Symmetry-based inhibitors of HIV-1 protease. Design, synthesis and preliminary structure-activity studies of acylated 2,3-diamino-1- hydroxypropanes and 2,4 diamino-1-hydroxybutanes

Article abstract of DOI:10.1016/S0223-5234(00)80034-3

Two series of peptidomimetics containing a novel C₂ pseudosymmetrical hydroxyalkyldiamino core structure were prepared from amino acid starting materials and tested for inhibitory activity against the HIV-1 protease (HIV- 1 Pt) and the virus in cell culture. In the 2,3-diamino-1-hydroxypropane series, compound 6a, containing P1/P1(I) benzyl and P2/P2(I) Fmoc substituents, displayed modest HIV-1 Pr inhibition (IC₅₀ = 430 nM). The corresponding 2,4-diamino-1-hydroxybutane derivative (6b) was the best inhibitor of the series (IC₅₀ = 160 nM). Interestingly, 6a and 6b showed satisfactory inhibition of HIV replication in cell culture. (ED₅₀ = 340 and 110 nM, respectively), a result which suggests good cell membrane penetration by this class of compounds.

Full text of DOI:10.1016/S0223-5234(00)80034-3

Eur. J. Med. Chem. 34 (1999) 651−657
© 1999 Éditions scientifiques et médicales Elsevier SAS. All rights reserved
651
Short communication
Symmetry-based inhibitors of HIV-1 protease.
Design, synthesis and preliminary structure-activity studies
of acylated 2,3-diamino-1-hydroxypropanes and 2,4 diamino-1-hydroxybutanes
Mauro Marastoni, Martina Bazzaro, Fabrizio Bortolotti, Severo Salvadori, Roberto Tomatis*
Department of Pharmaceutical Sciences and Biotecnology Centre, University of Ferrara, Ferrara, Italy
(Received 12 October 1998; accepted 3 March 1999)
Abstract – Two series of peptidomimetics containing a novel C2 pseudosymmetrical hydroxyalkyldiamino core structure were prepared from
amino acid starting materials and tested for inhibitory activity against the HIV-1 protease (HIV-1 Pr) and the virus in cell culture. In the
2,3-diamino-1-hydroxypropane series, compound 6a, containing P1/P1I benzyl and P2/P2I Fmoc substituents, displayed modest HIV-1 Pr
inhibition (IC50 = 430 nM). The corresponding 2,4-diamino-1-hydroxybutane derivative (6b) was the best inhibitor of the series (IC50 = 160
nM). Interestingly, 6a and 6b showed satisfactory inhibition of HIV replication in cell culture (ED50 = 340 and 110 nM, respectively), a result
which suggests good cell membrane penetration by this class of compounds. © 1999 Éditions scientifiques et médicales Elsevier SAS
aspartyl protease / HIV-1 inhibitors / peptidomimetic
1. Introduction
potent and selective inhibitors [8–10]. The C2-symmetry
of HIV-1 Pr, which functions as a dimer with each subunit
contributing an amino acid triad Asp-Thr-Gly to the
active site, stimulated the design of symmetry-based
inhibitors. On the basis of the hydroxyethylene core as a
tetrahedral transition-state replacement for the peptide
substrate, various potent and selective C2-symmetric
inhibitors have been designed [3, 8, 11, 12]. Recently,
following the Kempf strategy [10] and the aminodiol
inhibitors discovery [13], we disclosed a new class of
HIV-Pr inhibitor which incorporates a C2 symmetric
1,1-diamino-2-hydroxyethane core (gSer) as its key
structural feature [14]. A member of this class, repre-
sented by the lead compound LC (figure 1), displays
significant HIV-1 Pr inhibition (IC50 ≈ 450 nM).
Human immunodeficiency virus (HIV) has been iden-
tified as the probable causative agent of AIDS. The
different therapeutic strategies for intervention of this
disease have been reviewed recently [1]. A critical role in
the viral replication is played by the HIV protease (Pr), an
enzyme responsible for the processing of the polyproteins
to structural proteins and enzymes essential for the viral
maturation and infectivity [2]. This essential role makes it
a promising target for chemotherapy of AIDS [3–7].
Since the initial characterisation of this enzyme, rapid
progress has been achieved toward the development of
*Correspondence and reprints
Subsequently, dynamics simulations [15] indicated a
“pseudosymmetric” mode of binding, suggesting that LC
contains a core functionality capable of hydrogen bond-
ing (OH, and NH of gSer) to the catalytic aspartate (Asp
25I/Asp 25) residues of the enzyme and P1/P2 hydropho-
bic groups for placement in the S1/S2 subsites. Further-
more, these studies revealed that a lengthening of the
gSer core structure in LC, by one or two methylene
groups, would result in more potent inhibitors that would
Abbreviations: Standard abbreviations for amino acid derivatives
and peptides are according to IUPAC-IUB Biochemical Nomen-
clature [25]. Other abbreviations are: Ben, benzofuran-2-carbonyl;
Boc, tert-butyloxycarbonyl; DMF, dimethylformamide; DMSO,
dimethylsulfoxide; EtOAc, ethyl acetate; Et2O, diethyl ether;
Fmoc, fluorenylmethyloxycarbonyl; Ind, indol-2-carbonyl; MeOH,
methyl alcohol; NMM, N-methyl-morpholine; OSu, N-hydroxy-
succinimidyl ester; Pyp, pyperidin-2-carbonyl; THF, tetrahydrofu-
ran; TFA, trifluoroacetic acid; Z, benzyloxycarbonyl; WSC,
1-ethyl-3-(3’-dimethylaminopropyl)carbodiimide.

652
Figure 1. Chemical structures of KNI-272, symmetrical lead compound (LC) and its 2,4-diamino-1-hydroxybutane core structure
containing analogue 6b.
occupy, at least in part, the subsites of the enzyme and
which are capable of forming further H bonds to the
protein backbone. According to this prediction, pseudo-
symmetric analogues of LC (figure 2 and table I) con-
taining 2,3-diamino-1-hydroxypropane or 2,4-diamino-1-
hydroxybutane core units with different terminal groups
were designed, prepared, and their interaction with re-
combinant HIV-1 Pr and anti HIV-1 activity were evalu-
ated. For simplicity, the P1/P1I and P2/P2I nomenclature
is used throughout the paper, although there is only

653
Figure 2. Synthesis of 2,3-diamino-1-hydroxypropane (n = 1) or 2,4-diamino-1-hydroxybutane (n = 2) containing pseudotripeptides.
X = indol-2-carbonyl, benzofuran-2-carbonyl, pyridine-2-carbonyl.
preliminary evidence that these designated groups bind in
the corresponding subsites of the enzyme [15].
3b in very good yields and with retention of stereochem-
istry at the adjacent stereocenter. Deprotection of 3a and
3b followed by coupling of Z-OSu or Fmoc-OSu afforded
Z-pseudotripeptides 5a and 5b or Fmoc-analogues 6a and
6b, respectively. Finally, acylation of 4b with indol-2-
carboxylic acid, benzofuran-2-carboxylic acid or
pyperidin-2-carboxylic acid, in the presence of WSC and
HOBt, gave analogues 7b–9b.
2. Chemistry
The preparation of hydroxyalkyldiamino core structure
containing pseudopeptides 3a and b–6a and b is outlined
in figure 2. Coupling of two equivalents of Boc-Phe-OSu
with L-2,3-diaminopropanoic acid 1a or L-2,4-
diaminobutanoic acid 1b gave the key pseudotripeptide
intermediates 2a and 2b. Target alcohols 3a and 3b were
prepared by chemoselective reduction of the mixed an-
hydrides of the corresponding acids 2a and 2b. As
previously observed for the synthesis of homoserine
containing peptides [16], we found that a mixed anhy-
dride obtained from 2a and 2b by reaction with isobutyl
chloroformate, cleanly reacted with 3 M equivalents of
sodium borohydride, to lead to the corresponding 3a and
3. Biological activity
All compounds were tested for their inhibition of
purified recombinant HIV-1 Pr using the peptide substrate
His-Lys-Ala-Arg-Val-Leu-Phe(p-NO2)-Glu-Ala-Nle-Ser
(Bachem Bioscience) as previously described [14, 15].
The Leu-Phe(p-NO2) bond of the substrate is cleaved by
the enzyme, and substrate and products were separated by
reversed-phase HPLC, absorbance being measured at

654
Table I. Structure and inhibitory potencies of compounds.
MS[M + H+]
IC50 (nM)b ED50 (nM)b
a
Compound
X
(n)
Y
TLC
RF (A) (tR)
HPLC
M.p.
[°C]
[α]D
Calcula- Found
ted
2a
2b
3a
3b
5a
5b
6a
6b
7b
8b
9b
LC
C=O
C=O
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
CH2
1
2
1
2
1
2
1
2
2
2
2
Boc
Boc
Boc
Boc
Z
0.68
0.59
0.68
0.87
0.66
0.64
0.75
0.70
0.57
0.61
0.49
4.35
4.12
5.52
7.65
4.68
5.58
7.02
5.21
4.23
4.73
3.98
155–157 – 12.3
163–166 – 9.8
164–167 – 7.1
192–194 – 11.6
187–189 – 9.9
163–165 – 12.3
230–233 – 18.0
151–153 – 5.5
201–204 – 24.2
197–199 – 21.4
211–214 – 28.3
598
612
584
598
653
667
830
844
686
687
610
598
612
584
598
653
667
830
844
686
687
610
> 10 000
> 10 000
9 700
4 100
5 700
1 800
430
NDc
ND
ND
3 600
ND
2 300
340
110
3 800
ND
Z
Fmoc
Fmoc
Ind
Ben
Pyp
160
2 900
3 000
3 900
455
ND
290d
ac 1.0, methanol; temperature 21 °C.
bValues are the average of at least two determinations (n = 2) unless otherwise noted.
cNot determined.
dED50 values ranged from 130–350 nM (n = 8).
220 nm. Inhibitory activities are expressed as IC50 (con-
centration of compound which inhibits 50% cleavage of
substrate). Some compounds were also tested for their
ability to inhibit the replication of the HIV-1 virus in cell
culture (See experimental section).
A potent pseudopeptidic HIV-1 Pr inhibitor, KNI-272
(figure 1), was used as a standard (17), yielding an
average IC50 = 4.8 nM against the isolated enzyme and an
ED50 = 50 nM in cell culture.
potent KNI-272 [17] and inhibitors approved for the
treatment of AIDS [18–20]. As expected, the presence of
the carboxylic group in the core structure of pseudopep-
tides 2a and 2b is not tolerated, suggesting that this
functionality is incapable of hydrogen bonding to the
catalytic aspartate residues of the enzyme.
The 2,4-diamino-1-hydroxybutane containing com-
pounds (3b, 5b and 6b) are slightly more active than the
corresponding diamino hydroxypropane derivatives (3a,
5a and 6a) suggesting that the latter spacer can allow less
effective interactions between the inhibitors and protease.
4. Results and discussion
Comparison of Boc (3a and 3b), Z (5a and 5b) and
Fmoc (6a and 6b) functionalities at P2/P2I shows that a
bulky tricyclic system is preferred over a single aromatic
ring or a branched aliphatic group, confirming the trend
previously observed for symmetrical LC related ana-
logues [14, 21]. This preference seems to be supported by
the inhibitory potencies of the bis-acyl derivatives 7b–9b
which are at least 15-fold less potent than the Fmoc
analogue 6b. Nevertheless, it should be pointed out that
The HIV-1 Pr inhibition activity of new pseudotripep-
tides 2a–9b is reported in table I, in comparison with LC.
Analysis of the results from the HIV-1 Pr tests showed
weak inhibition for almost all compounds, displaying
IC values ≥ 2.8 µM. Only compounds 6a and 6b dis-
50
played significant activities comparable to aminodiol
derivatives [13] even if they are at least 50–100 times less
active than structurally more complex molecules such as

655
in comparison to carbamate derivatives (e.g. 5b and 6b),
in the bis-hetero acyl compounds 7b–9b the oxy groups
have been deleted and this modification might be critical
for enzyme-inhibitor interactions.
a gradient made up of two solvents: A, 10% (v/v)
acetonitrile in water; B, 60% (v/v) acetonitrile in water,
both containing 0.1% TFA. The gradient program used
was as follows: linear gradient from 0–100% B in 25 min
at a flow of 1 mL/min. All analogues showed purity
greater than 99% following analytical HPLC monitored
at 220 nm. Preparative reversed-Phase HPLC was carried
out with a Waters Delta Prep 3 000 using a Delta Pack C
18–300 A (30 mm × 30 cm, 15 mm, spherical). The gra-
dient used was identical to that of analytical determina-
tions. Chromatography was performed at a flow rate of 30
Five LC analogues were tested for their ability to
inhibit the replication of the HIV-1 virus in cell culture.
As can be readily observed (table I), the measured ED50
values are quite close to the IC50 values for enzyme
inhibition as had been noted previously for LC [22] and
aminodiol inhibitors [13, 23]. Then, in spite of their
modest ability to inhibit the enzyme, 6a and 6b showed
satisfactory inhibition of HIV replication in cell culture
(ED50 = 340 and 130 nM, respectively), a result that may
indicate good cell membrane penetration by this class of
compounds.
1
mL/min. H-NMR spectra were obtained using a Varian
Gemini 300 (Palo Alto, CA, USA). Amino acid analyses
were carried out using PITC methodology as the amino
acid derivatization reagent (Pico-Tag, Waters-Millipore,
Waltham, MA, USA). Lyophilised samples of peptides
(50–l00 pmol) were placed in heat treated borosilicate
tubes (50 × 4 mm), sealed and hydrolysed using 200 mL
6 N HCl containing 1% phenol in the Pico-Tag work
station for 1 h at 150 °C. A Hypersil ODS column (250 ×
4.6 mm, 5 mm particle size) was employed to separate
the PITC-amino acid derivatives. TLC used precoated
plates of silica gel F254 (E. Merk, Darmstadt, FRG) in
the following solvent system: (A) 1-butanol/acetic
acid/H2O (3:1:1), (B) Et0Ac/pyridine/acetic acid/H2O
(12:4:1.2:2.2), (C) CH2Cl2/MeOH/toluene (8.5:1:O.5),
(D) CHCl3/MeOH/benzene/H2O (8:8:8:1). Ninhydrin
1%, fluorescamine and chlorine spray reagents were
employed to detect the peptides.
5. Conclusion
Previous efforts in this laboratory resulted in the
synthesis, from amino acid starting materials, of a series
of HIV-1 protease inhibitors containing a novel hydroxy-
alkyl gem-diamino core structure as exemplified by
LC [14, 21]. In spite of its low ability to inhibit the
isolated enzyme, LC showed satisfactory inhibition of
HIV replication in cell culture [22]. This attractive prop-
erty prompted us to design, through computational stud-
ies, LC analogues with improved potency against the
isolated protease while maintaining the lipophilic charac-
ter of the parent molecule. By lengthening, with an
ethylene group, the core structure of LC we have been
successful in designing the pseudotripeptide 6b which
has increased potency against the enzyme and the virus in
cell culture.
6.2. Chemistry
6.2.1. Coupling and deprotection procedures
The main procedures used for the preparation and
deprotection of the pseudopeptides of figure 2 varied little
in the individual steps and are therefore summarised in
the following general form:
6. Experimental protocols
A. Active ester coupling procedure. The activated
carboxy component (OSu) (2 mmol) and the protonated
amino component (1 mmol) were dissolved in DMF
(10 mL) containing N-methylmorpholine (2 mmol). The
mixture was allowed to react at room temperature for 8 h
and then diluted with EtOAc (100 mL). The solution was
washed consecutively with brine, 0.5 N KHSO4, and
brine. The organic phase was dried (MgSO4), filtered and
evaporated to dryness. The residue was crystallised from
appropriate solvents or purified by HPLC.
B. WSC/HOBt coupling procedure. To a solution of the
carboxy component (2 mmol) in DMF (10 mL) were
added the amino acid component (1 mmol), NMM
(2 mmol), HOBt (2 equiv.) and WSC (2.1 mmol) in the
above order at 0 °C. The reaction mixture was stirred for
6.1. General
Melting points were determined by a Kofler apparatus
and are uncorrected. Optical rotations were determined
with a Perkin Elmer 141 polarimeter with a 10 cm water
jacketed cell. Matrix-assisted laser desorption ionisation
time-of-flight (MALDI-TOF) mass spectra were obtained
using the HPG2025A mass spectrometer. HPLC analysis
was performed on a Bruker liquid chromatograph
LC21-C instrument using a Vydac C 18 218 TP 5415
column (175 × 4.5 mm, 5 mm particle size) equipped
with a Bruker LC 313 UV variable wavelength detector;
recording and quantification were accomplished with a
chromatographic data processor (Epson computer
FX80X7). Analytical determinations were carried out by

656
1 h at 0 °C and 14 h at room temperature; then the
solution was diluted with EtOAc (100 mL) and worked
up as described in (B).
C. Trifluoroacetic acid deprotection. Boc group was
removed by treating the peptide with aqueous 90% TFA
(1:10, w/v) containing anisole (0.2 mL) for 30–40 min.
The solvent was evaporated in vacuo at 0 °C and the
residue was triturated with Et2O; the resulting solid
peptide was collected and dried.
4b were recrystallised from MeOH-Et2O (89%) and,
according to coupling procedure A, were treated with
Z-OSu or Fmoc-OSu (two-fold excess) to lead to the
corresponding Z- or Fmoc-title pseudopeptides
(65–70%). All compounds gave correct elemental analy-
sis. Characterisation of 5a and 5b, 6a and 6b are
summarised in table I.
6.2.5. Preparation of pseudotripeptides 7a–9a
According to general procedure B, 4a (1 mmol) was
treated with the appropriate arylcarboxylic acid
(2 mmol). Crude bis-acyl derivatives 7a–9a were purified
by HPLC (70–75%). All pseudopeptides gave correct
6.2.2. Preparation of tripeptides 2a and 2b
According to the general coupling procedure (A),
L-2,3-diamino propanoic acid or L-2,4-diamino butanoic
acid (2 mmol) was treated with Boc-Phe-OSu (4 mmol).
Crude compounds 2a and 2b were crystallised from
MeOH-Et2O (70–75%). The tripeptides gave correct
amino acid analysis and their physicochemical properties
are reported in table I.
1
elemental analysis and expected H-NMR spectra. The
physicochemical properties of 7a–9a are reported in
table I.
6.2.6. Test for the inhibition of HIV-1 protease
For determination of IC50 values, affinity-purified
HIV-1 protease (Bachem Bioscience), 1.1 nM final con-
centration, was added to a solution (100 mL final volume)
containing inhibitor, 4 mM peptide substrate (His-Lys-
Ala-Arg-Val-Leu-p-nitro-Phe-Glu-Ala-Nle-Ser, Bachem
Bioscience), and 1.0% DMSO in assay buffer: 1.0 mM
dithiothreitol, 0.1% glycerol, 80 mM sodium acetate,
160 mM sodium chloride, 1.0 mM EDTA, all at pH 4.7.
The solution was mixed and incubated for 25 min at
37 °C and the reaction quenched by the addition of
trifluoroacetic acid, 2% final concentration. The Leu-
Phe(p-NO2) bond of the substrate was cleaved by the
enzyme. The cleavage products and substrate were sepa-
rated by reversed-phase HPLC. Absorbance was mea-
sured at 220 nm, peak areas were determined, and percent
conversion to product was calculated using relative peak
areas. The data were plotted as percent control (the ratio
of percent conversion in the presence and absence of
inhibitor) vs. inhibitor concentration and fit with the
equation Y = 100/1 + (x/IC50)A, where IC50 is the inhibi-
tor concentration at 50% inhibition and A is the slope of
the inhibition curve. A norstatin-peptidic HIV-1 protease
inhibitor (KNI-272) [17] was used as a standard, yielding
an average IC50 = 4.8 nM in this assay. Inhibitor
concentrations of 0.1, 1 and 10 mM were initially evalu-
ated to aid in choosing the concentrations of inhibitor
used for determination of the IC50.
6.2.3. Preparation of pseudotripeptides 3a and 3b
To a stirred solution of tripeptide 2a or 2b (1 mmol) in
THF (10 mL, distilled over CaH2) at –15 °C, NMM
(0.11 mL, 1 mmol) was added, followed by isobutylchlo-
roformate (0.136 mL, 1 mmol). After 20 min the precipi-
tated NMM hydrochloride was removed by filtration,
washed with cold THF (3 × 0.5 mL) and the filtrate was
added dropwise to a cold (–10 °C) solution of NaBH4
(0.113 g, 3 mmol) in EtOH (2 mL). The solution was
stirred for 10 min at 0 °C and for an additional 15 min at
room temperature and then evaporated under reduced
pressure. The residue was extracted with EtOAc (35 mL)
and washed with 0.5 N KHSO4, brine, 5% NaHCO3 and
brine. The organic phase was dried (MgSO4), filtered and
evaporated. The resulting crude alcohols were purified by
HPLC (80–83%).
1
3a: H-NMR (CDCl3) δ 8.41 (d, 2H, J = 8.18 Hz),
7.48–7.24 (m, 10H), 5.91 (d, 1H, J = 7.81 Hz), 5.87 (d,
1H), 5.69 (t, 1H), 4.67 (m, 2H), 3.88–3.72 (m, 2H), 3.61
(m, 1H), 3.43 (m, 2H), 3.31–3.22 (m, 4H), 1.48 (s, 9H),
1.45 (s, 9H). Anal. C31H44N4O7: C, 63.68; H, 7.58; N,
9.58. Found: C, 63.51; H, 7.62; N, 9.37.
3b: 1H-NMR (CDCl3) δ 8.51–8.30 (dd, 2H), 7.57–7.26
(m, 10H), 6.01 (d, 1H, J = 7.78 Hz), 5.85 (d, 1H), 5.61 (t,
1H), 4.75 (m, 2H), 3.91–3.80 (m, 2H), 3.60 (m, 1H), 3.48
(m, 2H), 3.20–3.04 (m, 4H), 1.52 (s, 9H), 1.47 (s, 9H),
1.28 (m, 2H). Anal. C32H46N4O7: C, 64.19; H, 7.74; N,
9.36. Found: C, 63.98; H, 7.86; N, 9.41.
6.2.7. Cell culture activity against HIV-1 IIIB
HIV-1 IIIB was obtained from HIV-1 IIIB chronically
infected Molt-4 cells as a supernatant fluid. The 50%
tissue culture infection dose (TC ID50) was determined
by an endpoint titration procedure [24]. CEM cells
(5 000/mL) were exposed to HIV-1 IIIB fluid at a
multiplicity of infection (m.o.i.) 0.001 TC ID50 (mL).
Characterisation of 3a and 3b are summarised in table I.
6.2.4. General preparation of pseudotripeptides 5a and
5b, 6a and 6b
According to deprotection procedure C, 3a and 3b
were treated with TFA. The resulting deprotected 4a and

657
[6] Darke P.L., Huff J.R., Adv. Pharmacol. 25 (1994) 399–404.
Aliquots (0.2 mL) of cells were placed in 96 well
microtitre plates with 2 mL of the appropriate concentra-
tions of inhibitors dissolved in DMSO. After incubation
for 6 d in RPMI-1640 medium containing 10% foetal calf
serum, the p24 antigen of HIV in the supernatant was
determined by an ELISA assay kit (RETRO-TEK, Cel-
lular Products Inc., Buffalo, USA). The ED50 values were
calculated as the dose of the inhibitor that resulted in a
50% reduction in p24 levels as compared to those in
control wells.
[7] Barrish J.C., Zahler R., Annu. Rep. Med. Chem. 28 (1993)
131–140.
[8] Erickson J., Kempf D.J., Arch. Virol. 9 (1994) 19–29.
[9] Clare M., Perspect. Drug Discov. Des. 1 (1993) 49–68.
[10] Wlodawer A., Erickson J.W., Annu. Rev. Biochem. 62 (1993)
543–585.
[11] Kempf D.J., Methods Enzymol. 241 (1994) 334–354.
[12] Lam P.Y.S., Jadhav P.K., Eyermann C.J., Hodge C.N., Ru Y.,
Bacheler L.T. et al., Science 263 (1994) 280–284.
[13] Barrish J.C., Gordon E., Alam M., Lin P.F., Bisacchi G.S., Chen P,
et al., J. Med. Chem. 37 (1994) 1758–1768.
[14] Marastoni M., Salvadori S., Bortolotti F., Tomatis R., J. Pept. Res.
49 (1997) 538–544.
Acknowledgements
[15] Marastoni M., Bergonzoni M., Bortolotti F., Tomatis R., Arzneim.-
Forsch./Drug Res. 47 (1997) 889–892.
[16] Marastoni M., Salvadori S., Balboni G., Spisani S., Reali E.,
Traniello S., Tomatis R., Int. J. Pept. Protein Res. 41 (1993)
447–454.
Financial support for this work by Ministero dell’
Università e della Ricerca Scientifica e Tecnologica
(Cofinanziamento Progetto Chimica dei Composti Or-
ganici di Interesse Biologico) is gratefully acknowledged.
We warmly thank Prof. Y. Kiso of Kyoto Pharmaceutical
University (Department of Medicinal Chemistry) for a
kind gift of HIV-1 protease inhibitor, KNI-272.
[17] Mimoto T., Imai J., Kisanuki S., Enomoto H., Hattori N., Akaji K.,
Kiso Y., Chem. Pharmacol. Bull. 40 (1992) 2251–2253.
[18] Kempf D.J., Sham H.L., Marsh K.C., Flentge C.A., Betebenner D.,
Green B.E. et al., J. Med. Chem. 41 (1998) 602–617.
[19] Roberts N.A., Martin J.A., Kitchington D., Broadhurst A.V., Craig
J.C., Duncan I.B. et al., Science 248 (1990) 358–361.
[20] Vacca J.P., Dorsey B.D., Schleif W.A., Levin R.B., McDaniel S.L.,
Darke P.L. et al., Proc. Natl. Acad. Sci. USA 91 (1994) 4096–4100.
References
[21] Marastoni M., Bortolotti F., Salvadori S., Tomatis R., Arzneim.-
Forsch. Drug Res. 48 (I) (1998) 709–712.
[1] De Clercq E., J. Med. Chem. 38 (1995) 2491–2517.
[22] Marastoni M., Bortolotti F., Tomatis R., Arzneim.-Forsch. Drug Res.
in press.
[2] Gottlinger H.G., Sodroski J.G., Haseltine W.A., Proc. Natl. Acad.
Sci. USA. 86 (1989) 5781–5785.
[23] Chen P., Cheng P.T.W., Alam M., Beyer B.D., Bisacchi G.S.,
Dejneka T. et al., J. Med. Chem. 39 (1996) 1991–2007.
[3] Kempf D., Sham H., Curr. Pharm. Des. 2 (1996) 225–246.
[4] West M.L., Fairlie D.P., Trends Pharmacol. Sci. 16 (1995) 67–75.
[5] Thaisrivongs S., Annu. Rep. Med. Chem. 29 (1994) 133–144.
[24] Ho D.D., Moudgil T., Alam M., New Engl. J. Med. 321 (1989)
1621–1625.