Expedient solid-phase synthesis of both symmetric and asymmetric diol libraries targeting aspartic proteases

Article abstract of DOI:10.1016/j.bmcl.2009.03.041

C₂-symmetric diols have been shown to be highly potent against HIV-1 protease (PR). However, gaining access to these compounds has been hampered by the need of multistep solution-phase reactions which are often tedious and inefficient. In this Letter, we have disclosed a solid-phase strategy for rapid preparation of small molecule-based, symmetric and asymmetric diols as potential HIV-1 protease inhibitors. Upon biological screening, we found one of them, SYM-5, to be a potent and selective inhibitor (K_i = 400 nM) against HIV-1 protease.

Full text of DOI:10.1016/j.bmcl.2009.03.041

Bioorganic & Medicinal Chemistry Letters 19 (2009) 3945–3948
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Expedient solid-phase synthesis of both symmetric and asymmetric diol
libraries targeting aspartic proteases
Haibin Shi a, Kai Liu b, Wendy W. Y. Leong a, Shao Q. Yao a,b,
*
a Department of Chemistry, National University of Singapore, 3 Science Drive 3, Singapore 11754, Singapore
b Department of Biological Sciences, National University of Singapore, 3 Science Drive 3, Singapore 11754, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
C2-symmetric diols have been shown to be highly potent against HIV-1 protease (PR). However, gaining
access to these compounds has been hampered by the need of multistep solution-phase reactions which
are often tedious and inefficient. In this Letter, we have disclosed a solid-phase strategy for rapid prep-
aration of small molecule-based, symmetric and asymmetric diols as potential HIV-1 protease inhibitors.
Upon biological screening, we found one of them, SYM-5, to be a potent and selective inhibitor
(Ki = 400 nM) against HIV-1 protease.
Received 5 February 2009
Revised 5 March 2009
Accepted 9 March 2009
Available online 17 March 2009
Keywords:
Ó 2009 Elsevier Ltd. All rights reserved.
Combinatorial chemistry
Solid-phase synthesis
Protease inhibitor
Drug discovery
A key challenge in current drug discovery is the development of
high-throughput (HT) amenable chemical reactions that allow ra-
pid synthesis of diverse chemical libraries of enzyme inhibitors.
In recent years, one reaction that has received much attention is
the amide-bond formation between an amine and a carboxylic acid
using suitable activating/coupling reagents.1 The reaction is highly
efficient in that it often generates desired products in nearly quan-
titative yields, thus allowing direct in situ biological screening to
be carried out. This strategy has been successfully adopted, in both
solution- and solid-phase, by several groups for the rapid discovery
of small molecule inhibitors.2,3 The solution phase amide-forming
strategy is comparatively easier to implement but has the follow-
ing problems:2 (1) undesired chemicals (starting materials,
reagents, by-products, etc) are present during the biological
screening and therefore often cause false results; (2) multi-step
reactions cannot be carried out without proper purification of the
intermediates; and (3) excessive starting materials/reagents nor-
mally cannot be used to drive an otherwise-difficult coupling to
completion. Solid-phase synthesis could potentially solve many
of these problems, and has been widely adopted in the synthesis
of biopolymers (e.g., oligonucleotides and peptides) as well as
small molecules (e.g., in combinatorial chemistry). In most cases,
however, low-quality products are generated and need to be labo-
riously purified before screening. We have recently become inter-
ested in developing highly efficient solid-phase strategies that
enable large-scale synthesis of high-quality compound collections
which are suitable for direct in situ biological screening.3,4 With
these new chemical tools, we have successfully developed small
molecule inhibitors targeting different classes of proteases, includ-
ing cysteine and metallo-proteases. Herein, by making use of the
amide-forming reaction, we report the solid-phase synthesis of
both symmetric and asymmetric diols as putative inhibitors of
aspartic proteases. Many aspartic proteases are known therapeutic
targets.5,6 For example, HIV-1 protease is one of the main targets of
AIDS.5 Renin is currently being used to treat hypertension.6a
Cathepsin D is involved in breast cancer metastasis.6b ß-Secretase
is related to Alzheimer disease.6c Plasmepsins, key aspartic prote-
ases from malaria involved in the hemoglobin degrading pathway
in parasite-infected macrophages, are being pursued to treat
malaria.6d
HIV-1 protease (PR) is one of the best known aspartic proteases,
and an attractive target for the treatment of AIDS.5 The enzyme
consists of two identical, non-covalently associated subunits of
99 amino acid residues formed in an C2-symmetric fashion. The
highly symmetrical active site is formed at the dimer interface.
Over the years, many HIV-1 protease inhibitors (PI) have been
developed.5a Two of the most effective PI are shown in Figure 1
(top two): they contain a C2-symmetric, diol-containing moiety
which mimics the transition state of amide hydrolysis. Wong’s
peptidic inhibitor developed in 1998,5b for example, has a Ki of
1.5 nM against HIV-1 PR. Its poor bioavailability (e.g., peptide-
based, large MW) however makes it ineffective in cell-based assay.
Abbott’s Ritonavir,5c an FDA-approved drug and improved version
of Wong’s inhibitor, is nearly C2-symmetric, and much more effec-
tive against AIDS due to its smaller size (MW = 820) and fewer
* Corresponding author. Tel.: +65 65161683; fax: +65 6779169.
E-mail address: chmyaosq@nus.edu.sg (S.Q. Yao).
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.03.041

3946
H. Shi et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3945–3948
corresponding amino alcohols 2 was oxidized to the aldehydes 3
using Dess–Martin reagent. Subsequently, pinacol homocoupling
reaction between two equivalents of 3 in the presence of VCl3 affor-
ded diastereomeric diols 4 in a single-step transformation (ꢀ20%
yield). Subsequent protection of the dihydroxyl group with
2,2-dimethoxypropane, flash chromatography to remove minor
diastereomeric impurities, and deprotection of the Cbz groups using
H2 (in Pd/C) gave the resulting enantiomerically pure diaminodiols,
6a–c, in moderate yield (50% in 2 steps).
O
OH
O
O
H
N
H
N
H
N
O
N
H
N
H
N
H
O
O
O
OH
O
Wong's inhibitor (Ki = 1.5 nM)
(MW: 909.08)
O
OH
OH
O
N
In the subsequent solid-phase reactions (Scheme 2), the alde-
hyde-containing FMP resin was used for capturing of 6 by reductive
amination, followed by amide bond-forming reaction with acids. As
previously shown,3 key advantages of this method include (i) chem-
ical modification of the library building blocks (i.e., diol core &
acids) is unnecessary; (ii) it’s solid-phase, enabling a large library
to be constructed efficiently; (iii) it’s robust, giving high-quality
products. The IRORITM directed sorting technology was used in the
synthesis of the 75-member library.4 First, the three diaminodiols,
6a–c, were treated with FMP resin in the presence of
Na(OAc)3BH/2% glacial acetic acid in DCE to give the resin-bound
amines, 7, which contain an 1° and 2° amine each. Initially, simul-
taneous acylation of both the 1° and 2° amines with the same acid
building blocks was met with some difficulties under most coupling
conditions (e.g., DIC/HBTU/TBTU/HATU); in most cases, only the
monoacylated product (at 1° amine position) was generated. This
was clearly due to the chemical and steric difference of the two
amines in 7. We therefore took advantage of this to make both sym-
metric and asymmetric diols (Scheme 2). An exhaustive testing of
(7 + acid) coupling under a variety of different coupling conditions
finally gave rise to the following two sets of optimized conditions:
(I) to make symmetric diols, Route I was used in which PyBrOP/
DIEA coupling with the acid (10 equiv) was carried out to gave 8;
(II) to make asymmetric diols, Route II was used in which PyBOP/
HOAt coupling with the first acid (5 equiv) was performed, giving
9, followed by PyBrOP/DIEA coupling with the second acid (5 equiv)
to give 10. Finally, cleavage of the products from the resin using an
optimized TFA cocktail (5:4:1 TFA/DCM/water) gave a total of 44
C2-symmetric and 31 asymmetric diol inhibitors (representative
compounds are shown in Scheme 2; see Supplementary data for
complete list). To ensure the crude products generated from our
strategy were sufficiently pure for direct in situ screening, the com-
pounds were further characterized by LCMS and NMR; in most
cases the desired products were obtained with good purity (>90%
and 60–95% for symmetric and asymmetric diols, respectively).8
Next, the inhibitory activity of these 75 diol-based inhibitors
was determined against HIV-1 PR, plasmepsin I (PM I) and plasm-
epsin II (PM II) using a standard fluorescence microplate assay
method. First, an inhibitor fingerprint of the library against the
aspartic proteases was obtained, from which six potential hits
(SYM-5, -21, -35, -43 and ASM-16, -29) were identified. Detailed
inhibition studies were then carried out to obtain the correspond-
ing IC50/Ki of these compounds, and the results are summarized in
Table 1. The best inhibitor against HIV-1 PR was found to be SYM-5,
with IC50 and Ki values of 395.4 and 400 nM, respectively (Fig. 2).9
Significantly, it showed a very poor inhibition against PM I and PM
H
N
H
N
S
N
O
F
N
H
N
H
N
S
O
O
Abbott's Ritonavir
(MW: 820.08)
Cl
O
O
H
N
N
H
OH
Cl
F
SYM-5 (this study; K i = 400 nM)
(MW: 613.48)
Figure 1. Structures of two known diol-containing HIV-1 protease inhibitors (top
two) and the inhibitor identified from this study (bottom). The core diol structures
are highlighted (Red). Top two are C2-symmetric or near symmetric. The newly
identified one was shown to have good inhibition against HIV-1 PR (Ki = 400 nM).
amide bonds. The rapid emergence of drug-resistant HIV-1 PR,
however, has rendered many of these PI ineffective. Thus, there
is an urgent need to develop chemistry that permits rapid and
efficient synthesis of new PI. Another paradigm shift in HIV PI
research is the introduction of asymmetric PI which could be more
effective against some of the drug-resistant HIV strains.5d In the
current study, we have developed a solid-phase strategy for rapid
synthesis of potential symmetric and asymmetric PI (Schemes 1
and 2). As a proof of concept, the strategy was successfully used
to synthesize 75 diol-containing compounds, which, upon direct
in situ screening, revealed a small (MW = 613) and potent HIV-1
PI (Ki = 400 nM).
In our strategy, we adopted a solution-cum-solid phase strategy,
in which the most important component of the inhibitors—the dia-
minodiol core group, was synthesized in solution (Scheme 1) and
purified to homogeneity before being installed onto the commer-
cially available 4-formyl-3-methoxyphenoxy (FMP) resin,7 followed
by diversification with a variety of acids (Scheme 2). As shown in
Scheme 1, following previously published procedures,5c three differ-
ent C2-symmetric diaminodiols, 6a–c, were made from commer-
cially available Cbz-protected amino acids 1. Upon reduction, the
O
O
a, b
c
CbzHN
CbzHN
CbzHN
CbzHN
OH
H
OH
R1
R1
d
R1
3
2
1
O
P1
OH P1
e
CbzHN
NHCbz
NHCbz
II (>25
towards HIV-1 PR. None of the asymmetric compounds was able
to inhibit 50% of the HIV-1 PR activity at a concentration of 25 M,
lM), indicating that this inhibitor is highly specific only
P1
f
O
P1 OH
5
4
l
O
P1
which is expected since the assay was done with the C2-symmetric
wildtype HIV-1 PR. However, two of these compounds exhibited
CH3
H2N
P1:
NH2
moderate activity against PM II (IC50 = 12.6 and 11.7 lM for
P1
O
6a-c
c
a
b
ASM-16and -29, respectively), which may be developed into potent
inhibitors in future. SYM-21 and SYM-43 were identified to be
moderate and selective inhibitors of PM II and PM I, respectively.
These results thus validate our strategy as a feasible method for
future discovery of other aspartic protease inhibitors.
Scheme 1. Solution-phase synthesis of the diol warheads. (a) Isochloroformate,
NMM, DCE, À15 °C. (b) NaBH4, THF, 0 °C, 70–77%. (c) Dess–Martin, DCM, rt, 60–70%.
(d) VCl3, 1,3-dimethylimidazole, Zn, dry THF, reflux, 10–20%. (e) 2,2-dimethoxy-
propane, p-TsOH, acetone, rt, 60–70%. (f) H2, Pd/C, MeOH/EA, rt, 60–70%.

H. Shi et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3945–3948
3947
Route II
c
P1
P1
OMe
OMe
OMe
-
PF6
NH
O
O
O
R1
O
NH2
a
N
N
N
HN
O
H
+
HN
O
O
PH
P1
P1
9
7
Route I
b
d
PyBOP
P1
P1
OMe
OMe
O
NH
O
NH
O
-
O
O
PF6
R1
R1
N
O
N
O
+
P1
P1
Br PH
O
8
10
R1
R2
PyBrOP
e
Representative assymmetric diols
Representative symmetric diols
Cl
F
F
F
F
Cl
O
OH
O
H3C
H
N
OH
OH
O
H
N
OH
O
OH
OH
O
H
N
H
N
N
H
N
N
H
N
N
H
O
OH
O
H
OH
O
O
Cl
F
ASM-01
ASM-16
SYM-05
OH
SYM-21
O
OMe
O
O
O
H
N
OH
O
Cl
OH
H
O
OH
OH
O
H
N
H
N
N
N
O
N
N
N
O
H
MeO
O
OH
H
H
H
O
OH
O
OH
O
F
SYM-35
SYM-43
ASM-29
ASM-31
Scheme 2. Solid-phase synthesis of both symmetric and asymmetric diols. (a) Na(OAc)3BH, 2% AcOH in DCE, rt. (b) acid (10 equiv), PyBrOP/DIEA in DMF, rt. (c) acid 1
(5 equiv), PyBOP/HOAt/DIEA in DMF, rt. (d) acid 2 (5 equiv), PyBrOP/DIEA in DMF, rt. (e) TFA/DCM/H2O (5/4/1), rt.
In summary, we have developed a solid-phase amide-forming
approach for rapid assembly of both symmetric and asymmetric
diol-containing small molecules as potential aspartic protease
inhibitors. One potent and selective inhibitor of HIV-1 PR was dis-
covered which should possess better bioavailability properties
(fewer amide bonds, lower MW) than its parental compounds.
The clear advantage in our strategy lies in its feasibility for solid-
phase synthesis of asymmetric diols. When compared with other
Table 1
Inhibition of the six selected inhibitors
Enzyme
IC50 (Ki) in nM
SYM-5
SYM-21
SYM-35
SYM-43
ASM-16
ASM-29
HIV-1 PR
PM I
PM II
395.4 (400)
>25,000
>25,000
>25,000
4391
3908
1313
>25,000
>25,000
>25,000
2642
>25,000
>25,000
>25,000
12,657
>25,000
>25,000
11,773
F
Cl
OH
O
H
N
N
H
O
OH
F
Cl
SYM-05
0.8
100
Ki = 400 nM
2.5 µM
5.0 µM
7.5 µM
10 µM
0.6
0.4
0.2
IC50 = 395.4 nM
50
0
-3000
1000
5000
9000
0
2
4
[I], nM
Log[Conc/nM]
Figure 2. (a) Structure of SYM-5. (b) IC50 (left) and Ki (right) of SYM-5 against HIV-1 PR.

3948
H. Shi et al. / Bioorg. Med. Chem. Lett. 19 (2009) 3945–3948
Lee, L. V.; Wong, C.-H. ChemBioChem 2004, 5, 811; (d) Wu, C.-Y.; Chang, C.-F.;
Chen, J. S.-Y.; Lee, S.-T.; Wong, C.-H.; Lin, C.-H. Angew. Chem., Int. Ed. 2003, 42,
4661; (e) Chang, C.-F.; Ho, C.-W.; Wu, C.-Y.; Chao, T.-A.; Wong, C.-H.; Lin, C.-H.
Chem. Biol. 2004, 11, 1301; (f) Wu, C.-Y.; King, K.-Y.; Kuo, C.-J.; Fang, J.-M.; Wu,
Y.-T.; Ho, M.-Y.; Liao, C.-L.; Shie, J.-J.; Liang, P.-H.; Wong, C.-H. Chem. Biol. 2006,
13, 261.
potent HIV-1 protease inhibitors, our compound pales in terms of
its inhibitory activity. Nevertheless, we believe the chemical strat-
egy discolsed herein may be readily expanded in future for the
construction of much bigger compound libraries which may lead
to candidate compounds with improved biological activities.
3. (a) Srinivasan, R.; Tan, L. P.; Wu, H.; Yao, S. Q. Org. Lett. 2008, 10, 2295; (b) Yang,
P.-Y.; Wu, H.; Lee, M. Y.; Xu, A.; Srinivasan, R.; Yao, S. Q. Org. Lett. 2008, 10, 1881.
4. (a) Uttamchandani, M.; Wang, J.; Li, J.; Hu, M.; Sun, H.; Chen, K. Y.-T.; Liu, K.; Yao,
S. Q. J. Am. Chem. Soc. 2007, 129, 7848; (b) Uttamchandani, M.; Lee, W. L.; Wang,
J.; Yao, S. Q. J. Am. Chem. Soc. 2007, 129, 13110.
5. (a) Mastrolorenzo, A.; Rusconi, S.; Scozzafava, A.; Barbaro, G.; Supuran, C. T. Curr.
Med. Chem. 2007, 14, 2734; (b) Lee, T.; Laco, G. S.; Torbett, B. E.; Fox, H. S.; Lerner,
D. L.; Elder, J. H.; Wong, C.-H. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 939; (c) Kempf,
D. J. In Protease Inhibitors in AIDS Therapy; Ogden, R. C., Flexner, C. W., Eds.;
Marcel Dekker: New York, 2001; p 49; (d) Wilkerson, W. W.; Dax, S.; Cheatham,
W. W. J. Med. Chem. 1997, 40, 4079.
Acknowledgments
Funding was provided by the National University of Singapore
(R-143-000-280-112 and R-154-000-281-112) and the Agency for
*
Science, Technology, and Research (A STAR) of Singapore (R-154-
000-274-305).
6. (a) Jensen, C.; Herold, P.; Brunner, H. R. Nat. Rev. Drug Disc. Today 2008, 7, 399;
(b) Dash, C.; Kulkarni, A.; Dunn, B. Crit. Rev. Biochem. Mol. Biol. 2008, 38, 89; (c)
Hills, I. D.; Vacca, J. P. Curr. Opin. Drug. Disc. Dev. 2007, 10, 383; (d) Ersmark, K.;
Samuelsson, B.; Hallberg, A. Med. Res. Rev. 2006, 26, 626.
7. Jensen, K. J.; Alsina, J.; Songster, M. F.; Barany, G. J. Am. Chem. Soc. 1998, 120,
5441.
8. The purity of the compounds were estimated based on the LC profile of each
compound (see Supplementary data) obtained under the UV channel of 254 nm
(all 75 compounds in this study contain aromatic moieties).
Supplementary data
Supplementary data (detailed experimental procedures and re-
sults) associated with this article can be found, in the online ver-
sion, at doi:10.1016/j.bmcl.2009.03.041.
References and notes
9. The inhibition experiments were carried out using commercially available HIV-1
protease kit (see Supplementary data for details), and the screening conditions
were adjusted to ensure IC50 and Ki values obtained can be directly compared
with Wong’s and Abbott’s compounds.
1. Brik, A.; Wu, C.-Y.; Wong, C.-H. Org. Biomol. Chem. 2006, 4, 1446.
2. (a) Brik, A.; Lin, Y.-C.; Elder, J.; Wong, C.-H. Chem. Biol. 2002, 9, 891; (b) Lee, S.-G.;
Chmielewski, J. Chem. Biol. 2006, 13, 421; (c) Best, M. D.; Brik, A.; Chapman, E.;