C2-Symmetric azobenzene-amino acid conjugates and their inhibition of Subtilisin Kexin Isozyme-1

Article abstract of DOI:10.1016/j.bmcl.2010.04.101

C₂-Symmetric azobenzene-amino acid/peptide hybrids containing stable E-azo moiety have been synthesized. Upon irradiation with long wavelength UV, these compounds isomerized to the Z-form, whose thermal reisomerization to the E isomer slowed do

Full text of DOI:10.1016/j.bmcl.2010.04.101

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3977–3981
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
C2-Symmetric azobenzene-amino acid conjugates and their inhibition
of Subtilisin Kexin Isozyme-1
a
b
c
c,
*
Amit Basak a, , Debarati Mitra , Amit K. Das , Dayani Mohottalage , Ajoy Basak
*
a Department of Chemistry, Indian Institute of Technology, Kharagpur 721 302, India
b Department of Biotechnology, Indian Institute of Technology, Kharagpur 721 302, India
c Chronic Disease Program, Regional Protein Chemistry Center, Ottawa Hospital Research Institute, Department of Biochemistry, Microbiology and Immunology,
University of Ottawa, Ottawa, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
C2-Symmetric azobenzene-amino acid/peptide hybrids containing stable E-azo moiety have been synthe-
sized. Upon irradiation with long wavelength UV, these compounds isomerized to the Z-form, whose
thermal reisomerization to the E isomer slowed down considerably. These compounds exhibited
in vitro moderate to strong inhibition of mammalian cellular protease Subtilisin Kexin Isozyme-1, also
called Site 1 Protease, which plays vital roles in cholesterol synthesis, lipid metabolism, bone formation,
and viral infections.
Received 30 December 2009
Revised 21 April 2010
Accepted 22 April 2010
Available online 6 May 2010
Keywords:
C2-Symmetric
Azobenzene
Amino acid
Peptide
Ó 2010 Elsevier Ltd. All rights reserved.
Hybrid
Inhibitor
SKI-1
Symmetry has a crucial role in many chemical and biological
processes.1 In asymmetric synthesis, C2-symmetric catalysts play
a crucial role.2,3 There are many proteins and enzymes which are
biologically active in homo dimeric form possessing C2-symmetry.
Presence of C2-symmetry is a common feature of restriction en-
zymes4 (e.g., HindIII) which enables them to recognize palindrome
sequence of ds DNA. HIV-protease is another example of C2-sym-
metric aspartyl protease enzyme which functionally active in
homo dimeric form.5 Other proteolytic enzymes have been shown
to recognize substrates with C2-symmetry or pseudo C2-symmetry
thus pointing out complementarity with the active site of the en-
zyme. One such family is the Proprotein Convertases (PCs), cur-
rently known as Proprotein Convertase Subtilisin/Kexin-type
(PCSKs).6,7 So far there are nine members in this family, of which
seven, namely PC1/PC3 (PCSK1), PC2 (PCSK2), Furin (PCSK3), PC4
(PCSK4), PACE4 (PCSK5), PC5/PC6 (PCSK6), and PC7/PC8/LPC
(PCSK7) are Kexin-type and cleave a variety of inactive precursor
proteins at the consensus sequence (K/R)-Xn-(K/R);, where X is
any amino acid except Cys and n = 0, 2, 4, 6. The eighth member
is a pyrolysin type and is called Subtilisin Kexin-Isozyme-1 (SKI-
1)8/Site 1 Protease (S1P)9/(PCSK8) while the ninth or the last mem-
ber is known as Neural Apoptosis-Regulated Convertase (NARC-1)/
PCSK9 is a Proteinase K subtype.10 While SKI-1 has been shown to
cleave inactive proproteins at the motif R-X-(hydrophobic)-Z;, Z
being mostly L. In rare cases, P1 residue Z can be His (CREB-H),
Arg (Luman), Lys (SKI-1, Machupo and Tacaribe viruses GPC)
(REF). PCSK9 has been shown only to cleave its own prodomain
at V-F-A-Q;S-I-P site. While most PCSK enzymes exist in mono-
meric form, a few of them have been shown to exist also in dimeric
form with enhanced enzymatic and/or functional activity in most
cases.11–13 Owing to various functional and physiological roles,
PCs including SKI-1 enzyme are considered as important targets
for intervention of a number of diseases including viral infections
and cancer.14,15 Since there is a pseudo C2-symmetry around the
active site (motif: Arg/Lys/His-X-Leu/Ile-Leu)12,16 of SKI-1, sub-
strates with C2-symmetry are attractive targets for studying their
interactions with these enzymes. Based on this idea and the pref-
erence of hydrophobic residue at P1/P2 by SKI-1, we have designed
synthesized azobenzene-amino acid (or peptide) hybrids with C2-
symmetry (Fig. 1) and studied their protease inhibitory activities
in vitro against SKI-1. In this paper we describe our results.
The syntheses of the target compounds 1a–d are depicted in
Scheme 1. The key steps involved the linking of the protected
bromoacylated amino acid benzyl esters 3a–c to the azobenzene
via bis O-alkylation. The resulting products 2a–c were hydrolyzed
by LiOH to the corresponding free acids. The other compound 1d
containing a bis dipeptide unit was similarly prepared (Scheme 2)
* Corresponding authors. Tel.: +91 3222283300; fax: +91 3222282252 (A.B.).
E-mail address: absk@chem.iitkgp.ernet.in (A. Basak).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.101

3978
A. Basak et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3977–3981
absorption between wave length 354–361 nm. Moreover, the
H
appearance of the aromatic hydrogens at relatively downfield (well
above d 7.2) as compared to the known cis (Z) azo compounds17
supported the E-configuration of N@N moiety. One of the com-
pounds 1a gave a single crystal which allowed us to confirm the
structure by X-ray crystallography. It was elaborated in a bis ened-
iyne derivative as reported earlier.17
HOOC
N
O
R
O
N
N
O
R
O
N
COOH
H
The dibenzyl esters 2a–d were then examined for their thermal
reisomerization kinetics. The photoisomerization studies were per-
formed in dichloromethane with UV-irradiation at 365 nm for 3–
4 h. The thermal reisomerization for Z to E conversion of all the ser-
ies at 30 °C showed first order kinetic as shown in Table 1 which
reveals that the half lives for various azobenzene esters follow
the order: 2c > 2b > 2a > 2d. That the peptide bonds on the two
arms of the azobenzene have rendered some stability to the Z-iso-
mer is clear from the larger t1/2 values as compared to the azoben-
zene derivatives without such unit. The stability is, however,
maximum for glycine substituted compound 2c, followed by the
phenyl alanine and valine-based compounds 2b and 2a, respec-
tively, clearly demonstrating the higher stability for compounds
with minimum steric strains. The incorporation of phenyl or iso-
propyl substituent, led to an increase in steric strain thus lowering
the half life of the Z-isomer. The greater stability of the phenyl
For a, R = -iPr, for b, R = CH2Ph, for c, R = H
1a-c
R
O
H
N
HOOC
N
O
H
R
O
N
N
O
R
H
O
N
COOH
N
H
O
R
R = -CH2Ph
1d
Figure 1. Target azobenzene-amino acid/peptide hybrids.
by O-alkylation with the bromoacylated dipeptide 3d. The com-
pounds 2a–d when isolated by column chromatography generally
remain in the trans (E) form which was deduced from strong UV
H
N
PhH COOC
2
a
O
R
O
Br
R
O
N
N
COOCH2Ph
N
O
R
H
2a-c
O
3a-c
N
H
COOCH Ph
2
H
N
HOOC
b
O
R
O
N
N
O
R
O
N
H
COOH
1a-c
a) 2,2'-dihydroxyazobenzene, anh. Cs2CO3, dry CH3CN, 40-50º C, 12 h, 50%; b)
LiOH, THF/H2O, 18 h, 60%; for a, R = -iPr; for b, R = -CH2Ph; for c, R = H
Scheme 1. Synthesis of azobenzene-amino acid hybrids.
R
H
O
O
R
g
H
N
N
COOCH2Ph
Br
COOCH2Ph
TfOH3N
N
O
R
H
O
R
4
3d
R
O
H
N
N
O
PhH COOC
2
H
R
O
N
h
i
N
O
R
H
N
O
COOCH2Ph
N
H
2d
O
R
R
O
H
N
HOOC
N
O
H
R
O
N
N
O
R
H
N
O
COOH
R = CH 2Ph
N
H
1d
O
R
g) BrCH2COCl, Et3N, DCM, 12-15 m; h) 2, 2'-dihydroxy azo benzene, Cs2CO3, CH3CN (dry), 12 h,
60%; i) LiOH, THF/ H2O, 16-18 h, 50%
Scheme 2. Synthesis of azobenzene-amino acid derivatives.

A. Basak et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3977–3981
3979
Table 1
terminal peptide fragment per microliter of enzyme sample in
one hour following incubation at 37 °C.18 The progress of enzyme
inhibition was monitored by stop time assay.
The inhibition assay was performed only with the thermally
stable E-form because of instability of Z-form.19 End point assay
was conducted by measuring relative fluorescence units (RFU)
(kex = 320 nm and kem = 420 nm) at various times starting from 0
Rate constants and half lives of Z-azo isomers
Compound
K (hꢀ1
t1/2 (h)
2a
0.022
31.5
2b
0.010
69.3
2c
0.009
77.0
2d
0.024
28.9
)
alanine derivative over the valine counterpart may be due to the p-
to 6 h after incubation of SKI-1 (10 lL) with Q-GPC251–263
stacking interaction between the two phenyl rings in the Z-form.
The incorporation of more peptide bonds, however, caused a low-
ering of half life. Whether or not this is a general phenomenon is
still to be addressed because of lack of number examples.
Table 2
Measured inhibition constant values
With the compounds in hands, inhibition studies in vitro
against human (h) recombinant soluble SKI-1 enzyme12,16 were
carried out. For this study, the fluorogenic substrate Q-GPC251–263
(Abz-251Asp-Ile-Tyr-Ile-Ser-Arg-Arg-Leu-Leu-;-Gly-Thr-Phe
Thr263-Tyx-Ala-CONH2 where Abz = 2-amino benzoic acid, Tyx = 3-
nitrotyrosine) was used. For each study the activity of the enzyme
used was nearly identical, releasing 10 pmol of Abz containing N-
Compound
Inhibition constant Ki (lm)
2a
2b
1a
1b
1c
1d
125 15
11.8 1.5
75
8
225 18
265 11
140 12
2b
2a
80
50
0.1 mM QGPC
0.05 mM QGPC
40
30
20
10
60
0.025 mM QGPC
40
20
-0.05
0
0.05
0.1
0.15
0.2
-0.15 -0.1 -0.05
0
0.05 0.1 0.15 0.2
Inhibitor Concentration (mM)
Inhibitor Concentration (mM)
1a
80
1b
100
60
40
20
80
60
40
20
-0.1 -0.05
0
0.05
0.1
0.15
0.2
-0.3
-0.2
-0.1
0
0.1
0.2
Inhibitor Concentration (mM)
Inhibitor Concentration (mM)
80
80
1c
1d
60
40
20
60
40
20
-0.15 -0.1 -0.05
0 0.05 0.1 0.15 0.2
-0.3
-0.2
-0.1
0
0.1
0.2
Inhibitor Concentration (mM)
Inhibitor Concentration (mM)
Figure 2. Inhibition kinetics of various azo compounds against SKI-1 enzyme using Dixon plots.

3980
A. Basak et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3977–3981
Figure 3. Models showing SKI-1 catalytic domain (right) and the docking of compound 2b with SKI-1 catalytic region (left).
(100
l
M) in buffer [25 mM Tris + 25 mM Mes + 2 mM CaCl2, pH
L in 96-well plate. For determination
Acknowledgments
7.4] in a total volume of 100
l
of Ki values, each inhibitor concentration was varied over a range
wide enough to yield residual activities of 20–80% of control value
following incubation with SKI-1 enzyme. Typically h-SKI-1 sample
The authors Amit B. and D.M. are grateful to CSIR, Government
of India for research funding and fellowship, respectively. D.S.T. is
thanked for the funds for 400 MHz NMR facility under the IRPHA
program. Ajoy B. sincerely thanks National Sciences and Engineer-
ing Research Council, Government of Canada for a research grant
under discovery program.
(10
ume 100
l
L) and substrate (10
l
L) were incubated in buffer (total vol-
L) of varying concentrations
lL) with each inhibitor (2
l
in 96-well micro-titer plate at 37 °C. The rate of substrate hydroly-
sis was obtained from the changes in fluorescence readings and the
values transformed into amounts of lmol/h of peptide cleaved by
References and notes
using standard curve and the measured quenching corrections as
we did previously.16 Non-linear regression analysis of plots of the
hydrolysis rate versus the inhibitor concentration was used. In all
cases the inhibition is mostly competitive in nature as determined
by Dixon plots (Fig. 2). The inhibition constant Ki (Table 2) was
measured using three different concentrations (100, 50, and
1. Richards, F. M.; Wyckoff, H. W. The Enzymes, 3rd ed.; Academic press: New
York, 1971. p 647.
2. Raines, R. T. Chem. Rev. 1998, 98, 1045.
3. Fett, J. W.; Strydom, D. J.; Lobb, R. R.; Alderman, E. M.; Bethune, J. L.; Riordan, J.
F.; Vallee, B. L. Biochemistry 1985, 20, 5480.
4. Sorrentino, S.; Glitz, D. G.; Hamann, K. J.; Loegering, D. A.; Checkel, J. L.; Gleich,
G. J. J. Biol. Chem. 1992, 267, 14859.
25
l
M) of Q-GPC251–263
.
5. Matousek, J. Comp. Biochem. Phys. C 2001, 129, 175.
From this table we find that all the compounds have inhibition
6. Seidah, N. G.; Benjannet, S.; Hamelin, J.; Mamarbachi, A. M.; Basak, A.;
Marcinkiewicz, J.; Mbikay, M.; Chrétien, M.; Marcinkiewicz, M. Ann. N.Y. Acad.
Sci. 1999, 885, 57.
7. Seidah, N. G.; Mayer, G.; Zaid, A.; Rousselet, E.; Nassoury, N.; Poirier, S.;
Essalmani, R.; Prat, A. Int. J. Biochem. Cell Biol. 2008, 40, 1111.
8. Seidah, N. G.; Mowla, S. J.; Hamelin, J.; Mamarbachi, A. M.; Benjannet, S.; Touré,
B. B.; Basak, A.; Munzer, J. S.; Marcinkiewicz, J.; Zhong, M.; Barale, J. C.; Lazure,
C.; Murphy, R. A.; Chrétien, M.; Marcinkiewicz, M. Proc. Natl. Acad. Sci. U.S.A.
1999, 96, 1321.
9. Cheng, D.; Espenshade, P. J.; Slaughter, C. A.; Jaen, J. C.; Brown, M. S.; Goldstein,
J. L. J. Biol. Chem. 1999, 274, 22805.
10. Seidah, N. G.; Benjannet, S.; Wickham, L.; Marcinkiewicz, J.; Jasmin, S. B.;
Stifani, S.; Basak, A.; Prat, A.; Chretien, M. Proc. Natl. Acad. Sci. U.S.A. 2003, 100,
928.
11. Nagahama, M.; Taniguchi, T.; Hashimoto, E.; Imamaki, A.; Mori, K.; Tsuji, A.;
Matsuda, Y. FEBS Lett. 1998, 434, 155.
12. Mohottalage, D.; Goto, N.; Basak, A. Adv. Exp. Med. Biol. 2009, 611, 83.
13. Fan, D.; Yancey, P. G.; Qiu, S.; Ding, L.; Weeber, E. J.; Linton, M. F.; Fazio, S.
Biochemistry 2008, 47, 1631.
14. Bontemps, Y.; Scamuffa, N.; Calvo, F.; Khatib, A. M. Med. Res. Rev. 2007, 27, 631.
15. Chrétien, M.; Seidah, N. G.; Basak, A.; Mbikay, M. Expert Opin. Ther. Targets
2008, 12, 1289.
16. Basak, A.; Mohottalage, D.; Basak, A. Protein Pept. Lett. 2006, 13, 863.
17. (a) Basak, A.; Mitra, D.; Kar, M.; Biradha, K. Chem. Commun. 2008, 3067; (b) Kar,
M.; Basak, A. Chem. Rev. 2007, 107, 2861.
with IC50 or Ki in the micromolar range with compound 1b being
the most potent inhibitor. Unlike the highly reactive peptidyl-
cmk (chloromethyl ketone) derivatives,20 the azo-compound deriv-
atives did not exhibit any significant inhibition of Furin at 100 lM
or lower concentrations. We have also performed the docking
study so as to compare the in vitro and in silico results. For that,
we used Autodock version 3.0.5 program. It was observed from
the Swiss PDB Viewer that no direct H-bonding interactions oc-
curred with SKI-1 active site residues, but other stabilizing interac-
tions occurred in the active site region especially with the catalytic
218Asp, 249His, 414Ser, and 338Asn residues (Fig. 3). H-bonding and
other polar interactions occurred with nearby residues (not shown
in the figure). As docking occurred in the active site region of SKI-1
(in case of compound 2b as an example), it suggested to a compet-
itive inhibition. In reality also, this compound showed competitive
inhibition. Thus our docking experiment supported in vitro study.
Figure 3 represents theoretical active site structure of the SKI-1 en-
zyme (right) and the docking result with compound 2b (left).
Thus we have successfully synthesized novel C2-symmetric azo-
benzene-amino acid conjugates which showed modest to good
inhibitory activities against SKI-1 enzyme in a competitive manner.
Efforts are currently underway to evaluate their ex vivo activities
using specific cell lines. Previously it was demonstrated that simi-
lar C2-symmetric bis-azo compounds such as 2,20-azobis(2-amidi-
nopropane hydrochloride can cause peroxidation of rat liver
microsomes21 but no report of their protease inhibitory activity to-
wards SKI-1 was ever reported. Therefore this may provide an
alternate strategy for design of more selective and potent inhibi-
tors of SKI-1 that may find useful therapeutic and biochemical
applications.
18. Dreostic, I. E.; Wargovich, M. J.; Yang, C. S. Crit. Rev. Food Sci. 1997, 37, 761.
19. Initially we decided to go with the E-forms only for comparison purpose but
other isoform (Z-form) will also be tested in future study.
20. Pasquato, A.; Pullikotil, P.; Asselin, M. C.; Vacatello, M.; Paolillo, L.; Ghezzo, F.;
Basso, F.; Di Bello, C.; Dettin, M.; Seidah, N. G. J. Biol. Chem. 2006, 281, 23471.
21. Cai, Y.; Ma, L.; Hou, L.; Zhou, B.; Yang, L.; Liu, Z. Chem. Phys. Lipids 2002, 120,
109.
Spectral data of selected compounds (IR data taken in KBr pellet and expressed in
cmꢀ1 1H NMR and 13C NMR were recorded at 400 and 100 MHz, respectively,
,
in CDCl3 unless stated otherwise):
1a: max 3376, 2966, 1728, 1646, 1542, 1285, 1056; dH (d6-DMSO) 12.87 (2H, br
t
s), 7.83 (2H, d, J = 8.8 Hz), 7.57 (2H, d, J = 7.6 Hz), 7.42 (2H, t, J = 7.6 Hz), 7.00
(2H, d, J = 8.4 Hz), 6.95 (2H, d, J = 7.6 Hz), 4.76 (4H, s), 4.23 (2H, m), 1.99 (2H,
m), 0.76 (6H, d, J = 6.8 Hz), 0.71 (6H, d, J = 6.8 Hz); dC (d6-DMSO) 173.0, 168.0,
155.6, 142.1, 133.3, 121.9, 117.0, 115.4, 67.9, 56.8, 30.6, 19.4, 17.9; mass (ES+)

A. Basak et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3977–3981
3981
m/z 529.31 (MH+); HRMS Anal. Calcd for C26H32N4O8 529.2298. Found
529.2300.
121.4, 119.5, 67.8, 43.0; mass (ES+) m/z 445.13 (MH+); HRMS Anal. Calcd for
C20H20N4O8 445.1359. Found 445.1362.
1b: tmax 3401, 1723, 1661, 1529, 1486, 1439, 1233, 1120, 759; dH (d6-DMSO)
13.0 (2H, br s), 8.26 (2H, m), 7.49 (2H, d, J = 8.0 Hz), 7.41 (2H, t, J = 8.0 Hz), 7.18
(10H, m), 7.01 (4H, m), 4.72 (4H, s), 4.58 (2H, m), 3.09 (2H, dd, J = 13.2, 4.8 Hz) ,
2.90 (2H, dd, J = 13.2, 4.8 Hz); dC (d4-MeOH) 172.5, 169.1, 155.3, 142.5, 136.0,
132.5, 129.1, 127.9, 126.4, 122.3, 117.4, 115.3, 68.5, 52.9, 37.0; mass (ES+) m/z
625.23 (MH+); HRMS Anal. Calcd for C34H32N4O8 645.2278. Found 625.2282..
2c: tmax 2963, 1745, 1660, 1376, 759; dH (200 MHz) 7.80–7.61 (4H, m,
aromatic-H), 7.44–7.21 (10H, m, aromatic-H), 7.11–7.02 (6H, m, aromatic-H, –
CONH), 5.12 (4H, m, –CO2CH2Ph), 4.76 (4H, s, –OCH2CO), 4.11 (4H, d, J = 5.2 Hz,
–NHCH2COOCH2Ph); dC (50 MHz) 169.2, 168.8, 155.2, 135.0, 132.9, 128.6,
128.5, 128.4, 128.3, 128.3, 128.2, 123.0, 67.1, 40.8, 29.6; mass (ES+) m/z 625.25
(MH+); HRMS Anal. Calcd for C34H32N4O8 625.2278. Found 625.2283.
1c: tmax 2966, 1728, 1662, 1217, 1056, 735; dH (d6-DMSO) 8.39–8.33 (2H, m),
7.64–7.62 (2H, m), 7.50 (2H, t, J = 8.0 Hz), 7.30–7.23 (2H, m), 7.12–7.08 (2H, m),
4.79 (4H, s), 3.85 (4H, s); dC (d6-DMSO) 172.9, 168.7, 155.5, 138.1, 131.7, 123.7,
2d: tmax 2965, 1748, 1662, 1578, 759; dH 7.43 (2H, m), 7.36–7.35 (6H, m,
aromatic-H), 7.30–7.24 (12H, aromatic-H), 7.18–7.15 (4H, aromatic-H), 7.10–
6.89 (16H, m, aromatic-H, –CONH), 6.25 (2H, d, J = 8.0 Hz, –CONH), 5.10 (4H, s,
–CO2CH2Ph), 4.78 (2H, m,
a-H), 4.66–4.62 (6H, m, a-H, –OCH2CO), 3.04–2.92
(8H, m, 2ꢁ –CH2Ph); dC (CDCl3): 170.7, 169.8, 168.3, 155.2, 136.3, 135.4, 135.0,
133.0, 129.2, 129.2, 129.1, 128.8, 128.7, 128.6, 128.6, 128.5, 127.1, 126.9, 123.1,
118.0, 115.4, 67.3, 54.1, 53.4, 37.8, 37.7, 29.7; mass (ES+) m/z 1099.5 (MH+);
HRMS Anal. Calcd for C66H62N6O10 1099.4605. Found 1099.4608.
1d:
tmax 2976, 1728, 1667, 1217, 1056, 735; dH (d6-DMSO): 7.52 (2H, br s), 7.33
(2H, s), 7.18–7.14 (16H, m), 6.99–6.97 (4H, m), 6.86 (4H, m), 4.59 (6H, m,
a-H),
4.33 (2H, br s), 3.01–2.83 (8H, m); dC (d6-DMSO) 170.2, 170.0, 167.2, 155.3,
138.5, 138.0, 137.7, 132.7, 129.5, 129.4, 129.2, 128.3, 128.1, 127.9, 126.3, 126.0,
121.5, 67.6, 55.0, 53.4, 37.9, 37.4; mass (ES+) m/z 919.4 (MH+); HRMS Anal.
Calcd for C52H50N6O10 919.3666. Found 919.3670.