Lysine sulfonamides as novel HIV-protease inhibitors: Nε-disubstituted ureas

Article abstract of DOI:10.1016/j.bmcl.2004.05.049

A series of lysine sulfonamide analogues bearing a Nε-benzylic ureas was synthesized using both solution-phase and solid-phase approaches. A novel synthetic route of Nα-(alkyl)-Nα-(sulfonamides)lysinol using α-amino-caprolactam was developed. Evaluation of these novel protease inhibitors revealed compounds with high potency against wild-type HIV virus.

Full text of DOI:10.1016/j.bmcl.2004.05.049

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3971–3974
Lysine sulfonamides as novel HIV-protease inhibitors:
Ne-disubstituted ureas
*
ꢀ
ꢀ
Brent R. Stranix, Gilles Sauve, Abderrahim Bouzide, Alexandre Cote, Guy Sevigny,
ꢀ
ꢀ
Jocelyn Yelle and Valerie Perron
Procyon Biopharma Inc., 1650 Trans-Canada Highway, Suite 200, Dorval, Qc, Canada, H9P 1H7
Received 17 March 2004; revised 21 May 2004; accepted 21 May 2004
Available online 17 June 2004
Abstract—A series of lysine sulfonamide analogues bearing a Ne-benzylic ureas was synthesized using both solution-phase and
solid-phase approaches. A novel synthetic route of Na-(alkyl)-Na-(sulfonamides)lysinol using a-amino-caprolactam was developed.
Evaluation of these novel protease inhibitors revealed compounds with high potency against wild-type HIVvirus.
Ó 2004 Elsevier Ltd. All rights reserved.
Human immunodeficiency virus (HIV) aspartyl protease
inhibitors are a major component of anti-HIVchemo-
therapy, the current treatment for acquired immunode-
ficiency syndrome (AIDS).1–3 This class of compounds
inhibits the formation of mature viral particles and thus
the infectious process. Although protease inhibitors
have radically improved the life of AIDS patients and
Figure 1. General structure of lysine sulfonamide HIVprotease
inhibitors.
contributed in large part to the success of highly active
Moreover several types of Ne-linkages led to different
anti-retroviral therapy (HAART), new problems have
classes of compounds, which retained excellent poten-
recently been identified. The rapid emergence of several
cies.
viral strains resistant to one or more of the drugs cur-
rently available for the treatment of AIDS has now
One such class, Ne-ureas will be discussed in this letter.
become the most important issue in the treatment of
A synthetic route was devised using Ne-(Cbz)-lysine
methyl ester 3 as a starting point (Scheme 1). Reductive
4
HIVinfection.
Most currently available drugs are
peptidomimetics containing the hydroxyethylene moi-
ety, which mimic the hydrolytic transition state of the
protease substrate.5;6 We recently discovered HIVpro-
tease inhibiting compounds devoid of the hydroxyeth-
ylene moiety of general structure as shown in Figure 1,
which also demonstrated interesting anti-viral activ-
ity.7–9 We had previously found that lysine or lysinol
could serve as an efficient backbone scaffold for the
synthesis of such inhibitors.10
Keywords: Lysine; Ureas; HIVprotease inhibitors.
* Corresponding author at present address: Procyon Biopharma Inc.,
6100 Royalmount, Montreal, Qc, Canada, H4P 2R2. Tel.: +1-514-
496-1735; fax: +1-514-496-1702; e-mail: bstranix@procyonbio-
pharma.com
Scheme 1. Reagents and conditions: (a) RCHO, NaCNBH3, MeOH,
pH 4; (b) TsCl, DIPEA, DCM; (c) H2 Pd/C, EtOH; (d) CDI, DMF, 17
or H2NR1; (e) NaOH 1 M/THF 1:1, HCl 1 M; (f) LiAlH4 THF.
Present address: INRS-Institut Armand-Frappier, 531, des Prairies,
Blvd, Building 27, Office D-103, Laval, Quebec, Canada, H7V1B7.
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.05.049

3972
B. R. Stranix et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3971–3974
alkylation and a slow sulfonamidation with toluene-
sulfonyl chloride (TsCl) gave intermediate 4. Flash
chromatography purification permitted isolation of this
intermediate in moderate yield. The elimination of the
Cbz protecting group was effected using catalytic
hydrogenation, which yielded the desired free amino
ester 5 in quantitative yields. This intermediate was then
converted to an isocyanate using carbonyldiimidazole
(CDI) in DMF,11 which could be kept stable in solution
for several weeks. The urea portion was then simply
formed by adding primary or secondary amines 17 to
the pendant isocyanate solution to give the corre-
sponding urea esters in moderate to good yields. Finally,
the ester was selectively hydrolyzed with NaOH or
reduced with LiAlH4 to yield the desired protease
inhibitors 6 or 7. However, the final hydrolysis step with
NaOH was often accompanied by partial or complete
racemization of the chiral centre. An alternative syn-
thetic route was devised to eliminate certain time con-
suming reactions, low yields, protecting groups,
chromatography and frequent racemization of the
intermediates.
to avoid the use of LiAlH4 to generate each final com-
pound. A Fmoc protected13 derivative of 11 was stirred
with a suspension of freshly activated trityl resin
(1 mmol/g) and reacted with the pendant alcohol to yield
a resin 12 with 0.4 mmol/g degree of substitution.
Standard deprotection13 of the amine followed by acti-
vation with CDI or thiocarbonyldiimidazole, gave pen-
dant isocyanates11;14 13 or isothiocyanates, which were
reacted with primary and secondary amines to yield,
after mild cleavage, the desired Ne-ureas 15–16. Acyl-
ation of the pendant aniline was not observed. The sec-
ondary benzylic amines 17, were synthesized using two
standard procedures, reductive amination of aldehydes
and nucleophilic substitution of alkyl or benzylic halides
both of which gave good yields of desired products
(Scheme 3). The final products were then subjected to a
preparative RPHPLC purification followed by LC/MS
and NMR characterization. The products were then
evaluated as protease inhibitors in an in vitro enzyme
assay and in cell based assays.15–17 Table 1 describes the
inhibition of the compounds on the activity of purified
HIVprotease. The library produced using the above
methodologies revealed that N-benzyl urea of the lysine
sulfonamide 6a inhibits HIVprotease with sixfold
greater efficacy than the longer phenethyl urea 6b or the
heterocyclic 2-picolyl urea 6c. The addition of a second
substituent such as the aliphatic methyl (6d) and iso-
propyl (6g) groups gave no improvement. Ureas
substituted with two aliphatic substituents such as 6i or
with the phenethyl group 6e, gave inactive compounds.
However, much more effective, was the N,N-dibenzyl
urea system 6f, which improved the potency by fivefold.
Also noteworthy is the N-(benzyl)-N-4-picolyl urea 6h,
which showed some twofold improvement in inhibition
over 6f. Compound 7a, the lysinol based analogue of 6f
showed a twofold loss of potency in the enzyme based
assay. However, these lysine-based compounds did not
show any significant anti-viral activity in the whole-cell
based assay. The substituted N-benzyl urea system was
explored further using the Na-alkyl-Na-(4-amino-
This procedure used a-amino-caprolactam 8 as a start-
ing material (Scheme 2). Reductive alkylation and
sulfonamidation of the free amine gave excellent yields
of enantiomerically pure crystalline intermediate 9.
Hydrolysis of the lactam proceeded quantitatively and
without any detectable racemization.12 Reduction of the
resulting carboxyl was effected by esterification to form
10 and reduction with LiAlH4 to yield 11. Once again,
crystalline intermediates were obtained without race-
mization. The amino alcohol obtained could be selec-
tively acylated with a variety of activated acids,
isocyanates, or chloroformates. When required the
amino ester obtained was used to form urea compounds
as in Scheme 1, on larger scale (1–10 g). We then
adapted a solid-phase technique for a rapid generation
of compounds bearing ureas and thioureas of both pri-
mary and secondary amines. This route also enabled us
O
R
R
R
N
e
c,d
O
N
N
NH2
NH
a,b
NH2
O2S
NH
O2S
H2N
O2S
OH
f,g
O
O
H2N
NH2
AcHN
8
9
10
11
= trityl resin
O
h,i
j
O
k
O
R
N
R1
N
R
O
R
R
N
HN
SO2
Fmoc
N
R2
N
O2S
N
H
R1
SO2
O
C
SO2
CH2OH O
N
H
N
N
H2N
R2
H2N
H2N
NH2
15-16
14
13
12
Scheme 2. Reagents and conditions: (a) RCHO, STAB, DCM 89%; (b) 4-(AcNH)PhSO2Cl, TEA, EtOAc 88%; (c) 6 M HCl reflux 99%; (d) TMS-Cl,
MeOH rt 95%; (e) LiAlH4, ether 85–95%; (f) Fmoc-Osu, THF 1 M, K2CO3 1:1 75%; (g) trityl chloride resin, DMF, pyridine, then methanol 40%
conversion; (h) 30% piperidine DMF rt quant; (i) CDI, (ThioCDI) DMF, TEA 80 °C, 0.5 h quant; (j) 17 excess, DMF; (k) 5% TFA, DCM 10–60%
isolated yield.

B. R. Stranix et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3971–3974
Table 1. Inhibition constants of compounds for HIVaspartyl protease
3973
Compound
R1
R2
Ki (nM)
6a
6b
6c
6d
6e
6f
H
C6H5CH2
C6H5CH2CH2
2-Picolyl
32
204
>300
55
H
H
C6H5CH2
C6H5CH2
C6H5CH2
C6H5CH2
C6H5CH2
(CH3)2CH
C6H5CH2
CH3
C6H5CH2CH2
C6H5CH2
(CH3)2CH
4-Picolyl
>300
7.2
19
6g
6h
6i
3.7
(CH3)2CH
C6H5CH2
300
15
7a
Table 2. Inhibition constants of compounds for HIVaspartyl protease and Wild type virus in anti-viral assays
Compound
R1
R2
15 Ki (nM)
EC50 (nM)
16 Ki (nM)
EC50 (nM)
a
b
c
d
e
f
C6H5CH2
C6H5CH2
C6H5CH2
3-Picolyl
4-Picolyl
12
34
3000
8500
>104
5000
>104
3200
>104
1400
4000
5000
620
5.8
>104
C6H5CH2
C6H5CH2
90
11
2-Thiophene-CH2
2,3-(OCH2O)C6H3CH2
2,4-F2C6H3CH2
4-FC6H4CH2
C6H5CH2
C6H5CH2
5.4
12
4.0
>104
g
h
i
C6H5CH2
C6H5CH2
3.3
2.6
2.0
6.3
2.0
2.0
2.5
2.2
1.7
3,4-(OCH2O)C6H3CH2
2-Thiophene-CH2
3-Thiophene-CH2
4-NO2C6H4CH2
4-FC6H4CH2
4.1
1.4
871
800
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
j
k
l
2.2
4.9
3.0
2.9
845
488
1200
>104
700
m
n
o
4-CF3C6H4CH2
4-CH3OC6H4CH2
3,4-(OCH2O)C6H3CH2
1600
500
150
whole-cell anti-viral assay. Since the piperonyl group
appeared to be a better pharmacophore, a series of Ne-
(piperonyl-arylsubstituted benzyl)ureas were prepared.
Although this series of compounds gave inhibition
constants on the purified enzyme ranging from 1.7 to
6.3 nM, the antiviral assay gave EC50 values ranging
from 5 to 0.15 lM. Thiophene bearing derivatives such
as 15i–j showed poor antiviral properties, whereas
Scheme 3. Reagents and conditions: (a) RCHO, STAB, DCM; (b) R–
Cl (Br) EtOH reflux.
benzenesulfonyl)-lysinol moiety, which was shown to
be more effective than the lysine based compounds.10
A
library of N,N-dibenzyl compounds 15 (Na-isobutyl)
and 16 (Na-isovaleryl) containing various aromatic
substituents and heterocycles were then synthesized in
parallel. Table 2 reveals the dibenzyl urea analogue 15a
based upon the Na-isobutyl-Na-(4-aminobenzenesulfo-
nyl) lysinol scaffold showed a similar Ki to the Na-
isobutyl Na-(tosyl) analogue 7a. Heterocycles such as
picolines 15b–c and thiophene 15d did not significantly
improve the activity against the enzyme as compared to
the N,N-dibenzyl compound 15a. The addition of elec-
tron donating substituents such as piperonyl 15h, gave
the best activities, increasing the enzyme inhibition
sevenfold. Extending the isobutyl group of 15h to an
Na-isovaleryl group 16h, showed an improvement in the
12
10
8
15g
6
15j
4
15l
15a
2
15o
15m
0
3.8
4
4.2
4.4
4.6
log P(octanol/water)
Figure 2. Whole-cell assay EC50 (lM) versus log P.

3974
B. R. Stranix et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3971–3974
Table 3. The effect of urea versus thioureas on the inhibition constants of compounds for HIVaspartyl protease
Compound
R1
R2
X
18 Ki
19 Ki
a
b
c
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(OCH2O)C6H3CH2
3,4-(O(CH2)2O)C6H3CH2
3-CH3OC6H4CH2
4-CH3OC6H4CH2
3,4-(CH3O)C6H3CH2
4-FC6H4CH2
2.2
2.9
3.0
5.4
2.6
3.6
2.9
9.9
d
Na-isovaleryl derivatives 16i–j, were an improvement
over 15h. Compounds 15–16m bearing the para-trifluoro
benzyl group also gave poor antiviral inhibition. Other
electron withdrawing substituents such as para-fluoro
and para-nitro groups as in 15–16k–l in contrast to the
previously mentioned compounds 15–16m, gave good
antiviral inhibition. By comparison with 15h, com-
pounds bearing electron rich moieties such as the para-
methoxy benzyl 15–16n and piperonyl 15o showed an
increase in anti-viral potency of three, two and ten fold,
respectively. The role of the piperonyl substituent in
improving the anti-protease activity is unclear, although
we can speculate dioxolane oxygen atoms may form an
additional hydrogen bond to the enzyme active site. The
increase in anti-viral potency in the whole cell assay can
also be partially explained by the improved water solu-
bility of the molecules bearing substituents such as
A. Ahmed for their analyses of compound purity and
log P determination.
References and notes
1. Eron, J. J., Jr. Clin. Infect. Dis. 2000, 30, S160–S170.
2. Vella, S. Handb. Exp. Pharmacol. 2000, 140, 23–32.
3. Tomasselli, A. G.; Heinrikson, R. L. Biochim. Biophys.
Acta 2000, 1477, 189–214.
4. Swanstrom, R.; Erona, J. Pharmacol. Ther. 2000, 86, 145–
170.
5. Babine, R.; Bender, S. L. Chem. Rev. 1997, 97, 1359–1472.
6. Roberts, N. A.; Martin, J. A.; Kinchington, D.; Broad-
hurst, A. V.; Craig, J. C.; Duncan, I. B.; Galpin, S. A.;
Handa, B. K.; Kay, J., et al. Science 1990, 248, 358–361.
ꢀ
7. Bouzide, A.; Sauve, G.; Stranix, B. R.; Sevigny, G.; Yelle,
J. U.S. Patent No. 6,455,587; Pharmacor Inc: U.S., 2002.
ꢀ
ꢀ
^ ꢀ
ꢀ
ꢀ
8. Stranix, B. R.; Sauve, G.; Bouzide, A.; Cote, A.; Berube,
G.; Soucy, P.; Zhao, Y.; Yelle, J. U.S. Patent No.
6,506,786; Pharmacor Inc: U.S., 2002.
piperonyl, which lowered the log P
value.
ðoctanol=waterÞ
Figure 2 relates the effect of log P on the whole cell assay
EC50 of a few representative compounds, and clearly
show the beneficial effect of better water solubility.
Furthermore, the replacement of urea (Series 18) by a
thiourea (Series 19) showed no improvement (Table 3).
In summary, a series of lysine sulfonamide analogues
bearing a Ne-benzylic ureas was synthesized using both
solution-phase and solid-phase approaches. A novel
synthesis of Na-(alkyl), Na-(sulfonamides)lysinol using
a-amino-caprolactam was developed. Evaluation of
these novel protease inhibitors revealed highly potent
compounds against wild-type HIVvirus. Further studies
are ongoing to assess the inhibitory activity against
resistant HIVstrains.
9. Stranix, B. R.; Bouzide, A.; Sauve, G. U.S. Patent No.
6,528,532, 2002.
ꢀ
^ ꢀ
ꢀ
10. Stranix, B. R.; Sauve, G.; Bouzide, A.; Cote, A.; Sevigny,
G.; Yelle, Y. Bioorg. Med. Chem. Lett. 2003, 13, 4289–
42922.
11. Staab, H. A.; Benz, W. Angew. Chem. 1961, 73, 66.
12. Boyle, W. J.; Sifniades, S.; Van Peppen, J. F. J. Org.
Chem. 1979, 44, 4841–4847.
13. Atherton, E.; Sheppard, R. C. In The Fluorenylmethoxy-
carbonyl Amino Protecting Group; Academic: New York,
1987; Vol. 9.
14. A dried portion of resin 13 was analyzed by FT-IR and
showed a characteristic IR absorption band at 2235 cmꢀ1
15. Matayoshi, E. D.; Wang, G. T.; Krafft, G. A.; Erickson,
J. /contribution> Science 1990, 247, 954–958.
.
16. Japour, A. J.; Mayers, D. L.; Johnson, V. A.; Kuritzkes,
D. R.; Beckett, L. A.; Arduino, J. M.; Lane, J.; Black,
R. J.; Reichelderfer, P. S.; D’Aquila, R. T. Antimicrob.
Agents Chemother. 1993, 37, 1095–1101.
17. Pauwels, R.; Balzarini, J.; Baba, M.; Snoeck, R.; Schols,
D.; Herdewijn, P.; Desmyter, J.; Clercq, E. D. J. Virol.
Methods 1988, 20, 309–321.
Acknowledgements
We gratefully acknowledge A. Dubois, P. Soucy, Y.
Zhao and B. Tian for their participation in the biolog-
ical analyses. We also acknowledge Drs. F. Dahani and