (C(O)O or C(O)NH) ppm;
MS, (CI+) (rel. int.) 392 (M++1, 92), 374 (13), 255 (13),
254 (100), 236 (12), 196 (41); Mr (+ CI) 392.242 91
[M++1] (calcd for C22H34NO5 392.243 69).
.
for C13H17NO3 0.5 H2O: C, 63.93; H, 7.38; N, 5.74.
Found: C, 64.23; H, 7.14; N, 5.69.
Synthesis of (S)-N-benzyl 2,2-dimethyl-1,3-dioxolane-4-
carboxamide ((S)-4). Compound (S)-4 was prepared
using the preceding procedure and beginning with (S)-
129 (2.86 g, 15.5 mmol), tri¯uoroacetic acid (1.77 g, 15.5
mmol), 1,10-carbonyldiimidazole (2.60 g, 16 mmol) and
benzylamine (1.81 mL, 16.5 mmol): yield, 3.00 g (82%);
mp 84±87ꢁC; [a]d23 17.2ꢁ (c 0.1, MeOH); Rf 0.70 (5%
MeOH±CHCl3); IR (KBr) 3336, 1649, 1540, 1220, 1090,
735, 502 cm 1; 1H NMR (CDCl3) d 1.39 (s, CCH3), 1.45
(s, CCH3), 4.16 (dd, J=5.4, 9.0 Hz, OCHH0), 4.32 (dd,
J=7.6, 9.0 Hz, OCHH0), 4.49 (d, J=6.0 Hz, CH2Ph),
4.55 (dd, J=5.4, 7.6 Hz, OCH), 6.80±6.95 (m, NH),
7.20±7.41 (m, 5 PhH); 13C NMR (CDCl3) 25.0 (CCH3),
26.2 (CCH3), 42.9 (CH2Ph), 67.8 (CH2), 75.1 (CH),
110.9 (C(CH3)2), 127.6, 128.8, 137.9 (Ph), 171.2 (C(O))
ppm, the remaining aromatic signal was not detected
and is believed to overlap with nearby peaks; MS
(CI+) (rel. int.) 236 (M++1, 100), 208 (7), 178 (20); Mr
(+ CI) [M++1] 236.128 44 (calcd for C13H18NO3
Synthesis of N-benzyl 2-hydroxy-3-(2-(S)-phenylbutyryl-
oxy)propionamide (10). A THF solution (2 mL) of (S)-
phenylbutyric acid (0.05 g, 0.3 mmol) and 1,10-carbo-
nyldiimidazole (0.05 g, 0.3 mmol) was heated at re¯ux (1
h), allowed to cool to room temperature and then a THF
solution (1.0 mL) of (R,S)-7 (0.05 g, 0.3 mmol) was added
and the solution heated at re¯ux (3 h). The solvent was
evaporated in vacuo and the residue puri®ed by pre-
parative TLC (SiO2, 10% MeOH±CHCl3) to a€ord an
oily ester (0.05 g, 54%) as an ꢀ1:1 mixture of diaster-
eomers: Rf 0.62 (10% MeOH±CHCl3);19 IR (liquid ®lm)
1
3386, 2966, 2876, 1738, 1652, 1538, 1455, 1124, 699 cm
;
1H NMR (CDCl3) d 0.86 (t, J=7.2 Hz, CH2CH3 dia-
stereomer a or b), 0.88 (t, J=7.2 Hz, CH2CH3 diastereo-
mer a or b), 1.70±1.86 (m, CH3CHH0), 2.00±2.20 (m, CH3
CHH0), 3.47 (t, J=7.5 Hz, CH2CHPh diastereomer a or
b), 3.48 (t, J=7.5 Hz, CH2CHPh diastereomer a or b),
4.25±4.60 (m, CH2CH(OH), CH2Ph), 6.85±7.10 (m, NH),
7.15±7.40 (m, 10 PhH); 13C NMR (CDCl3) 12.0 (CH3
CH2), 26.3 (CH3CH2 diastereomer a or b), 26.4 (CH3CH2
diastereomer a or b), 43.0 (CH2Ph diastereomer a or b),
43.1 (CH2Ph diastereomer a or b), 53.2 (CH2CHPh), 66.5
(C(O)OCH2 diastereomer a or b), 66.7 (C(O)OCH2 dia-
stereomer a or b), 71.2 (CHOH diastereomer a or b), 71.3
(CHOH diastereomer a or b), 127.4, 127.5, 127.6, 127.7,
127.8, 128.6, 137.6, 138.4, 138.6 (Ph diastereomer a and b),
170.2 (C(O)NH), 175.1 (C(O)O) ppm; MS, (CI+) (rel.
int.) 342 (M++1, 100), 196 (18), 178 (15), 165 (11), 162
(13); Mr (+ CI) 342.170 10 [M++1] (calcd for
C20H24NO4 342.170 53).
.
236.128 67); Anal. calcd for C13H17NO3 0.25 H2O: C,
65.14; H, 7.31; N, 5.85. Found: C, 65.33; H, 7.32; N,
5.81.
Synthesis of (R)-N-benzyl 2,3-dihydroxypropionamide ((R)
-7). A 50% aqueous acetic acid solution (30 mL) con-
taining (R)-4 (1.22 g, 5.19 mmol) was heated at re¯ux
(30 min) and then concentrated in vacuo. The residue
was puri®ed by column chromatography (SiO2, 10%
MeOH±CHCl3) to obtain (R)-7 as a white solid (0.86 g,
85%); mp 83±84ꢁC; [a]d23 +35.4ꢁ (c 0.2, MeOH); Rf 0.35
(10% MeOH±CHCl3); IR (KBr) 3336, 1649, 1623, 1542,
1110, 1049, 972, 739, 697 cm 1; 1H NMR (DMSO-d6) d
3.42±3.56 (m, CHH0OH), 3.58±3.64 (m, CHH0OH),
3.89±3.95 (m, CHOH), 4.28 (d, J=6.0 Hz, CH2Ph), 4.74
(t, J=5.7 Hz, CH2OH), 5.57 (d, J=5.4 Hz, CHOH),
7.19±7.35 (m, 5 PhH), 8.24 (t, J=6.0 Hz, NH); 13C
NMR (DMSO-d6) 41.7 (CH2Ph), 64.0 (CH2OH), 73.1
(CHOH), 126.6, 127.2, 128.2, 139.6 (Ph), 172.2
(C(O)NH) ppm; MS (CI+) (rel. int.) 196 (M++1, 100);
Mr (+ CI) 196.098 03 [M++1] (calcd for C10H14NO3
196.097 37); Anal. calcd for C10H13NO3: C, 61.54; H,
6.67; N, 7.13. Found: C, 61.45; H, 6.69; N, 7.13.
Synthesis of (R)-N-benzyl 2,2-dimethyl-1,3-dioxolane-4-
carboxamide ((R)-4). To an N2-blanketed suspension of
(R)-129 (1.31 g, 7.1 mmol) in dry CH2Cl2 (50 mL) was
added tri¯uoroacetic acid (0.81 g, 7.1 mmol). The mix-
ture was stirred at room temperature (30 min) and then
1,10-carbonyldiimidazole (1.15 g, 7.1 mmol) was added
and the mixture was heated at re¯ux until CO2 evolu-
tion ceased (ꢀ45 min). The mixture was cooled to room
temperature, treated with benzylamine (0.78 mL, 7.1
mmol) and then stirred (18 h). The mixture was washed

S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
2387
Synthesis of (S)-N-benzyl 2,3-dihydroxypropionamide ((S)-
7). The preceding procedure was employed to prepare
(S)-7 using (S)-4 (3.49 g, 14.8 mmol) and 50% aqueous
Synthesis of (R,S)-N-benzyl 2,2-dimethyl-1,3-dioxolane-
4-carboxamide ((R,S)-4). Employing the procedure for
(R)-4 and using (R,S)-129,10 (2.63 g, 19 mmol), tri-
¯uoroacetic acid (1.1 mL, 19 mmol), 1,10-carbonyldi-
imidazole (2.4 g, 21 mmol), and benzylamine (1.6 mL,
19 mmol), compound (R,S)-4 was produced in 78% yield
(2.83 g): mp 83±85ꢁC; Rf 0.70 (5% MeOH±CHCl3); IR
acetic acid (86 mL): yield, 2.26 g (78%); mp 83±84ꢁC;
23
d
‰aŠ
35.1ꢁ (c 0.1, MeOH); Rf 0.35 (10% MeOH±
CHCl3); IR (KBr) 3337, 1649, 1623, 1546, 1109, 1049,
971, 739, 697 cm 1; H NMR (DMSO-d6) d 3.45±3.59
1
(m, CHH0OH), 3.60±3.66 (m, CHH0OH), 3.90±3.98 (m,
CHOH), 4.29 (d, J=6.2 Hz, CH2Ph), 4.65±4.85 (m,
OH), 5.40±5.65 (m, OH), 7.15±7.40 (m, 5 PhH), 8.25 (t,
J=6.2 Hz, NH); 13C NMR (DMSO-d6) 41.8 (CH2Ph),
64.1 (CH2OH), 73.2 (CHOH), 126.7, 127.2, 128.3, 139.7
(Ph), 172.4 (C(O)NH) ppm; MS (CI+) 196 (M++1);
Mr (+ CI) 196.097 09 [M++1] (calcd for C10H14NO3
196.097 37). Anal. calcd for C10H13NO3: C, 61.54; H,
6.67; N, 7.13. Found: C, 61.25; H, 6.59; N, 7.01.
(KBr) 3336, 1649, 1540, 1220, 1090, 735, 502 cm 1; H
1
NMR (CDCl3) d 1.39 (s, CCH3), 1.45 (s, CCH3), 4.16
(dd, J=5.4, 9.0 Hz, OCHH0), 4.32 (dd, J=7.6, 9.0 Hz,
OCHH0), 4.49 (d, J=6.0 Hz, CH2Ph), 4.55 (dd, J=5.4,
7.6 Hz, OCH), 6.80±6.95 (m, NH), 7.20±7.41 (m, 5 PhH);
13C NMR (CDCl3) 25.0 (CCH3), 26.2 (CCH3), 42.9
(CH2Ph), 67.8 (CH2), 75.1 (CH), 110.9 (C(CH3)2), 127.6,
128.8, 137.9 (Ph), 171.2 (C(O)) ppm, the remaining aro-
matic signal was not detected and is believed to overlap
with nearby peaks; MS (CI+) (rel. int.) 236 (M++1,
100), 178 (28); Mr (+CI) 236.129 30 [M++1] (calcd for
C13H18NO3 236.128 67); Anal. calcd for C13H17NO3: C,
66.38; H, 7.23; N, 5.97. Found: C, 66.33; H, 7.34; N,
5.97.
Synthesis of (R)-N-benzyl 2,3-dimethoxypropionamide ((R)-
1). An acetonitrile solution (44 mL) of (R)-7 (1.35 g,
6.9 mmol) was stirred at room temperature in the pre-
sence of Ag2O (8.05 g, 35 mmol) and MeI (4.4 mL, 70
mmol) (2 days). The salts were ®ltered, and the ®ltrate
was evaporated in vacuo to obtain an oily residue that
was puri®ed by column chromatography (SiO2, EtOAc)
to furnish (R)-1 (1.29 g, 83%) as a clear oil: [a]2d3 +33.6ꢁ
Synthesis of (R,S)-N-benzyl 2-hydroxy-3-methanesulfo-
natopropionamide ((R,S)-13). To a cold (0ꢁC), N2-blan-
keted solution of (R,S)-7 (4.52 g, 23.2 mmol) and
triethylamine (3.3 mL, 23 mmol) in dry tetrahydrofuran
(100 mL) was added dropwise methanesulfonyl chloride
(2.7 g, 23 mmol). The mixture was allowed to warm to
room temperature and then stirred (1 h). The solvent
was evaporated in vacuo, and the residue was puri®ed
by column chromatography (SiO2, ethyl ether) to obtain
pure (R,S)-13 (2.5 g, 40%) and (R,S)-N-benzyl-2,3-
dimethanesulfonatopropionamide (R,S)-14 (0.41 g,
5%).
(c 0.1, MeOH); Rf 0.43 (EtOAc); IR (liquid ®lm) 3320,
1
2930, 1662, 1528, 1455, 1107, 700 cm
;
1H NMR
(CDCl3) d 3.41 (s, CH2OCH3), 3.49 (s, CHOCH3), 3.71
(dd, J=4.5, 10.5 Hz, CHH0OCH3), 3.80 (dd, J=2.7,
10.5 Hz, CHH0OCH3), 3.88 (dd, J=2.7, 4.5 Hz,
CHOCH3), 4.51 (d, J=6.0 Hz, CH2Ph), 6.95±7.05 (m,
NH), 7.28±7.40 (m, 5 PhH), addition of excess (R)-
mandelic acid gave only one signal (d 3.38) for the
CH2OCH3 protons; 13C NMR (CDCl3) 43.0 (CH2Ph),
58.6 (CH2OCH3), 59.4 (CHOCH3), 72.3 (CH2OCH3),
81.8 (CHOCH3), 127.5, 127.6, 128.7, 138.0 (Ph), 170.0
(C(O)) ppm; MS (CI+) (rel. int.) 224 (M++1, 100),
119 (3); Mr (+ CI) 224.127 99 [M++1] (calcd for
C12H18NO3 224.128 67); Anal. calcd for C12H17
(R,S)-13: mp 139±141ꢁC; Rf 0.68 (EtOAc); IR (KBr)
3392, 3243, 1653, 1558, 1404, 1347, 1172, 1127, 989,
1
966, 721, 529 cm 1; H NMR (DMSO-d6) d 3.12 (s,
CH3), 4.22±4.40 (m, CH2Ph, CH2OMs, CHOH), 6.28
(d, J=5.4 Hz, CHOH), 7.17±7.35 (m, PhH), 8.54 (t,
J=6.2 Hz, NH); 13C NMR (DMSO-d6) 36.7 (CH3),
41.9 (CH2Ph), 69.7 (CH2OMs), 72.4 (CHOH), 126.7,
127.1, 128.2, 139.4 (Ph), 170.1 (C(O)) ppm; MS (CI+)
(rel. int.) 274 (M++1, 100), 178 (27); Mr (+CI) 274.074
90 [M++1] (calcd for C11H16NO5S 274.074 92); Anal.
calcd for C11H15NO5S ꢃ 0.25 H2O: C, 47.58; H, 5.59; N,
5.05. Found: C, 47.80; H, 5.62; N, 5.02.
.
NO3 0.15 H2O: C, 63.80; H, 7.67; N, 6.20. Found: C,
63.85; H, 7.53; N, 6.17.
Synthesis of (S)-N-benzyl 2,3-dimethoxypropionamide ((S)-
1). Using the preceding procedure, (S)-1 was prepared
in 70% yield (1.89 g) beginning with (S)-7 (2.26 g, 11.6
mmol), Ag2O (13.4 g, 58 mmol), MeI (7.4 mL, 116
mmol) and acetonitrile (74 mL): ‰aŠ
23
d
33.2ꢁ (c 0.1,
MeOH); Rf 0.43 (EtOAc); IR (liquid ®lm) 3324, 2931,
1667, 1528, 1455, 1108, 701 cm 1; H NMR (CDCl3) d
(R,S)-14: mp 104±105ꢁC; Rf 0.83 (EtOAc); IR (KBr)
3341, 3034, 2939, 1658, 1540, 1356, 1182, 1069, 959,
1
3.41 (s, CH2OCH3), 3.49 (s, CHOCH3), 3.72 (dd,
J=4.5, 10.5 Hz, CHH0OCH3), 3.80 (dd, J=2.7, 10.5
Hz, CHH0OCH3), 3.88 (dd, J=2.7, 4.5 Hz, CHOCH3),
4.51 (d, J=6.0 Hz, CH2Ph), 6.85±7.03 (m, NH), 7.25±
7.38 (m, 5 PhH), addition of excess (R)-mandelic acid
gave only one signal (d 3.39) for the CH2OCH3 protons;
13C NMR (CDCl3) 42.8 (CH2Ph), 58.5 (CH2OCH3),
59.2 (CHOCH3), 72.2 (CH2OCH3), 81.7 (CHOCH3),
127.3, 127.5, 128.5, 138.0 (Ph), 169.9 (C(O)NH) ppm;
MS (CI+) (rel. int.) 224 (M++1, 100), 222 (11), 191
(2); Mr (+ CI) 224.129 29 [M++1] (calcd for
C12H18NO3 224.128 67); Anal. calcd for C12H17NO3
ꢃ 0.25 H2O: C, 63.30; H, 7.69; N, 6.15. Found: C, 63.32;
H, 7.76; N, 6.19.
1
903, 787, 703, 529 cm 1; H NMR (CDCl3) d 3.00 (s,
CH3), 3.20 (s, CH3), 4.43 (dd, J=5.7, 15.3 Hz,
CHH0Ph), 4.57 (dd, J=6.6, 15.3 Hz, CHH0Ph), 4.64
(dd, J=5.1, 11.7 Hz, CHH0OMs), 4.70 (dd, J=3.0, 11.7
Hz, CHH0OMs), 5.32 (dd, J=3.0, 5.1 Hz, CHOMs),
6.94±7.00 (m, NH), 7.25±7.40 (m, PhH); 13C NMR
(CDCl3) 37.8 (CH3), 39.3 (CH3), 43.9 (CH2Ph), 68.6
(CH2OMs), 76.5 (CHOMs), 127.8, 128.1, 129.1, 137.1
(Ph), 164.7 (C(O)) ppm; MS (CI+) (rel. int.) 352
(M++1, 100), 257 (12), 256 (97), 178 (22); Mr (+CI)
352.052 00 [M++1] (calcd for C12H18NO7S2 352.052
47); Anal. calcd for C12H17NO7S2: C, 41.02; H, 4.84; N,
3.99. Found: C, 41.60; H, 4.80; N, 3.94.

2388
S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
1
Synthesis of (R,S)-N-benzyl 2,3-epoxypropionamide ((R,S)
-5). To a dry tetrahydrofuran solution (100 mL) of
mesylate (R,S)-13 (2.02 g, 7.4 mmol) was added NaH
(60% suspension in mineral oil, 0.30 g, 7.5 mmol) and
stirred at room temperature (1 h). The mixture was
poured into H2O (150 mL) and the ``pH'' adjusted to 7
(5% aq H2SO4) and then extracted with EtOAc (3Â75
mL). The organic extracts were combined, dried
(Na2SO4), ®ltered and evaporated in vacuo. The oily
residue obtained was puri®ed by column chromatography
(SiO2, EtOAc) to give epoxide (R,S)-5 as a white crystal-
line solid (1.25 g, 95%): mp 61±63ꢁC; Rf 0.80 (EtOAc); IR
(KBr) 3262, 3085, 2989, 2358, 1654, 1559, 1493, 1426,
1453, 1332, 1265, 1073, 1026, 739, 699 cm 1; H NMR
(CDCl3) d 1.19 (d, J=7.2 Hz, CH3), 2.10±2.40 (m, OH),
2.45±2.58 (m, CH), 3.74 (d, J=6.0 Hz, CH2OH), 4.46 (d,
J=5.4 Hz, CH2Ph), 6.10±6.30 (m, NH), 7.20±7.40 (m,
PhH); 13C NMR (DMSO-d6) 14.4 (CH3), 41.8 (CH2Ph),
42.8 (CH), 63.8 (CH2OH), 126.6, 127.0, 128.2, 139.7 (Ph),
174.4 (C(O)) ppm, the 13C NMR assignments were con-
sistent with the APT spectrum; MS (CI+) (rel. int.) 194
(M++1, 100); Mr (+CI) 194.118 13 [M++1] (calcd for
C11H16NO2 194.118 10); Anal. calcd for C11H15NO2: C,
68.40; H, 7.80; N, 7.25. Found: C, 68.27; H, 7.82; N, 7.15.
Synthesis of (S)-N-benzyl 3-hydroxy-2-methylpropion-
amide ((S)-16). Following the preceding procedure and
using (S)-15 (2.00 g, 17 mmol) and benzylamine (10
mL), (S)-16 was obtained as a white crystalline solid (1.82
g, 56%): mp 120±122ꢁC; [a]d24 +25.6ꢁ (c 1.3, MeOH); Rf
0.53 (EtOAc); IR (KBr) 3372, 3269, 3054, 1650, 1562,
1
1364, 1268, 1238, 1041, 888, 735, 696 cm 1; H NMR
(CDCl3) d 2.78 (dd, J=2.6, 5.8 Hz, CHH0O), 3.00 (dd,
J=4.8, 5.8 Hz, CHH0O), 3.50 (dd, J=2.6, 4.8 Hz, CH),
4.35±4.50 (m, CH2Ph), 6.35±6.45 (m, NH), 7.21±7.39 (m,
PhH); 13C NMR (CDCl3) 43.0 (CH2Ph), 47.8 (CH2O),
49.8 (CH), 127.9, 128.9, 137.7 (Ph), 168.6 (C(O)) ppm, the
remaining aromatic signal was not detected and is believed
to overlap with nearby peaks. MS (CI+) (rel. int.) 178
(M++1, 100); Mr (+CI) 178.086 42 [M++1] (calcd for
C10H12NO2 178.086 80); Anal. calcd for C10H11NO2: C,
67.80; H, 6.21; N, 7.91. Found: C, 67.64; H, 6.22; N, 7.98.
1
1445, 1340, 1268, 1070, 1025, 742, 695 cm 1; H NMR
(CDCl3) d 1.18 (d, J=6.9 Hz, CH3), 2.40±2.60 (m, OH,
CH), 3.73 (d, J=5.7 Hz, CH2OH), 4.45 (d, J=5.7 Hz
CH2Ph), 6.20±6.35 (m, NH), 7.20±7.40 (m, PhH); 13C
NMR (DMSO-d6) 14.3 (CH3), 41.8 (CH2Ph), 42.8 (CH),
63.8 (CH2OH), 126.6, 127.0, 128.2, 139.7 (Ph), 174.4
(C(O)) ppm, the 13C NMR assignments were consistent
with the APT spectrum; MS (CI+) (rel. int.) 194 (M++1,
100); Mr (+CI) 194.118 32 [M++1] (calcd for
C11H16NO2 194.118 10). Anal. calcd for C11H15NO2: C,
68.40; H, 7.80; N, 7.25. Found: C, 68.17; H, 7.79; N, 7.14.
Synthesis of (R,S)-N-benzyl 2-hydroxy-3-methoxypropion-
amide ((R,S)-6). An anhydrous MeOH (150 mL) solu-
tion containing epoxide (R,S)-5 (1.52 g, 8.6 mmol) and
concentrated H2SO4 (1 mL) was stirred at re¯ux (2 h)
under N2. The solution was allowed to cool to room
temperature and the ``pH'' was adjusted to 7.0 (satu-
rated aq NaHCO3). The mixture was concentrated in
vacuo without the use of heat and the resulting aqueous
concentrate was extracted with EtOAc (3Â100 mL). The
organic extracts were combined, dried (Na2SO4), ®ltered
and evaporated in vacuo to obtain a clear viscous oil.
Further puri®cation was achieved by column chromato-
graphy (SiO2, EtOAc) to a€ord (R,S)-6 as a clear viscous
Synthesis of (R)-N-benzyl 3-methoxy-2-methylpropion-
amide ((R)-8). To an acetonitrile solution (50 mL) of
(R)-16 (1.15 g, 6.0 mmol) was successively added Ag2O
(2.10 g, 9.0 mmol) and MeI (0.6 mL, 9.0 mmol) and the
mixture was stirred at room temperature (18 h). The
insoluble salts were removed by ®ltration, and the ®l-
trate was evaporated in vacuo. The oily residue was
puri®ed by column chromatography (SiO2, ethyl ether:
oil (1.27 g, 71%): Rf 0.64 (EtOAc); IR (liquid ®lm) 3337,
1
pet. ether, 1:1) to a€ord pure (R)-8 as a clear oil (1.23 g,
5.7ꢁ (c 3.2, MeOH); Rf 0.40 (ethyl ether: pet.
24
d
2929, 1653, 1540, 1455, 1260, 1194, 1129, 1029, 699 cm
;
60%): ‰ꢀŠ
1H NMR (DMSO-d6) d 3.26 (s, OCH3), 3.46 (dd, J=6.0,
9.9 Hz, CHH0OCH3), 3.53 (dd, J=3.3, 9.9 Hz,
CHH0OCH3), 4.02±4.09 (m, CHOH), 4.28 (d, J=6.0 Hz,
CH2Ph), 5.69 (d, J=5.4 Hz, CHOH), 7.19±7.40 (m, PhH),
ether, 1:1); IR (liquid ®lm) 3311, 2932, 2830, 1653, 1553,
1455, 1388, 1258, 1205, 1109, 1029, 955, 732, 699 cm 1; 1H
NMR (CDCl3) d 1.17 (d, J=6.9 Hz, CH3), 2.49±2.61 (m,
CH), 3.34 (s, OCH3), 3.38±3.55 (m, CH2OCH3), 4.46 (d,
J=5.7 Hz, CH2Ph), 6.42±6.55 (m, NH), 7.20±7.39 (m,
PhH); 13C, NMR (CDCl3) 14.1 (CH3), 41.2 (CH), 43.4
(CH2Ph), 59.0 (OCH3), 75.0 (CH2OCH3), 127.3, 127.6,
128.7, 138.7 (Ph), 174.8 (C(O)) ppm, the 13C NMR
assignments were consistent with the APT spectrum; MS
(CI+) (rel. int.) 208 (M++1, 100); Mr (+CI) 208.134 37
[M++1] (calcd for C12H18NO2 208.133 75); Anal. calcd
1
8.27 (t, J=6.0 Hz, NH), the H NMR assignments were
consistent with the COSY spectrum; 13C NMR (DMSO-
d6) 41.7 (CH2Ph), 58.4 (OCH3), 71.1 (CH), 74.6
(CH2OCH3), 126.6, 127.0, 128.1, 139.6 (Ph), 171.6 (C(O))
ppm; MS (CI+) (rel. int.) 210 (M++1, 100), 121 (16),
120 (51); Mr (+CI) 210.113 86 [M++1] (calcd for C11H16
.
NO3 210.113 01); Anal. calcd for C11H15NO3 0.45 H2O:
C, 60.80; H, 7.12; N, 6.45. Found: C, 60.95; H, 7.11; N, 6.47.
.
for C12H17NO2 0.25 H2O: C, 68.09; H, 8.27; N, 6.62.
Found: C, 68.28 H, 8.20; N, 6.52.
Synthesis of (R)-N-benzyl 3-hydroxy-2-methylpropion-
amide ((R)-16). A solution containing (R)-15 (2 g, 17
mmol) and benzylamine (10 mL) was stirred neat at
100ꢁC (18 h). The solution was allowed to cool to room
temperature and then triturated with ethyl ether (150
mL) to give a white solid, which was isolated by ®ltration
and washed with additional amounts (50 mL) of ethyl
Synthesis of (S)-N-benzyl 3-methoxy-2-methylpropion-
amide ((S)-8). The procedure utilized for the preparation
of (R)-8 was repeated using (S)-16 (1.47 g, 7.62 mmol),
acetonitrile (50 mL), Ag2O (3.20 g, 13.7 mmol) and MeI
(0.91 mL, 13.7 mmol) to a€ord pure (S)-8 as a clear oil
24
(1.32 g, 84%): ‰ꢁŠ +5.3ꢁ (c 4.7, MeOH); Rf 0.40 (ethyl
d
ether to a€ord (R)-16 as a white crystalline solid (1.75 g,
54%): mp 120±121ꢁC; ‰aŠ
ether: pet. ether, 1:1); IR (liquid ®lm) 3315, 2927, 2835,
1650, 1550, 1457, 1390, 1255, 1198, 1097, 1030, 950, 740,
701 cm 1; 1H NMR (CDCl3) d 1.17 (d, J=7.2 Hz, CH3),
24
d
24.9ꢁ (c 1.5, MeOH); Rf
0.53 (EtOAc); IR (KBr) 3369, 3266, 3088, 1649, 1559,

S. V. Andurkar et al. / Bioorg. Med. Chem. 7 (1999) 2381±2389
2389
2.50±2.61 (m, CH), 3.34 (s, OCH3), 3.40±3.55 (m,
CH2OCH3), 4.46 (d, J=5.7 Hz, CH2Ph), 6.40±6.52 (m,
NH), 7.20±7.40 (m, PhH); 13C NMR (CDCl3) 14.2
(CH3), 41.33 (CH), 43.5 (CH2Ph), 59.2 (OCH3), 75.1
(CH2OCH3), 127.5, 127.7, 128.9, 138.8 (Ph), 174.8 (C(O)),
the 13C NMR assignments were consistent with the APT
spectrum; MS (CI+) (rel. int.) 208 (M++1, 100); Mr
(+CI) 208.134 36 [M++1] (calcd for C12H18NO2 208.133
compound until at least two points were determined
between 100 and 0% protection and minimal motor
impairment. The dose of candidate substance required
to produce the desired endpoint (abolition of hindlimb
tonic extensor component) in 50% of the animals in
each test, and the 95% con®dence interval were calcu-
lated by a computer program based on methods descri-
bed by Finney.23
.
75); Anal. calcd for C12H17NO2 0.25 H2O: C, 68.09; H,
8.27; N, 6.62. Found: C, 68.11; H, 8.19; N, 6.47.
References and Notes
Pharmacology. Compounds were screened under the
auspices of the National Institutes of Health's Anti-
convulsant Screening Project. Experiments were per-
formed in male rodents, speci®cally, albino Carthworth
Farms No. 1 mice (intraperitoneal route), and albino
Sprague±Dawley rats (oral route). The mice weighed
between 18 and 25 g while rats were between 100 and
150 g. All animals had free access to feed and water
except during the actual testing period. Housing, hand-
ling and feeding were all in accordance with recom-
mendations contained in the ``Guide for the Care and
Use of Laboratory Animals''. All of our compounds
were administered in suspensions of 0.5% (w/v) of
methylcellulose in water. The volumes administered
were 0.01 mL/g of body weight for mice and 0.2 mL/10
g for rats. Anticonvulsant activity was established using
the electrical (maximal electroshock of MES) test.20 For
the MES test, a drop of electrolyte solution with an
anesthetic (0.5% butacaine hemisulfate in 0.9% sodium
chloride) was placed in the eyes of the animals prior to
positioning the corneal electrodes and delivery of a non-
lethal current. A 60-cycle alternating current was
administered for 0.2 s in both species, utilizing 50 mA in
mice and 150 mA in rats.21 Protection endpoints were
de®ned as the abolition of the hind limb tonic extensor
component of the induced seizure.22 In mice, e€ects of
compounds on forced spontaneous motor activity were
determined using the rotorod test. The inability of
experimental mice to maintain their balance for 1 min
on a 1 in diameter knurled rod rotating at 6 rpm in three
successive trials was interpreted as a demonstration of
motor impairment. Under these conditions, mice can
normally maintain their balance inde®nitely. Motor
impairment in rats was assessed by observing the overt
evidence of ataxia, abnormal gait and stance, and/or
loss of placing response and muscle tone. In the mouse
identi®cation screens all compounds were administered
at three dose levels (30, 100, 300 mg/kg) and two time
periods (0.5 and 4 h). Typically, in the MES seizure test
one animal was used at 30 and 300 mg/kg, and three
animals at 100 mg/kg. In the rotorod toxicity test four
animals were used at 30 and 300 mg/kg, and eight ani-
mals at 100 mg/kg (Table 1). Oral rat identi®cation
screening was performed using four animals at a ®xed
dose of 30 mg/kg for both the MES and the rotorod
toxicity tests over ®ve time periods ranging from one-
quarter to four hours post drug administration. The
quantitative determination of the median e€ective
(ED50) and toxic doses (TD50) were conducted at pre-
viously calculated times of peak e€ect using the ip
route in mice and oral route for rats. Groups of at least
eight animals were tested using di€erent doses of test
1. Choi, D.; Stables, J. P.; Kohn, H. Bioorg. Med. Chem. 1996,
4, 2105.
2. Porter, R. J.; Cereghino, J. J.; Gladding, G. D.; Hessie, B.
J.; Kupferberg, H. J.; Scoville, B.; White, B. G. Cleveland Clin.
Q. 1984, 51, 293.
3. Conley, J. D.; Kohn, H. J. Med. Chem. 1987, 30, 567.
4. Kohn, H.; Conley, J. D. Chem. Brit. 1988, 24, 231.
5. Choi, D.;Stables, J. P.; Kohn, H. J. Med. Chem. 1996, 39, 1907.
6. Kohn, H.; Sawhney, K. N.; LeGall, P.; Robertson, D. W.;
Leander, J. D. J. Med. Chem. 1991, 34, 2444.
7. Kohn, H.; Conley, J. D.; Leander, J. D. Brain Res. 1988,
457, 371.
8. Kohn, H.; Sawhney, K. N.; Bardel, P.; Robertson, D. W.;
Leander, J. D. J. Med. Chem. 1993, 36, 3350.
9. Gluchowski, C.; Bischo€, T. E.; Garst, M. E.; Kaplan, L. J.;
Dietrich, S. W.; Aswad, A. S.; Ga€ney, M. A.; Aoki, K. R.;
Garcia, C.; Wheeler, L. A. J. Med. Chem. 1991, 34, 392.
10. Tanaka, A.; Yamashita, K. Agric. Biol. Chem. 1980, 44, 199.
11. For comparable procedures for resolving stereoisomers,
see: Weisman, G. R. In Asymmetric SynthesisÐAnalytical
Methods, Morrison, J. D., Ed.; Academic: New York, 1983;
Vol. 1, pp 153±171.
12. For comparable procedures resolving stereoisomers, see:
Dyllick-Brenzinger, R.; Roberts, J. D. J. Am. Chem. Soc.
1980, 102, 1166.
13. For comparable procedures resolving stereoisomers, see:
Parker, D.; Taylor, R. J. Tetrahedron 1987, 43, 5431.
14. Stables, J. P.; Kupferberg, H. J. In Molecular and Cellular
Targets for Antiepileptic Drugs, Avanzini, G.; Tanganelli, P.;
Avoli, M., Eds.; John Libbey: London, 1997, pp 191±198.
15. Swinyard, E. A.; Woodhead, J. H.; White, M. S.; Franklin,
M. R. In Antiepileptic Drugs, 3rd ed., Levy, R. H.; Dreifuss, F.
E.; Mattson, R. H.; Meldrum, B. S.; Penry, J. K., Eds.; Raven
Press: New York, 1989, p 97.
16. Dunham, N. W.; Miwa, T. S. J. Am. Pharm. Assoc. 1957,
46, 208.
17. White, H. S.; Woodhead, J. H.; Franklin, M. R. In Anti-
epileptic Drugs, 4th ed., Levy, R. H.; Mattson, R. H.; Mel-
drum, B. S., Eds.; Raven Press: New York, 1995, pp 99±110.
18. Andurkar, S. V.; Stables, J. P.; Kohn, H. Tetrahedron:
Asymmetry 1998, 9, 3841.
19. Attempted separation of the diastereomers using other
blends of TLC solvents (e.g. ethyl acetate±petroleum ether,
ethyl acetate±hexanes, ethyl acetate±acetone, chloroform±
methanol) were unsuccessful.
20. Swinyard, E. A.; Woodhead, J. H.; White, H. S.; Franklin, M.
R. General Principles: Experimental Selection, Quanti®cation,
and Evaluation of Anticonvulsants. In Antiepileptic Drugs, 3rd
edn; Levy, R. H.; Driefuss, F. E.; Mattson, R. H.; Meldrum, B. S.;
Penry, J. K., Eds.; Raven Press: New York, 1989; pp 85±102.
21. Woodbury, L. A.; Davenport, V. D. Arch. Int. Phar-
amacodyn. Ther. 1952, 92, 97±104.
22. White, H. S.; Woodhead, J. H.; Franklin, M. R. In Anti-
epileptic Drugs, Levy, R. H.; Mattson, R. H.; Meldrum, B. S.
Eds.; Raven: New York, 1998, pp 99±100.
23. Finney, D. J. Probit Analysis, 3rd ed. Cambridge Uni-
versity Press: London, 1971.