Metal mediated protease inhibition: Design and synthesis of inhibitors of the human cytomegalovirus (hCMV) protease

Article abstract of DOI:10.1016/S0960-894X(00)00462-5

A versatile synthetic route to a novel series of bis-imidazolemethanes designed to inhibit the hCMV protease has been developed and a series of potential metal binding inhibitors has been identified. In selectivity assays, the compounds were highly specific for CMV protease and showed no inhibition (IC₅₀ >100 μM) of other prototypical serine proteases such as trypsin, elastase, and chymotrypsin. Although the presence of free zinc ions was found to be an absolute requirement for the in vitro biological activity of this class of inhibitor, the potency of the inhibitors could not be improved beyond the micromolar level. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00462-5

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2279±2282
Metal Mediated Protease Inhibition: Design and Synthesis of
Inhibitors of the Human Cytomegalovirus (hCMV) Protease
Dashyant Dhanak,a,* George Burton,a Lisa T. Christmann,a Michael G. Darcy,a
Kyle C. Elrod,b Arun Kaura,a Richard M. Keenan,a John O. Link,b
Catherine E. Peisho€ a and Dinubhai H. Shaha
aSmithKline Beecham Pharmaceuticals, 1250 South Collegeville Road, PO Box 5089, Collegeville, PA 19426-0989, USA
bAxys Pharmaceuticals, 180 Kimball Way, S. San Francisco, CA 94080, USA
Received 8 June 2000; accepted 1 August 2000
AbstractÐA versatile synthetic route to a novel series of bis-imidazolemethanes designed to inhibit the hCMV protease has been
developed and a series of potential metal binding inhibitors has been identi®ed. In selectivity assays, the compounds were highly
speci®c for CMV protease and showed no inhibition (IC50>100 mM) of other prototypical serine proteases such as trypsin, elastase,
and chymotrypsin. Although the presence of free zinc ions was found to be an absolute requirement for the in vitro biological
activity of this class of inhibitor, the potency of the inhibitors could not be improved beyond the micromolar level. # 2000 Elsevier
Science Ltd. All rights reserved.
The serine protease of the human cytomegalovirus
(hCMV) has become established as an attractive thera-
peutic target and has been the subject of intense research
aimed at discovering potent inhibitors that are e€ective
antiviral agents in vivo.1 We have recently reported2 the
inhibition of the protease by a series of benzothiopyr-
anone derivatives and other workers3 have successfully
utilized alternative templates to achieve potent inhibition
of the enzyme. A number of these compounds have also
been shown to be active in models of CMV infection.4
mediated inhibition concept to the design of novel
hCMV protease inhibitors. We report herein the suc-
cessful realization of this goal and disclose a novel,
metal mediated mode of inhibition of hCMV protease.
Examination of the hCMV protease crystal structure
suggested that the bis-benzimidazolemethane based
metal binding templates (BABIM) could not be readily
accommodated within the CMV protease active site.
However, we reasoned that excision of the benzo portion
of the benzimidazole nucleus would be more successful
in generating a geometrically viable ternary complex of
enzyme, zinc and inhibitor. In a further design element,
we wished to incorporate the ability to display a variety
of groups capable of interacting with the protein to
allow both for variations in the binding mode of the
template and also to allow access to binding pockets on
the enzyme. These design requirements could most
easily be incorporated into the bis-imidazolemethane
(BIM) template 1.
The X-ray crystal structure of the protease5 has revealed
a number of unique features noteworthy amongst which
is the presence of an active site catalytic triad composed
of a serine and two histidine residues. Recently, Katz et
al. have reported a conceptually novel approach to the
inhibition of serine proteases such as trypsin in which a
zinc ion mediated the formation of a ternary inhibitory
complex of a low molecular weight metal chelate, metal
ion and enzyme.6 X-ray crystallographic analysis of the
complex revealed that the zinc was tetrahedrally coor-
dinated at the enzyme active site and engaged the cata-
lytically important serine and histidine residues of the
protease.7 In view of the presence of an additional zinc
ligating histidine residue in the CMV protease active
site, we were interested in the application of the metal
We envisaged that alkylation at sulfur of the thioimida-
zolone followed by amide formation at the carboxylic
*Corresponding author. E-mail: dash_dhanak-1@sbphrd.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00462-5

2280
D. Dhanak et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2279±2282
Scheme 1. Reagents: (A) 1. BuOH, Pr2EtN; (B) 1. TFA, CH2Cl2; 2. Na/Hg, NH4SCN, EtOH/H2O, 0±5 ꢀC, pH 1.5±3.5; (C) 1. R1X, DMF; 2.
CH2N2; 3. HCl, Á; (D) R2NH2, EDC, HOBt, Et3N, DMF.
t
i
acid could be carried out in a tandem manner and
would enable the rapid synthesis of a diverse set of
potential metal binding inhibitors of hCMV protease. In
addition, we chose to pursue the 4,40-bisimidazole
methane system in preference to the isomeric 2,20-linked
system in order to minimize the potential for oxidation
at the linking methylene moiety.
Table 1. A selection of bis(imidazole)methanes 4 prepared according
to Scheme 1
Compound
R1
R2
7
8
9
10
11
12
n-Octyl
n-Octyl
CH2CO(4-Ph)Ph
CH2-5-Pyrimidinyl
CH2CN
l-Asp(OBn)2
l-Phe-OBn
n-Octyl
l-Val(OBn)
3,4-(MeO)2CH2Ph
3,4-(MeO)2CH2Ph
Chemistry
Our approach to 1, based on that reported by Dreyton
and Frew,8 involved regioselective N-alkylation9 of histi-
dine methyl ester followed by thioimidazolone formation
from the resulting pendant a-amino acid (Scheme 1).
CH2-5-Pyrimidinyl
Careful monitoring of the reaction pH during the
sodium amalgam mediated reduction of the ester 2 was
critical in obtaining good conversion to the desired pro-
duct. However, the need to isolate the highly water solu-
ble 1 from a complex mixture of inorganic byproducts
made the overall process inecient and the isolated
yields varied considerably. Fortunately, sulfur alkyla-
tion proceeded smoothly with a variety of electrophiles
the aldehyde followed by thioimidazolone formation
proceeded uneventfully to furnish the modi®ed template
6 in good overall yield. Tandem functionalization of 6
using the protocol established above gave access to a
second series of potential zinc chelating CMV protease
inhibitors (Scheme 2 and Table 2).
1
and, on the basis of chemical shift di€erences in the H
Results and Discussion
NMR spectra, the regiochemistry of the thioimidazo-
lone alkylation was readily determined to be at sulfur
and not at either of the N1 or N3 alternatives.10 The
resulting crude (thioalkyl)bis-imidazole acids 3 could be
coupled with various amines using EDCI as the con-
densing reagent (Table 1) to give the desired amides 4.11
The hCMV protease inhibitory activity of the bis-imi-
dazolemethanes was determined13 by following the
increase in ¯uorescence resulting from the enzymatic
processing of a quenched ¯uorescent peptide substrate.
The e€ect of zinc ion on the potency of the compounds
was determined by carrying out the assay with and
without 1 mM EDTA. As shown in Table 3 and con-
sistent with the proposed mechanism of action of these
compounds, the hCMV protease inhibitory activity was
strongly dependent on the presence of free metal ion. In
the absence of free zinc, the bis-imidazoles had little or
no measurable inhibitory activity towards the protease
but upon the addition of 10 mM zinc ion, signi®cant
inhibition of the proteolytic activity was observed for
some of the compounds, consistent with the formation
of a ternary complex of enzyme, zinc, and inhibitor.
However, despite the introduction of an extensive range
of substituents onto the BIM template, the inhibitory
potency of the compounds could not be improved sub-
stantially beyond the micromolar level (data not shown).
Having demonstrated the viability of the tandem alky-
lation±acylation strategy, we next focused on improving
the synthetic route with a view to altering the capricious
amalgam reduction and thereby allowing for a more
facile isolation of the desired product. We considered that
the incorporation of a more lipophilic benzyl linker to
display the carboxylic acid would allow for both a
selective reduction of the amino acid ester in 5 and also
considerably improve the overall organic solvent solubi-
lity of the template.
These expectations were realized in practice and treat-
ment of 5 with NaBH4/LiCl in methanol:THF12 and
subsequent oxidation of the resulting primary alcohol to

D. Dhanak et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2279±2282
2281
Scheme 2. Reagents: (A) tBuOH, iPr2EtN, Á; (B) 1. NaBH4, LiCl, 2:1 MeOH:THF; 2. (COCl)2, DMSO, CH2Cl2, 78 ꢀC; (C) 1. 4 M HCl/dioxane,
10 ꢀC; 2. NaSCN, H2O, 100 ꢀC; (D) 1. NaOH, MeOH: H2O, Á; 2. R1X, DMF; 3. R2NH2, EDC, HOBt, Et3N, DMF.
Table 2. Bis-(imidazole)methanes prepared according to Scheme 2
bifunctionalization of the template has been demon-
strated and the chemistry used to prepare a range of
substituted derivatives. The presence of free zinc ions
was found to be an absolute requirement for the in vitro
CMV protease inhibitory activity of this class of inhi-
bitor. In selectivity assays, the compounds were gen-
erally highly speci®c for CMV protease and showed no
inhibition (IC50 >100 mM) of other prototypical serine
proteases such as trypsin, elastase and chymotrypsin.15
Compound
R2
13
14
15
16
OMe
References and Notes
NH CH2cC3H5
NH CH2-4-MeOPh
NH CH2CH2OH
1. Holwerda, B. G. Antiviral Res. 1997, 35, 1.
2. Dhanak, D.; Keenan, R. M.; Burton, G.; Kaura, A.;
Darcy, M. G.; Shah, D. H.; Ridgers, L. H.; Breen, A.; Lavery,
P.; Tew, D. G.; West, A. Bioorg. Med. Chem. Lett. 1998, 8,
3677.
Table 3. hCMV protease inhibition of bis-imidazolemethanes
3. (a) Jarvest, R. L.; Connor, S. C.; Gorniak, J. G.; Jennings,
L. J.; Sera®nowska, H. T.; West, A. Bioorg. Med. Chem. Lett.
1997, 7, 1733. (b) Bonneau, P. R.; Grand-Maitre, C.; Green-
wood, D. J.; Lagace, L.; LaPlante, S. R.; Massariol, M.-J.;
Ogilvie, W. W.; O'Meara, J. A.; Kawai, S. H. Biochemistry
1997, 36, 12644. (c) Ogilvie, W.; Bailey, M.; Poupart, M.-A.;
Abraham, A.; Bhavsar, A.; Bonneau, P.; Bordeleau, J.; Bous-
quet, Y.; Chabot, C.; Duceppe, J.-S.; Fazal, G.; Goulet, S.;
Grand-Maitre, C.; Guse, I.; Halmos, T.; Lavallee, P.; Leach,
M.; Malenfant, E.; O'Meara, J.; Plante, R.; Plou€e, C.; Poir-
ier, M.; Soucy, F.; Yoakim, C.; Deziel, R. J. Med. Chem. 1997,
40, 4113. (d) Pinto, I. L.; West, A.; Debouck, C. M.; DiLella,
A. G.; Gorniak, J. G.; O'Donnell, K. C.; O'Shannessy, D. J.;
Patel, A.; Jarvest, R. L. Bioorg. Med. Chem. Lett. 1996, 6,
2467. (e) Yoakim, C.; Ogilvie, W. W.; Cameron, D. R.; Cha-
bot, C.; Guse, I.; Hache, B.; Naud, J.; O'Meara, J. A.; Plante,
R.; Deziel, R. J. Med. Chem. 1998, 41, 2882. (f) Borthwick, A.
D.; Weingarten, G.; Haley, T. M.; Tomaszewski, M.; Wang,
W.; Hu, Z.; Bedard, J.; Jin, H.; Yuen, L.; Mansour, T. S.
Bioorg. Med. Chem. Lett. 1998, 8, 365. (g) Pinto I. L.; Jarvest
R. L.; Clarke B.; Dabrowski C. E.; Fenwick A.; Gorzyca, M.
M.; Jennings J. L.; Lavery P.; Sternberg E. J.; Tew D. G.;
West A. Bioorg. Med. Chem. Lett. 1999, 9, 449.
Compound
hCMV protease IC50 (mM)
(+) Zinc ) Zinc
(
7
8
13
14
5
41
7
15
125
>150
>150
15
Presumably, the inhibitor template is not involved in
making suciently extensive contacts with the protein
and consequently, the substituents introduced also do
not interact optimally with binding site(s) on the
enzyme. This may be a consequence of the high anity
of the protease active site for zinc14 such that additional
contacts of the template with the bound zinc are weak
and disfavour the formation of a tight, well de®ned
ternary inhibitory complex.
4. (a) Abood, N. A.; Schretzman L. A.; Flynn D. L.; House-
man K. A.; Wittwer A. J.; Dilworth V. M.; Hippenmeyer P. J.;
Holwedra B. C. Bioorg. Med. Chem. Lett. 1997, 7, 2105. (b)
Jarvest R. L.; Pinto I. L.; Ashman S. M.; Dabrowski C. E.;
Fernandez A. V.; Jennings J. L.; Lavery P.; Tew D. G. Bioorg.
Med. Chem. Lett. 1999, 9, 443.
Summary
A versatile synthetic route to a novel series of bis-imi-
dazolemethanes designed to inhibit the hCMV protease
has been developed. The potential for rapid, tandem

2282
D. Dhanak et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2279±2282
5. (a) Qui, X.; Culp, J. S.; DiLella, A. G.; Hellmig, B.; Hoog, S.
S.; Janson, C. A.; Smith, W. W.; Abdel-Meguid, S. S. Nature
1996, 383, 275. (b) Chen, P.; Tsuge, H.; Almassy, R. J.; Gribs-
kov, C. L.; Katoh, S.; Vanderpool, D. L.; Margosiak, S. A.;
Pinko, C.; Matthews, D. A.; Kan, C.-C. Cell 1996, 86, 835. (c)
Tong, L.; Qian, C.; Massariol, M.-J.; Bonneau, P. R.; Cording-
ley, M. G.; Lagace, L. Nature 1996, 383, 272. (d) Shieh, H.-S.;
Kurumball, R. G.; Stevens, A. M.; Stegeman, R. A.; Sturman,
E. J.; Pak, J. Y.; Wittwer, A. J.; Palmier, M. O.; Wiegand, R.
C.; Holwerda, B. C.; Stallings, W. C. Nature 1996, 383, 279.
6. Clark, J. M.; Jenkins, T. E.; LaHonda, B. A.; Katz, R. M.
US Patent 5,693,515, 1997; Chem. Abstr. 1997, 772267.
7. Katz, B. A.; Clark, J. M.; Finer-Moore, J. S.; Jenkins, T.
E.; Johnson, C. R.; Ross, M. J.; Luong, C.; Moore, W. R.;
Stroud, R. M. Nature 1998, 391, 608.
12. Hamada, Y.; Kohfuku, K.; Shioiri, T. Tetrahedron Lett.
1984, 25, 5303.
13. Assay conditions: CMV protease (50 nM) was combined
with varying concentrations of inhibitor in 50 mM HEPES
(pH 7.5), 4.5 mM oxalate, 150 mM NaCl, 0.01% PEG-3350,
30% sucrose and incubated 30 min prior to the addition
of substrate. Zinc chloride and EDTA reactions were sup-
plemented with 10 mM zinc chloride or 1 mM EDTA,
respectively. Following incubation, substrate (10 mM dabsyl-
RGVV NASSRLAKK-dansyl) was added and the rate
of substrate hydrolysis was measured using a ¯uorescence
plate reader (ex 355, em 510). The resulting progress curves
were ®t to the Morrison equation to obtain Ki apparent
values.
14. Sideris, S.; Ruppert, J.; Qiu, X.; O'Donnell, K.; Elrod, K.
C. FASEB 1999, 13, A1536.
8. Drey, C. N. C.; Fruton, J. S. Biochemistry 1965, 4, 1.
9. Chivikas, C. J.; Hodges, J. C. J. Org. Chem. 1987, 52, 3591.
10. The chemical shift for the protons of the methyl group in
2-(methylthio)imidazole is reported16 to be 2.52 ppm, whereas
in N-methyl-2-mercaptoimidazole, the methyl protons reso-
15. The compounds were evaluated for their inhibitory activ-
ity under conditions of optimal enzyme activity and with the
appropriate substrates. None of the compounds showed sig-
ni®cant activity below 100 mM concentration in the presence of
an excess of free zinc ion.
1
nate at 3.6 ppm.17 In the H NMR of 14, the corresponding
methylene protons resonate at 2.95 ppm consistent with
attachment to sulfur.
11. All compounds gave spectroscopic and analytical data
consistent with their assigned structure.
16. Imbach, J. L.; Gosselin, G.; Panciza, R. P.; Townsend, L.
B. J. Heterocycl. Chem. 1979, 16, 1185.
17. Pouchert, C. J.; Behnke, J. In The Aldrich Library of 13C
1
and H FT NMR Spectra; 1st ed.; 1993; Vol. III, p. 81A.