New anti-HIV derivatives: Synthesis and antiviral evaluation

Article abstract of DOI:10.1016/S0968-0896(00)00055-9

A small focused library of 18 compounds incorporating the motif 1,3-(N,N'-dibenzyl)diamino-2-propanol has been synthesized, using adapted synthetic methodologies. These series of compounds were evaluated for their in vitro anti-HIV activity on infected MT₄ cells (syncytium formation observation). Some of the new synthesized compounds show potent anti-HIV activities. EC₅₀ values for compounds (31, 40, 34, 37 and 46Scheme 3(i) Na₂SO₄, CH₂Cl₂, rt; (ii) NaBH₄, EtOH, 0°C; (iii) Boc₂O, CH₂Cl₂, 0°C; (iv) DMSO, TFAA, Et₃N, CH₂Cl₂, -60°C; (v) TFA, CH₂Cl₂, rt; (vi) BOP, RCOOH, Et₃N, CH₂Cl₂, rt.) range from 0.1 to 1μM. In order to determine at which level these new derivatives interfere with the HIV replicative cycle, inhibition assays on recombinant HIV protease and HIV integrase have been performed. None of the compounds were found active on these two enzymatic targets. Experiments are in progress in order to identify their biological target within the HIV replicative cycle. Copyright (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00055-9

Bioorganic & Medicinal Chemistry 8 (2000) 1253±1262
New Anti-HIV Derivatives: Synthesis and Antiviral Evaluation
C. De Michelis,a,b L. Rocheblave,a,b G. Priem,a,b J. C. Chermannb and J. L. Krausa,b,
*
a
    Â
Laboratoire de Chimie Biomoleculaire, Faculte des Sciences de Luminy, case 901, Universite de la Mediterranee,
Â
70 route Leon Lachamp, 13288 Marseille cedex 9, France
 Â
INSERM U322 `Retrovirus et Maladies associees', Campus Universitaire de Luminy, BP33, 13273 Marseille cedex 9, France
b
Received 9 November 1999; accepted 10 January 2000
AbstractÐA small focused library of 18 compounds incorporating the motif 1,3-(N,N0-dibenzyl)diamino-2-propanol has been syn-
thesized, using adapted synthetic methodologies. These series of compounds were evaluated for their in vitro anti-HIV activity on
infected MT4 cells (syncytium formation observation). Some of the new synthesized compounds show potent anti-HIV activities.
EC50 values for compounds (31, 40, 34, 37 and 46) range from 0.1 to 1 mM. In order to determine at which level these new deriva-
tives interfere with the HIV replicative cycle, inhibition assays on recombinant HIV protease and HIV integrase have been per-
formed. None of the compounds were found active on these two enzymatic targets. Experiments are in progress in order to identify
their biological target within the HIV replicative cycle. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
anti-HIV candidates would be to randomly screen
collections of compounds through a primary screening
test, which translates any cell protection against HIV
cell infection, whatever the viral step at which the
inhibitor acts in the course of the infection. Observation
of MT4 syncytium formation12 constitutes a suitable test
to screen the inhibitory activity of numerous com-
pounds, since whatever the step of the replicative cycle
they target, the appearance of syncytia will be delayed.
This approach o€ers strengths and weaknesses. This
method overcomes the problem that a compound could
be very active on an enzymic in vitro test (antiprotease
or anti-RT) but found inactive on infected cell cultures
because the compound cannot penetrate the cell mem-
brane. Moreover this approach allows the discovery of
totally new chemical structures as potential anti-HIV
compounds. The weakness of this test is that it gives no
information on the mechanism of action of these new
primary analogues, and consequently no information on
the biological target.
Progress in HIV biology has provided detailed knowl-
edge of molecular events in the replicative cycle of HIV-
1. Current understanding of these molecular events in
the HIV life cycle proposes seven steps: viral entry,
reverse transcription, integration, gene expression, gene
assembly, budding and maturation. In theory every
stage in the viral life cycle could serve as a potential
target for designing anti-HIV agents and therapies.1
Currently a quite large number of drugs are known to
act as: virus adsorption inhibitors,2,3 virus-cell fusion
inhibitors,4 virus fusion uncoating inhibitors,5 reverse
transcriptase inhibitors,6 integrase inhibitors,7 Tat and
Rev inhibitors,8 protease inhibitors,9 myristoylation and
glycosylation inhibitors.10 What is remarkable among
these drugs is the large chemical diversity of their mol-
ecular structure. Compounds as diverse as polymers,
peptides, pseudopeptides, modi®ed nucleosides, bis poly-
azamacrocycles, oligonucleotides and polyheterocycles
have been found to be active in vitro on various HIV
infected culture cells and some of them are clinically
used for anti-HIV therapy.11 Usually the design of HIV
compounds is based on the knowledge of well-de®ned
targets such as HIV protease, reverse transcriptase or
chemokine receptors, which allows the design of tailor-
made molecules. Another approach to discover new
It has been shown13 that compounds incorporating the
symmetrical motif 1,3-(N,N0-dibenzyl)diamino-2-propa-
nol elicited anti-HIV activities. Moreover, another rea-
son to employ this motif was that in a random screening
syncytium formation assay a compound from our library
incorporating the motif 1,3-(N,N0-dibenzyl)diamino-2-
propanol slightly protected HIV infected MT4 cells
against syncytium formation. Furthermore the reason
which strengthened the use of this symmetrical structural
*Corresponding author. Tel.: +33-9126-9141; fax: +33-4-9182-9416;
e-mail: kraus@luminy.univ-mrs.fr
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00055-9

1254
C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
motif in the design of anti-HIV derivatives was the
results reported by Kemp et al. which show that com-
pounds incorporating this motif allow anti-HIV activity.
In order to explore more deeply the anti-HIV properties
of this type of compounds, we have synthesized a small
focused library based on this structural motif, sub-
stituted at the two end positions by various mono- or
polyaminated substituted chains as indicated in Figure
1. It is known that aminated chains allow a reduction in
the peptidic character of bioactive molecules by repla-
cing peptidic residues by alkyl diamine without impair-
ing the biological activity of the resulting derivatives.14
Our research approach is to discover new lead com-
pounds for potential anti-HIV agents and to modify
these compounds to ®nd still more potent anti-HIV
agents. The present paper describes the synthesis and
the antiviral evaluation of a new family of compounds,
using observation of syncytium formation as screening
test.
Results and Discussion
Chemistry
As shown in Schemes 1, 2 and 3, the synthesis of poly-
aminated inhibitors required the preliminary preparation
of speci®c synthons. The N-Boc and N-benzyl-N-Boc
protected carboxylic acid synthons 6, 7, 11, 12, 13 and
Figure 1. General structure of the synthesized compounds.
Scheme 1. (i) Boc2O, CH2Cl2, 0 ꢀC; (ii) PDC, DMF, rt.
Scheme 2. (i) Na2SO4, CH2Cl2, rt; (ii) NaBH4, EtOH, 0 ꢀC; (iii) Boc2O, CH2Cl2, 0 ꢀC; (iv) PDC, DMF, rt; (v) TsCl, Et2O/Et3N, rt; (vi) various
o-aminoalcohols, K2CO3, CH3CN, rt.

C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
1255
21, 22, 23, 24 were prepared from commercial o-amino
alcohols 1, 2 according to a synthetic route shown in
Schemes 1 and 2. After Boc protection, the resulting
alcohols 4, 5 and 8, 9, 10 were oxidized into their cor-
responding carboxylic acids 6, 7 and 11, 12, 13 using
PDC in dry DMF (Schemes 1 and 2).
N-Boc protected amino alcohols 8, 9, 10 were ®rst tosy-
lated in order to obtain the corresponding derivatives 14,
15, 16. These were then condensed on various o-amino
alcohols. After a protection step of the N-diaminoalkyl
derivatives, the resulting compounds 17, 18, 19, 20 were
oxidized into the ®nal carboxylic synthons 21, 22, 23, 24
using PDC as catalyst in dry DMF.
The fully protected ketone 28 synthon was obtained
from the following sequences: condensation of benzal-
dehyde on 2-hydroxy-1,3-diaminopropane, followed by
a subsequent reduction with NaBH4 and Boc protection
of the resulting amine, led to the hydroxy derivative 27,
which was oxidized into the corresponding ketone 28
using Swern oxidation conditions15 (Scheme 3).
All the obtained carboxylic acid intermediates 6, 7, 11,
12, 13, 21, 22, 23 and 24 were coupled using BOP
reagent to the deprotected ketone synthon 28. Using
NaBH4 in ethanol (Scheme 3) the resulting analogues
29, 30, 31, 32, 33, 34, 35, 36, 37 were then reduced into
the corresponding alcohol derivatives 38, 39, 40, 41, 42,
43, 44, 45, 46.
Protected diaminoalkyl carboxylic acids 21, 22, 23, 24
were prepared in 7 steps from o-amino alcohols 1, 2, 3
according to a standard route summarized in Scheme
2.
Both series of ketones (29, 30, 31, 32, 33, 34, 35, 36, 37)
and alcohols (38, 39, 40, 41, 42, 43, 44, 45, 46) were then
submitted to anti-HIV screening.
Scheme 3. (i) Na2SO4, CH2Cl2, rt; (ii) NaBH4, EtOH, 0 ꢀC; (iii) Boc2O, CH2Cl2, 0 ꢀC; (iv) DMSO, TFAA, Et3N, CH2Cl2, 60 ꢀC; (v) TFA, CH2Cl2,
rt; (vi) BOP, RCOOH, Et3N, CH2Cl2, rt.

1256
C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
Table 1. HIV activities of the new derivatives on MT4 HIV infected cells
SIc
Log Pd
a
b
No.
R
X
EC50
CC50
29
ˆO
1
>1
>1
8.28‹0.77
38
30
39
31
40
-OH
ˆO
Inactive
1
10
Ð
7.82‹0.72
9.13‹0.75
9.13‹0.75
9.58‹0.74
9.12‹0.70
>50
Toxic
>50
50
>50
Ð
-OH
ˆO
Inactive
0.1±1
0.1±1
50±500
50±500
-OH
32
41
33
42
34
ˆO
-OH
ˆO
1
10
10
10
Ð
4.94‹0.63
4.48‹0.57
6.43‹0.59
5.97‹0.52
10.75‹0.92
Inactive
NTe
NTe
10
Ð
-OH
ˆO
Inactive
0.1±1
Ð
50
50±500
43
35
44
36
45
37
-OH
ˆO
10
Inactive
NTe
>50
Ð
>5
Ð
10.29‹0.88
13.16‹0.90
12.70‹0.86
11.57‹0.91
11.11‹0.87
14.15‹0.90
-OH
ˆO
NTe
50
Ð
10±50
NTe
1±50
Ð
-OH
ˆO
NTe
50
0.1
500
(continued on next page)

C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
1257
Table 1 (continued)
a
b
No.
R
X
EC50
CC50
10
SIc
Log Pd
46
-OH
0.1±1
10±100
13.69‹0.85
aEC50 was de®ned as the concentration of the compound at mM concentrations that inhibited 50% of the viral replication.
bCC50 was de®ned as the concentration of the compound at mM concentrations that reduced the viability of the cells to 50% compared to untreated
cells.
cSI: Selective index: SI=CC50/EC50
.
dLog P: partition coecient (octanol±water) calculated from ACD software.
eNT: not tested.
Virology
rather high lipophilicity should confer them a similar dif-
fusion rate into the cells. Such observations have been
reported by Agrawal et al.16 in the case of various lipo-
philic AZT prodrugs. It can be con®rmed that as shown
in Table 2, the structural motif 2-hydroxy- or oxo-1,3-
[N1,N3 -dibenzyl]diaminopropane A (Fig. 1) is required for
HIV activity since polyaminated carboxylic chains alone
(compounds 13, 19) do not show any anti-HIV activity.
Syncytium formation. The newly synthesized analogues
were evaluated for their inhibitory e€ects on HIV repli-
cation in MT4 cell cultures (Table 1). Under the assay
conditions, some of the new compounds elicited potent
anti-HIV activities, with EC50 values ranging from 0.1
to 10 mM (29, 30, 31, 32, 34, 37, 40, 43, 46). It can also
be observed that cytotoxicity values (CC50) of the new
analogues are relatively high. Their Selective Index
values range from 1 to 50 (Table 1). This observation
suggests that the analogues permeated cell membranes
quite easily. One explanation could be the relatively
high lipophilicity of these derivatives. Indeed log P
values of the studied compounds (Table 1) calculated
using ACD Software (Advanced Chemistry Develop-
ment, Inc) range from 4 to 14. Derivatives like 31, 34,
37, 40, 46 whose EC50 ranged between 0.1 and 1 mM
could represent interesting hits which could be opti-
mized, since we have found that, in a similar assay (MT4
syncytium formation), the anti-RT inhibitor 3TC and
the antiprotease Saquinavir elicited EC50 values of 0.1
and 0.05 mM respectively. Saquinavir was the most
potent compound under our testing conditions.
0
Antiprotease assay. Both series of derivatives ketones
(29, 30, 31, 32, 33, 34, 35, 36, 37) and alcohols (38, 39,
40, 41, 42, 43, 44, 45, 46) were also screened for their
HIV-1 protease inhibitory activity according to a classi-
cal HIV protease assay17 described in the Experimental.
We found that both series of compounds tested at 10 mM
do not demonstrate any signi®cant HIV-1 protease inhi-
bitory activity. In identical assay conditions, the HIV-1
protease inhibitor Saquinavir18 elicited enzyme inhibi-
tion in the nanomolar range (IC50=5 nM). These results
clearly indicate that under these assay conditions, the
new synthesized derivatives do not interfere at the HIV
protease level. These results are in good agreement with
the HIV protease molecular modelling predictions, since
none of the derivatives studied presented favourable ®t-
ting energies (ÁEc) within the HIV protease active-site.
However, these results should be examined cautiously
since it has been suggested that HIV protease inhibitors
not only interfered with the virus maturation in the late
phase of the HIV replication cycle but can also interfere
From the results reported in Table 1, several comments
can be made:
1. The presence of an hydroxyl group or a ketone at
the 2 position of the diamino propyl motif (Fig. 1)
led to active anti-HIV compounds.
2. As far as cytotoxicity of the new analogues is
concerned, it can be observed that some of the
active compounds are relatively cytotoxic on
infected MT4 cells (30) while other compounds
(31, 34, 36, 37, 40) show selectivity index values
(SI) ranging from 50 to 500. This observation is
quite intriguing since the coecient partition
values (log P) calculated for the whole compounds
are quite high (log P values ranging from 4 to 14).
Table 2. Anti-HIV activities of compounds 5d, 7h and 8h
No.
Compound
EC50 CC50
Log Pc
a
b
13
Inactive 50 3.24‹0.56
Inactive 50 5.09‹0.62
19
From the results reported in Table 2, the fact that
compounds 13, 19 are found to be inactive on HIV-
replication in MT4 cells suggests that both moieties
polyaminated or mono- and 1,3-diamino-2-propanol or
ketone represented in Figure 1 are required for anti-HIV
activity. If we make the assumption that these analogues
cross the cell membranes by non-facilitated di€usion, the
aEC50 was de®ned as the concentration of the compound at mM con-
centrations that inhibited 50% of the viral replication.
bCC50 was de®ned as the concentration of the compound at mM con-
centrations that reduced the viability of the cells to 50% compared to
untreated cells.
cLog P: partition coecient (octanol±water) calculated from ACD
software.

1258
C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
in an early phase of the cycle, before the formation of
the provirus. This result strengthened the argument that
these derivatives are probably not transition state HIV
protease inhibitors. Indeed very often HIV protease
inhibitors are characterized by an hydroxyethylamine
moiety which mimics the hydrolysis transition state.
All reagents were of commercial quality (Aldrich Com-
pany) from freshly opened containers.
The described compounds were obtained using the fol-
lowing procedures.
General procedure for !-N-tert-butyloxycarbonylamino
carboxylic acids 6, 7. To a solution of o-amino alcohol 1,
2 (1equiv) in CH2Cl2 (40 mL) cooled to 0 ꢀC, Boc2O was
added slowly (1.1 equiv). The reaction was followed by
TLC. After 3 h, the solvent was removed and the crude
product (1 equiv) obtained was directly dissolved in dry
DMF (52.5 mL). PDC was added (3.5 equiv) and the solu-
tion was stirred overnight under nitrogen. The reaction was
stopped by adding 10 volumes (525 mL) of a solution:
H2O:Brine (1:1). The solution was extracted with Et2O
(2Â200 mL). The organic phase was washed with Brine
solution (2Â150 mL), 1N NaOH solution (2Â200 mL).
The aqueous layer was acidi®ed by 1N HCl solution
(2Â200 mL), then extracted with CH2Cl2 (2Â150 mL). The
organic layer was dried over Na2SO4 and evaporated.
Integrase inhibition assay. The most active compounds
(13b, 15c, 16c, 21d, 27g, 28g) were also tested as possible
integrase inhibitors according to a screening protocol
described by Pommier et al.7 None of these compounds
were found to be integrase inhibitors.
Conclusion
Through this study it can be observed that some of the
new analogues bearing a ketone or an hydroxyl function
(11a, 13b, 15c, 16c, 17a, 21d, 22d, 25e, 27g, 28g) elicited
potent in vitro anti-HIV activity on infected MT4 cells.
Enzymatic inhibition studies on recombinant HIV pro-
tease and on integrase reveal that these new series of
compounds appear to act neither at the HIV protease
level of the HIV replication cycle nor at the integrase
level. One way to determine at what stage of the HIV
replicative cycle these compounds interfere would be to
perform time-of-addition experiments, whereby the new
derivatives are added at di€erent times after infection.
However it seems that this series of compounds do not
have speci®c anti-HIV activities.
3-N-tert-Butyloxycarbonylamino propanoic acid (6)
1
(yield: 48%). H NMR (CDCl3): d 1.45 (11H, m, Boc,
CH2-CH2-CH2); 2.46 (2H, t, CH2COOH); 2.91 (2H, m,
CH2N); 4.63 (1H, m, NH).
6-N-tert-Butyloxycarbonylamino hexanoic acid (7)
1
(yield: 43%). H NMR (CDCl3): d 1.46 (13H, m, Boc,
CH2-(CH2)2-CH2); 2.47 (2H, t, CH2COOH); 2.92 (2H,
m, CH2N); 4.68 (1H, m, NH).
This result suggests that both approaches for the
discovery of new anti-HIV drugs are complemen-
tary. The rational approach, allowing tailor-made
molecules to address to well-de®ned targets is well
adapted for optimization process, while random screen-
ing using a multiple target screening model (syncytium
formation) appears to be suitable for new anti-HIV
drug discovery.
General procedure for !-N-[N-(tert-butyloxycarbonyl)-
benzylamino]-carboxylic acids 11, 12, 13. To a solution
of o-amino alcohol 1, 2, 3 (1 equiv) in CH2Cl2 (150 mL),
benzaldehyde (1.1 equiv) and Na2SO4 (1 equiv) were
added at room temperature. The solution was stirred
overnight under nitrogen. Na2SO4 was ®ltered and the
solvent evapored under vacuum. The crude product 8,
9, 10 was directly used without further puri®cation. To
a solution of the previous product 8, 9, 10 (1 equiv) dis-
solved in EtOH (15 mL) and cooled to 0 ꢀC, NaBH4 was
added slowly (1.5 equiv). After complete reaction (2 h),
the reaction was quenched by adding H2O (10 mL). The
solvent was removed. The residue was dissolved in
EtOAc (40 mL) and washed with 5% NaHCO3 solution
(1Â20 mL). The organic layer was dried over Na2SO4,
®ltered and evaporated. The crude product was directly
used for the next step without puri®cation.
Experimental
Chemistry
Nuclear magnetic resonance spectra were recorded with
a Bruker AC-250 (1H NMR); chemical shifts are
expressed as d units (part per million) down®eld from
TMS. Fast atom bombardment mass spectral analyses
were obtained by Dr. Astier (Laboratoire de Mesures
Physiques-RMN, USTL, Montpellier, France) on a
Jeol DX-100 using a cesium ion source and glycerol:
thioglycerol (1:1) (GT) or m-nitrobenzyl alcohol
(NOBA) as matrix. Mass calibration was performed
using cesium iodide. Thin layer chromatography (TLC)
and preparative layer chromatography (PLC) were per-
formed using silica gel plates 0.2, 1 or 2 mm thick
(60F254 Merck). Preparative ¯ash column chromato-
graphy was carried out on silica gel (230±240 mesh, G60
Merck). Analytical HPLC was performed on a Waters
600E instrument with M991 photodiode detector using
the following conditions: 4.6Â150 mm Column (Waters
Spherisorb S5 ODS2, 5 mM).
To a solution of the previous product (1 equiv) in
CH2Cl2 (70 mL), Boc2O was added slowly at 0 ꢀC
(2.5 equiv). Once the reaction was complete (3 h), the
solvent was evaporated and the crude Boc residue was
isolated in very good yield. 3.7 mmol of the previous
residue were directly dissolved in dry DMF (9.8 mL)
before PDC was added (3.5 equiv). After 3 h, the reaction
was stopped by adding 10 volumes (98 mL) of a solution:
H2O:Brine (1:1) and the solution was extracted with
Et2O (3Â50 mL). The organic phase was consecutively
washed with Brine solution (2Â30 mL) and 1N NaOH
solution (2Â30 mL). The aqueous layer was acidi®ed by
addition of 1N HCl solution (1Â30 mL), then extracted

C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
1259
with CH2Cl2 (3Â30 mL). The organic layer was dried
tert-Butyl-N-benzyl-N-[3-N-tert-butyl(benzylamino)]pro-
1
over Na2SO4 before evaporation.
pyl-N-(3-hydroxypropyl)carbamate (17) (yield: 90%). H
NMR (CDCl3): d 1.20 (13H, m, Boc, 2 CH2); 2.95 (2H,
m, NCH2); 3.44 (2H, t, CH2OH); 4.24 (2H, s, PhCH2);
7.10 (5H, m, Ph).
3-N-[N(tert-Butyloxycarbonyl)benzylamino]-propanoic acid
(11) (yield: 44%). 1H NMR (CDCl3): d 1.36 and 2.31
(11H, m, CH2, Boc), 2.75 (2H, t, CH2COOH), 3.05 (2H, s,
CH2N), 4.25 (2H, s, CH2Ph), 7.05 (5H, m, Ph).
tert-Butyl-N-benzyl-N-[4-N-tert-butyl(benzylamino)]pro-
1
pyl-N-(4-hydroxypentyl)carbamate (18) (yield: 94%). H
4-N-[N(tert-Butyloxycarbonyl)benzylamino]-pentanoic
1
NMR (CDCl3): d 1.22 (17H, m, Boc, 4 CH2); 2.96 (2H,
m, NCH2); 3.45 (2H, t, CH2OH); 4.25 (2H, s, PhCH2);
7.11 (5H, m, Ph).
acid (12) (yield: 46%). H NMR (CDCl3): d 1.30 and
2.34 (13H, m, 2 CH2, Boc), 2.70 (2H, t, CH2COOH), 3.05
(2H, s, CH2N), 4.25 (2H, s, CH2Ph), 7.05 (5H, m, Ph).
tert-Butyl-N-benzyl-N-[5-N-tert-butyl(benzylamino)]pentyl
1
6-N-[N(tert-Butyloxycarbonyl)benzylamino]-hexanoic acid
1
- N - (5 - hydroxypentyl)carbamate (19) (yield: 89%). H
(13) (yield: 42%). H NMR (CDCl3): d 1.32 and 2.32
(17H, m, 4 CH2, Boc), 2.70 (2H, t, CH2COOH), 3.05
(2H, s, CH2N), 4.25 (2H, s, CH2Ph), 7.05 (5H, m, Ph).
NMR (CDCl3): d 1.20 (19H, m, Boc, 5 CH2); 2.95 (2H,
m, NCH2); 3.48 (2H, t, CH2OH); 4.27 (2H, s, PhCH2);
7.12 (5H, m, Ph).
General procedure for ! - [benzyl(tert - butoxycarbonyl)-
amino]-!-alkyl-1-benzenesulfonate 14, 15, 16. To a solu-
tion of o-amino alcohol 1, 2, 3 (1 equiv) in Et2O:Et3N (1:1),
tosyl chloride was added to the solution (1.1 equiv). After
3 h, the solvent was evaporated. The product dissolved in
EtOAc (1Â20 mL) was washed with 5% citric acid solution
(1Â15 mL), H2O (1Â15 mL), 5% NaHCO3 solution
(1Â15 mL), then the organic layer was dried over Na2SO4,
®ltered and evaporated. The crude product (1.74 g) was
puri®ed by ¯ash chromatography with EtOAc:Hexane as
eluent (3:7) to give pure compounds 14, 15, 16.
tert-Butyl-N-benzyl-N-[6-N-tert-butyl(benzylamino)]hex-
1
yl-N-(6-hydroxyhexyl)carbamate (20) (yield: 88%). H
NMR (CDCl3): d 1.21 (23H, m, Boc, 7 CH2); 2.92 (2H,
m, NCH2); 3.46 (2H, t, CH2OH); 4.22 (2H, s, PhCH2);
7.10 (5H, m, Ph).
General procedure for !-[!-N-tert-butyl(benzylamino)
alkyl](tert-butoxycarbonyl)amino] carboxylic acids 21, 22,
23, 24. PDC (3.5 equiv) was added to a solution of 17,
18, 19, 20 (1 equiv) in dry DMF (3.6 mL). The solution
was stirred overnight under nitrogen and the solvent
was evaporated. After 3 h, the reaction was stopped by
adding 10 volumes (40 mL of a (1:1) H2O:Brine solu-
tion). The solution was extracted with Et2O (3Â15 mL)
and the organic layer was washed with Brine solu-
tion (2Â10 mL), 1N NaOH solution (2Â10 mL). The
aqueous layer was acidi®ed by addition of a solution of
1N HCl (1Â10 mL), then extracted with CH2Cl2
(2Â15 mL), dried over Na2SO4, and evaporated. The
pure compounds 21, 22, 23, 24 were isolated in good
yield.
3-[Benzyl(tert-butoxycarbonyl)amino]propyl-3-propyl-1-
1
benzenesulfonate (14) (yield: 50%). H NMR (CDCl3): d
1.61 (11H, m, CO-CH2-CH2-CH2-OTs, Boc); 2.62 (3H,
m, CH3-Ph); 3.20 (2H, m, CH2Ph); 3.74 (2H, t, CH2OTs);
4.91 (2H, s, CH2Ph) 7.66 (9H, m, Ph, CH3Ph).
4-[Benzyl(tert-butoxycarbonyl)amino]pentyl-4-pentyl-1-
1
benzenesulfonate (15) (yield: 48%). H NMR (CDCl3): d
1.62 (15H, m, CO-CH2-(CH2)3-CH2-OTs, Boc); 2.63 (3H,
m, CH3-Ph); 3.22 (2H, m, CH2Ph); 3.71 (2H, t, CH2OTs);
4.94 (2H, s, CH2Ph) 7.67 (9H, m, Ph, CH3Ph).
3-[3-N-tert-Butyl(benzylamino)propyl](tert-butoxycarbo-
nyl)amino]propanoic acid (21) (yield: 75%). 1H NMR
(CDCl3): d 1.22 (20H, m, Boc, CH2); 2.10 (2H, t,
CH2COOH); 2.98 (2H, s, NCH2); 4.22 (2H, m, CH2Ph);
7.16 (5H, m, Ph).
6-[Benzyl(tert-butoxycarbonyl)amino]hexyl-6-hexyl-1-ben-
zenesulfonate (16) (yield: 53%). 1H NMR (CDCl3): d
1.61 (17H, m, CO-CH2-(CH2)4-CH2-OTs, Boc); 2.62 (3H,
m, CH3-Ph); 3.21 (2H, m, CH2Ph); 3.70 (2H, t, CH2OTs);
4.94 (2H, s, CH2Ph) 7.62 (9H, m, Ph, CH3Ph).
4-[4-N-tert-Butyl(benzylamino)propyl](tert-butoxycarbo-
nyl)amino]pentanoic acid (22) (yield: 72%). 1H NMR
(CDCl3): d 1.20 (24H, m, Boc, 3 CH2); 2.12 (2H, t,
CH2COOH); 2.94 (2H, s, NCH2); 4.23 (2H, m, CH2Ph);
7.14 (5H, m, Ph).
General procedure for tert-butyl-N-benzyl-N-[!-N-tert-
butyl(benzylamino)]alkyl-N-(!-hydroxyalkyl)carbamates
17, 18, 19, 20. To a solution of product 14, 15, 16
(1 equiv) in CH3CN (20 mL), o-amino alcohol 1, 2, 3
(1.2 equiv) and K2CO3 (3 equiv) were added. After 5
days at room temperature, K2CO3 was ®ltered o€ and
the solvent evaporated. The remaining crude product
was directly used for the next step without puri®cation.
6-[6-N-tert-Butyl(benzylamino)pentyl](tert-butoxycarbo-
1
nyl)amino]pentanoic acid (23) (yield : 70%). H NMR
(CDCl3): d 1.22 (28H, m, Boc, 5 CH2); 2.10 (2H, t,
CH2COOH); 2.96 (2H, s, NCH2); 4.24 (2H, m, CH2Ph);
7.15 (5H, m, Ph).
To a solution of the previous product (1 equiv) in
CH2Cl2 (15 mL) cooled to 0 ꢀC, Boc2O was added
(1.2 equiv). After 2 h, the solvent was evaporated and
the compound was puri®ed by ¯ash chromatography
with EtOAc:Hexane as eluent (5:5). The pure products
17, 18, 19, 20 were isolated in very good yield.
6-[6-N-tert-Butyl(benzylamino)hexyl](tert-butoxycarbonyl)
amino]hexanoic acid (24) (yield: 69%). 1H NMR
(CDCl3): d 1.22 (32H, m, Boc, 7 CH2); 2.14 (2H, t,
CH2COOH); 2.94 (2H, s, NCH2); 4.25 (2H, m, CH2Ph);
7.19 (5H, m, Ph).

1260
C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
1,3-di[N-(N0-tert-Butyloxycarbonyl)benzylamino]-propan-2-
ol (27). To a solution of 1,3-diaminopropan-2-ol (1
equiv: 33.30 mmol, 3 g) in CH2Cl2 (100 mL), Na2SO4
(2 equiv, 66.60 mmol, 9.50 g) then slowly benzaldehyde
were added (2 equiv, 66.60 mmol, 6.80 mL). The reaction
was stirred overnight under nitrogen. After ®ltration of
Na2SO4, the solvent was evaporated. The crude product
was puri®ed by recrystallization in hot hexane (250 mL).
White crystals of diimine product were isolated after
®ltration (3.31 g, 20 mmol, 60% yield).
tered and evaporated. The crude product (330 mg) was
puri®ed by ¯ash chromatography with EtOAc:hexane
(5:5) as eluent to give pure compounds 29, 30, 31, 32,
33, 34, 35, 36, 37.
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-3-amino
propanoyl benzylamino]-propan-2-one (29) (yield: 85%).
1H NMR (CDCl3): d 1.30 (30H, m, COCH2-CH2-CH2,
Boc); 2.22 (4H, t, CH2CON); 3.01 (4H, m, CH2CH2N);
4.05 (4H, m, NCH2CO); 4.20 (4H, m, PhCH2NCO); 4.41
(4H, m, PhCH2NBocCH2); 7.10 (20H, m, Ph). Mass
spectrum (NOBA, FAB+): 833 (M+H)+. Rf=0.5 in
EtOAc:hexane (2:3).
The diimine (1 equiv: 19 mmol, 5.08 g) was dissolved in
EtOH (100 mL), cooled to 0 ꢀC and NaBH4 was added
slowly (2.5 equiv: 48 mmol, 1.80 g). The reaction was
quenched by adding H2O (48 mL). The solvent was
removed and the residue dissolved in EtOAc (40 mL)
was washed with 5% NaHCO3 solution (1Â10 mL).
The organic layer was washed with 1N HCl solution
(2Â10 mL). Then, the aqueous layer was basi®ed with
1N NaOH solution (2Â10 mL) and extracted with
CH2Cl2 (2Â20 mL). The resulting organic layer was
dried over Na2SO4, ®ltered and evaporated. The pure
product (3.67 g, 13.70 mmol) was isolated in 72% yield.
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-4-amino
pentanoyl benzylamino]-propan-2-one (30) (yield: 87%).
1H NMR (CDCl3): d 1.31 (30H, m, COCH2-(CH2)2-CH2,
Boc); 2.20 (4H, t, CH2CON); 3.08 (4H, m, CH2CH2N);
4.10 (4H, m, NCH2CO); 4.23 (4H, m, PhCH2NCO);
4.44 (4H, m, PhCH2NBocCH2); 7.16 (20H, m, Ph).
Mass Spectrum (NOBA, FAB+): 847 (M+H)+.
Rf=0.4 (EtOAc:Hexane: 2:3).
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-3-amino
hexanoyl benzylamino]-propan-2-one (31) (yield: 85%).
1H NMR (CDCl3): d 1.33 (30H, m, CO-CH2-(CH2)4-
CH2, Boc); 2.21 (4H, t, CH2CON); 3.30 (4H, m,
CH2CH2N); 4.00 (4H, m, NCH2CO); 4.21 (4H, m,
PhCH2NCO); 4.40 (4H, m, PhCH2NBocCH2); 7.11
(20H, m, Ph). Mass spectrum (NOBA, FAB+): 861
(M+H)+. Rf=0.5 (EtOAc:Hexane: 2:3).
To a solution of 1,3-di(N-benzylamino)-2-propanol
(1 equiv: 3.70 mmol, 1 g) in CH2Cl2 (1Â20 mL) cooled to
0ꢀC, Boc2O was added slowly (2.2 equiv: 8.15 mmol,
17.80 g). After 30 min, the solvent was evaporated. The
crude product was puri®ed by ¯ash chromatography
with EtOAc:Hexane (1:4) as eluent. The pure product
27 (1.55 g, 3.30 mmol) was isolated in 89% yield. H
NMR (CDCl3): d 2.61 (4H, m, 2 CH2), 3.83 (3H, 2s,
CH2Ph and CHOH), 7.32 (10H, m, Ph).
1
1,3-di[N0-(tert-Butyloxycarbonyl)-3-amino
propanoyl
1
benzylamino]-propan-2-one (32) (yield: 89%). H NMR
(CDCl3): d 1.43 (22H, m, 2Boc, 2CH2); 2.61 (4H, m,
CH2CON); 3.50 (4H, m, BocNHCH2); 4.01 (4H, m,
CH2COCH2); 4.61 (4H, m, PhCH2); 7.30 (10H, m, Ph).
Mass spectrum (NOBA, FAB+): 611 (M+H)+. Rf=0.35
(EtOAc:Hexane: 1:1).
1,3-di-N-[N0-(tert-Butyloxycarbonyl)benzylamino]ace-
tone (28). A solution of DMSO (2 equiv: 7.40 mmol,
530 mL) in dry CH2Cl2 (15 mL) was cooled to 60 ꢀC
under nitrogen before tri¯uoroacetic anhydride was
added (1.5 equiv: 5.60 mmol, 795 mL). After 10 min, the
compound 27 (1 equiv: 3.70 mmol, 1.73 g) was added to
the solution. After 45 min, Et3N (4.4 equiv: 16.30 mmol,
2.30 mL) was added. The reaction was quenched with
H2O (1Â10 mL). The organic phase was washed with
H2O (2Â15 mL), dried over Na2SO4, ®ltered and evapo-
rated. The crude product (2.08 g) was puri®ed by ¯ash
chromatography with EtOAc:Hexane: (15:85) as eluent.
Pure compound 28: (1.10 g, 2.85 mmoles, 64% yield). 1H
NMR (CDCl3): d 1.31 (18H, 2s, 2Boc); 3.79 (4H, q,
BocNCH2CO); 4.36 (4H, s, CH2Ph); 7.15 (10H, m, Ph).
1,3-di[N0-(tert-Butyloxycarbonyl)-6-amino hexanoyl ben-
zylamino]-propan-2-one (33) (yield: 88%). 1H NMR
(CDCl3): d 1.43 (30H, m, 2Boc, 6CH2); 2.63 (4H, m,
CH2CON); 3.55 (4H, m, BocNHCH2); 4.02 (4H, m,
CH2COCH2); 4.65 (4H, m, PhCH2); 7.36 (10H, m, Ph).
Mass spectrum (NOBA, FAB+): 695 (M+H)+ Rf=0.4
(EtOAc:Hexane: 9:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-Butyloxycarbonyl)-3-pro-
pyl]-N0-tert-Butyloxycarbonyl-3-amino propanol benzyla-
General procedure for 1,3-di[N-[N0-(N-benzyl-N0-tert-bu-
tyloxycarbonyl)-!-amino alkanoyl benzylamino]-propan-
2-one 29, 30, 31, 32, 33, 34, 35, 36, 37. Tri¯uoroacetic
acid was added (30 equiv) to a solution of compound 28
(1 equiv) in CH2Cl2 (2 mL). After 20 min, the solvent
was evaporated. The dried crude remaining product was
dissolved in CH2Cl2 (2 mL) before the addition of BOP
(2.4 equiv) compound 6, 7, 11, 12, 13, 21, 22, 23 or 24
(2.4 equiv) and Et3N (12 equiv). After 2 h, the solvent
was removed. The residue dissolved in EtOAc (3 mL)
was washed with 5% NaHCO3 solution (2Â4 mL), then
Brine solution (1Â4 mL), and 5% citric acid solution
(2Â4 mL). The organic phase was dried (Na2SO4), ®l-
mino]-propan-2-one (34) (yield: 85%). H NMR (CDCl3):
1
d ;1.42 (40H, m, 4Boc, 2CH2); 2.41 (4H, 2t, COCH2); 3.11
(12H, m, PhCH2BocNCH2, CH2NBocCH2); 4.10 (4H, s,
CH2COCH2); 4.45 (4H, s, PhCH2BocNCH2); 4.66 (4H, s,
PhCH2BocNCH2); 7.30 (20H, m, Ph). Mass spectrum
(NOBA, FAB+): 1105 (M+H)+. Rf=0.4 (EtOAc:
Hexane: 1:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-5-pen-
tyl]-N0-tert-butyloxycarbonyl-5-amino pentanol benzyl-
1
amino]-propan-2-one (35) (yield: 83%). H NMR (CDCl3):
d 1.40 (56H, m, 4Boc, 10CH2); 2.44 (4H, 2t, COCH2);
3.11 (12H, m, PhCH2BocNCH2, CH2NBocCH2); 4.12

C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
1261
(4H, s, CH2COCH2); 4.45 (4H, s, PhCH2BocNCH2);
4.60 (4H, s, PhCH2BocNCH2); 7.30 (20H, m, Ph). Mass
spectrum (NOBA, FAB+) 1219 (M+H)+. Rf=0.4
(EtOAc:Hexane: 2:3).
1,3-di[N0-(tert-Butyloxycarbonyl)-3-amino propanoil ben-
zylamino]-propan-2-ol (41) (yield: 64%). 1H NMR
(CDCl3): d 1.43 (22H, m, 2Boc, 2CH2); 2.62 (4H, m,
CH2CON); 2.98 (4H, m, BocNHCH2); 3.17 (4H, m,
CH2CHO HCH2); 3.50 (1H, m, CHOH); 4.64 (4H, m,
PhCH2); 7.31 (10H, m, Ph). Mass spectrum (NOBA,
FAB+) : 613 (M+H)+. Rf=0.4 (EtOAc:Hexane: 1:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-3-pro-
pyl]-N0-tert-butyloxycarbonyl-5-amino pentanol benzyl-
amino]-propan-2-one (36) (yield: 81%). 1H NMR
(CDCl3): d 1.47 (56H, m, 4Boc, 10CH2); 2.44 (4H, 2t,
COCH2); 3.12 (12H, m, PhCH2BocNCH2, CH2NBoc-
CH2); 4.10 (4H, s, CH2COCH2); 4.43 (4H, s,
PhCH2BocNCH2); 4.65 (4H, s, PhCH2BocNCH2); 7.36
(20H, m, Ph). Mass spectrum (NOBA, FAB+): 1163
(M+H)+. Rf=0.4 (EtOAc:Hexane: 2:3).
1,3-di[N0-(tert-Butyloxycarbonyl)-6-amino hexanoyl ben-
zylamino]-propan-2-ol (42) (yield: 65%). 1H NMR
(CDCl3): d 1.43 (30H, m, 2Boc, 6CH2); 2.64 (4H, m,
CH2CON); 2.89 (4H, m, BocNHCH2); 3.10 (4H, m,
CH2CHOHCH2); 3.51 (1H, m, CHOH); 4.67 (4H, m,
PhCH2); 7.30 (10H, m, Ph). Mass spectrum (NOBA,
FAB+): 697 (M+H)+. Rf=0.5 (EtOAc:Hexane: 3:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-6-hex-
yl]-N0-tert-butyloxycarbonyl-6-amino hexanol benzyl-
amino]-propan-2-one (37) (yield: 82%). 1H NMR (CDCl3):
d 1.42 (68H, m, 4Boc, 16CH2); 2.40 (4H, 2t, COCH2);
3.15 (12H, m, PhCH2BocNCH2, CH2NBocCH2); 4.12
(4H, s, CH2COCH2); 4.46 (4H, s, PhCH2BocNCH2);
4.69 (4H, s, PhCH2BocNCH2); 7.37 (20H, m, Ph). Mass
spectrum (NOBA, FAB+): 1261 (M+H)+. Rf=0.4
(EtOAc:Hexane: 4:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-3-pro-
pyl]-N0-tert-butyloxycarbonyl-3-amino propanol benzyl-
amino]-propan-2-ol (43) (yield: 60%). H NMR (CDCl3):
d 1.40 (40H, m, 4Boc, 2CH2); 2.42 (4H, 2t, COCH2);
3.14 (12H, m, PhCH2BocNCH2, CH2NBoc CH2); 3.28
(4H, m, CH2CHOHCH2); 3.55 (1H, m, CHOH); 4.41
(4H, s, PhCH2BocNCH2); 4.64 (4H, s, PhCH2Boc
NCH2); 7.37 (20H, m, Ph). Mass spectrum (NOBA,
FAB+): 1107 (M+H)+. Rf=0.4 (EtOAc:Hexane: 2:3).
1
General procedure for 1,3-di[N-[N0-(N-Benzyl-N0-tert-bu-
tyloxycarbonyl-!-amino alkanoyl benzylamino]-propan-
2-ol 38, 39, 40, 41, 42, 43, 44, 45, 46. To a solution of
the previous products 29, 30, 31, 32, 33, 34, 35, 36, 37
(1 equiv) in EtOH (5 mL) cooled to 0 ꢀC, NaBH4 was
added (1.2 equiv). After 10 min, the solvent was evapo-
rated. The product dissolved in EtOAc (6 mL) was
washed in brine solution (2Â4 mL), dried over Na2SO4,
®ltered and evaporated. The crude product was puri®ed
by PLC with EtOAc:hexane as eluent: (2:3) to give pure
compounds 38, 39, 40, 41, 42, 43, 44, 45, 46.
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-5-pen-
tyl]-N0-tert-butyloxycarbonyl-5-amino pentanol benzyl-
amino]-propan-2-ol (44) (yield: 66%). H NMR (CDCl3):
d 1.44 (46H, m, 4Boc, 10CH2); 2.42 (4H, 2t, COCH2);
3.11 (12H, m, PhCH2BocNCH2, CH2NBocCH2); 3.22
(4H, m, CH2CHOHCH2); 3.55 (1H, m, CHOH); 4.43
(4H, s, PhCH2BocNCH2); 4.6 7(4H, s, PhCH2Boc
NCH2); 7.35 (20H, m, Ph). Mass spectrum (NOBA,
FAB+): 1221 (M+H)+. Rf=0.4 (EtOAc:Hexane: 2:3).
1
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-3-pro-
pyl]-N0-tert-butyloxycarbonyl-5-amino pentanol benzyl-
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-3-ami-
no]propanoyl benzylamino]-propan-2-ol (38) (yield:
1
amino]-propan-2-ol (45) (yield: 62%). H NMR (CDCl3):
1
63%). H NMR (CDCl3): d 1.30 (18H, 2s, Boc); 2.13
d 1.44 (46H, m, 4Boc, 5CH2); 2.45 (4H, 2t, COCH2);
3.12 (12H, m, PhCH2BocNCH2, CH2NBocCH2); 3.27
(4H, m, CH2CHOHCH2); 3.56 (1H, m, CHOH); 4.43
(4H, s, PhCH2BocNCH2); 4.67 (4H, s, PhCH2Boc
NCH2); 7.35 (20H, m, Ph). Mass spectrum (NOBA,
FAB+): 1165 (M+H)+. Rf=0.4 (EtOAc:Hexane: 1:1).
(4H, t, CH2CON); 2.96 (4H, m, CH2CH2N); 3.51 (1H,
m, CHOH); 4.05 (4H, m, NCH2CO); 4.22 (4H, m,
PhCH2NCO); 4.41 (4H, m, PhCH2NBocCH2); 7.14
(20H, m, Ph). Mass spectrum (NOBA, FAB+): 835
(M+H)+. Rf=0.4 (EtOAc:Hexane: 1:1).
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-4-ami-
no]pentanoyl benzylamino]-propan-2-ol (39) (yield: 62%).
1H NMR (CDCl3): d 1.32 (26H, m, CO-CH2-(CH2)2-
CH2, Boc); 2.14 (4H, t, CH2CON); 2.90 (4H, m, CH2
CH2N); 3.53 (1H, m, CHOH); 4.02 (4H, m, NCH2CO);
4.25 (4H, m, PhCH2NCO); 4.41 (4H, m, PhCH2N
BocCH2); 7.17 (20H, m, Ph). Mass spectrum (NOBA,
FAB+): 849 (M+H)+. Rf=0.35 (EtOAc:Hexane: 9:1).
1,3-N-{N0[N00(N-Benzyl-N0-tert-butyloxycarbonyl)-6-hex-
yl]-N0-tert-butyloxycarbonyl-6-amino hexanol benzyl-
amino]-propan-2-ol (46) (yield: 65%). H NMR (CDCl3):
d 1.42 (56H, m, 64 Boc, 14CH2); 2.42 (4H, 2t, COCH2);
3.13 (12H, m, PhCH2BocNCH2, CH2NBocCH2); 3.21
(4H, m, CH2CHOHCH2); 3.53 (1H, m, CHOH); 4.44
(4H, s, PhCH2BocNCH2); 4.68 (4H, s, PhCH2BocNCH2);
7.32 (20H, m, Ph). Mass spectrum (NOBA, FAB+):
1263 (M+H)+. Rf=0.5 (EtOAc:Hexane: 1:1).
1
1,3-di[N-[N0-(N-Benzyl-N0-tert-butyloxycarbonyl)-6-ami-
no]hexanoyl benzylamino]- propan-2-ol (40) (yield: 61%).
1H NMR (CDCl3): d 1.33 (34H, m, CO-CH2-(CH2)4-
CH2, Boc); 2.10 (4H, t, CH2CON); 2.98 (4H, m, CH2
CH2N); 3.54 (1H, m, CHOH); 4.06 (4H, m, NCH2CO);
4.24 (4H, m, PhCH2NCO); 4.40 (4H, m, PhCH2N
BocCH2); 7.14 (20H, m, Ph). Mass spectrum (NOBA,
FAB+) : 863 (M+H)+. Rf=0.4 (EtOAc:Hexane: 2:3).
Virology
Anti-HIV activity was monitored by the eciency of
drug compounds to inhibit syncytium formation after
HIV-infection of MT4, as already described.19 Brie¯y,
3Â105 MT4 cells were ®rst pre-incubated with 100 mL of
various concentrations of drug compounds dissolved in

1262
C. De Michelis et al. / Bioorg. Med. Chem. 8 (2000) 1253±1262
DMSO or in H2O then diluted in phosphate bu€er sal-
ine solution for one hour at 37 ꢀC. Then 100 mL of an
appropriate virus dilution was added to the mixture and
incubated for another hour at 37 ꢀC. After three washes,
cells were resuspended in culture medium in the pre-
sence or absence of the drug compounds. Cultures were
then continued for 7 days at 37 ꢀC, under 5% CO2
atmosphere and culture medium replaced at day 3 post-
infection (supplemented or not with drug compounds).
Each culture condition was carried out in duplicate. The
appearance of syncytia was followed each day with an
inverted optical microscope. Typically, the virus dilu-
tion used in this assay (multiplicity of infection of 0.1
TCID50/cell) allowed syncytium formation at day 5 post
infection. The inhibitory concentration of drug com-
pounds was expressed as the concentration that caused
50% inhibition of syncytium formation (EC50) without
direct toxicity for the cells. Cytotoxic concentration
(CC50) of drug compounds was monitored on growth of
non-infected cells by trypan blue exclusion assay and
corresponded to the concentration required to cause
50% of cell death.
References
1. (a) Dormont, J. AIDS 1994, 8, S31. (b) Sarin, P.; Goldstein,
A. L. Immunopharmacol. Immunotoxicol. 1995, 17, 217.
2. De Clercq, E. In Carbohydrates and Carbohydrate Polymers,
Analysis, Biotechnology, Modi®cation. Antiviral, Biomedical and
Other Applications, ATL Press: Mt. Prospect, 1993; pp 87±100.
3. De Clercq, E. In HIV Molecular Organization, Pathogeni-
city and Treatment; Morrow, W. J. W., Haigwood, N. L.,
Eds.; Elsevier Science Publishers: Amsterdam, 1993; pp 267±
292.
4. Balzarini, J.; Neyts, J.; Schols, D.; Hosaya, M.; Van
Damma, E.; Peumans, W.; De Clercq, E. Antiviral Res. 1992,
18, 191.
5. De Clercq, E.; Yamamoto, N.; Pauwels, R.; Baba, M.;
Schols, D.; Nakashima, H.; Balzarini, J.; Debyser, Z.; Murrer,
B. A.; Schwartz, D.; Thornton, D.; Bridger, G.; Fricker, S.;
Henson, G.; Abrams, M.; Picker, D. Proc. Natl. Acad. Sci
U.S.A. 1992, 89, 5286.
6. De Clercq, E. J. Med. Chem. 1995, 38, 2491.
7. Neamati, N.; Hong, H.; Owen, J.; Sunder, S.; Winslow, H.;
Christensen, J.; Zhao, H.; Burke, T.; Hilne, G.; Pommier, Y.
J. Med. Chem, 1998, 41, 3202. Neamati, N.; Turpin, J. A.;
Winslow, H. E.; Christensen, J. L.; Williamson, K.; Orr, A.;
Rice, W. G.; Pommier, Y.; Garofalo, A.; Brizzi, A.; Campiani,
G. J. Med. Chem. 1999, 42, 3334.
HIV-1 protease inhibition assay. The HPLC based assay
for inhibition of HIV-1 protease was performed by using
cloned HIV-1 protease and HIV-1 protease substrate (H-
Hys-Lys-Ala-Arg-Val-Leu-pnitro-Phe-Glu-Ala-Ser-NH2,
BACHEM S.A.). Potential inhibitor was solubilized in
DMSO at various concentrations. Fifteen microlitres of
this solution were diluted in 100 mL of bu€er (sodium
acetate 50 nM pH 4.9, sodium chloride 200 nM, dithio-
threitol 5 nM and glycerol 10%, V/V) to obtain a [I]bu€ered
concentration. 40mL of [I]bu€ered were mixed with 15mL of
the HIV-1 substrate, then with 25cdmL of rec-HIV-1
protease. The mixture was incubated at 37ꢀC during 20
minutes and quenched by 10mL of a 10% TFA solution.
Twenty microlitres of the pre-incubated assay was injected
in reverse phase HPLC (Waters C18) with CH3CN:H2O
(5:95 to 100:0) in 30 min. Inhibition was determined by
relative percent of the peak areas corresponding to the
HIV-1 protease substrate cleavage.
8. Hsu, M. C.; Schutt, A. D.; Holly, M.; Slice, L. W.; Sher-
man, M. I.; Richman, D. D.; Potash, M. J.; Volsky, D. J.
Science 1991, 254, 1799.
9. Chrusciel, R. A.; Romines, K. R. Exp. Opin. Ther. Patents
1997, 7, 111.
10. Karpas, A.; Fleet, G. W.; Dwek, R. A.; Patursson, S.;
Namgoong, S. K.; Ramsden, N. G.; Jacob, G. S.; Rade-
macher, T. W. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 9229.
11. Kimura, T.; Matsumoto, H.; Matsuda, T.; Hamawaki, T.;
Akaji, K.; Kiso, Y. Bioorg. Med. Chem. Lett. 1999, 9, 803.
12. Rey, F.; Barre-Sinoussi, F.; Schmidtmayerova, H.; Cher-
mann, J. C. J. Virol. Methods 1987, 16, 239.
13. Kempf, D.; Codacovi, L.; Wang, X.; Kohlbrenner, W.;
Wideburg, N.; Saldivar, A.; Vasavanonda, S.; Marsh, K.;
Bryant, P.; Sham, H.; Green, B.; Betebenner, D.; Erickson, J.;
Norbeck, D. J. Med. Chem. 1993, 36, 320.
14. Callahan, J. F.; Ashton-Shue, D.; Bryan, W. M.; Heck-
man, G. D.; Kinter, L. B.; McDonald, J. E.; Moore, M. L.;
Schmidt, D. B.; Silvestri, J. S.; Stassen, F. L.; Sulat, L.; Yim,
N. C. F.; Hu€man, W. F. J. Med. Chem. 1989, 32, 391.
15. Mancuso, A. J.; Swern, D. Synthesis 1981, 165.
16. Aggarwal, S. K.; Gogu, S. R.; Rangan, S. R. S.; Agrawal,
K. C. J. Med. Chem. 1990, 33, 1505.
Ten micromolar was the optimal concentration at which
the new analogues were tested.
17. Betageri, J. C.; Hopkins, J. L.; Thibeault, D.; Emmanuel,
M. S.; Chow, G. C.; Skoog, M. T.; Dedreu, P.; Cohen, K. A.
J. Biochem. Biophys. Methods 1993, 27, 191.
Acknowledgements
18. Pettit, G. R.; Herald, C. L.; Boyld, M. R.; Leet, J. E.;
Dufresne, C.; Doubek, D. L.; Schmidt, J. M.; Cerny, R. L.;
Hooper, J. M. A.; Rutzler, K. C. J. Med. Chem. 1991, 34,
3340.
19. Rey, F.; Donker, G.; Hirch, I.; Chermann, J. C. Virology
1991, 181, 165.
This research was supported ®nancially by the Institut
National de la Recherche Medicale (INSERM). We are
indebted to U-322 members for technical assistance
in antiviral evaluation. We are indebted to Dr. T.
Williamson for English correction of the manuscript.