Design of inhibitors against guanase: Synthesis and biochemical evaluation of analogues of azepinomycin

Article abstract of DOI:10.1016/j.bmcl.2006.08.033

As part of a program to design rational, mechanism-based inhibitors of guanase, we report here the synthesis and biochemical screening of two analogues of azepinomycin (1 and 2), a naturally occurring inhibitor of guanase, known to mimic the transition-state of the enzyme-catalyzed reaction. Our biochemical results show that compounds 1 and 2 are competitive inhibitors with K_i of 2.01 ± 0.16 × 10^-5 and 5.36 ± 0.14 × 10^-5 M, respectively.

Full text of DOI:10.1016/j.bmcl.2006.08.033

Bioorganic & Medicinal Chemistry Letters 16 (2006) 5551–5554
Design of inhibitors against guanase: Synthesis and biochemical
evaluation of analogues of azepinomycin
Ravi K. Ujjinamatada, Anila Bhan and Ramachandra S. Hosmane*
Laboratory for Drug Design and Synthesis, Department of Chemistry and Biochemistry,
University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, MD 21250, USA
Received 8 July 2006; revised 4 August 2006; accepted 7 August 2006
Available online 22 August 2006
Abstract—As part of a program to design rational, mechanism-based inhibitors of guanase, we report here the synthesis and bio-
chemical screening of two analogues of azepinomycin (1 and 2), a naturally occurring inhibitor of guanase, known to mimic the
transition-state of the enzyme-catalyzed reaction. Our biochemical results show that compounds 1 and 2 are competitive inhibitors
with Ki of 2.01 0.16 · 10ꢀ5 and 5.36 0.14 · 10ꢀ5 M, respectively.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Guanine deaminase (GDA) or guanase (EC 3.5.4.3) is
an enzyme that catalyzes the hydrolytic deamination
of guanine to xanthine. This enzyme has been found in
human liver, brain, and kidney.1 There have been re-
ports of abnormally high levels of serum guanase activ-
ity in patients with liver diseases,2–4 and so, the elevated
enzyme activity has been suggested as a marker of hep-
atitis and hepatoma.3 Furthermore, such a high guanase
activity is believed to be a biochemical indicator of rejec-
tion in liver transplant recipients.5 Increased levels of
guanase have also been detected in cancerous kidney6–8
and breast cancer tissues.7,9 In addition, patients with
multiple sclerosis exhibit significantly elevated levels of
guanase activity in their cerebral spinal fluids.10 These
observations suggest that a potent guanase inhibitor is
necessary for exploring the biochemical mechanisms of
the above metabolic disorders as well to understand
the specific physiological role played by guanase, and
not to mention its potential therapeutic use in treating
these disorders. While many studies on guanase inhibi-
tion have been reported in the literature,11–36 a potent
guanase inhibitor with a submicromolar or nanomolar
Ki has yet to be discovered. We report here the design,
synthesis, and guanase inhibitory activity of two
analogues (1 and 2) of the natural product azepinomy-
cin, a moderate inhibitor of guanase, which is believed
to mimic the transition state of the enzyme-catalyzed
deamination reaction.37
O
O
1
1
HN
8
N
HN
6
5
N
8
7
7
2
6
5
2
HO
HO
3
3
4
4
N
N
N
N
H
H
O
H
R
1, R = -CH2 C6 H5
2, R = H
Azepinomycin
Guanase catalyzes the hydrolysis of guanine (3) to xan-
thine (4) (Scheme 1) via the tetrahedral intermediate
(5).38 The recently solved crystal structure of guanase39
from Bacillus subtilis suggests that the enzyme-catalyzed
reaction is assisted by an active site zinc metal ion
(Zn2+), which forms a tetrahedral complex with His-
53, Cys-83, and Cys-86 of the protein, as well as with
an isolated water molecule. Glutamate-55 serves as a
proton shuttle, abstracting a proton from the zinc-acti-
vated water to form the necessary hydroxide nucleo-
phile, while also enabling protonation at the N-3 site
of guanine, thus resulting in the formation of the inter-
mediate 5, as shown. Glu-55 also assists in protonation
of the NH2 group of 5, facilitating the elimination of a
molecule of ammonia to form the final product
Keywords: Guanase (guanine deaminase); Inhibitors; Azepinomycin
analogues; Synthesis; Biochemical studies.
*
Corresponding author. Tel.: +1 410 455 2520; fax: +1 410 455
1148; e-mail: hosmane@umbc.edu
Present address: G.E Healthcare, BioSciences, Protein Separations,
800 Centennial Ave, Piscataway, NJ 08855, USA.
0960-894X/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.08.033

5552
R. K. Ujjinamatada et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5551–5554
O
shown to be bonded to the metal are only arbitrary).
O
7
6
With this rationale, we set out to synthesize our target
inhibitors 1 and 2. The substitution of a benzyl group
at position-3 of 1 was to explore if a hydrophobic moi-
ety in that position would enhance, lower, or unaffect
the strength of enzyme binding. In our earlier guanase
inhibition studies involving planar, aromatic ring-ex-
panded purine bases, we had noticed that a benzyl group
at the N-3 position often potentiated the enzyme
inhibition.30,32
1
HN
5
N
N
HN
8
2
3
9
N
N
H
H2N
N
4
O
N
H
H
Guanine (3)
Xanthine (4)
H2O
- NH3
O
Cys-86
Cys-83
Zn+2
The target compound 140 was synthesized (Scheme 3) by
two different methods, both by reduction mediated by a
mixture of lithium borohydride and super hydride in
THF at 0 ꢁC. The first method employed 3-benzyl-
4,5,7,8-tetrahydro-6-methoxy-6-methoxycarbonyl-6H-
imidazo[4,5-e][1,4]diazepine-5,8-dione (6)20 as the starting
material, while the second method used 3-benzyl-
4,5,7,8-tetrahydro-6-methoxycarbonyl-6H-imidazo[4,5-
e][1,4]diazepine-5,8-dione (7).41 The detailed multi-step
synthesis of both compounds 6 and 7 has been reported
by us several years ago,20,41 employing a common start-
ing material, 1-benzyl-5-nitroimidazole-4-carboxylic
acid.20 The target compound 240 was prepared by deben-
zylation of 1 using catalytic hydrogenation with palladi-
um hydroxide in acetic acid at 40 psi. The target
N
HN
His-53
O
O
O
H
HO
Glu55
H
N
N
H
H
H2N
(5)
Scheme 1. Zn2+-assisted hydrolysis of guanine to xanthine, catalyzed
by guanase.
xanthine.39 Azepinomycin may interrupt this hydrolytic
process via coordination of its own OH group at posi-
tion –6 with the active site zinc, displacing the crucial ac-
tive site water molecule involved in hydrolysis.
The above zinc-mediated hydrolysis of guanine to xan-
thine, catalyzed by guanase, led us to the design of the
title inhibitors of this paper. In view of only a moderate
inhibition of guanase by azepinomycin,37 which suggest-
ed somewhat weak binding of the inhibitor to the en-
zyme via zinc metal coordination, we hypothesized
that the additional metal coordination sites on the inhib-
itor would strengthen its binding to the protein, and
hence enhance its inhibitory potential. The introduction
of a carbonyl group at position-5 of the heterocycle,
coupled with movement of the 6-hydroxy group away
from the ring by an additional carbon atom, would al-
low excellent coordination of the inhibitor with Zn2+
to form a stable, 6-membered ring structure. The latter
would use two of the four metal coordination sites,
while the other two would be occupied by two of the
three original amino acid residues at the enzyme active
site, as depicted in Scheme 2 (Note: the two amino acids
1
compounds were characterized by H and 13C NMR,
and mass spectral data, along with elemental
microanalyses.
Guanase from rabbit liver (purchased from Sigma–Al-
drich) was employed in the biochemical studies. All
studies were carried out at 25 ꢁC and pH 7.4 by spectro-
scopic measurements of the rate of hydrolysis of the sub-
strate guanine at kmax 245 nm. The change in optical
density at kmax 245 nm per unit time is a measure of
the guanase activity. A total of seven different concen-
trations of the substrate, ranging from 5 to 20 lM,
was employed for each inhibitor concentration that
was either 8 or 25 lM, while the amount of enzyme in
each assay was 7.67 · 10ꢀ3 unit. The Lineweaver–Burk
plots (1/V vs 1/S) (see Figs. 1A and B) were used to
O
O
O
LiBH4 / THF
Super Hydride
0 oC - reflux
1
N
HN
HN
N
N
N
N
HN
8
MeOOC
7
His-53
Cys-83
H2O
2
6
5
HO
3
N
HO
4
N
H
MeO
Zn+2
N
H
N
H
51 %
O
O
Ph
O
R
Ph
Cys-86
6
1
1
2
Guanine
N
Pd (OH)2 / H2
LiBH4 / THF
Super Hydride
0 oC - reflux
O
3
N
CH3COOH,
40 psi
63 %
N
8
6
R
4
7
HN
O
X
Xanthine
42 %
NH
5
Guanase
O
O
O
HN
N
N
HN
MeOOC
Zn+2
HO
N
H
N
H
N
O
O
Ph
H
Cys-86
Cys-83
7
2
Scheme 2. The proposed pathway for inhibition of guanase by the title
analogues of azepinomycin.
Scheme 3. Synthesis of the target compounds 1 and 2.

R. K. Ujjinamatada et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5551–5554
5553
A
B
90
75
60
45
30
15
72
60
48
36
24
12
1/V
1/V
μ
μ
I = 0
I = 8
M
M
I = 0 μM
I = 8 μM
I = 25 μM
I = 2 5
M
μ
-0.125 -0.050 0.025 0.100 0.175
1/[S] (
M-1)
-0.125 -0.050 0.025 0.100 0.175
1/[S] (mM-1)
μ
Figure 1. (A) Lineweaver–Burk plot of target compound 1. (B) Lineweaver–Burk plot of target compound 2.
calculate Km, Vmax
,
and Ki. Our biochemical
as in pentostatin because of possibly imperfect relative
placements of the ligand and the metal at the enzyme
active site. The best answer may come from the X-ray
structure of a complex of azepinomycin or one of its
analogues with a mammalian guanase.
results showed that Km of enzyme with guanine as
substrate was 0.966 · 10ꢀ5 M. The target compounds 1
and 2 both showed competitive inhibition with Ki of
2.01 0.16 · 10ꢀ5 and 5.36 0.14 · 10ꢀ5 M, respective-
ly. These results suggest that the presence of a benzyl
group at position-3 (compound 1) enhances the inhibito-
ry activity by greater than twofold as compared to that
without the benzyl group (compound 2).
Acknowledgments
The research was supported in part by grants (# 5 RO1
AI55452 & #1 R21 AI071802) from the National Insti-
tute of Allergy and Infectious Diseases (NIAID) of the
National Institutes of Health, Bethesda, Maryland,
and by an unrestricted grant from Nabi Biopharmaceu-
ticals, Rockville, Maryland.
Inhibitory activities (Ki) of the target compounds 1 and
2 are roughly comparable to that of azepinomycin
(IC50 = 0.5 · 10ꢀ5 M),37 especially considering that the
inhibition for the latter is reported only as IC50, which
being specific to the conditions of the assay such as con-
centrations of the substrate and the enzyme is often
somewhat lower than the actual Ki value.42 The ob-
served inhibition data suggest that the introduction of
an additional Zn2+ coordination site on azepinomycin
nucleus has little effect on the enzyme inhibition. There
are multiple possibilities to explain these observed re-
sults, including but not limited to: (a) The mechanism
of GDA inhibition by either azepinomycin or its ana-
logues 1 and 2 may not involve the active site Zn2+ coor-
dination. This is a distinct possibility in view of their
moderate Ki values as compared to, for example, that
of the structurally analogous pentostatin (20-deoxyco-
formycin), another 5:7-fused imidazo[4,5-d][1,3]diaze-
pine ring system containing an exocyclic hydroxy
group.43 Pentostatin, unlike azepinomycin, is an
extremely tight-binding, totally irreversible inhibitor of
adenosine deaminase (ADA) with a Ki ranging from
10ꢀ11 to 10ꢀ13 M,43 and whose X-ray structure com-
plexed with the protein clearly shows coordination of
its hydroxy group with an active site zinc metal
ion.44,45 (b) Another possibility is that both azepinomy-
cin and its analogues 1 and 2 coordinate with the active
site zinc but the ligand–metal interaction is not as strong
References and notes
1. Knights, E. M., Jr.; Whitehouse, J. L.; Hue, A. C.; Santos,
C. L. J. Lab. Clin. Med. 1965, 65, 355.
2. Nishikawa, Y.; Fukumoto, K.; Watanabe, F.; Yoshihara,
H. Rinsho. Byori. 1989, 37, 1392.
3. Shiota, G.; Fukada, J.; Ito, T.; Tsukizawa, M.; Yamada,
M., et al. Jpn. J. Med. 1989, 28, 22.
4. Ito, S.; Tsuji, Y.; Kitagawa, N.; Akihiko, I.; Syundo, J.,
et al. Hepatology 1988, 8, 383.
5. Crary, G. S.; Yasmineh, W. G.; Snover, D. C.; Vine, W.
Transplant. Proc. 1989, 21, 2315.
6. Durak, I.; Beduk, Y.; Kavutcu, M.; Suzer, O.; Yaman, O.,
et al. Cancer Invest. 1997, 15, 212.
7. Yuan, G.; Bin, J. C.; McKay, D. J.; Snyder, F. F. J. Biol.
Chem. 1999, 274, 8175.
8. Paletzki, R. F. Neuroscience 2002, 109, 15.
9. Canbolat, O.; Durak, I.; Cetin, R.; Kavutcu, M.; Demirci,
S., et al. Breast Cancer. Res. Treat. 1996, 37, 189.
10. Kaneshige, Y.; Matsumoto, H.; Chiba, S.; Hashimoto, S.;
Noro, H. Rinsho. Shinkeigaku. 1989, 29, 854.
11. Baker, B. R. J. Med. Chem. 1967, 10, 59.
12. Baker, B. R.; Santi, D. V. J. Med. Chem. 1967, 10, 62.

5554
R. K. Ujjinamatada et al. / Bioorg. Med. Chem. Lett. 16 (2006) 5551–5554
13. Baker, B. R.; Siebeneick, H. U. J. Med. Chem. 1971, 14, 802.
14. Baker, B. R.; Siebeneick, H. U. J. Med. Chem. 1971, 14,
799.
15. Baker, B. R.; Wood, W. F. J. Med. Chem. 1967, 10, 1106.
16. Baker, B. R.; Wood, W. F. J. Heterocycl. Chem. 1967, 10,
1101.
17. Baker, B. R.; Wood, W. F. J. Med. Chem. 1968, 11, 650.
18. Baker, B. R.; Wood, W. F. J. Med. Chem. 1968, 11, 644.
19. Baker, B. R.; Wood, W. F. J. Med. Chem. 1969, 12, 216.
20. Bhan, A.; Hosmane, R. S. Tetrahedron Lett. 1994, 35,
6831.
21. Bhan, A.; Hosmane, R. S. Nucleosides Nucleotides 1995,
14, 455.
22. Dietrich, L. S.; Shapiro, D. M. J. Biol. Chem. 1953, 203,
89.
23. Kanzawa, F.; Hoshi, A.; Kuretani, K. Chem. Pharm. Bull.
1971, 19, 1737.
24. Kanzawa, F.; Hoshi, A.; Nishimoto, T.; Kuretani, K.
Chem. Pharm. Bull. 1970, 18, 392.
mixture was refluxed with stirring for 5 h. It was slowly
cooled to room temperature. Methanol (5 mL) was added
and the pH of the mixture was carefully brought to 7.0
with 1 N HCl. The reaction mixture was adsorbed on silica
gel (1 g), loaded onto a flash silica gel column, and eluted
with a mixture of chloroform–methanol (6:1). The appro-
priate fractions were pooled and evaporated to obtain a
white solid. Yield 23 mg, 51% ; Rf = 0.24 (chloroform–
methanol, 5:1); mp 169–171 ꢁC; 1H NMR (400 MHz,
DMSO-d6): d 10.95 (s, 1H, CONH, D2O exchangeable),
7.68 (s, 1H, imidazole CH), 7.51 (s, 1H, CONH, D2O
exchangeable), 7.35–7.13 (m, 5H, C6H5), 5.26 (s, 2H,
CH2), 4.75 (s, 1H, OH, D2O exchangeable), 3.72 (m, 1H,
CH), 3.62 (m, 2H, CH2); 13C NMR (100 MHz, DMSO-
d6): d 169.1, 163.8, 136.31, 136.0, 131.7, 128.7, 127.7,
126.9, 122.7, 58.4, 55.1, 46.6; HRMS (FAB) calcd for
C14H14N4O3, 286.1065 (M+); observed m/z 287.1058
(MH+). Anal. Calcd for C14H14N4O3: C, 58.73; H, 4.93;
N, 19.57. Found: C, 58.55; H, 5.05; N, 19.28.
25. Kumar, S.; Josan, V.; Sanger, K. C. S.; Tewari, K. K.;
Krishnan, P. S. Biochem. J. 1967, 102, 691.
26. Lucacchini, A.; Bazzichi, L.; Biagi, G.; Livi, O.; Segnini,
D. Ital. J. Biochem. 1982, 31, 153.
27. Lucacchini, A.; Montali, U.; Segnini, D. Ital. J. Biochem.
1977, 26, 27.
Method B: The title compound was prepared in 42% yield
from
3-benzyl-4,5,7,8-tetrahydro-6-methoxy-carbonyl-
6H-imidazo[4,5-e][1,4]diaze-pine-5,8-dione (7),41 using
the procedure described above in Method A. The isolated
product was identical in all aspects, including m.p., Rf, 1H
NMR, 13C NMR, and Mass spectral data.
28. Lucacchini, A.; Segnini, D.; Da Settimo, A.; Livi, O. Ital.
J. Biochem. 1979, 28, 194.
29. Meyer, R. B., Jr.; Skibo, E. B. J. Med. Chem. 1979, 22,
944.
30. Rajappan, V.; Hosmane, R. S. Bioorg. Med. Chem. Lett.
1998, 8, 3649.
31. Rajappan, V. P.; Hosmane, R. S. Nucleosides Nucleotides
1998, 17, 1141.
4,5,7,8-Tetrahydro-6-hydroxymethyl-6H-imidazo[4,5-e] [1,4]-
diazepine-5,8-dione (2): A mixture of 1 (46 mg, 0.16 mmol)
and palladium hydroxide (30 mg, 0.2 mmol) in glacial
acetic acid (5 mL) was hydrogenated in a Parr hydroge-
nator at 40 psi for 18 h. The reaction mixture was filtered
through Celite. The Celite cake was washed with methanol
and the filtrate was evaporated to dryness. The residue was
triturated with ether. The precipitated solid was filtered
and washed with ether. Yield 21 mg (63 %); Rf = 0.15
32. Rajappan, V. P.; Hosmane, R. S. Nucleosides Nucleotides
1999, 18, 835.
(methanol–chloroform,
7:3);
light
brown
solid;
33. Saxena, A. K.; Ahmad, S.; Shanker, K.; Bhargava, K. P.;
Kishor, K. Pharmazie 1980, 35, 16.
34. Saxena, A. K.; Ahmad, S.; Shanker, K.; Kishor, K.
Pharmacol. Res. Commun. 1984, 16, 243.
mp>250 ꢁC; 1H NMR (400 MHz, DMSO-d6): d 0.72 (s,
1H, CONH, D2O exchangeable), 7.6 (s, 1H, imidazole
CH), 7.51 (s, 1H, CONH, D2O exchangeable), 4.75 (s, 1H,
OH, D2O exchangeable), 3.65 (m, 3H, CH and CH2); 13C
NMR (100 MHz, DMSO-d6): d 168.9, 162.2, 137.9, 125.4,
112.5, 59.4, 56.5; HRMS (FAB), calcd. for C7H8N4O3,
196.0596 (M+); obsd. m/z 197.0587 (MH+); Anal. Calcd
for C7H8N4O3: C, 42.86; H, 4.11; N, 28.56. Found: C,
42.97; H, 4.16; N, 28.45.
35. Silipo, C.; Hansch, C. J. Med. Chem. 1976, 19, 62.
36. Wang, L.; Hosmane, R. S. Bioorg. Med. Chem. Lett. 2001,
11, 2893.
37. Isshiki, K.; Takahashi, Y.; Iinuma, H.; Naganawa, H.;
Umezawa, Y., et al. J. Antibiot. (Tokyo) 1987, 40, 1461.
38. Lewis, A. S.; Glantz, M. D. J. Biol. Chem. 1974, 249, 3862.
39. Liaw, S.-H.; Chang, Y.-J.; Lai, C.-T.; Chang, H.-C.;
Chang, G.-G. J. Biol. Chem. 2004, 279, 35479.
40. Preparation and physicochemical properties of the target
compounds are as follows:
41. Bhan, A.; Hosmane, R. S. J. Heterocycl. Chem. 1993, 30,
1453.
42. Robers, D. A.; Ward, W. H. J. Enzyme Inhibitors. In
Medicinal Chemistry: Principles and Practice, 2nd ed.; The
Royal Society of Chemistry: Cambridge, UK, 2002; pp
70–71.
43. Agarwal, R. P.; Cha, S.; Crabtree, G. W.; Parks, R. E., Jr.
Coformycin and deoxycoformycin: tight-binding inhibi-
tors of adenosine deaminase. In Chemistry and Biology of
Nucleosides and Nucleotides; Academic Press: New York,
1978; pp 159–197.
44. Hosmane, R. S.; Hong, M. Biochem. Biophys. Res.
Commun. 1997, 236, 88.
45. Wang, Z.; Quiocho, F. A. Biochemistry 1998, 37, 8314.
3-Benzyl-4,5,7,8-tetrahydro-6-hydroxymethyl-6H-imidazo-
[4,5-e][1,4]diazepine-5,8-dione (1). Method A: 3-Benzyl-
4,5,7,8-tetrahydro-6-methoxy-6-methoxycarbonyl-6H-imi-
dazo [4,5-e][1,4]diazepine-5,8-dione (6).20 (50 mg,
0.159 mmol) was placed in a flame-dried three-necked
flask. Anhydrous THF (10 mL) and LiBH4 (10 mg,
0.45 mmol) were added. The reaction mixture was stirred
at 0 ꢁC for 10 min, and then 1 M super hydride (0.05 mL)
was added. The ice bath was removed and the reaction