Synthesis and dopamine receptor modulating activity of substituted bicyclic thiazolidine lactam peptidomimetics of L-prolyl-L-leucyl-glycinamide

Article abstract of DOI:10.1021/jm990140n

6-Substituted bicyclic thiazolidine lactam peptidomimetics of Pro-Leu- Gly-NH₂ (1) were synthesized to test the hypothesis that incorporation of a hydrophobic side chain into the bicyclic thiazolidine lactam scaffold would further enhance the d

Full text of DOI:10.1021/jm990140n

J . Med. Chem. 1999, 42, 2977-2987
2977
Syn th esis a n d Dop a m in e Recep tor Mod u la tin g Activity of Su bstitu ted Bicyclic
Th ia zolid in e La cta m P ep tid om im etics of L-P r olyl-L-leu cyl-glycin a m id e
Ehab M. Khalil,† Ashish Pradhan,§ William H. Ojala,‡ William B. Gleason,‡ Ram K. Mishra,§ and
Rodney L. J ohnson*,†
Department of Medicinal Chemistry, The Biomedical Engineering Center, and Department of Laboratory Medicine and
Pathology, University of Minnesota, Minneapolis, Minnesota 55455, and Department of Psychiatry and Behavioural
Neuroscience, McMaster University, Hamilton, Ontario L8N 3Z5, Canada
Received March 25, 1999
6-Substituted bicyclic thiazolidine lactam peptidomimetics of Pro-Leu-Gly-NH2 (1) were
synthesized to test the hypothesis that incorporation of a hydrophobic side chain into the bicyclic
thiazolidine lactam scaffold would further enhance the dopamine receptor modulating activity
of such peptidomimetics. The substituents employed were the isobutyl, butyl, and benzyl groups
to give peptidomimetics 3-5, respectively. These peptidomimetics were evaluated in vivo as
modulators of apomorphine-induced rotational behavior in the 6-hydroxydopamine-lesioned
rat model of hemiparkinsonism and were compared with the unsubstituted bicyclic thiazolidine
lactam Pro-Leu-Gly-NH2 peptidomimetic 2. Peptidomimetics 3-5 each affected rotational
behavior in a bell-shaped dose-response relationship producing maximal increases of 44% (1
µg/kg, ip), 56% (0.1 µg/kg, ip), and 30% (1 µg/kg, ip), respectively. In comparison, unsubstituted
peptidomimetic 2 increased rotational behavior by only 23% at a dose of 0.1 µg/kg, ip.
In tr od u ction
The endogenous tripeptide L-prolyl-L-leucyl-glycin-
amide (1, PLG), also known as melanocyte-stimulating
hormone release-inhibiting factor (MIF-1),1 is able to
modulate dopaminergic neurotransmission in the CNS.
This modulatory activity has been demonstrated through
in vitro and in vivo studies.2-7 In vitro receptor binding
studies have shown that 1 and its analogues induce an
increase in agonist binding to dopamine receptors in the
striatum, which is due to an increase in the affinity of
found in 1, it proved to be one of the most effective
the dopamine receptor for agonists.3,6 Peptide 1 and its
analogues tested thus far. Nevertheless, since earlier
SAR studies with linear tripeptide analogues of 1
active analogues also are capable of preventing the 5′-
guanylylimidodiphosphate [Gpp(NH)p]-induced conver-
modified at the leucine position showed that a hydro-
sion of D2 dopamine receptors from a high-affinity to a
phobic residue at this position was important for activ-
low-affinity state, suggesting that modulation of dopam-
ity,18 we postulated that incorporation of the hydrophilic
ine receptors by 1 may involve guanine nucleotide
side chain into the structure of 2 would enhance further
regulatory proteins (G-proteins).6
the activity of this peptidomimetic. To test this hypoth-
In the rat nigrostriatal 6-hydroxydopamine lesion
esis, the synthesis of the 6-substituted bicyclic thiazo-
rotational model,8 1 has been shown to potentiate the
lidine lactams 3-5 was undertaken. In addition to the
contralateral rotational behavior induced by apomor-
isobutyl side chain, the butyl and benzyl side chains
phine.2,5,7,9 Further evidence supporting the ability of
were also chosen because the SAR studies with linear
1 to modulate dopamine receptor activity is shown in
tripeptide analogues of 1 showed that replacing the
studies of neuroleptic induced dopamine receptor su-
leucyl residue with either the norleucine or phenylala-
persensitivity where it antagonizes the up-regulation
nine residue provided analogues equal in potency and
of dopamine receptors produced by chronic haloperidol
efficacy to that of the parent peptide.
treatment.4,10,11
Numerous conformationally constrained analogues of
1 have been synthesized and tested in an effort to
elucidate the peptide’s bioactive conformation.12-17 One
such analogue is the bicyclic thiazolidine lactam 2.14
Although this analogue lacked the side chain that
corresponds to the isobutyl group of the leucine residue
Syn th esis
The R,R-disubstituted amino acids 8, 17, and 18
served as the starting materials for the synthesis of
peptidomimetics 3-5. Two different routes were em-
ployed in obtaining these compounds. In the synthesis
of 8, a modification of the method by Karady et al.19
was used as depicted in Scheme 1. L-Cbz-Phe-OH was
converted to oxazolidinones 6a and 6b, which were
separated by a combination of chromatography and
crystallization. Consistent with previously published
† Department of Medicinal Chemistry.
‡ The Biomedical Engineering Center and Department of Laboratory
Medicine and Pathology.
§ Department of Psychiatry and Behavioural Neuroscience.
10.1021/jm990140n CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/02/1999

2978 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15
Khalil et al.
Sch em e 1
oxazolidinone was determined on the basis of NOE
difference experiments which showed a reciprocal NOE
between the H2 and H4 hydrogens. The results were in
agreement with those of Seebach and Fadel21 and Smith
et al.22 The potassium enolate of 11 was diastereose-
lectively alkylated with allyl iodide to give 13 which was
hydrolyzed to give the allyloxycarbonyl R-allyl amino
acid 15 in an excellent overall yield. The same sequence
of reactions was carried out on L-norleucine. Since the
allyloxycarbonyl group of 15 and 16 was incompatible
with the rest of the synthesis, it was removed24 in each
case, and the corresponding amino acids 17 and 18 were
reprotected to give the tert-butoxycarbamate derivatives
19 and 20, respectively.
The conversion of the R-allyl amino acids 8, 19, and
20 to peptidomimetics 3-5, respectively, was carried out
as outlined in Scheme 3. Oxidative cleavage of each of
the R-allyl amino acids to the corresponding aldehydes
was accomplished with OsO4 and NaIO4. However, in
each case the product actually isolated after workup was
a diastereoisomeric mixture of γ-hydroxy lactones 21-
23. Condensation of each lactone with D-cysteine methyl
ester under conditions previously described by Genin
and J ohnson25 afforded, in each case, a diastereoiso-
meric mixture of thiazolidines 24-26. Ring closure of
24-26 to form the epimeric mixture of bicyclic lactams
27a /27b-29a /29b, respectively, was carried out with
Mukaiyama’s reagent.26 In all three cases, the isomer
with the bridgehead stereochemistry of 7a-R (a ) was the
major isomer isolated. The ratio of a :b ranged from 2-3:
1. Separation of each pair of isomers was possible by
chromatographic techniques. However, in the case of
compounds 27a /27b and 28a /28b, the separation was
very laborious and it proved more expedient to ac-
complish this in the subsequent carboxamidation step
where a differential rate of reaction was observed. Upon
treatment of 27a /27b and 28a /28b with methanolic
ammonia, the desired isomers in each case, 27a and
28a , were found to undergo a much more rapid reaction
than their diastereoisomeric counterparts. Thus, it was
possible to readily obtain the carboxamides 30 and 31
in pure form after a simple flash chromatographic step.
In the case of 29a and 29b, the two isomers were easily
separated from one another, and the desired isomer 29a
was converted to the carboxamide derivative 32 with
NH3/MeOH.
Sch em e 2
The stereochemistry at the bridgehead carbon (7a-C)
of the 6-substituted bicyclic thiazolidine lactams was
assigned on the basis of extensive NOE difference
experiments performed on the methyl ester (27-29) and
carboxamide (30-32) derivatives (Figure 1). For ex-
ample, in the experiment performed on 28b, weak
reciprocal NOEs (∼1%) were observed between H3 and
work by Cheng et al.,20 the major isomer obtained was
the cis-oxazolidinone 6a . This compound was stereose-
lectively alkylated with potassium hexamethyldisilazide
and allyl iodide to give the R-allyloxazolidinone 7. Basic
hydrolysis of the lactone furnished the benzyloxycarbo-
nyl-protected amino acid 8 quantitatively.
H
7a, with stronger reciprocal NOEs seen between H3 and
H2a and between H7a and H2a. These data support the
assignment of an S-configuration at 7a-C. In support
of this assignment, NOEs observed for the bicyclic
lactam derived from the 7a-epimer of 27b, 31, are
consistent with an R-configuration at 7a-C. As indicated
in Figure 1 similar results were seen with bicyclic
lactams 27b, 29b, 30, and 32.
Initially, the above approach was tried for the syn-
thesis of 17. However, the mixture of cis/trans-oxazoli-
dinones that was obtained proved very difficult to
separate. As an alternative, the procedure of Seebach
and Fadel21 as modified by Smith et al.22,23 was applied
and found to work. As shown in Scheme 2, the sodium
salt of L-leucine was condensed with pivalaldehyde to
form imine 9. This Schiff base was cyclized to the
corresponding cis-oxazolidinone 11 by N-acylation with
allyl chloroformate. The relative configuration of this
The X-ray crystal structure of lactam 29a was solved,
and an ORTEP representation is shown in Figure 2.27
In addition to confirming the NOE stereochemical

Bicyclic Thiazolidine Lactam Peptidomimetics
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15 2979
Sch em e 3
F igu r e 2. ORTEP plot of 6-benzyl bicyclic thiazolidine lactam
29a . Hydrogen atoms are drawn as spheres of arbitrary radius.
nonconstrained angles (φ2 ) -53° and ψ3 ) -26°) fall
within the acceptable range of an ideal type II â-turn.28
The final conversions of the dipeptide â-turn mimics
30-32 to 3-5 were carried out with standard coupling
methods as outlined in Scheme 3. In removing the tert-
butoxycarbonyl group from 33 and 34, epimerization
was observed at the bridgehead carbon (7a-C). This is
in accord with a previously proposed mechanism for
epimerization involving acid-catalyzed ring opening.29
This problem was overcome by carrying out the reaction
at 0 °C for short periods of time (<45 min).
F igu r e 1. Graphical summary of the NOEs observed for the
bicyclic thiazolidine lactams 27b-29b and 30-32.
assignment, the solid-state structure of 29a provides
detailed information on the accessible peptide backbone
conformation of this bicyclic thiazolidine lactam scaffold.
A comparison of the φ, ψ dihedrals defining the turn
region of 29a with those of the classical type II â-turn
reveals that the two torsion angles constrained by the
heterocycle (ψ2 ) 123° and φ3 ) 113°) as well as the
P h a r m a cology
The substituted bicyclic lactams 2-5 were evaluated

2980 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15
Khalil et al.
for the unsubstituted bicyclic thiazolidine lactam scaf-
fold.17 Both of these X-ray crystallography studies
confirm the prediction of molecular modeling14 that the
bicyclic thiazolidine lactam scaffold is a good mimic of
a type II â-turn as an idealized â-turn has values of 120°
( 30° for the ψ2 torsion angle and 80° ( 30° for the φ3
torsion angle.28
Pro-Leu-Gly-NH2 peptidomimetic 2 previously was
shown to possess very good activity in enhancing the
binding of the dopamine receptor agonist 2-amino-6,7-
dihydroxy-1,2,3,4-tetrahydronaphthalene to isolated bo-
vine dopamine receptors even though it lacked the
isobutyl side chain present in the parent peptide.14
Previous work had shown this side chain to be impor-
tant for the activity of Pro-Leu-Gly-NH2 as replacing
the side chain with lower alkyl moieties resulted in
analogues that posessed little or no activity.18 In con-
trast, replacing the isobutyl side chain with the butyl
or benzyl group yielded analogues that retained the
ability to modualte dopamine receptors.
F igu r e 3. Effect of ip administered peptidomimetics 2-5 on
contralateral behavior of ip administered apomorphine (0.25
mg/kg) in rats (n ) 4) with unilateral lesions. Number of
contralateral rotations in 10 min expressed as a percentage
change from control rotational response. Dosage expressed as
log (mg/kg) from -5 or 0.00001 mg/kg to -2 or 0.01 mg/kg.
Each point represents the mean percent change ( SEM.
Statistical significance: *p < 0.05.
in the rat nigrostriatal 6-hydroxydopamine lesion rota-
tional model. When administered in the absence of
apomorphine, peptidomimetics 2-5 did not produce any
contralateral rotations or other observable changes in
behavior. Peptidomimetics 3-5 each produced a bell-
shaped dose-response curve when administered with
apomorphine (Figure 3). The maximal effect for 3-5 was
a 44%, 56%, and 30% increase in rotations, respectively,
compared to apomorphine administration alone. For
peptidomimetics 3 and 5, the maximal response was
observed at a dose of 1 µg/kg ip. In the case of 4, the
maximal response was observed at a dose of 0.1 µg/kg,
ip. The comparison compound, peptidomimetic 2, pro-
duced a maximal dose response at 0.1 µg/kg ip and
resulted in a nonsignificant 23% increase in the number
of rotations over a 10-min period compared to apomor-
phine alone.
We postulated that placement of lipophilic side chains
on the bicyclic thiazolidine lactam scaffold would en-
hance the activity of these Pro-Leu-Gly-NH2 peptido-
mimetics by virture of their ability to access the binding
site with which the leucyl side chain interacts. Thus,
peptidomimetics 3-5 were synthesized in which posi-
tion 6 of the bicyclic thiazolidine lactam ring scaffold
was substituted with the isobutyl, butyl, and benzyl
groups, respectively. As illustrated in Figure 3, all three
6-substituted peptidomimetics possessed activity greater
than that seen with 2 in terms of enhancing rotational
behavior by apomorphine in 6-hydroxydopamine-
lesioned rats. The relative ranking of activity was 4 >
3 > 5 > 2.
Previously, Pro-Leu-Gly-NH2 was found to increase
rotational behavior by 30 ( 7% with this maximal
response occurring at a dose of 1.0 mg/kg, ip.7 The
results obtained in this study show that the substituted
bicyclic thiazolidine lactams 3-5 are much more potent
than Pro-Leu-Gly-NH2, since the maximal responses for
these compounds occur at doses that are 1000-10000
times lower. These results support the hypothesis that
incorporation of a hydrophobic substituent on the
6-position of the bicyclic thiazolidine lactam scaffold
leads to enhanced dopamine receptor modulatory activ-
ity through the ability of these peptidomimetics to
access the binding site with which the leucine side chain
interacts.
Discu ssion
The numerous pharmacological studies that have
been conducted on 1 have clearly shown that this
tripeptide enhances dopaminergic neurotransmission
within the CNS.30 Studies have shown that 1 does not
modulate dopaminergic neurotransmission by affecting
either dopamine synthesis, uptake, or metabolism.31-34
Rather, biochemical and pharmacological studies indi-
cate that this modulation is brought about by a mech-
anism in which 1 renders the dopamine receptor more
responsive to agonists.3,6 The ability of 1 and its
analogues to enhance apomorphine-induced rotations in
6-hydroxydopamine-lesioned rats is believed to be due
to this enhanced responsiveness of the dopamine recep-
tor to the agonist apomorphine.2,5,7,9
Compound 2 was originally synthesized as a pepti-
domimetic of Pro-Leu-Gly-NH2 wherein the bicyclic
thiazolidine lactam scaffold was designed to mimic the
type II â-turn conformation that had been postulated
to be the bioactive conformation of Pro-Leu-Gly-NH2.14
Molecular modeling of the bicyclic thiazolidine lactam
ring system had predicated that this template would
constrain the ψ2 and φ3 torsion angles of a type II â-turn
to values that were in the range of the idealized values
for such a turn. The X-ray structure obtained for 29a
in the current study shows that the ψ2 and φ3 torsion
angles possess values of 123° and 113°, respectively.
This result resembles the ψ2 and φ3 torsion angle values
of 142.7° and 120.5°, respectively, that were obtained
Exp er im en ta l Section
Gen er a l Asp ects. Thin-layer chromatography was per-
formed on Analtech 250-µm silica gel HLF Uniplates visualized
by UV, I2, ninhydrin spray (amines), 2,6-dichlorophenol in-
dophenol spray (acids), and 2,4-dinitrophenylhydrazine (alde-
hydes). Chromatographic purification on silica gel (Merck,
grade 60, 240-400 mesh, 60 Å) was done by flash or gravity
methods and semipreparative HPLC was done with a Waters
Associates 25- × 100-mm PrepPak cartridge. Optical rotations
were measured on a Rudolph Research Autopol III polarimeter
at the 589-nm Na D-line. Melting points were obtained on a
Thomas-Hoover melting point apparatus and are uncorrected.
Elemental analyses were performed by M-H-W Laboratories,
Phoenix, AZ. 1H and 13C NMR spectra were measured in CDCl3
at 300 and 75.5 MHz, respectively, with CDCl3 as the internal
reference for 1H (δ 7.26) and 13C (δ 77.06).
(2S,4S)- a n d (2R,4S)-2-P h en yl-3-N-(ben zyloxyca r bo-
n yl)-4-ben zyloxa zolid in on e (6a a n d 6b). Oxazolidinone 6a

Bicyclic Thiazolidine Lactam Peptidomimetics
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15 2981
was synthesized by a modification of the Karady method19 that
was developed by Cheng et al.20 Cbz-L-Phe-OH (30 g, 100
mmol) and benzaldehyde dimethyl acetal (14.4 mL, 96 mmol)
in Et2O (500 mL) were cooled to -78 °C. BF3‚Et2O (60.2 mL,
489 mmol) was added dropwise, and the mixture was then
allowed to warm to room temperature where it was stirred
for 4 days under an atmosphere of N2. The mixture was cooled
to 0 °C, and excess BF3‚Et2O was quenched by adding
saturated NaHCO3, initially in 1-mL portions until the strong
bubbling ceased. The aqueous layer was separated and ex-
tracted with Et2O (3 × 150 mL). The combined Et2O extracts
were dried with MgSO4 and concentrated under vacuum to
give a crude solid that was crystallized from EtOAc/Et2O to
give a total of 13.8 g (37%) of 6a . The mother liquor contained
a mixture of 6a and its (2R,4S)-epimer which was difficult to
separate cleanly by silica gel chromatography alone (Et2O/
EtOAc/hexanes, 5:1:20). Fractions were obtained from the
column that when combined and concentrated gave a tan solid
which was highly enriched with the (2R,4S)-trans-epimer 6b
(9 g, 24%). The pure trans-epimer was then obtained by
recrystallization from EtOAc/hexanes.
Et2O (200 mL) and then extracted with EtOAc (3 × 200 mL)
after its pH was adjusted to 1 with 1 N HCl. The combined
organic extracts were dried with MgSO4, and the solvents were
evaporated in vacuo to give 3.7 g (100%) of 8 as a tan oil which
became a glassy solid on standing: [R]D +41.5 (c 2.2, MeOH);
1H NMR (CDCl3) δ 2.74 (dd, J ) 14.7 and 7.5 Hz, 1 H, CH2-
CHdCH2), 3.23-3.33 (m, 2 H, CH2CHdCH2 and CH2Ph), 3.67
(d, J ) 13.5 Hz, 1 H, CH2Ph), 5.11-5.31 (m, 4 H, OCH2Ph
and CHdCH2), 5.66 (s, 1 H, NH), 5.70-5.79 (m, 1 H, CHd
CH2), 7.09-7.54 (m, 15 H, Ph), 10.76 (br s, 1 H, COOH); 13C
NMR (CDCl3) δ 39.7 and 40.4 (CH2Ph and CH2CHdCH2), 64.8
and 66.5 (R-C and OCH2Ph), 119.6, 126.9, 128.1, 128.2, 128.4,
128.9, 129.8, 130.2, 131.6, 134.0, 135.6, and 136.4 (Ph), 154.5
(Cbz CdO), 177.4 (COOH); FAB MS m/z 340 [M + H]+, 250
[M - PhCH2]+. Anal. (C20H21NO4) C, H, N.
(2S,4R)-2-ter t-Bu tyl-3-(a llyloxyca r bon yl)-4-(2-m eth yl-
p r op yl)-4-a llyloxa zolid in on e (13). Oxazolidinone 11 (10.8
g, 38.1 mmol), prepared according to a procedure by Smith et
al.,22 was transferred to an oven-dried 500-mL round-bottom
flask that was purged with Ar. Dry THF (180 mL) was
introduced via a cannula, and the flask was cooled to -78 °C.
A solution of KN(TMS)2 in toluene (0.5 M, 91.4 mL, 45.7 mmol)
was added to the solution dropwise over 15 min under Ar. After
another 15 min of stirring, allyl iodide (7.0 mL, 76.2 mmol)
was added dropwise. The reaction was stirred for 3 h and then
quenched with 10% NaHSO4 at -78 °C. The reaction mixture
was extracted with EtOAc (2 × 300 mL), and the combined
extracts were washed with saturated NaHCO3 (200 mL) and
brine, dried (MgSO4), and evaporated under vacuum. The
resulting residue was loaded onto a 5.5- × 58-cm silica gel
column and eluted with 10% EtOAc in hexanes to give 11.1 g
6a : mp 126-127 °C (lit.20 mp ) 123-125 °C); [R]D +45.6 (c
1.6, CHCl3); EI MS m/z 387 [M]+.
6b: mp 136-137 °C; [R]D +205.9 (c 2.0, CHCl3); 1H NMR
(CDCl3, 323 K) δ 3.24 (d, J ) 14 and 2.4 Hz, 1 H, CH2Ph), 3.7
(m, 1 H, CH2Ph), 4.82-4.83 (m, 1 H, 4-CH), 5.03-5.06 (m, 2
H, OCH2Ph), 5.57 (s, 1 H, 2-CH), 7.12-7.42 (m, 15 H, Ph); 13
C
NMR (CDCl3, 323 K) δ 34.4 (CH2Ph), 57.9 (4-C), 67.4 (OCH2-
Ph), 89.9 (2-C), 126.5, 127.4, 127.9, 128.0, 128.3, 128.5, 128.6,
129.6, 129.8, 134.4, 135.2, and 137.4 (Ph), 151.7 (Cbz CdO),
170.7 (5-CdO); EI MS m/z 387 [M]+. Anal. (C24H21NO4) C, H,
N.
1
(90% yield) of 13 as a clear oil: [R]D +4.8 (c 1.6, CHCl3); H
NMR (CDCl3) δ 0.84-0.89 (m, 15 H, (CH3)3 and (CH3)2), 1.77
(dd, J ) 14.6 and 4.9 Hz, 1 H, CH2CH(CH3)2), 1.85 (dd, J )
14.6 and 6.1 Hz, 1 H, CH2CH(CH3)2), 1.96-2.04 (m, 1 H,
CH(CH3)2), 2.37 (dd, J ) 13.4 and 6.1 Hz, 1 H, CH2CHdCH2),
3.04-3.12 (m, 1 H, CH2CHdCH2) 4.44 (dd, J ) 12.2 and 6.1
Hz, 1 H, OCH2CHdCH2), 4.56 (dd, J ) 12.2 and 6.1 Hz, 1 H,
OCH2CHdCH2), 4.99-5.05 (m, 2 H, CH2CHdCH2), 5.16-5.29
(m, 2 H, OCH2CHdCH2), 5.34 (s, 1 H, 2-CH), 5.36-5.48 (m, 1
H, CH2CHdCH2), 5.79-5.88 (m, 1 H, OCH2CHdCH2); 13C
NMR (CDCl3) δ 23.4 (CH(CH3)2), 24.4 and 24.6 (CH(CH3)2),
25.1 ((CCH3)3), 37.7 ((CCH3)3), 40.1 (CH2CHdCH2), 45.9 (CH2-
CH(CH3)2), 66.3 (OCH2CHdCH2), 66.7 (4-C), 94.8 (2-C), 119.0
(OCH2CHdCH2), 121.1 (CH2CHdCH2), 130.4 (CH2CHdCH2),
131.5 (OCH2CHdCH2), 154.4 (Alloc CdO), 173.9 (5-CdO); FAB
MS m/z 324 [M + H]+. Anal. (C18H29NO4) C, H, N.
(2S,4S)-2-P h en yl-3-(ben zyloxyca r bon yl)-4-ben zyl-4-a l-
lyloxa zolid in on e (7). Oxazolidinone 6a (12.8 g, 33 mmol) was
dissolved in dry THF (175 mL) and the solution cooled to -78
°C. A solution of KN(TMS)2 in toluene (0.5 M, 73 mL, 36.5
mmol) was added to the solution dropwise under an atmo-
sphere of Ar. After 30 min of stirring, allyl iodide (4.5 mL,
49.5 mmol) was added dropwise over 5 min. The reaction
changed color from deep orange after addition of KN(TMS)2
to yellow after addition of allyl iodide. The mixture was stirred
for another 3 h at -78 °C and then allowed to warm to room
temperature over 4 h. The reaction was worked up at -78 °C
by addition of saturated NaHCO3 followed by removal of the
THF in vacuo. The remaining solution was extracted with
EtOAc (3 × 200 mL). The combined extracts were washed
twice with 200 mL of each of the following: H2O, saturated
NaHCO3, and brine. Drying and removal of solvents under
vacuum gave 13.3 g of a crude oil which was purified using
flash silica gel chromatography (10% EtOAc in hexanes). The
pure product, 7, was obtained in 54% yield (7.6 g) as a tan
solid which was crystallized from CH2Cl2/hexanes mixture: mp
108-109 °C; [R]D +69.8 (c 1.6, CHCl3); 1H and 13C NMR show
the presence of two rotamers about the carbamate bond in a
ratio of 3:1; 1H NMR (CDCl3) δ 2.69 (dd, J ) 13.4 and 6.1 Hz,
1 H, CH2CHdCH2), 3.01 and 3.29-3.37 (dd, J ) 13.4 and 9.8
Hz and m, 2 H, CH2CHdCH2 and CH2Ph), 3.66 (d, J ) 13.4
Hz, 1 H, CH2Ph), 4.89 (d, J ) 12.2, 1 H, OCH2Ph), 5.08 (d, J
) 12.2 Hz, 1 H, OCH2Ph), 5.20-5.39 (m, 2 H, CHdCH2), 5.69-
5.89 (m, 1 H, CHdCH2), 6.05 (s, 1 H, 2-CH), 6.12 and 6.30 (d,
J ) 7.3 Hz, 1 H, Ph), 6.78 (d, J ) 7.3 Hz, 1 H, Ph), 6.97-7.06
(m, 2 H, Ph), 7.14-7.48 (m, 12 H, Ph); 13C NMR (CDCl3) δ
40.6, 41.9, 42.5, and 42.6 (CH2Ph and CH2CHdCH2), 67.1,
67.9, 68.9, and 69.1 (4-C and OCH2Ph), 89.8 and 90.0 (2-C),
121.9, 127.1, 127.3, 127.4, 127.7, 127.9, 128.1, 128.8, 129.1,
129.2, 129.3, 129.9, 130.5, 135.2, 135.8, and 135.9 (Ph), 152.1
and 153.3 (Cbz CdO), 172.9 and 173.1 (5-CdO); FAB MS m/z
428 [M + H]+. Anal. (C27H25NO4) C, H, N.
(2S,4R)-2-ter t-Bu tyl-3-(a llyloxyca r bon yl)-4-bu tyl-4-a l-
lyloxa zolid in on e (14). Oxazolidinone 12 (3.0 g, 10.2 mmol)
was treated in the same manner as described above for the
synthesis of 13. The product of the reaction was chromato-
graphed on a 4- × 45-cm silica gel column (10% EtOAc in
hexanes) to give 2.8 g (85% yield) of 14 as a clear oil: [R]D
1
+9.1 (c 4.4, CHCl3); H NMR (COSY assignment, CDCl3, 323
K) δ 0.88 (t, J ) 7.3 Hz, 3 H, (CH2)3CH3), 0.94 (s, 9 H, (CH3)3),
1.24-1.41 (m, 3 H, (CH2)2CH2CH3 and CH2CH2CH2CH3),
1.61-1.74 (m, 1 H, CH2CH2CH2CH3), 1.88-2.09 (m, 2 H,
CH2(CH2)2CH3), 2.48 (dd, J ) 13.4 and 6.1 Hz, 1 H, CH2CHd
CH2), 3.13-3.48 (dd, J ) 13.4 and 8.6 Hz, 1 H, CH2CHdCH2),
4.50 (dd, J ) 12.2 and 6.1 Hz, 1 H, OCH2CHdCH2), 4.64 (dd,
J ) 12.2 and 4.9 Hz, 1 H, OCH2CHdCH2), 5.07-5.13 (m, 2 H,
CH2CHdCH2), 5.23-5.36 (m, 2 H, OCH2CHdCH2), 5.42 (s, 1
H, 2-CH), 5.44-5.57 (m, 1 H, CH2CHdCH2), 5.86-5.99 (m, 1
H, OCH2CHdCH2); 13C NMR (DEPT assignment, CDCl3, 323
K) δ 13.7 (CH2)3CH3), 22.9 and 26.4 (CH2(CH2)2CH3), 25.6
((CCH3)3), 37.5 (CH2(CH2)2CH3), 38.0 (C(CH3)3), 39.5 (CH2CHd
CH2), 66.4 (OCH2CHdCH2), 66.8 (4-C), 95.05 (2-C), 119.0 and
121.1 (OCH2CHdCH2 and CH2CHdCH2), 130.8 and 131.9
(OCH2CHdCH2 and CH2CHdCH2), 154.6 (Alloc CdO), 174.2
(5-CdO); FAB MS m/z 324 [M + H]+. Anal. (C18H29NO4) C, H,
N.
(2S)-N-(Ben zyloxyca r bon yl)-2-a llylp h en yla la n in e (8).
Oxazolidinone 7 (4.69 g, 11 mmol) was dissolved in a mixture
of MeOH and 1 N NaOH (50 mL each) and the reaction
mixture then heated at reflux for 16 h. The reaction mixture
was concentrated under vacuum, and the resulting residue was
redissolved in H2O (100 mL). This solution was washed with
(2R)-N-(Allyloxyca r bon yl)-2-a llylleu cin e (15). Oxazoli-
dinone 13 (10.5 g, 32.5 mmol) was dissolved in a mixture of
MeOH and 1 N NaOH (200 mL, 1:1) and the reaction mixture

2982 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15
Khalil et al.
then heated at reflux for 24 h. The reaction mixture was
concentrated under vacuum to about one-half its volume, and
the resulting solution was washed with Et2O (150 mL). The
aqueous layer was acidified to pH 1 with 1 N HCl and then
extracted with EtOAc (3 × 300 mL). The combined organic
extracts were dried with MgSO4, and solvents were evaporated
in vacuo to give 8.3 g (100% yield) of 15 as a tan oil. This
material was used without further purification: [R]D -15.0 (c
1.6, MeOH); 1H NMR (CDCl3) δ 0.83 (d, J ) 7.3 Hz, 3 H,
(CH3)2), 0.90 (d, J ) 7.0 Hz, 3 H, (CH3)2), 1.59-1.77 (m, 2 H,
CH2CH(CH3)2 and CH2CH(CH3)2), 2.35 (dd, J ) 13.4 and 6.1
Hz, 1 H, CH2CHdCH2), 2.47 (dd, J ) 14.6 and 7.3 Hz, 1 H,
CH2CH(CH3)2), 3.11 (dd, J ) 13.4 and 7.3 Hz, 1 H, CH2CHd
CH2), 4.53 (d, J ) 6.1 Hz, 1 H, OCH2CHdCH2), 5.05-5.11 (m,
2 H, CH2CHdCH2), 5.18-5.32 (m, 2 H, OCH2CHdCH2), 5.55-
5.69 (m, 1 H, CH2CHdCH2), 5.83-5.97 (m, 1 H, OCH2CHd
CH2), 5.86 (s, 1 H, NH), 11.41 (br s, 1 H, COOH); 13C NMR
(CDCl3) δ 22.6, 23.6, and 24.6 (CH(CH3)2 and CH(CH3)2), 40.6
and 43.5 (CH2CHdCH2 and CH2CH(CH3)2), 63.2 and 65.2
(OCH2CHdCH2 and R-C), 117.5 and 119.2 (OCH2CHdCH2 and
CH2CHdCH2), 131.7 and 132.7 (CH2CHdCH2 and OCH2CHd
CH2), 154.01 (Alloc CdO), 178.9 (COOH); FAB MS m/z 256
[M + H]+. Anal. (C13H21NO4) C, H, N.
CH2), 2.63 (dd, J ) 14.0 and 6.7 Hz, 1 H, CH2CHdCH2), 5.18-
5.24 (m, 2 H, CH2CHdCH2), 5.72-5.86 (m, 1 H, CH2CHdCH2);
13C NMR (CD3OD) δ 14.2 ((CH2)3CH3), 23.9 and 26.8 (CH2-
(CH2)2CH3), 37.2 and 42.3 (CH2CHdCH2 and CH2(CH2)2CH3),
65.3 (R-C), 121.0 (CH2CHdCH2), 132.1 (CH2CHdCH2), 175.4
(COOH); FAB MS m/z 172 [M + H]+. Anal. (C9H17NO2) C, H,
N.
(2R)-N-(ter t-Bu toxyca r bon yl)-2-a llylleu cin e (19). (2R)-
2-Allylleucine (17; 2.1 g, 12.26 mmol), tetramethylammonium
hydroxide pentahydrate (TMAH; 2.2 g, 12.26 mmol), and Boc2O
(5.3 g, 24.5 mmol) were added to a 4:1 mixture of CH3CN/DMF
(150 mL). A clear solution was obtained after about 30 min,
and the reaction was stirred for 2 days at room temperature.
On the third day, another 0.5 equiv of Boc2O (1.3 g, 6.1 mmol)
was added, and the mixture was stirred for another day. The
CH3CN and DMF were removed in vacuo, and the resulting
residue was redissolved in H2O (100 mL). This solution was
washed with Et2O (2 × 100 mL). The aqueous layer was
acidified with solid citric acid to pH 3 and then extracted with
EtOAc (3 × 150 mL). The combined organic extracts were
washed with H2O and dried with MgSO2, and the EtOAc was
removed in vacuo to give 19 as a tan oil (2.7 g, 82% yield) which
was used without further purification: [R]D -13.1 (c 1.4,
1
MeOH); H NMR (CDCl3) δ 0.84 (d, J ) 6.1 Hz, 3 H, (CH3)2),
(2R)-N-(Allyloxyca r bon yl)-2-a llyln or leu cin e (16). Ox-
azolidinone 14 (2.7 g, 8.3 mmol) was hydrolyzed with the same
procedure as that described above for the synthesis of 15. The
product was obtained as a tan oil in a yield of 2.1 g (100%
yield). This material was used without further purification:
1H NMR (CDCl3) δ 0.83 (t, J ) 6.7 Hz, 3 H, (CH2)3CH3), 0.93-
1.33 (m, 4 H, CH2(CH2)2CH3), 1.72-1.80 (m, 1 H, CH2(CH2)2-
CH3), 2.22-2.29 (m, 1 H, CH2(CH2)2CH3), 2.48-2.54 (m, 1 H,
CH2CHdCH2), 2.99-3.03 (m, 1 H, CH2CHdCH2), 4.49-4.50
(m, 2 H, OCH2CHdCH2), 5.03-5.29 (m, 4 H, CH2CHdCH2 and
OCH2CHdCH2), 5.55-5.68 (m, 1 H, CH2CHdCH2), 5.81-5.92
(m, 1 H, OCH2CHdCH2), 5.80 (s, 1 H, NH), 11.23 (br s, 1 H,
0.92 (d, J ) 6.1 Hz, 3 H, (CH3)2), 1.43 (s, 9 H, Boc CH3), 1.58-
1.74 (m, 2 H, CH2CH(CH3)2 and CH2CH(CH3)2), 2.32 (dd, J )
13.2 and 3.6 Hz, 1 H, CH2CHdCH2), 2.47 (dd, J ) 13.2 and
7.5 Hz, 1 H, CH2CH(CH3)2), 3.10 (dd, J ) 13.2 and 7.2 Hz, 1
H, CH2CHdCH2), 5.06-5.12 (m, 2 H, CH2CHdCH2), 5.58 (s,
1 H, NH), 5.6-5.7 (m, 1 H, CH2CHdCH2), 10.34 (br s, 1 H,
COOH); 13C NMR (CDCl3) δ 22.95, 23.67, and 24.65 (CH(CH3)2
and CH(CH3)2), 40.54 and 43.50 (CH2CHdCH2 and CH2CH-
(CH3)2), 63.08 (R-C), 79.37 (Boc C-O), 119.13 (CH2CHdCH2),
132.12 (CH2CHdCH2), 153.92 (Boc CdO), 178.81 (COOH);
FAB MS m/z 272 [M + H]+, 216 [M-(CH3)3C]+, 172 [M - (CH3)3-
COCO]+. Anal. (C14H25NO4) C, H, N.
COOH); 13C NMR (DEPT assignment, CDCl3)
δ 13.7
((CH2)3CH3), 22.3 and 26.0 (CH2(CH2)2CH3), 34.8 and 39.4
(CH2CHdCH2 and CH2(CH2)2CH3), 63.5 and 65.2 (OCH2CHd
CH2 and R-C), 117.4 and 118.9 (OCH2CHdCH2 and CH2CHd
CH2), 131.9 and 132.5 (CH2CHdCH2 and OCH2CHdCH2),
154.1 (Alloc CdO), 177.2 (COOH); FAB MS m/z 256 [M + H]+.
Anal. (C13H21NO4) C, H, N.
(2R)-N-(ter t-Bu t oxyca r b on yl)-2-a llyln or leu cin e (20).
(2R)-2-Allylnorleucine (18; 0.6 g, 3.5 mmol) was converted to
20 under the same reaction conditions described above for the
synthesis of 19. The product was obtained as a clear oil (0.84
g, 89% yield): [R]D +2.8 (c 1.6, MeOH); 1H NMR (CDCl3) δ
0.88 (t, J ) 6.7 Hz, 3 H, (CH2)3CH3), 1.15-1.35 (m, 4 H, CH2-
(CH2)2CH3), 1.43 (s, 9 H, Boc CH3), 1.75-1.83 (m, 1 H,
CH2(CH2)2CH3), 2.2-2.3 (m, 1 H, CH2(CH2)2CH3), 2.54-2.61
(m, 1 H, CH2CHdCH2), 2.9-3.1 (m, 1 H, CH2CHdCH2), 5.08-
5.14 (m, 2 H, CH2CHdCH2), 5.41 (br s, 1 H, NH), 5.63-5.72
(m, 1 H, CH2CHdCH2), 11.51 (br s, 1 H, COOH); 13C NMR
(CDCl3) δ 13.9 ((CH2)3CH3), 22.5 and 26.1 (CH2(CH2)2CH3),
28.3 (Boc CH3), 34.9 and 39.5 (CH2CHdCH2 and CH2(CH2)2-
CH3), 63.3 (R-C), 79.5 (Boc C-O), 119.1 (CH2CHdCH2), 132.4
(CH2CHdCH2), 154.1 (Boc CdO), 178.4 (COOH); FAB MS m/z
272 [M + H]+, 216 [M - (CH3)3C]+, 172 [M - (CH3)3COCO]+.
Anal. (C14H25NO4) C, H, N.
(2RS,4S)-2-[2(R)-[N-(ter t-Bu toxycar bon yl)am in o]-2-car -
boxy-4-m eth yl-1-p en tyl]th ia zolid in e-4-ca r boxylic Acid
Meth yl Ester (24). The procedures described for the prepara-
tion of 26 were followed. Reaction of 19 (2.3 g, 8.48 mmol) with
OsO4 (107 mg, 0.42 mmol) and NaIO4 (4.5 g, 21.2 mmol) in
THF/H2O (4:1, 150 mL) afforded 2.4 g of (2RS,4R)-4-[N-(tert-
butoxycarbonyl)amino]-4-(2-methylpropyl)-2-hydroxy-5-oxotet-
rahydrofuran (21) as a tan oil: 13C NMR (CDCl3) δ 96.8 (2-C);
FAB MS m/z 274 [M + H]+, 218 [M - (CH3)3C]+.
Reaction of the γ-hydroxybutyrolactone 21 (2.2 g, 8.05 mmol)
with NaHCO3 (0.67 g, 8.05 mequiv) and D-cysteine methyl
ester hydrochloride (1.4 g, 8.05 mmol) in H2O/95% EtOH (1:1,
60 mL) afforded 2.6 g (84% yield) of the crude thiazolidine
mixture 24 as a pink foam which was used without further
purification: FAB MS m/z 391 [M + H]+, 335 [M - (CH3)3C]+.
(2RS,4S)-2-[2(R)-[N-(ter t-Bu toxycar bon yl)am in o]-2-car -
boxy-1-h exyl]th ia zolid in e-4-ca r boxylic Acid Meth yl Es-
ter (25). The procedures described for the preparation of 26
were followed. Reaction of 20 (0.63 g, 2.3 mmol) with OsO4
(0.031 g, 0.12 mmol) and NaIO4 (1.3 g, 6.0 mmol) in THF/H2O
(4:1, 50 mL) afforded 1.0 g of crude (2RS,4R)-4-[N-(tert-
(2R)-2-Allylleu cin e (17). In a round-bottom flask equipped
with a stir bar, 15 (4.47 g, 17.5 mmol), PdCl2(PPh3)2 (0.246 g,
0.35 mmol), and H2O (1.6 mL, 97 mmol) were dissolved in CH2-
Cl2 (200 mL). To that mixture was added n-Bu3Sn (5.6 mL,
21.0 mmol) rapidly via a syringe, and the reaction was stirred
for 2 h at room temperature under Ar. The solvent was
removed under vacuum, and the residue was triturated with
Et2O producing a white precipitate. Filtration of the mixture
afforded 1.1 g (37% yield) of pure 17 as a white solid. The
filtrate was extracted with H2O (5 × 100 mL), and the
combined aqueous extracts were evaporated in vacuo. The
resulting solid was triturated with Et2O to give another 1.1 g
of product as a tan solid (74% combined yield): mp 266-267
1
°C dec; [R]D +51.6 (c 1.26, MeOH); H NMR (CD3OD) δ 0.94
(s, 3 H, (CH3)2), 0.95 (s, 3 H, (CH3)2), 1.62-1.68 (m, 1 H, CH2-
CH(CH3)2), 1.75-1.85 (m, 2 H, CH2CH(CH3)2 and CH2-
CH(CH3)2), 2.40 (dd, J ) 14.0 and 7.9 Hz, 1 H, CH2CHdCH2),
2.63 (dd, J ) 14.0 and 6.7 Hz, 1 H, CH2CHdCH2), 5.18-5.25
(m, 2 H, CH2CHdCH2), 5.73-5.87 (m, 1 H, CH2CHdCH2); 13
C
NMR (CD3OD) δ 23.3 (CH(CH3)2), 25.0 and 25.2 (CH(CH3)2),
43.4 and 45.9 (CH2CHdCH2 and CH2CH(CH3)2), 64.7 (R-C),
121.2 (CH2CHdCH2), 132.6 (CH2CHdCH2), 175.6 (COOH);
FAB MS m/z 172 [M + H]+. Anal. (C9H17NO2) C, H, N.
(2R)-2-Allyln or leu cin e (18). (2R)-N-(Allyloxycarbonyl)-2-
allylnorleucine (16; 2.1 g, 8.2 mmol) was converted to product
by the same procedure as that described above for 17 to give
the product as a tan solid in a total yield of 61%. This material
was successfully crystallized from MeOH/Et2O: mp 254-256
°C dec; [R]D +27.9 (c 1.4, MeOH); 1H NMR (CD3OD) δ 0.92 (t,
J ) 7.3 Hz, 3 H, (CH2)3CH3), 1.22-1.42 (m, 4 H, CH2(CH2)2-
CH3), 1.61-1.71 (m, 1 H, CH2(CH2)2CH3), 1.79-1.88 (m, 1 H,
CH2(CH2)2CH3), 2.42 (dd, J ) 14.0 and 7.9 Hz, 1 H, CH2CHd

Bicyclic Thiazolidine Lactam Peptidomimetics
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15 2983
butoxycarbonyl)amino]-4-butyl-2-hydroxy-5-oxotetrahydrofu-
ran (22) as a yellow semisolid: 13C NMR (CDCl3) δ 96.6 (2-C);
FAB MS m/z 274 [M + H]+, 218 [M - (CH3)3C]+.
Reaction of the γ-hydroxybutyrolactone 22 (0.63 g, 2.3) with
NaHCO3 (0.19 g, 2.3 mequiv) and D-cysteine methyl ester
hydrochloride (0.395 g, 2.3 mmol) in H2O/95% EtOH (1:1, 20
mL) afforded 0.59 g (66% yield) of the crude thiazolidine
mixture 25 as a clear oil which was used without further
purification: FAB MS m/z 391 [M + H]+, 335 [M - (CH3)3C]+.
CDCl3) δ 0.87 (d, J ) 6.1 Hz, 3 H, (CH3)2), 0.95 (d, J ) 6.1 Hz,
3 H, (CH3)2), 1.42 (s, 9 H, Boc CH3), 1.61 (dd, J ) 14.6 and 4.9
Hz, 1 H, CH2CH(CH3)2), 1.72-1.92 (m, 1 H, CH2CH(CH3)2),
2.14 (dd, J ) 14.6 and 7.3 Hz, 1 H, CH2CH(CH3)2), 2.56 (dd, J
) 12.2 and 8.6 Hz, 1 H, pro-S 7-CH2), 3.02 (dd, J ) 12.2 and
4.9 Hz, 1 H, pro-R 7-CH2), 3.51 (d, J ) 12.2, 1 Hz, pro-R 2-CH2),
3.67 (dd, J ) 12.2 and 7.3 Hz, 1 H, pro-S 2-CH2), 3.80 (s, 3 H,
OCH3), 4.32 (d, J ) 7.3 Hz, 1 H, 3-CH), 5.10 (dd, J ) 8.6 and
4.9 Hz, 1 H, 7a-CH), 5.47 (s, 1 H, NH); 13C NMR (CDCl3) δ
23.8, 23.9 and 24.5 (CH(CH3)2 and(CH(CH3)2), 28.3 (Boc CH3),
38.9 (CH2CH(CH3)2), 42.9 (7-C), 44.2 (2-C), 52.8 (OCH3), 56.9
(3-C), 64.7 and 65.3 (6-C and 7a-C), 79.4 (Boc C-O), 154.2 (Boc
CdO), 168.4 and 172.8 (5-CdO and CO2CH3); FAB MS m/z
373 [M + H]+, 317 [M - (CH3)3C]+, 273 [M - (CH3)3COCO]+.
Meth yl [3S-(3R,6R,7a R)]-6-[N-(Ben zyloxyca r bon yl)a m i-
n o]-6-ben zyl-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r -
boxyla te (29a ) a n d Meth yl [3S-(3R,6R,7a â)]-6-[N-(Ben zy-
loxyca r b on yl)a m in o]-6-b en zyl-5-oxo-(5H )-p yr r olo[2,1-
b]th ia zolid in e-3-ca r boxyla te (29b). To a solution of the
thiazolidine mixture 26 (0.5 g, 1.1 mmol) in freshly distilled
CH2Cl2 (100 mL) was added 2-chloro-1-methylpyridinium
iodide (0.27 g, 1.2 mmol). This was followed by addition of NEt3
(0.34 mL, 2.4 mmol) in 2 mL of CH2Cl2. The resulting solution
was heated at reflux for 8 h. The reaction mixture was allowed
to cool and then was washed with 10% citric acid (30 mL), 1
N NaHCO3 (30 mL), and brine (30 mL). The organic layer was
dried (MgSO4) and then stripped of solvent under vacuum to
give a crude mixture of the bridgehead carbon (7a-C) epimers
29a and 29b in a ratio of about 3:1, respectively. These epimers
were separated by silica gel chromatography (EtOAc/hexanes,
1:3f1:1).
(2RS,4S)-2-[2(R)-[N-(Ben zyloxyca r bon yl)a m in o]-2-ben -
zyl-2-ca r boxyeth yl]th ia zolid in e-4-ca r boxylic Acid Meth -
yl Ester (26). To a solution of compound 8 (3.26 g, 9.6 mmol)
in THF/H2O (4:1, 150 mL) was added OsO4 (170 mg, 0.67
mmol) under Ar. After 5 min of stirring, NaIO4 (5.1 g, 24 mmol)
was added as a suspension in H2O (8 mL) over 10 min. The
reaction mixture was stirred under Ar for 24 h and then
filtered to remove a white solid. The filtrate was concentrated
under vacuum and the residue partitioned between Et2O (100
mL) and 1 N NaHCO3 (200 mL). The aqueous layer was
acidified with 1 N HCl to pH 1 and then extracted with EtOAc
(3 × 250 mL). The extracts were dried (MgSO4) and then
stripped of solvent in vacuo to give 3.4 g of (2RS,4R)-4-[N-
(benzyloxycarbonyl)amino]-4-benzyl-2-hydroxy-5-oxotetrahy-
drofuran (23) as a white foam:13C NMR (CDCl3) δ 97.6 and
97.8 (2-C); FAB MS m/z 342 [M + H]+, 206 [M - PhCH2OCO]+.
The crude γ-hydroxybutyrolactone 23 (3.2 g, 9.37 mmol) in
95% EtOH (40 mL) was cooled to -20 °C in a salt/ice/water
bath. One equivalent of NaHCO3 (0.78 g, 9.37 mequiv) in H2O
(39 mL) was added followed by D-cysteine methyl ester
hydrochloride (1.6 g, 9.37 mmol). The pH of the resulting
solution was adjusted to around 6.5 with 2% NaHCO3. After
warming to room temperature, the mixture was stirred for 12
h. The reaction mixture was concentrated to about one-half
its volume, and the pH was adjusted to 6 with 1 N HCl. The
product was extracted into EtOAc (3 × 250 mL) with readjust-
ment of the pH to 6 between extractions. The combined
extracts were dried (MgSO4), and the solvent was evaporated
under vacuum to give 4.5 g of 26 as a pink foam which was
used without further purification: FAB MS m/z 459 [M + H]+.
Met h yl [3S-(3R,6R,7a R)]-6-[N-(ter t-Bu t oxyca r b on yl)-
a m in o]-6-(2-m eth ylp r op yl)-5-oxo-(5H)-p yr r olo[2,1-b]th ia -
zolid in e-3-ca r boxyla te (27a ) a n d Meth yl [3S-(3R,6R,7a â)]-
6-[N-(ter t-Bu toxyca r bon yl)a m in o]-6-(2-m eth ylp r op yl)-5-
oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r boxyla te (27b).
To a solution of the thiazolidine mixture 24 (2.6 g, 6.7 mmol)
in freshly distilled CH2Cl2 (400 mL) was added 2-chloro-1-
methylpyridinium iodide (1.9 g, 7.6 mmol). This was followed
by addition of NEt3 (2.1 mL, 15.2 mmol) in 15 mL of CH2Cl2.
The resulting solution was heated at reflux for 16 h. The
reaction mixture was allowed to cool and then was washed
with 200 mL of each of the following solutions: 10% citric acid,
1 N NaHCO3, and brine. The organic phase was dried (MgSO4),
and then the solvent was removed under vacuum to give a
crude mixture of the bridgehead carbon (7a-C) epimers 27a
and 27b in a ratio of 2-3:1, respectively. Chromatographic
separation of the epimers was accomplished with 3% MeOH
in benzene. The combined yield of both epimers after chroma-
tography was 2.3 g (92%).
29a : obtained 195 mg (41%) as a white solid which was
crystallized from EtOAc/hexanes; mp 142-143 °C; [R]D +69.9
1
(c 0.9, CHCl3); TLC Rf ) 0.52 (EtOAc/hexanes, 1:1); H NMR
(CDCl3) δ 2.56 (dd, J ) 14.7 and 2.4 Hz, 1 H, 7-CH2), 2.7-2.9
(m, 1 H, 7-CH2), 3.00-3.30 (m, 4 H, 2-CH2 and CH2Ph), 3.75
(s, 3 H, OCH3), 5.04-5.12 (m, 3 H, 3-CH and OCH2Ph), 5.29
(br s, 1 H, 7a-CH), 5.43 (s, 1 H, NH), 7.19-7.33 (m, 10 H, Ph);
13C NMR (CDCl3) δ 33.5 and 36.0 (7-C and CH2Ph), 42.4 (2-
C), 52.8 (OCH3), 57.8 (3-C), 62.4 (7a-C), 63.1 (6-C), 66.8 (OCH2-
Ph), 127.4, 128.1, 128.4, 128.6, 130.5, 134.4, and 135.8 (Ph),
154.6 (Cbz CdO), 169.6 and 174.1 (5-CdO and CO2CH3); FAB
MS m/z 441 [M + H]+. Anal. (C23H24N2O5S) C, H, N, S.
29b: obtained 66 mg (14%) as a clear oil which solidified in
EtOAc/hexanes; mp 147-148 °C; [R]D +15.2 (c 1.3, CHCl3);
1
TLC Rf ) 0.46 (EtOAc/hexanes, 1:1); H NMR (CDCl3) δ 2.66
(dd, J ) 13.4 and 8.6 Hz, 1 H, 7-CH2), 3.14-3.20 (m, 2 H, 7-CH2
and CH2Ph), 3.27-3.32 (apparent d, J ) 13.4 Hz, 2 H, pro-R
2-CH2 and CH2Ph), 3.42 (dd, J ) 13.4 and 7.3 Hz, 1 H, pro-S
2-CH2), 3.77 (s, 3 H, OCH3), 4.02 (dd, J ) 8.5 and 6.1 Hz, 1 H,
7a-CH), 4.17 (d, J ) 7.3 Hz, 1 H, 3-CH), 5.06-5.18 (m, 2 H,
OCH2Ph), 5.72 (s, 1 H, NH), 7.17-7.73 (m, 10 H, Ph); 13C NMR
(CDCl3) δ 38.8, 40.9, and 42.0 (7-C, CH2Ph, and 2-C), 52.8
(OCH3), 56.4 (3-C), 64.3 (7a-C), 66.6 (OCH2Ph), 66.9 (6-C),
127.4, 128.0, 128.1, 128.3, 128.5, 130.0, 134.7, and 136.2 (Ph),
154.8 (Cbz CdO), 168.2 and 171.0 (5-CdO and CO2CH3); FAB
MS m/z 441 [M + H]+. Anal. (C23H24N2O5S) C, H, N, S.
[3S-(3R,6R,7a R)]-6-[N-(ter t-Bu t oxyca r b on yl)a m in o]-6-
(2-m eth ylp r op yl)-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-
3-ca r boxa m id e (30). A mixture of bicyclic lactams 27a and
27b (1.9 g, 5.1 mmol) was treated with a concentrated solution
of methanolic ammonia (6 mL). The reaction vessel was sealed,
and the reaction was stirred for 1 h at room temperature. TLC
of the reaction showed complete conversion of the 7a-(R)-
epimer 27a to its amide, while the majority of the 7a-(S)-
epimer 27b remained unreacted. The solvents and excess
ammonia were evaporated in vacuo to give 1.8 g (95%) of a
white foam that was purified on a 3- × 42-cm flash silica gel
column (5% MeOH in benzene). Pure 30 was obtained as a
white foam in a 56% yield (1.1 g). From the column were also
obtained 0.48 g (25%) of 27b and a fraction corresponding to
a mixture of 30 and the amide conversion product of 27b (0.25
g, 25%): total mass balance 1.8 g (95%); mp 68-73 °C; [R]D
27a : clear oil; [R]D +128.0 (c 1.1, CHCl3); TLC Rf ) 0.30 (3%
1
MeOH in benzene); H NMR (CDCl3) δ 0.92 (d, J ) 6.0 Hz, 3
H, (CH3)2), 0.94 (d, J ) 6.0 Hz, 3 H, (CH3)2), 1.40 (s, 9 H, Boc
CH3), 1.62 (dd, J ) 14.7 and 6.0 Hz, 1 H, CH2CH(CH3)2), 1.73
(dd, J ) 14.7 and 4.8 Hz, 1 H, CH2CH(CH3)2), 1.81-1.89 (m,
1 H, CH2CH(CH3)2), 2.35-2.40 (m, 1 H, 7-CH2), 2.95 (m, 1 H,
7-CH2), 3.37 (apparent d, J ) 6.1 Hz, 2 H, 2-CH2), 3.74 (s, 3
H, OCH3), 4.89 (s, 1 H, NH), 4.99-5.02 (m, 1 H, 3-CH), 5.23-
5.35 (m, 1 H, 7a-CH); 13C NMR (C6D6) δ 23.9, 24.2 and 24.5
(CH(CH3)2 and CH(CH3)2), 28.3 (Boc CH3), 34.9 and 36.2 (7-C
and CH2CH(CH3)2), 46.1 (2-C), 52.0 (OCH3), 58.6 (3-C), 62.4
and 63.4 (6-C and 7a-C), 79.4 (Boc C-O), 154.6 (Boc CdO),
169.9 and 175.3 (5-CdO and CO2CH3); FAB MS m/z 373 [M +
H]+, 317 [M - (CH3)3C]+, 273 [M - (CH3)3COCO]+.
1
27b: clear oil; [R]D -46.9 (c 1.3, CHCl3); TLC Rf ) 0.36 (3%
+140.2 (c 1.0, CHCl3); H NMR (NOE difference assignment,
MeOH in benzene); 1H NMR (NOE difference assignment,
CDCl3) δ 0.90 (d, J ) 6.1 Hz, 3 H, (CH3)2), 0.97 (d, J ) 7.3 Hz,

2984 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15
Khalil et al.
3 H, (CH3)2), 1.36 (s, 9 H, Boc CH3), 1.52-1.69 (m, 1 H, CH2-
CH(CH3)2), 1.70-1.84 (m, 2 H, CH2CH(CH3)2 and CH2-
CH(CH3)2), 2.39 (dd, J ) 14.6 and 4.9 Hz, 1 H, pro-R 7-CH2),
2.62 (dd, J ) 14.6 and 8.6 Hz, 1 H, pro-S 7-CH2), 3.50 (dd, J
) 12.2 and 8.5 Hz, pro-S 2-CH2), 3.58 (dd, J ) 12.2 and 4.9
Hz, 1 H, pro-R 2-CH2), 4.79 (dd, J ) 8.6 and 4.9 Hz, 1 H, 3-CH),
5.14 (s, 1 H, NH), 5.22 (dd, J ) 8.6 and 4.9 Hz, 1 H, 7a-CH),
5.93 (br s, 1 H, cis NH2), 7.23 (br s, 1 H, trans NH2); 13C NMR
(DEPT assignment, CDCl3) δ 24.1 (CH(CH3)2), 23.1 and 24.4
(CH(CH3)2), 28.2 (Boc CH3), 36.1 and 36.7 (7-C and CH2CH-
(CH3)2), 45.6 (2-C), 57.2 (3-C), 62.7 (7a-C), 63.4 (6-C), 80.6 (Boc
C-O), 155.1 (Boc CdO), 171.8 and 172.7 (5-CdO and CON);
FAB MS m/z 358 [M + H]+, 302 [M - (CH3)3C]+, 258 [M -
(CH3)3COCO]+. Anal. (C16H27N3O4S) C, H, N, S.
128.1, 128.3, 128.5, 128.8, 130.3, 133.9, and 135.5 (Ph), 155.3
(Cbz CdO), 171.5 and 172.4 (CON); FAB MS m/z 426 [M +
H]+. Anal. (C22H23N3O4S) C, H, N, S.
[3S-(3R,6R,7a R)]-6-[[[1-(ter t-Bu toxyca r bon yl)-2(S)-p yr -
r olidin yl]car bon yl]am in o]-6-(2-m eth ylpr opyl)-5-oxo-(5H)-
p yr r olo[2,1-b]th ia zolid in e-3-ca r boxa m id e (33). Compound
30 (0.70 g, 1.96 mmol) was treated with excess HCl (4 N in
dioxane, 7 mL) for 18 h at room temperature. Excess HCl and
dioxane were removed under vacuum, and the resulting
residue was stored under high vacuum overnight and then
dissolved in DMF (20 mL). Boc-L-Pro-OH (2.1 g, 9.8 mmol) and
HOBt•H2O (1.3 g, 9.8 mmol) were added, and the solution was
cooled to -40 °C (CH3CN/CO2 bath). Et3N (0.4 mL, 2.9 mmol)
was added, followed by EDC‚HCl (1.9 g, 9.8 mmol) after 5 min
of stirring. The mixture was stirred for 4 days under an
atmosphere of Ar. Removal of DMF in vacuo gave a residue
that was partitioned between EtOAc (220 mL) and 10% citric
acid (150 mL). The organic layer was washed with 150 mL of
1 N NaHCO3 and brine. The organic layer was dried (MgSO4)
and then stripped of solvents under vacuum to give 1.8 g of a
white foam. The crude product was twice subjected to silica
gel chromatography (3- × 40-cm column, CH2Cl2/MeOH, 20:
1) to give 756 mg (85%) of pure product as a white foam which
was crushed to a white solid: mp 88-91 °C; [R]D +22.9 (c 1.2,
[3S-(3R,6R,7a R)]-6-[N-(ter t-Bu t oxyca r b on yl)a m in o]-6-
bu tyl-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r boxa m -
id e (31). To a solution of the thiazolidine mixture 25 (590 mg,
1.5 mmol) in freshly distilled CH2Cl2 (200 mL) was added
2-chloro-1-methylpyridinium iodide (375 mg, 1.7 mmol). This
was followed by addition of NEt3 (0.46 mL, 3.3 mmol) in 3 mL
of CH2Cl2. After 24 h of heating at reflux, the reaction mixture
was allowed to cool to room temperature. It was then washed
with 10% citric acid, 1 N NaHCO3, and brine. The organic
phase was dried over MgSO4. Removal of solvent under
vacuum gave a crude mixture of the 7a-C epimeric lactams
28a and 28b (Rf ) 0.55 and 0.50, respectively, 5% MeOH in
benzene). Chromatographic separation of these epimers with
3% MeOH in benzene was only partially successful. Fractions
from the column possessing both epimers were combined and
concentrated in vacuo, and the residue was treated with
concentrated ammonia in MeOH (5 mL) at room temperature.
After the reaction mixture stirred for 3 h, TLC showed virtual
disappearance of 28a , whereas 28b remained largely unre-
acted. Excess ammonia and MeOH were evaporated under
vacuum, and the resulting residue was chromatographed on
a 3- × 43-cm silica gel column (CH2Cl2/MeOH, 50:1f20:1)
affording 230 mg (43% over two steps) of pure 31 as a clear
oil: [R]D +128.1 (c 1.3, CHCl3); TLC Rf ) 0.46 (CH2Cl2/MeOH,
20:1), 0.22 (CH2Cl2/MeOH, 50:1); 1H NMR (NOE difference
assignment, CDCl3, 319 K) δ 0.89 (t, J ) 6.8 Hz, 3 H,
(CH2)3CH3), 1.27-1.37 (m, 4 H, CH2(CH2)2CH3), 1.35 (s, 9 H,
Boc CH3), 1.39-1.71 (m, 2 H, CH2(CH2)2CH3), 2.29 (dd, J )
14.6 and 4.3 Hz, 1 H, pro-R 7-CH2), 2.60 (dd, J ) 14.0 and 7.9
Hz, 1 H, pro-S 7-CH2), 3.48 (dd, J ) 11.6 and 8.5 Hz, pro-S
2-CH2), 3.58 (dd, J ) 11.6 and 4.9, 1 H, pro-R 2-CH2), 4.79
(dd, J ) 8.6 and 4.9 Hz, 1 H, 3-CH), 5.12 (br s, 1 H, NH), 5.20
(dd, J ) 7.9 and 4.3 Hz, 1 H, 7a-CH), 5.73 (br s, 1 H, cis NH2),
7.14 (br s, 1 H, trans NH2); 13C NMR (CDCl3) δ 13.7 ((CH2)3CH3),
22.6 and 25.2 (CH2(CH2)2CH3), 28.2 (Boc CH3), 36.1 and 36.3
(7-C and CH2(CH2)2CH3), 37.4 (2-C), 57.0 (3-C), 62.5 (7a-CH),
63.5 (6-C), 80.6 (Boc C-O), 155.1 (Boc CdO), 171.9 and 172.6
(5-CdO and CON); FAB MS m/z 358 [M + H]+, 302 [M -
(CH3)3C]+, 258 [M - (CH3)3COCO]+. Anal. (C16H27N3O4S) C,
H, N, S.
1
CHCl3); H NMR (CDCl3, 318 K) δ 0.93 (d, J ) 6.1 Hz, 3 H,
(CH3)2), 1.00 (d, J ) 6.1 Hz, 3 H, (CH3)2), 1.46 (s, 9 H, Boc
CH3), 1.54-1.69 (m, 1 H, CH2CH(CH3)2), 1.73-1.93 (m, 5 H,
CH2CH(CH3)2, CH2CH(CH3)2, Pro â-CH2, and Pro γ-CH2),
2.02-2.20 (m, 1 H, Pro â-CH2), 2.38 (dd, J ) 14.6 and 4.3 Hz,
1 H, 7-CH2), 2.56 (dd, J ) 14.6 and 7.8 Hz, 1 H, 7-CH2), 3.32-
3.36 (m, 2 H, Pro δ-CH2), 3.49 (dd, J ) 11.0 and 8.6 Hz, 1 H,
2-CH2), 3.61 (dd, J ) 11.6 and 4.9 Hz, 1 H, 2-CH2), 4.16 (m, 1
H, Pro R-CH), 4.80 (dd, J ) 8.6 and 4.9 Hz, 1 H, 3-CH), 5.23
(dd, J ) 7.8 and 4.3 Hz, 1 H, 7a-CH), 5.65 (br s, 1 H, NH),
7.25 (br s, 1 H, NH); 13C NMR (CDCl3) δ 23.1, 24.2, and 24.5
(CH(CH3)2 and CH(CH3)2), 27.3 (Pro γ-C), 28.3 (Boc CH3), 31.0
(Pro â-C), 36.2 and 37.8 (7-C and CH2CH(CH3)2), 45.6 (2-C),
47.1 (Pro δ-C), 57.3 (3-C), 59.5 (Pro R-C), 62.7 (7a-C), 63.4 (6-
C), 80.6 (Boc C-O), 156.0 (Boc CdO), 171.9, 172.3, and 172.4
(CON); FAB MS m/z 455 [M + H]+, 399 [M - (CH3)3C]+, 355
[M - (CH3)3COCO]+. Anal. (C21H34N4O5S) C, H, N, S.
[3S-(3R,6R,7a R)]-6-[[[1-(ter t-Bu toxyca r bon yl)-2(S)-p yr -
r olid in yl]ca r b on yl]a m in o]-6-b u t yl-5-oxo-(5H )-p yr r olo-
[2,1-b]th ia zolid in e-3-ca r boxa m id e (34). This compound
was prepared from bicyclic lactam 31 (0.15 mg, 0.42 mmol) in
the same manner as described for 33. The crude product was
twice subjected to silica gel chromatography (2.2- × 40-cm
column, CH2Cl2/MeOH, 20:1) to give 170 mg (89%) of pure
product as a white foam which was crushed to a white solid:
mp 93-96 °C; [R]D +20.2 (c 0.9, CHCl3); 1H NMR (CDCl3, 323
K) δ 0.92 (t, J ) 7.3 Hz, 3 H, (CH2)3CH3), 1.30-1.50 (m, 4 H,
Pro γ-CH2 and CH2(CH2)2CH3), 1.47 (s, 9 H, Boc CH3), 1.64-
1.87 (m, 5 H, CH2(CH2)CH3, Pro γ-CH2, and Pro â-CH2), 2.23
(m, 1 H, Pro â-CH2), 2.29 (dd, J ) 14.0 and 4.3 Hz, 1 H, 7-CH2),
2.52 (dd, J ) 14.0 and 7.9 Hz, 1 H, 7-CH2), 3.38 (m, 2 H, Pro
δ-CH2), 3.50 (dd, J ) 12.2 and 8.5 Hz, 1 H, 2-CH2), 3.61 (dd,
J ) 11.0 and 4.9 Hz, 1 H, 2-CH2), 4.23 (m, 1 H, Pro R-CH),
4.82 (dd, J ) 8.6 and 4.9 Hz, 1 H, 3-CH), 5.23 (dd, J ) 7.9 and
4.3 Hz, 1 H, 7a-CH), 5.57 (br s, 1 H, NH), 7.19 (br s, 1 H, NH);
13C NMR (DEPT assignment, CDCl3) δ 13.8 ((CH2)3CH3), 22.8,
24.7, and 25.2 (Pro γ-C and CH2(CH2)2CH3), 27.2 (Pro â-C),
28.3 (Boc CH3), 36.2, 36.9, and 37.4 (7-C, CH2(CH2)2CH3, and
2-C), 47.2 (Pro δ-C), 57.1 (3-C), 59.4 (Pro R-C), 62.6 (7a-C), 63.6
(6-C), 80.6 (Boc C-O), 156.1 (Boc CdO), 171.9, 172.2, and
172.4 (CON); FAB MS m/z 455 [M + H]+ 399 [M - (CH3)3C]+
355 [M - (CH3)3COCO]+. Anal. (C21H34N4O5S) C, H, N, S.
[3S-(3R,6R,7a R)]-6-[N-(Ben zyloxyca r b on yl)a m in o]-6-
ben zyl-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r boxa -
m id e (32). Bicyclic lactam 29a (165 mg, 0.375 mmol) in
CH2Cl2 (2 mL) was treated with a concentrated solution of
methanolic ammonia (4 mL). The reaction vessel was sealed,
and the reaction was stirred for 45 min at room temperature
at which time TLC showed complete disappearance of the
starting material. The solvents and excess ammonia were
evaporated in vacuo to give 160 mg (100%) of 32 as a white
foam. An analytical sample of 32 was obtained by silica gel
column chromatography (EtOAc/hexanes, 1:1f3:1): mp 138-
1
139 °C; [R]D +142.1 (c 2.1, CHCl3); H NMR (CDCl3) δ 2.53-
2.59 (m, 2 H, 7-CH2), 2.97 (d, J ) 13.4 Hz, 1 H, CH2Ph), 3.18
(d, J ) 13.4 Hz, 1 H, CH2Ph), 3.24-3.28 (m, 1 H, pro-S 2-CH2),
3.51 (dd, J ) 11.0 and 4.9 Hz, 1 H, pro-R 2-CH2), 4.83 (dd, J
) 8.5 and 4.9 Hz, 1 H, 3-CH), 4.98-5.09 (m, 2 H, OCH2Ph),
5.14 (m, 1 H, 7a-CH), 5.72 (s, 1 H, CbzNH), 6.10 (s, 1 H, cis
CONH2), 7.07 (s, 1 H, trans CONH2), 7.20-7.38 (m, 10 H, Ph);
13C NMR (CDCl3) δ 35.7 and 35.8 (7-C and CH2Ph), 42.8 (2-
C), 57.6 (3-C), 62.3 (7a-C), 64.1 (6-C), 67.2 (OCH2Ph), 127.7,
[3S-(3R,6R,7a R)]-6-[[[1-(ter t-Bu toxyca r bon yl)-2(S)-p yr -
r olid in yl]ca r bon yl]a m in o]-6-ben zyl-5-oxo-(5H)-p yr r olo-
[2,1-b]th ia zolid in e-3-ca r boxa m id e (35). The same general
procedure as that used to make 33 was used. The crude
product was twice subjected to silica gel chromatography (2.2-
× 40-cm column, CH2Cl2/MeOH, 20:1) to give 1.16 g (73%) of
pure 35: mp 232-233 °C; [R]D +93.8 (c 2.0, CHCl3); 1H NMR
(CDCl3, 328 K) δ 1.47 (s, 9 H, Boc CH3), 1.75-1.90 (m, 2 H,

Bicyclic Thiazolidine Lactam Peptidomimetics
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15 2985
Pro γ-CH2), 1.92-2.17 (m, 2 H, Pro â-CH2), 2.41 (dd, J ) 13.4
and 7.3 Hz, 1 H, 7-CH2), 2.54 (dd, J ) 13.4 and 4.9 Hz, 1 H,
7-CH2), 3.04 (d, J ) 13.4 Hz, 1 H, CH2Ph), 3.15 (d, J ) 13.4
Hz, 1 H, CH2Ph), 3.22-3.40 (m, 3 H, 2-CH2 and Pro δ-CH2),
3.57 (dd, J ) 12.2 and 4.9 Hz, 1 H, 2-CH2), 4.17 (m, 1 H, Pro
R-CH), 4.84 (dd, J ) 8.6 and 4.9 Hz, 1 H, 3-CH), 5.19 (dd, J )
7.3 and 4.9 Hz, 1 H, 7a-CH), 5.69 (br s, 1 H, NH), 7.22-7.38
(m, 7 H, CONH2 and Ph); 13C NMR (CDCl3, 328 K) δ 24.1 (Pro
γ-C), 28.3 (Boc CH3), 30.0 (Pro â-C), 35.9 and 37.2 (7-C and
CH2Ph), 43.3 (2-C), 47.0 (Pro δ-C), 57.7 (3-C), 60.6 (Pro R-C),
62.5 (7a-C), 63.6 (6-C), 80.9 (Boc C-O), 127.8, 129.0, 130.3,
and 134.2 (Ph), 155.0 (Boc CdO), 171.4, 171.5, and 172.8
(CON); FAB MS m/z 489 [M + H]+, 433 [M - (CH3)3C]+, 389
[M - (CH3)3COCO]+. Anal. (C23H31N4O5S) C, H, N, S.
7-CH2), 2.58 (dd, J ) 14.0 and 7.3 Hz, 1 H, 7-CH2), 2.86 (dd,
J ) 10.4 and 7.9 Hz, 1 H, Pro δ-CH2), 2.97-3.16 (m, 5 H, Pro
δ-CH2, 2-CH2, and CH2Ph), 4.22 (br s, 1 H, Pro R-CH), 4.60
(dd, J ) 7.3 and 3.7 Hz, 1 H, 3-CH), 5.07 (dd, J ) 7.3 and 3.7
Hz, 1 H, 7a-CH), 7.06 (s, 1 H, NH), 7.14-7.28 (m, 5 H, Ph),
7.41 (s, 1 H, NH), 8.61 (br s, 1 H, +NH2), 9.21 (s, 1 H, NH),
9.99 (br s, 1 H, +NH2); 13C NMR (DMSO-d6) δ 23.5 (Pro γ-C),
29.7 (Pro â-C), 35.1 and 35.5 (7-C and CH2Ph), 41.0 (2-C), 45.6
(Pro δ-C), 57.7 and 58.6 (3-C and Pro R-C), 62.4 (7a-C), 64.4
(6-C), 127.0, 128.2, 130.6, and 134.9 (Ph), 168.4, 170.5, and
171.6 (CON); FAB HRMS m/z 389.1666 (C19H24N4O3S + H+
requires 389.1650). Anal. (C19H24N4O3S‚HCl‚3H2O) C, H, N,
S, Cl.
X-r a y Diffr a ction . Colorless needles of 29a were grown
from EtOAc and hexanes. All measurements were made on a
Rigaku AFC6S diffractometer with graphite monochromated
Cu KR radiation (λ ) 1.54178 Å) with the ω - 2θ scan mode
at 173(1) K by using a Molecular Structure Corp. low-
temperature device. Intensities were corrected for Lorentz and
polarization effects. Equivalent reflections were merged, and
absorption effects were corrected using the psi-scan method.35
The structure was solved by direct methods using SHELXS8636
and refined using the TEXSAN structure analysis package.37
All non-hydrogen atoms were refined anisotropically. Hydro-
gens bonded to carbon atoms were placed in calculated
positions (0.95 Å) and were not refined. Hydrogens bonded to
heteroatoms (excluding the water atom) were located in
difference Fourier maps, and their positional parameters were
refined.
[3S-(3R,6R,7a R)]-6-[[2(S)-P yr r olid in ylca r bon yl]a m in o]-
6-(2-m eth ylp r opyl)-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolidin e-
3-ca r boxa m id e Hyd r och lor id e (3‚HCl). Peptidomimetic 33
(46 mg, 0.116 mmol) was treated with 4 N HCl in dioxane (3
mL) at 0 °C under an atmosphere of Ar. After the reaction
stirred for 45 min, TLC showed complete disappearance of
starting material. The solvent and excess HCl were removed
in vacuo, and the residue was twice suspended in CH2Cl2 and
evaporated to dryness under high vacuum. The resulting white
solid was dissolved in H2O (∼0.5 mL) and lyophilized to give
the pure product as a hygroscopic white solid (30 mg, 65%):
1
mp 131-135 °C; [R]D +94.3 (c 3.0, MeOH); H NMR (COSY
assignment, CDCl3) δ 0.93 (d, J ) 6.1 Hz, 3 H, (CH3)2), 0.97
(d, J ) 6.1 Hz, 3 H, (CH3)2), 1.35 (dd, J ) 14.6 and 6.1 Hz, 1
H, CH2CH(CH3)2), 1.63-1.72 (m, 1 H, CH2CH(CH3)2), 1.85-
193 (m, 2 H, Pro γ-CH2), 1.98 (dd, J ) 13.4 and 6.1 Hz, 1 H,
7-CH2), 2.04-2.21 (m, 2 H, Pro â-CH2 and Pro γ-CH2), 2.60-
2.69 (m,1 H, Pro â-CH2), 2.89-2.96 (m, 2 H, CH2CH(CH3)2 and
2-CH2), 3.31 (dd, J ) 13.4 and 6.1 Hz, 1 H, 7-CH2), 3.60-3.81
(m, 3 H, 2-CH2 and Pro δ-CH2), 4.37 (m, 1 H, Pro R-CH), 5.14
(d, J ) 6.1, 1 H, 3-CH), 5.78 (t, J ) 6.1, 1 H, 7a-CH), 7.17 (s,
1 H, NH), 7.84 (br s, 1 H, +NH), 8.00 (br s, 1 H, +NH), 8.28 (s,
1 H, NH), 9.81 (s, 1 H, NH); 13C NMR (CDCl3) δ 23.3, 23.5,
24.4, and 24.7 (CH(CH3)2, Pro γ-C, and CH(CH3)2), 31.2 and
32.5 (Pro â-C and CH2CH(CH3)2), 39.5 (2-C), 43.6 (7-C) 47.8
(Pro δ-C), 56.4 (3-C), 59.7 (Pro R-C), 60.4 (7a-C), 66.0 (6-C),
169.9, 170.8, and 170.9 (CON); FAB HRMS m/z 355.1829
(C16H26N4O3S + H+ requires 355.1806). Anal. (C16H26N4O3S•HCl)
C, H, N, S, Cl.
6-Hyd r oxyd op a m in e-Lesion ed Ra t Mod el of Hem ip a r -
k in son ism . Male Sprague-Dawley rats with weights ranging
from 275 to 300 g were procured from the Charles River
breeding facility at St. Constant, Quebec, Canada. Prior to
surgery, all animals were given 1 week to acclimate to a 12-h
light and 12-h dark photoperiodicity cycle, during which time
they were handled in order to minimize stress evoked from
naive human contact. Before any drug administration each
animal was accurately weighed to ensure all dosages were
precise. To spare noradrenergic pathways from the neurotoxic
effects of 6-hydroxdopamine, the rats were pretreated with
desipramine hydrochloride (15 mg/kg, ip) 30 min prior to
anesthesia with Somnotol (sodium pentobarbitol) given at 50
mg/kg, ip. Atropine (0.03 mg/kg, ip) was given to decrease
respiratory tract mucous secretions. The foot pinch reflex test
was used to determine the effectiveness of the anesthetic
(unresponsive animals were considered to be anesthetized).
Animals not passing this behavioral measure were adminis-
tered a mixture of 100 mg/kg ketamine and 20 mg/kg xylazine
in a volume of ∼0.01-0.05 mL/kg. Once full anesthesia of the
animals was reached, they were then prepared by shaving the
fur off the top of their scalps and disinfecting the area with
iodine. To prevent unnecessary moisture loss from the eyes,
the ocular lubricant Lacri-Lube (Allergan Inc., Markham,
Ontario) was applied topically to the eyes of each animal.
Animals were then secured in the stereoscopic apparatus, and
a single skull incision running form anterior to posterior was
made. The membrane covering the surface of the skull (peri-
osteum) was then clipped back exposing the surface of the
skull. A Hamilton syringe was centered on bregma, and three-
dimensional coordinates for bregma were determined in ac-
cordance with the atlas of Paxinos and Watson (A/P -4.8, L/M
+1.8, D/V -7.5).38 Once the coordinates were marked a hand
drill was used to make a small access hole into the surface of
the skull so that the 30-gauge Hamilton syringe could easily
enter into the brain. 6-Hydroxydopamine was infused at a dose
of 8 µg in 4 µL of 0.9% isotonic saline with 0.1% ascorbic acid
directly into the substantia nigra at a rate of 1 µL/min. The
syringe was removed after a 4-min latency period to ensure
that the placement of the drug was effective. The surgical
wound was then closed using surgical staples, and an antibiotic
agent Vetropolycin (J anssen, North York, Ontario) was applied
topically. Postoperative care included careful observation of
vital signs and the subcutaneous administration of 10 mL of
0.9% isotonic saline.
[3S-(3R,6R,7aR)]-6-[[(2(S)-P yr r olidin yl)car bon yl]am in o]-
6-bu tyl-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r boxa -
m id e Hyd r och lor id e (4•HCl). Peptidomimetic 34 (31 mg,
0.068 mmol) was deprotected by the same procedure used
above to make 3. The pure product was obtained as a
lyophilized hygroscopic white solid (25 mg, 91%): [R]D +109.8
(c 2.5, MeOH); 1H NMR (COSY assignment, CDCl3) δ 0.88 (t,
J ) 7.0 Hz, 3 H, (CH2)3CH3), 1.25-1.36 (m, 4 H, CH2(CH2)2-
CH3), 1.51-1.60 (m, 1 H, CH2CH(CH3)2), 1.89-1.96 (m, 2 H,
Pro γ-CH2 and 7-CH2), 2.01-2.12 (m, 1 H, Pro â-CH2) 2.14-
2.25 (m, 1 H, Pro γ-CH2), 2.63-2.72 (m, 1 H, Pro â-CH2), 2.82-
3.00 (m, 2 H, CH2(CH2)CH3 and 2-CH2), 3.17 (dd, J ) 12.8
and 4.9 Hz, 1 H, 7-CH2), 3.5-3.8 (m, 3 H, 2-CH2 and Pro
δ-CH2), 4.38 (m, 1 H, Pro R-CH), 5.11 (d, J ) 6.0, 1 H, 3-CH),
5.84 (t, J ) 6.0, 1 H, 7a-CH), 7.14 (s, 1 H, NH), 7.80 (br s, 1 H,
+NH), 8.09 (br s, 1 H, +NH), 8.30 (s, 1 H, NH), 9.78 (s, 1 H,
NH); 13C NMR (CD3OD) δ 14.1 ((CH2)3CH3), 23.8, 24.9, and
26.6 (Pro γ-C and CH2(CH2)2CH3), 31.2 (Pro â-C), 36.5, 37.2,
and 37.8 (7-C, CH2(CH2)2CH3, and 2-C), 47.6 (Pro δ-C), 59.1
(3-C), 60.9 (Pro R-C), 64.4 (7a-C), 65.5 (6-C), 170.1, 174.0, and
174.4 (CON); FAB HRMS m/z 355.1823 (C16H26N4O3S + H+
requires 355.1806). Anal. (C16H26N4O3S‚HCl‚H2O) C, H, N, S,
Cl.
[3S-(3R,6R,7a R)]-6-[[2(S)-P yr r olid in ylca r bon yl]a m in o]-
6-ben zyl-5-oxo-(5H)-p yr r olo[2,1-b]th ia zolid in e-3-ca r box-
a m id e Hyd r och lor id e (5‚HCl). Peptidomimetic 35 (154 mg,
0.32 mmol) was deprotected by the same procedure used above
to make 3. The pure product was obtained as a lyophilized
hygroscopic white solid (130 mg, 97%): mp 150-160 °C; [R]D
1
+167.8 (c 0.82, MeOH); H NMR (DMSO-d6) δ 1.71-1.88 (m,
3 H, Pro γ-CH2 and â-CH2), 2.24-2.30 (m, 2 H, Pro â-CH2 and

2986 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15
Khalil et al.
(11) Bhargava, H. N. Effects of Prolyl-leucyl-glycinamide and Cyclo-
(leucyl-glycine) on the Supersensitivity of Dopamine Receptors
in the Brain Induced by Chronic Administration of Haloperidol
to Rats. Neuropharmacology 1984, 23, 439-444.
(12) Yu, K.-L.; Rajakumar, G.; Srivastava, L. K.; Mishra, R. K.;
J ohnson, R. L. Dopamine Receptor Modulation by Conforma-
tionally Constrained Analogues of Pro-Leu-Gly-NH2. J . Med.
Chem. 1988, 31, 1430-1436.
(13) Sreenivasan, U.; Mishra, R. K.; J ohnson, R. L. Synthesis and
Dopamine Receptor Modulating Activity of Lactam Conforma-
tionally Constrained Analogues of Pro-Leu-Gly-NH2. J . Med.
Chem. 1993, 36, 256-263.
(14) Subasinghe, N. L.; Bontems, R. J .; McIntee, E.; Mishra, R. K.;
J ohnson, R. L. Bicyclic Thiazolidine Lactam Peptidomimetics of
the Dopamine Receptor Modulating Peptide Pro-Leu-Gly-NH2.
J . Med. Chem. 1993, 36, 2356-2361.
(15) Genin, M. J .; Mishra, R. K.; J ohnson, R. L. Dopamine Receptor
Modulation by a Highly Rigid Spiro Bicyclic Peptidomimetic of
Pro-Leu-Gly-NH2. J . Med. Chem. 1993, 36, 3481-3483.
(16) Baures, P. W.; Ojala, W. H.; Gleason, W. B.; Mishra, R. K.;
J ohnson, R. L. Design, Synthesis, X-ray Analysis, and Dopamine
Receptor-Modulating Activity of Mimics of the “C5” Hydrogen-
Bonded Conformation in the Peptidomimetic 2-Oxo-3(R)-[(2(S)-
pyrrolidinylcarbonyl)amino]-1-pyrrolidineacetamide. J . Med.
Chem. 1994, 37, 3677-3683.
To evaluate the effectiveness of the lesioning procedure,
testing of the animals began 10 days postoperatively. Apo-
morphine at a dose of 0.25 mg/kg, ip was administered to each
animal. Only those animals exhibiting 40 contralateral rota-
tions over a 10-min period were admitted as viable experi-
mental subjects. All rats were observed in the same round,
transparent, flat-bottom bowl. A 5-min habituation period in
the bowl was allocated to each animal before each trial.
Injections were administered after the 5-min habituation,
followed by a further 5-min latency period (from time of drug
administration) in order to ensure adequate time for physi-
ological distribution of the drugs to the target sites of action.
Rotational counts were then taken at 5-min intervals. This
process was repeated 4 times after at a 48-h washout period
between each injection. This established a stable individual
control measure for each animal.
Testing of peptidomimetics 2-5 was analogous to the
apomorphine protocol described above. Within this experi-
mental design, each individual peptidomimetic’s ability to
modify the rotational response to apomorphine was evaluated.
Injection of a peptidomimetic was administered directly before
apomorphine.
(17) Baures, P. W.; Ojala, W. H.; Costain, W. J .; Ott, M. C.; Pradhan,
A.; Gleason, W. B.; Mishra, R. K.; J ohnson, R. L. Design,
Synthesis, and Dopamine Receptor-Modulating Activity of Dike-
topiperazine Peptidomimetics L-Prolyl-L-leucylglycinamide. J .
Med. Chem. 1997, 40, 3594-3600.
(18) J ohnson, R. L.; Bontems, R. J .; Yang, K. E.; Mishra, R. K.
Synthesis and Biological Evaluation of Analogues of Pro-Leu-
Gly-NH2 Modified at the Leucyl Residue. J . Med. Chem. 1990,
33, 1828-1832.
(19) Karady, S.; Amato, J . S.; Weinstock, L. M. Enantioretentive
Alkylation of Acyclic Amino Acids. Tetrahedron Lett. 1984, 25,
4337-4340.
(20) Cheng, H.; Keitz, P.; J ones, J . B. Design and Synthesis of a
Conformationally Restricted Cysteine Protease Inhibitor. J . Org.
Chem. 1994, 59, 7671-7676.
To ensure results were not generated by an enhanced
sensitivity to increasing doses of drugs, a Latin Square design
was implemented. Rotational counts for each animal were then
generated into a percent change compared to the individual
control apomorphine-induced rotations. Results were analyzed
by a one-way analysis of variance (ANOVA), followed by a post
hoc comparison (Dunnett’s multiple comparison test). Apart
from producing an increase in the number of rotations over a
10-min period, administration of peptidomimetics 2-5 also
yielded a tendency toward stereotypical behaviors (e.g., groom-
ing of the contralateral side of body).8
Ack n ow led gm en t. This work was supported in part
by an NIH grant (NS20036) to R.L.J . The authors thank
Tom Crick for mass spectral analysis.
(21) Seebach, D.; Fadel, A. N,O-Acetals from Pivaldehyde and Amino
Acids for the R-Alkylation with Self-Reproduction of the Center
of Chirality. Enolates of 3-Benzoyl-2-(tert-butyl)-1,3-oxazolidin-
5-ones. Helv. Chim. Acta 1985, 68, 1243-1250.
(22) Smith, A. B., III; Guzman, M. C.; Sprengeler, P. A.; Keenan, T.
P.; Holocomb, R. C.; Wood, J . L.; Carroll, P. J .; Hirschmann, R.
De Novo Design, Synthesis, and X-ray Structures of Pyrrolinone-
Based â-Strand Peptidomimetics. J . Am. Chem. Soc. 1994, 116,
9947-9962.
Su p p or tin g In for m a tion Ava ila ble: X-ray crystallo-
graphic data including tables of positional parameters, bond
distances, and bond angles for 29a . This information is
available free of charge via the Internet at http://pubs.acs.org.
(23) Smith, A. B., III; Holocomb, R. C.; Guzman, M. C.; Keenan, T.
P.; Sprengeler, P. A.; Hirschmann, R. An Effective Synthesis of
Scalemic 3,5,5-Trisubstituted Pyrrolin-4-ones. Tetrahedron Lett.
1993, 34, 63-66.
(24) Dangels, O.; Guibe´, F.; Balavoine, G.; Lavielle, S.; Marquet, A.
Selective Cleavage of the Allyl and Allyloxycarbonyl Groups
through Palladium-Catalyzed Hydrostannolysis with Tributytin
Hydride. Application to the Selective Protection-Deprotection of
Amino Acid Derivatives and in Peptide Synthesis. J . Org. Chem.
1987, 52, 4984-4993.
Refer en ces
(1) Celis, M. E.; Taleisnik, S.; Walter, R. Regulation of Formation
and Proposed Structure of the Factor Inhibiting the Release of
Alpha-MSH. Proc. Natl. Acad. Sci. U.S.A. 1971, 68, 1428-1433.
(2) Kostrzewa, R. M.; Kastin, A. J .; Sobrain, S. K. Potentiation of
Apomorphine Action in Rats by L-Prolyl-L-leucyl-glycinamide.
Pharmacol. Biochem. Behav. 1978, 9, 375-378.
(3) Chiu, S.; Paulose, C. S.; Mishra, R. K. Effects of L-Prolyl-L-leucyl-
glycinamide (PLG) on Neuoleptic-induced Catalepsy and Dopam-
ine/Neuroleptic Receptor Bindings. Peptides 1981, 2, 105-111.
(4) Chiu, S.; Paulose, C. S.; Mishra, R. K. Neuroleptic Drug-induced
Dopamine Receptor Supersensitivity: Antagonism by L-Prolyl-
L-leucyl-glycinamide. Science 1981, 214, 1261-1262.
(25) Genin, M. J .; J ohnson, R. L. Design, Synthesis, and Conforma-
tional Analysis of a Novel Spiro-Bicyclic System as a Type II
â-Turn Peptidomimetic. J . Am. Chem. Soc. 1992, 114, 8778-
8783.
(5) Smith, J . R.; Morgan, M. The Effects of PLG on Drug Induced
Rotation in Rats. Gen. Pharmacol. 1982, 13, 203-207.
(6) Srivastava, L. K.; Bajwa, S. B.; J ohnson, R. L.; Mishra, R. K.
Interaction of L-Prolyl-L-Leucyl-Glycinamide with Dopamine D2
Receptor: Evidence for Modulation of Agonist Affinity States
in Bovine Striatal Membranes. J . Neurochem. 1988, 50, 960-
968.
(7) Ott, M. C.; Mishra, R. K.; J ohnson, R. L. Modulation of
Dopaminergic Neurotransmission in the 6-Hydroxydopamine
Lesioned Rotational Model by Peptidomimetic Analogues of
L-Prolyl-L-leucyl-glycinamide. Brain Res. 1996, 737, 287-291.
(8) Ungerstedt, U. Postsynaptic Supersensitivity After 6-Hydroxy-
dopamine Induced Degeneration of the Nigro-striatal Dopamine
System. Acta Physiol. Scand. 1971, 367 (Suppl.) 69-93.
(9) Mishra, R. K.; Marcotte, E. R.; Chugh, A.; Barlas, C.; Whan, D.;
J ohnson, R. L. Modulation of Dopamine Receptor Agonist-
Induced Rotational Behaviour in 6-OHDA-Lesioned Rats by a
Peptidomimetic Analogue of Pro-Leu-Gly-NH2 (PLG). Peptides
1997, 18, 1209-1217.
(26) Huang, H.; Iwasawa, N.; Mukaiyama, T. A. A Convenient
Method for the Construction of â-Lactam Compounds from
â-Amino Acids. Chem. Lett. 1984, 1465-1466.
(27) The authors have deposited atomic coordinates for this structure
with the Cambridge Crystallographic Data Centre. The deposi-
tion number for 29a is CCDC 127423. The coordinates can be
obtained, upon request, from the Director, Cambridge Crystal-
lographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
U.K., or via the Internet at http://www.ccdc.cam.ac.uk.
(28) Chou, P. Y.; Fasman, G. D. â-Turns in Peptides. J . Mol. Biol.
1977, 115, 135-175.
(29) Slusarchyk, W. A.; Robl, J . A.; Taunk, P. C.; Asaad, M. M.; Bird,
E.; DiMarco, J .; Pan, Y. Dual Metalloprotease Inhibitors. V.
Utilization of Bicyclic Azepinonethiazolidines and Azepinonetet-
rahydrothiazines in Constrained Peptidomimetics of Mercap-
toacyl Dipeptides. Bioorg. Med. Chem. Lett. 1995, 5, 753-758.
(30) Mishra, R. K.; Chiu, P.; Mishra, C. P. Pharmacology of L-Prolyl-
L-Leucyl-Glycinamide (PLG): A Review. Methods Find. Exp.
Clin. Pharmacol. 1983, 5, 203-233.
(10) Rajakumar, G.; Naas, F.; J ohnson, R. L.; Chiu, S.; Yu, K.-L.;
Mishra, R. K. Down Regulation of Haloperidol-induced Striatal
Dopamine Receptor Supersensitivity by Active Analogues of
L-Prolyl-L-leucyl-glycinamide (PLG). Peptides 1987, 8, 855-861.
(31) Kostrzewa, R. M.; Kastin, A. J .; Spirtes, M. A. R-MSH and MIF-1
Effects on Catecholamine Levels and Synthesis in Various Rat
Brain Areas. Pharmacol. Biochem. Behav. 1975, 3, 1017-1023.

Bicyclic Thiazolidine Lactam Peptidomimetics
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 15 2987
(32) Kostrzewa, R. M.; Spirtes, M. A.; Klava, M. W.; Christensen, C.
W.; Kastin, A. J .; J oh, T. H. Effects of L-Prolyl-L-leucyl-glycine
Amide (MIF-1) on Dopaminergic Neurons. Pharmacol. Biochem.
Behav. 1976, 5 (Suppl. 1), 125-127.
(35) North, A. C. T.; Phillips, D. C.; Mathews, F. S. A Semiempirical
Method of Absorption Correction. Acta Crystallogr. 1968, 24A,
351-359.
(36) Sheldrick, G. M. SHELXS86. Program for the Solution of Crystal
Structures; University of Go¨ttingen, Germany; 1986.
(37) TEXSAN-TEXRAY Structure Analysis Package; Molecular Struc-
ture Corp., 3200 Research Forest Dr., The Woodlands, TX 77381;
1985.
(33) Kostrzewa, R. M.; Fukushima, H.; Harston, C. T.; Perry, K. W.;
Fuller, R. W.; Kastin, A. J . Striatal Dopamine Turnover and
MIF-1. Brain Res. Bull. 1979, 4, 799-802.
(34) Torre, E.; Celis, M. E.; Chiocchio, S. A. R-MSH and Pro-Leu-
Gly-NH2 (PLG; MIF-1): Influence on Dopamine (DA) Uptake
in Crude Synaptosomal Preparations From Rat Mediobasal
Hypothalamus (MBH) and Caudate Putamen (CP). Peptides
1984, 5, 669-674.
(38) Paxinos, G.; Watson, C. The Rat Brain in Stereotaxic Coordi-
nates; Acedemic Press: Toronto, 1986.
J M990140N