Synthesis of di- and tripeptide analogues containing α-ketoamide as a new core structure for inhibition of HIV-1 protease

Article abstract of DOI:10.1016/S0223-5234(00)00187-2

Di- and tripeptide analogues containing α-ketoamide as a new core structure and incorporating allophenylnorstatine (Apns) as a transition state mimic, were designed and synthesized in the hope of obtaining a novel structural type of HIV-1 protease inhibitors. The immediate precursor, Apns-Thz-NHBu(t) was prepared by coupling of Boc-Apns with N-tert·butyl Thz-4-carboxamide hydrochloride. Removal of Boc group followed by coupling with the respective α-ketoacid residue (P2) gave the desired dipeptides (8-12) in almost quantitative yields. The α-keto tripeptides (18-21) were obtained by oxidation of the hydroxyl group of Apns (PI) in the appropriate tripeptide, iQOA-Val-Apns-(un)substituted Thz(Oxa)-NHBu₁ with DMSO/DCC. Preliminary evaluation of the activity of the synthesized derivatives was determined as percentage of enzyme inhibition at 5 μM and 50 nM levels of the di- and tripeptides respectively. The α-ketoamides displayed a significant enhanced potency relative to their parent isosteres as inhibitors of HIV-1 protease and are shown to be a promising new core structure for the development of enzyme inhibitors. A quantitative approach was attempted, using an LFE model, correlating the effect of structural modification and HIV-1 protease inhibition activity of the prepared dipeptides. The result indicates the contribution of the torsion angle by 84% to the activity of the inhibitors. (C) 2000 Editions scientifiques et medicales Elsevier SAS.

Full text of DOI:10.1016/S0223-5234(00)00187-2

Eur. J. Med. Chem. 35 (2000) 887–894
´
© 2000 Editions scientifiques et me´dicales Elsevier SAS. All rights reserved
Original article
PII: S0223-5234(00)00187-2/FLA
Synthesis of di- and tripeptide analogues containing a-ketoamide as a new core
structure for inhibition of HIV-1 protease
Mahmoud M. Shehaa, Nadia M. Mahfouza*, Hoda Y. Hassana, Adel F. Youssefa,
Tsutomu Mimotob, Yoshiaki Kisoc
aDepartment of Pharmaceutical Medicinal Chemistry, Faculty of Pharmacy, Assiut University, Assiut 71526, Egypt
bPharmaceutical & Biotechnology Laboratory, Japan Energy Co., Kyoto, Japan
cDepartment of Medicinal Chemistry Kyoto Pharmaceutical University, Yamashinaku, Kyoto 607-8414, Japan
Received 6 July 1999; revised 14 March 2000; accepted 18 March 2000
Abstract – Di- and tripeptide analogues containing a-ketoamide as a new core structure and incorporating allophenylnorstatine
(Apns) as a transition state mimic, were designed and synthesized in the hope of obtaining a novel structural type of HIV-1 protease
inhibitors. The immediate precursor, Apns-Thz-NHBut was prepared by coupling of Boc-Apns with N-tert·butyl Thz-4-carboxamide
hydrochloride. Removal of Boc group followed by coupling with the respective a-ketoacid residue (P2) gave the desired dipeptides
(8–12) in almost quantitative yields. The a-keto tripeptides (18–21) were obtained by oxidation of the hydroxyl group of Apns (PI)
in the appropriate tripeptide, iQOA-Val-Apns-(un)substituted Thz(Oxa)-NHBut with DMSO/DCC. Preliminary evaluation of the
activity of the synthesized derivatives was determined as percentage of enzyme inhibition at 5 mM and 50 nM levels of the di- and
tripeptides respectively. The a-ketoamides displayed a significant enhanced potency relative to their parent isosteres as inhibitors of
HIV-1 protease and are shown to be a promising new core structure for the development of enzyme inhibitors. A quantitative
approach was attempted, using an LFE model, correlating the effect of structural modification and HIV-1 protease inhibition
activity of the prepared dipeptides. The result indicates the contribution of the torsion angle by 84% to the activity of the inhibitors.
´
© 2000 Editions scientifiques et me´dicales Elsevier SAS
dipeptides / tripeptides / a-ketoamides / HIV-1 protease / inhibitory activity / QSAR
1. Introduction
type HIV-protease and synthetic HIV-protease ana-
logues [14, 15]. In addition, several small size
dipeptides containing HMC core of Apns, e.g. KNI-
413, KNI-549, KNI-727, KNI-577 and KNI-764 were
recently synthesized and exhibited excellent HIV-
protease inhibitory activity and were characterized by
low toxicity and good pharmacokinetic properties [14,
15].
Oxidation of the hydroxyl group of the HMC in
Phe-Pro resulted in the formation of an a-ketoamide
core. a-Ketoamide isosteres have been successively
used in the design of inhibitors for metalloprotein
(like calcium activated cysteine protease, calpains)
[16] and as a promising new core for the synthesis of
HIV-1 protease inhibitors [17, 18]. By analogy a-ke-
toamides were prepared by oxidation of the HMC
group in the norstatine-based HIV-1 protease in-
hibitors. The activity of the resultant a-ketoamides
A number of human immunodeficiency virus-1
protease (HIV-1 PR) inhibitors have been designed
on the concept of a transition state mimic. One of
these strategies was the insertion of hydroxymethyl-
carbonyl (HMC) core of allophenyl norstatine [Apns,
(2S, 3S)-3-ainino-2-liydroxy-4-phenylbutyric acid] as
a nonhydrolyzable isostere with the tetrahedral ge-
ometry at the Phe-Pro scissile amide bond. This de-
sign led to synthesis of many HIV-1 protease
inhibitors containing Apns [1–10]. Among them the
tripeptide KNI-272 (figure 1) is a highly selective and
superpotent HIV-1 protease inhibitor (Ki=5.5 PM)
currently in clinical study [11–15]. Also KNI-241
(figure 1) showed strong inhibition against both wild-
* Correspondence and reprints:
E-mail address: nmahfouz@aun.eun.eg (N.M. Mahfouz).

888
M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
Figure 1.
was either comparable with or improved upon the
corresponding HMC containing inhibitors [19–21].
The evident improvement of the activity of the a-ke-
toamides over the HMC precursors was found to be
due to the hydration of the ketone carbonyl of the
a-ketoamides to form gem-diol at the level of the
active site of the protease, as shown by NMR [22].
These results prompted us to synthesize a new series
of tripeptides incorporating Apns as a transition state
mimic, containing a-ketoamide at P1 site (18–21) and
to evaluate their activity as HIV-1 protease inhibitors.
Other a-ketoamides were also prepared at P2 site
(8–12) in the truncated dipeptide derivatives and their
activity was also tested. This study was carried out to
investigate the contribution of the a-ketoamide core
in the norstatine-based HIV-1 protease inhibitors to
their activity.
Figure 2.
provided 3:2 diastereomeric mixture of Boc-Apns
(2S,3S) and Boc-Pns (2R,3S), respectively, as moni-
tored by HPLC [9, 28]. The pure Boc-Apns (4) was
exclusively crystallized out from ether as dicyclohexy-
lamine (DCHA) salt. The purity of Boc-Apns (]
98%) was checked by HPLC and 1H-NMR. Its
1H-NMR spectrogram was superimposable on the
spectrogram of a reference sample.
The synthesis of the target dipeptides (figure 3) was
started by amidation of Boc thiazolidine carboxylic
acid (Boc-Thz-OH, 5) with tert-butylamine followed
by unmasking of the protected amino group to yield 6
which was coupled with C-activated Boc-Apns. Apns-
Thz-NHBut (7) was reacted with the appropriate a-
ketoacid to give the required compounds (8–12). The
dipeptide isostere, Val-Apns-Thz-NHBut (13) was
prepared by coupling of Apns-Thz-NHBut (7) with
C-activated Boc-Valine followed by N-deprotection
(figure 3). The target tripeptides (18–21) were ob-
tained (figure 4) by oxidation of the transition state
element, Apns of the previously synthesised, 3-{3(S)-
2. Chemistry
The preparation of the target peptide inhibitors
necessitate the synthesis of the transition state ele-
ment, allopheny1norstatine, Apns [3, 23–25]. The
most simple diastereoselective method for the prepa-
ration of N-tert-butyloxycarbonyl (Boc)-Apns (figure
2) was that analogous to the preparation of allocyclo-
[N-(5-isoquinolinyloxyacetyl)- -valinyl]-amino-2(S)-
L
hydroxy-4-phenylbutanoyl}-N(tert-butyl)-(un)substi-
tuted thiazolidine (oxazolidine)-4(R)-carboxamide,
IQOA-Val-Apns-[(un)substituted-Thz(Oxa)]NHBut,
(14–17) [29] to the corresponding a-ketoamides using
dimethyl sulfoxide and dicyclohexylcarbodiimide
(DMSO/DCC) as reported for similar compounds
[17].
hexylnorstatine [26]. Boc- -phenylalaninol (1) [27, 28]
L
was converted to the corresponding aldehyde, Boc-
phenylalaninal (2), by controlled oxidation with sul-
fur trioxide. The aldehyde was hydrocyanated and
hydrolyzed to provide 3-amino-2-hydroxy-4-phenyl-
butyric acid (AHPBA) as HCl salt (3). N-Protection
of AHPBA by ditert-butyl pyrocarbonate (Boc)2O,

M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
889
Figure 3. Reagents: a) i – DCC, HOBt, NH2-But; ii – 4N HCl dioxane. b) i – EDC·HCl, HOBt, Et3N·Boc-Apns; ii – 4N HCl
dioxane; c) i – EDC·HCl, Hobt, Et3N, Valine or a-keto acid.
3. Results and discussion
a route to the rational design of a low molecular weight
peptidomimitics lacking P3 ligand and the resultant
dipeptides (8–12) were tested at mM level of inhibitor
concentration. In this series the a-ketoacid residues im-
parted a significant enhancement of protease inhibition
activity (table II) as shown by the 4-fold improvement
3.1. In 6itro HIV-1 protease inhibitory acti6ity
As a primary screen, di- and tripeptide analogues
containing a-ketoamides as a new structural core were
initially evaluated in vitro for inhibitory activity against
HIV-1 protease according to reported technique [8].
Tripeptides containing an HMC core in P1 site were
found to inhibit HIV-1 protease activity and mimic the
Phe-Pro transition state in HIV-1 protease substrate
[29]. In our research to improve the HIV-1 protease
inhibitory activity, the hydroxymethyl group of the most
active tripeptide derivatives, (14–17) was oxidized to
a-keto analogues. The oxidized compounds 18–20 were
about 5–10% more active than the corresponding
hydroxymethyl containing compounds 14–16, respec-
tively. While oxidation of the hydroxymethyl group in
compound 17 resulted in an equipment a-ketoamide 21
(table I).
It was found that, the ketone carbonyl of the a-ke-
toamide of certain peptides hydrated and formed gem-
diol at the level of the active site of protease, as
indicated by NMR studies in aqueous DMSO solution
[22]. The observed improvement of the activity of the
target a-ketoamides over the HMC precursors suggests
that at the active site the formed gem-diol most proba-
bly enhances hydrogen bonding potentialities, which
places the inhibitor in a more proper manner to be
recognized by the enzyme.
The encouraging results of the a-ketoamides already
tested at P1 residue in the tripeptides (18–21) suggested
Figure 4.

890
M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
Table I. HIV PR activity (%) of tripeptide (14–17) and their a-ketoamide analogues (18–21) at 50 nM concentration.
Compd. no.
Y
Z
R1
R2
Activity (%)
14
18
15
19
16
20
17
21
O
O
S
CHOH (S)
CO
CHOH (S)
CO
CHOH (S)
CO
CHOH (S)
CO
H
H
H
H
CH3
CH3
CH3
CH3
H
H
H
H
H
H
CH3
CH3
71.3
76.3
88.5
95.5
86.3
96.4
97.7
97.7
S
O
O
S
S
of activity of 9 over its isostere, Val-Apns-Thz-NHBut
13. This observation is consistent with that already
signalled in the tripeptide derivatives (18–21). As seen
from table II, the activity gain goes in the order a-oxo-
propionoylBa-oxopentanoyl and isopentanoylBa-
oxo-b-methylpentanoylBa-oxoglutaroyl.
It is clear that elongation and branching of the chain
might increase the activity of 8 by 10–20%, however a
polar residue at the end of the chain doubled the enzyme
inhibition activity (compound 12).
n=5, r=0.921, SE=5.106, PB0.027
where n is the compound 8–12; r is the linear regression
coefficient and SE is the standard error at the confidence
level PB0.05.
Eq. (1) reveals that, in the given series of compounds
the torsion angle Ca–C at Z moiety contributed by 84%
to the inhibition activity. Other factors like branching at
Cb may also be of significant effect.
3.2. Quantitati6e structure acti6ity relationship of the
dipeptides
4. Conclusion
In this series of dipeptide-based HIV-1 protease in-
hibitors several factors may affect their activity. Among
these factors is the possible rotation of the a-keto group
about the Ca bond. The angle of rotation can be simu-
lated in a quantitative manner by the calculated torsion
angle (Tor) of the linking bond CaꢀC at Z moiety. A
semi-empirical calculation program (MMX) was used
for predicting the values of the torsion angle in com-
pounds 8–12 (table II). Linear free energy (LFE) ap-
proach was used as a statistical model to derive the
relation between HIV-1 protease inhibition activity and
the values (Tor), Eq. (1).
The a-ketoamide isostere was a good transition-
state mimetic in HIV-1 protease inhibitor design. Di-
and tripeptides containing a-ketoamide at P2 or P1
site, respectively, exhibited a significant enhanced
inhibitory activity relative to their parent isosteres
and are shown to be a promising core structure
for inhibition of HIV-1 protease for the develop-
ment of more potent derivatives. The study suggested
that the potency of the truncated dipeptide deriva-
tives are greatly affected by the nature of’ the incor-
porated a-ketoamide. QSAR study using LFE
indicated that, the torsion angle at the a-diketone
contributed by 84% to the activity in the dipeptide
derivatives.
HIV-1 protease inhibition activity
=3.689 (90.903) Tor −7.253
(1)

M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
891
using micro cover glass thickness No. 1, 0.12–0.17
mm and are uncorrected.
5. Experimental protocols
Allophenyl norstatine (reference sample) was ob-
tained from Nippon Mining Co., Japan. Chemicals
were of analytical grade and the solvents were dried
(if necessary) over molecular sieve 4. All coupling and
deprotection reactions were carried out under anhy-
drous conditions using an anhydrous silica drying
tube.
The standard workup procedures were applied as
follows, the reaction mixture was partitioned between
the indicated solvent and water, the combined organic
extracts were washed with suitable aqueous systems,
5% sodium bicarbonate (twice), 5% citric acid (twice),
water and brine, then dried over anhydrous sodium
sulphate, filtered, and evaporated in vacuo.
The optical rotations were measured on a union
automatic digital polarimeter PM-101. The purity of
the compounds was examined by TLC and HPLC.
Thin layer chromatography (TLC) was performed
on precoated Merck silica gel sheets (0.25 mm) 60
F254, and the spots were visualized by UV and/or by
ninhydrin reagent. The following solvent systems
were used: (a) chloroform–methanol 40:1 (b) chloro-
form–methanol 20:1, or (c) hexane–ethyl acetate 2:1
for the N-protected compounds; (d) chloroform–
methanol–water (lower layer) 8:3:1 system for the
amine hydrochloride salts and (e) chloroform–
methanol 10:1 system in the case of final compounds.
Column chromatography was carried out on Merck
silica gel 60 (particle size 0.063–0.200 mm).
Melting points were determined on a micro hot
plate of Yanagimoto micro melting point apparatus,
Preparative reversed-phase HPLC (Shimadzu Liq-
uid chromatograph LC-4A, spectrophotometer detec-
tor SPD-2AS) was performed utilizing a YMC
packed ODS-AM AM 12S05-2520 WT type SH-343-
5AM column (250×20 mm i.d, S-5 120A) with gradi-
ent elution using acetonitrile in 0.1% aqueous
trifluoroacetic acid solvent system. Reversed-phase
HPLC was performed on a Hitachi 655 A-11 utilizing
YMC packed ODS-AM AM-302 column (150×4.0
mm i.d S5 120A) thermostated at 38 °C. A gradient
elution sequence with acetonitrile from 20 to 80% v/v
(through 30 min) in 0.1% aqueous TFA was used at
flow rate 0.7 mL/min. Spectrophotometeric detection
at 230 nm was used and the retention time (Rt)
recorded in minutes.
Table II. Torsion angle of dipeptides (8–13) and their HIV-1
protease inhibitory activity
1H-NMR was recorded on a JEOL JNM-EX 270
MHz spectrometer, all chemical shifts are given in (l
ppm) relative to tetramethylsilane (TMS) as internal
standard.
FAB mass spectra (MS) and high resolution FAB
MS (HRMS) were recorded as (M+H)+ on a JEOL
JMS-SX 102 AQQ hybrid mass spectometer using
glycerol, thioglycerol, or magic bullet as internal ref-
erence. All the new compounds, obtained as
lyophilized materials, gave satisfactory high resolu-
tion mass spectra (error less than 3 ppm) and were
fully characterized spectroscopically.
Inhibition of HIV-1 protease activity was
quantified by using a peptidolytic assay with a syn-
thetic nona-peptide substrate Ac-Arg-Ala-Ser-Gln-
Asn-Tyr-Pro-Val-Val-NH2. The inhibitors were
dissolved in DMSO and assayed at 50 nM and 5 mM
concentrations of the target tripeptides and dipep-

892
M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
tides, respectively. The assay was performed at Bio-
science Research Laboratories, Nippon Mining Co.,
Japan.
reported [24], [a]D25 0.0° (c=1.3, MeOH). 1H-NMR
(DMSO-d6) l: 1.27, (s, 9H), 2.64 (d, 2H, J=8.3 Hz),
3.37 (bs, 1H), 3.94–3.98 (m, 1H), 4.11 (d, 1H, J=5.8
Hz), 6.72 (d, 1H, J=8.5 Hz), 7.12–7.25 (m, 5H), 12.31
(s, 1H).
5.1. General method for N-protection and
N-deprotection of amino acids
5.3. General coupling procedure
Synthesis of Boc-amino acids (N-protection) and re-
moval of the Boc group (N-deprotection) were carried
out according to reported methods [30–32].
Using 1-ethyl-3-(3-dimethylaminopropyl)carbodimide
hydrochloride, EDC·HCl, as reagent and 1-hydroxyben-
zotriazole monohydrate, HOBt·H2O as coupling additive
[33, 34].
5.2. (2S,3S)3-(tert-Butyloxycarbonyl)amino-2-
hydroxy-4-phenyl-butanoic acid (Boc-Apns-OH) (4)
5.3.1. 3-[3(S)-(Substituted)amino-2(S)-hydroxy-4-
phenyl-butanoyl]-N(tert-butyl)-thiazolidine-4(R)-
carboxamide, N-(Substituted)-Apns-Thz-NHBut (8–13)
The following compounds were prepared as described
under the general procedure for coupling using Apns-
Thz-NHBut (7) and the corresponding a-keto carboxylic
acid or the amino acid, valine.
To a stirred solution of Boc-Phe-ol (2) [27, 28] (22.5 g,
89.6 mmol) in DMSO (31.4 mL), benzene (31.4 mL) and
TEA (24.75 mL, 224 mmol) was added, portionwise, at
−15 °C, pyridine sulfur trioxide (28.57 g, 244 mmol) for
30 min. The reaction mixture was quenched with ice-wa-
ter, extracted with EtOAc and the organic layer washed
gently with water. Solution of potassium cyanide (13.15
g, 268.8 mmol) in a least amount of water (about 10 mL)
was added to the EtOAc extract and the stirring contin-
ued overnight. The organic layer was separated, washed
with water, brine, and evaporated under vacuum. To the
residue, a mixture of conc. HCl and dioxane (60 mL, 1:1)
was added and the reaction mixture was refluxed for 8 h,
cooled and washed twice with ether. The aqueous layer
was concentrated under vacuum at a temperature not
exceeding 40 °C and to the residue in water (35 mL),
acetone (200 mL) was added. The solution was cooled in
an ice-bath, neutralized with 4 N NaOH to pH 5.5, kept
overnight in a refrigerator and filtered to afford 6.45 g
(36.9%) of (3S)-3-amino-2-hydroxy-4-phenylbutyric acid
(AHPBA-OH) with m.p. 246–247 °C.
5.3.2. N-(2-Oxopropionyl)-Apns-Thz-NHBut (8)
25
Yield 81%, Rt=18.6, m.p. 109–111 °C, [a]D −157.0°
(c=0.50, MeOH). 1H-NMR (DMSO-d6) l: 1.25 (s,
9H), 2.24 (s, 3H), 2.79 (d, 2H, J=9.5 Hz), 2.99 (dd, 1H,
J=6.6, 5.0 Hz), 3.30 (dd, 1H, J=7.6, 5.0 Hz), 4.20–
4.27 (m, 2H), 4.45 (s, 1H), 4.74 (d, 1H, J=9.1 Hz), 4.79
(t, 1H, J=6.6 Hz), 4.95 (d, 1H, J=9.2 Hz), 7.15–7.26
(m, 5H), 7.65 (s, 1H), 8.41 (d, 1H, J=9.2 Hz). MS
(m/z): 436 (M+H)+, HRMS 436.2202 (calcd.
436.1906), elemental composition C21H30N3O5S.
5.3.3. N-(3-Methyl-2-oxobutanoyl)-Apns-Thz-NHBut (9)
Yield 89%, Rt=22.93, m.p. 131–133°C, [a]D25
1
−131.1° (c=0.45, MeOH). H-NMR (CDCl3) l: 0.97
(d, 3H, J=7.0 Hz), 1.02 (d, 3H, J=7.0 Hz), 1.32 (s,
9H), 2.00–2.18 (m, 1H), 2.75 (d, 2H, J=8.5 Hz), 3.13
(dd, 1H, J=6.6, 4.5 Hz), 3.33 (dd, 1H, J=7.0 Hz),
4.19–4.27 (m, 2H), 4,46 (bs, 1H), 4.69 (d, 1H, J=9,5
Hz), 4.79 (t, 1H, J=7.2 Hz), 4.96 (d, 1H, J=9.2 Hz),
7.15–7.26 (m, 5H), 7.67 (s, 1H), 8.42 (d, 1H, J=8.8 Hz).
MS (m/z): 464 (M+H)+, HRMS 464.1902 (calcd.
464.1882), elemental composition C23H34N3O5S.
A solution of AHPBA-OH (6.5 g, 333 mmol) in a
mixture of THF (10 mL), water (30 mL) and TEA (6.94
mL, 50 mmol) was stirred and cooled in an ice-water
bath. Solution of (Boc)2O (8 g, 36.63 mmol) in THF (7
mL) was added and the reaction proceeded as reported
for preparation of Boc amino acid [31]. The resulting
residue of Boc-AHPBA-OH diastereomeric mixture was
dissolved in ether, dicyclohexylamine (DCHA) was
added dropwise until pH 7–8. The solution was stirred
by glass rod, kept for 1 day in refrigerator, and filtered.
The residue was dissolved in water, covered with a layer
of EtOAc (40 mL) and acidified with 10% solution of
citric acid to pH 2–3. The aqueous phase was worked up
(EtOAc, water, and brine) and yielded 3.2 g (32.5%) of
Boc-Apns-OH (4) as white crystals, m.p. 145–147 °C as
5.3.4. N-(2-Oxopentanoyl)-Apns-Thz-NHBut (10)
25
Yield 90%, Rt=25.34, m.p. 119–120°C, [a]D −115.8°
1
(c=0.50, MeOH). H-NMR (DMSO-d6) l: 0.82 (t, 3H,
J=7.3 Hz), 0.94–1.07 (m, 2H) 1.25 (s, 9H), 1.43 (t, 2H,
7.3 Hz), 2.79 (d, 2H, J=9.6 Hz) 2.98 (dd, 1H, J=
6.6, 5.3 Hz), 3.30 (dd, 1H, J=7.6, 4.0 Hz), 4.13–

M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
893
4.20 (m, 2H), 4.44 (s, 1H), 4.55 (d, 1H, J=8.9 Hz), 4.75
(t, 1H, J=7.0 Hz), 4.96 (d, 1H, J=8.9 Hz), 7.13–7.29
(m, 5H), 7.60 (s, 1H), 8.41 (d, 1H, J=8.9 Hz). MS
(m/z): 464 (M+H)+, HRMS 464.1899 (calcd.
464.1882), elemental composition C23H34N3O5S.
mixture of DCC (66.4 mg, 0.332 mmol), pyridine (8.0
mL, 0.108 mmol), and DMSO (320 mL) in benzene (700
mL) at 0°C. The mixture was stirred for 22 h at room
temperature, then EtOAc (1 mL) was added. The result-
ing insoluble substance was filtered off. The filtrate was
worked up (EtOAc, NaHCO3) and purified by prepara-
tive HPLC to give the corresponding a-ketoamide as a
white powder.
5.3.5. N-(3-Methyl-2-oxopentanoyl)-Apns-Thz-
NHBut (11)
25
Yield 80%, Rt=27.30, m.p. 121–124°C, [a]D −69.5°
1
(c=0.50, MeOH). H-NMR (DMSO-d6) l: 0.65 (t, 3H,
5.3.9. iQOA-Val-Apns(CꢁO)-Oxa-NHBut (18)
J=8.1 Hz), 0.74–0.81 (m, 2H), 0.91 (d, 3H, J=6.9
Hz), 1.25 (s, 9H), 1.50–1.57 (m, 1H), 2.76 (d, 2H,
J=10.0 Hz), 2.99 (dd, 1H, J=6.6, 5.0 Hz), 3.03–3.11
(m, 1H), 4.15–4.22 (m, 2H), 4.45 (bs, 1H), 4.59 (d, 1H,
J=8.9 Hz), 4.76 (t, 1H, J=7.0 Hz), 4.97 (d, 1H, J=9.2
Hz), 7.12–7.26 (m, 5H), 7.61 (s, 1H), 8.48 (d, 1H,
J=8.9 Hz). MS (m/z): 478 (M+H)+, HRMS 478.2303
(calcd. 478.2301), elemental composition C24H36N3O5.
The titled compound was obtained by oxidation of
iQOA-Val-Apns-Oxa-NHBut (14): Yield 34%, m.p.
142–144°C, [a]2D5 −120.3° (c=0.32, MeOH), Rt=
14.63. 1H-NMR (DMSO-d6) l: 0.83 (dd, 6H, J=6.6
Hz), 1.28 (s, 9H), 1.90–2.03 (m, 1H), 2.83 (d, 2H,
J=8.5 Hz), 3.42 (dd, 1H, J=6.6, 4.9 Hz), 3.72 (dd, 1H,
J=7.5, 3.8 Hz), 4.22 (dd, 1H, J=7.0 Hz), 4.35–4.41
(m, 1H), 4.80 (t, 1H, J=6.7 Hz), 4.85 (s, 2H), 5.06 (d,
1H, J=8.3 Hz), 5.11 (d, 1H, J=8.5 Hz), 7.07 (d, 1H,
J=7.2 Hz), 7.16–7.32 (m, 6H), 7.73 (s, 1H), 7.79 (d,
1H, J=7.0 Hz), 7.89 (t, 1H, J=6.1 Hz), 7.92 (d, 1H,
J=8.6 Hz), 8.48 (d, 1H, J=7.0 Hz), 8.59 (d, 1H,
J=7.0 Hz), 9.65 (s, 1H). MS (m/z): 632 (M+H)+,
HRMS 632.3082 (calcd. 632.3084) elemental composi-
tion C34H42N5O7.
5.3.6. N-(4-Carboxy-2-oxobutanoyl)-Apns-Thz-
NHBut (12)
25
Yield 90%, Rt=16.36, m.p. 114–117°C, [a]D −135.0°
1
(c=0.50, MeOH). H-NMR (CDC13) l: 1.31 (s, 9H),
2.72 (d, 2H, J=8.5 Hz), 2.79 (t, 2H, J=6.3 Hz), 2.98 (t,
2H, J=5.8 Hz), 3.12 (dd, 1H, J=6.5, 5.5 Hz), 3.43 (dd,
1H, J=6.9, 5.7 Hz), 4.29–4.37 (m, 2H), 4.48 (s, 1H),
4.69 (d, 1H, J=9.2 Hz), 4.79 (t, 1H, J=7.1 Hz), 5.00
(d, 1H, J=9.3 Hz), 7.18–7.27 (m, 5H), 7.68 (s, 1H),
8.42 (d, 1H, J=7.9 Hz), 12.03 (bs, 1H). MS (m/z): 494
(M+H)+, HRMS 494.1871 (calcd. 494.1882), elemental
composition C23H32N3O7S.
5.3.10. iQOA-Val-Apns(CꢁO)-Thz-NHBut (19)
The titled compound was obtained by oxidation of
iQOA-Val-Apns-Thz-NHBut (15). Yield 40%, m.p.
25
135–137°C, [a]D 120.5° (c=0.50, MeOH), Rt=18.41.
1H-NMR (DMSO-d6) l: 0.82 (dd, 6H, J=7.0 Hz), 1.27
(s, 9H), 1.92–2.03 (m, 1H), 2.84 (d, 2H, J=8.6 Hz),
3.02 (dd, 1H, J=6.6, 5.6 Hz), 3.42 (dd, 1H, J=7.5, 4.2
Hz), 4.33 (dd, 1H, J=6.8 Hz), 4.48–4.54 (m, 1H), 4.70
(t, 1H, J=8.2 Hz), 4.83 (s, 2H), 4.88 (d, 1H, J=8.1
Hz), 5.02 (d, 1H, J=8.6 Hz), 7.15–7.24 (m, 6H), 7.64
(d, 1H, J=6.8 Hz), 7.67 (s, 1H), 7.79 (d, 1H, J=7.9
Hz), 7.87 (t, 1H, J=6.3 Hz), 8.13 (d, 1H, J=6.6 Hz),
8.58 (d, 1H, J=6.7 Hz), 8.67 (d, 1H, J=7.3 Hz), 9.43
(s, 1H). MS (m/z): 648 (M+H)+, HRMS 648.2847
(calcd. 648.2855) elemental composition C34H42N5O6S.
5.3.7. Val-Apns-Thz-NHBut·HC1 (13)
25
Yield 76%, m.p. 153–166°C, [a]D −118.0° (c=1.0,
1
MeOH). H-NMR (DMSO-d6) l: 0.76 (dd, 6H, J=6.4,
11.5 Hz), 1.27 (s, 9H), 1.93–2.01 (m, 1H), 2.75 (d, 2H,
J=7.8 Hz), 3.0 (dd, 1H, J=5.5, 4.2 Hz), 3.36 (dd, 1H,
J=5.5, 5.0 Hz), 3.72–3.80 (h, 2H), 4.31–4.36 (m, 3H),
4.43 (bs, 1H), 4.66 (d, 1H, J=7.6 Hz), 4.79 (t, 1H,
J=6.9 Hz), 4.95 (d, 1H, J=8.2 Hz), 7.09–7.20 (m, 5H)
and 8.24 (d, 1H, J=6.9 Hz).
5.3.8. {3(S)-[N-(5-Isoquinolinyloxyacetyl)-L-6alinyl]-
amino-2-oxo-4-phenyl-butanoyl}-N(tert-butyl)-(un)-
substituted thiazolidine (oxazolidine)-4(R)-carboxamide
IQOA-Val-Apns(CꢁO)-((un)substituted-Thz(Oxa))-
NHBut, (18–21) General procedure for oxidation of
Apns in tripeptides (14–17) [17]
5.3.11. iQOA-Val-Apns(CꢁO)-[5-(R)-methyl-Oxa]-
NHBut (20)
The titled compound was obtained by oxidation of
iQOA-Val-Apns-[5-(R)-methyl-Oxa-NHBut (16): Yield
25
55.9%, Rt=16.37, m.p. 142–143°C, [a]D −23.50° (c=
1
0.50, MeOH), H-NMR (CDCl3) l: 0.82 (dd, 6H, J=
Trifluoroacetic acid (4.0 mL, 0.054 mmol) and the
appropriate tripeptide (0.108 mmol) were added to a
6.7 Hz), 1.32 (s, 9H), 1.47 (d, 3H, J=6.4 Hz), 1.95–2.06
(m, 1H), 2.81 (d, 2H, J=8.6 Hz), 3.36–3.42 (m, 1H),

894
M.M. Sheha et al. / European Journal of Medicinal Chemistry 35 (2000) 887–894
[10] Komai T., Higashida S., Sakurai M. et al., Bioorg. Med. Chem.
4 (1996) 1365–1377.
4.22 (dd, 1H, J=7.0 Hz), 4.31 (t, 1H, J=6.5 Hz), 4.81
(d, 1H, J=6.7 Hz), 4.85 (s, 2H), 4.98 (d, 1H, J=6.3
Hz), 5.08 (d, 1H, J=7.5 Hz), 7.09 (d, 1H, J=7.2 Hz),
7.16–7.32 (m, 6H), 7.73 (s, 1H), 7.79 (d, 1H, J=7.0
Hz), 7.89 (t, 1H, J=6.1 Hz), 7.92 (d, 1H, J=8.6 Hz),
8.48 (d, 1H, J=7.0 Hz), 8.59 (d, 1H, J=7.0 Hz), 9.65
(s, 1H). MS (m/z): 646 (M+H)+, HRMS 646.3242
(calcd. 646.3240) elemental composition C35H44N5O7.
[11] Baldwin E.T., Bhat T.N., Gulnik S. et al., Structure 3 (1995)
581–590.
[12] Wany Y.X., Freedberg D.I., Yamazaki T. et al., Biochemistry
35 (1996) 9945–9950.
[13] Kiso Y., Biopolymers 40 (1996) 235–244.
[14] Kiso Y., Yamaguchi S, Matsumoto H. et al., Arch. Pharm.
Pharm. Med. Chem. 331 (1998) 87–89.
[15] Kiso Y., J. Synth. Org. Chem. Jpn. 56 (1998) 896–907.
5.3.12. iQOA-Val-Apns(CꢁO)-(5,5-dimethyl-Thz)-
NHBut (21)
[16] Harbenson S.L., Abelleira S.M., Akiyama A. et al., J. Med.
Chem. 37 (1994) 2918–2929.
[17] Kitazaki T., Asano T., Kato K., Kishimoto S., Itoh K., Chem.
Pharm. Bull. 42 (1994) 2636–2640.
The titled compound was prepared by oxidation of
iQOA-Val-Apns-(5,5-dimethyl-Thz)-NHBut (17): Yield
25
[18] Munoz B., Gian C.-Z., Wong C.-H., Bioorg. Med. Chem. 2
(1994) 1085–1090.
35%, m.p. 130–132°C, [a]D −73.0° (c=0.45, MeOH),
1
Rt=19.80. H-NMR (DMSO-d6) l: 0.77 (dd, 6H, J=
[19] To¨zse´r J., Weber I.T., Gustchina A., Blaha I., Copeland T.D.,
Louis J.M., Oroszlan S., Biochemistry 31 (1992) 4793–4800.
6.5 Hz), 1.29 (s, 9H), 1.40 (d, 3H, J=9.5 Hz), 1.48 (d,
3H, J=9.5 Hz), 1.93–1.98 (m, 1H), 2.76 (d, 2H, J=9.3
Hz), 4.28 (dd, 1H, J=6.5 Hz), 4.30–4.36 (m, 1H), 4.77
(s, 2H), 4.84 (s, 1H), 4.84 (d, 1H, J=7.7 Hz), 5.00 (d,
1H, J=8.5 Hz), 7.09–7.26 (m, 6H), 7.71 (s, 1H), 7.76 (t,
1H, J=6.5 Hz), 7.97–8.09 (m, 2H), 8.27 (d, 1H, J=8.2
Hz), 8.35 (d, 1H, J=9.8 Hz), 8.60 (d, 1H, J=9.9 Hz),
9.57 (s, 1H). MS (m/z): 676 (M+H)+, HRMS 676.3164
(calcd. 676.3168) elemental composition C36H46N5O6S.
[20] Meek T.D., Larnbert D.M., Metcalf B.W., Petteway S.R.,
Dreyer G.B., in: De Clercq E. (Ed.), Design of Anti-AIDS
Drugs, Elsevier Science Publishing Company, 1990, pp. 225–
255.
[21] Debouck C., AIDs Res. Hum. Retrovir. 8 (1992) 153–164.
[22] Rich D.H., Bernatowiez M.S., Schmidt P.G., J. Am. Chem.
Soc. 104 (1982) 3535–3536.
[23] Bunnage M.E., Davies S.G., Goodwin C.J., SYNLETT (1993)
731–732
[24] Herranz R., Castro-Pichel J, Garsia-Lo´pez T., Synthesis (1989)
703–706.
References
[25] Herranz R., Castro-Pichel J., Vinuesa S., Garsia-Lo´pez M.T., J.
Org. Chem. 55 (1990) 2232–2234.
[1] Mimoto T., Imai J., Tanaka S. et al., Chem. Pharm. Bull. 38
(1991) 2465–2467.
[26] Harada H., Tsubaki A., Kamijo T., Iizuka K., Kiso Y., Chem.
Pharm. Bull. 37 (1989) 2570–2572.
[2] Mimoto T., Imai J., Tanaka S., Hattori N., Kisanuki S., Akaji
K., Kiso Y., Chem. Pharm. Bull. 39 (1991) 3088–3090.
[27] McCorkindale N.J., Baxter R.L., Roy T.P., Shields H.S., Stew-
art R.M, Huthinson S.A., Tetrahedron 34 (1978) 2791–2795.
[3] Raju B., Deshpande M.S., Biochem. Biophys. Res. Commun.
180 (1991) 181–186.
[28] Shelia M.M., Ph.D. Thesis, Submitted to Faculty of Pharmacy
Assiut University, Assiut, Egypt, 1996.
[4] Tam T.F., Carrie`re J., MacDonald I.D. et al., J. Med. Chem.
35 (1992) 1318–1320.
[29] Sheha M., Mimoto T., Mahfouz N.M., Hassan H.Y., Akaji K.,
Kiso Y., Youssef A.F., Pharm. Soc. of Eygpt XXV Conference,
24–26 Dec. (1996) 69.
[5] Kiso Y., Adv. Exp. Med. Biol. 362 (1995) 413–423.
[6] Sheha M., Nakata S., Enomoto H., in: Ohno M. (Ed.), Peptide
Chemistry, Protein Research Foundation, Osaka, 1995, pp.
349–352.
[30] Bodanszky M., Bodanszkv A., The Practical of Peptide Synthe-
sis, Springer-Verlag, Berlin, 1994, p. 20.
[31] Kiso Y., Isawa H, Kitagawa K., Akita T., Chem. Pharm. Bull.
26 (1978) 2562–2564.
[7] Abdel-Rahman H.M., Matsumoto H., Bekhit A.A., in: Shini-
anishi Y. (Ed.), Peptide Science—Present and Future, Kluwer
Academic, UK, 1998, pp. 683–685.
[32] Kiso Y., Ito K., Nakamura S., Kitagawa K., Akita T., Mori-
toki H., Chem. Pharm. Bull. 27 (1979) 1472–1475.
[8] Komai T., Yagi R., Suzuki-Sunagawa H. et al., Biol. Pharm.
Bull. 20 (1997) 175–180.
[33] Sheehan J.C., Hlavka J.J., J. Org. Chem. 21 (1956) 439–441.
[9] Sakurai M., Higashida S., Sugano M. et al., Bioorg. Med.
Chem. 2 (1994) 807–825.
[34] Sheehan J.C., Cruickshank P.A., Boshart G.L., J. Org. Chem.
26 (1961) 2525–2528.
.