Structure-based design, synthesis, and structure-activity relationship studies of novel non-nucleoside adenosine deaminase inhibitors

Article abstract of DOI:10.1021/jm0306374

We disclose herein optimization efforts around the novel, highly potent non-nucleoside adenosine deaminase (ADA) inhibitor, 1-[(R)-1-hydroxy-4-(6-(3-(1- methylbenzimidazol-2-yl)propionylamino)indol-1-yl)-2-butyl]imidazole-4- carboxamide 1 (K_i =

Full text of DOI:10.1021/jm0306374

3730
J . Med. Chem. 2004, 47, 3730-3743
Str u ctu r e-Ba sed Design , Syn th esis, a n d Str u ctu r e-Activity Rela tion sh ip
Stu d ies of Novel Non -n u cleosid e Ad en osin e Dea m in a se In h ibitor s
Tadashi Terasaka,*,† Takayoshi Kinoshita,‡ Masako Kuno,§ Nobuo Seki,§ Kohichiro Tanaka,| and
Isao Nakanishi*,‡,#
Medicinal Chemistry Research Laboratories, Basic Research Laboratories, Medicinal Biology Research Laboratories, and
Biopharmaceutical and Pharmacokinetic Research Laboratories, Fujisawa Pharmaceutical Co., Ltd.,
2-1-6, Kashima, Yodogawa-ku, Osaka 532-8514, J apan
Received December 22, 2003
We disclose herein optimization efforts around the novel, highly potent non-nucleoside adenosine
deaminase (ADA) inhibitor, 1-[(R)-1-hydroxy-4-(6-(3-(1-methylbenzimidazol-2-yl)propionyl-
amino)indol-1-yl)-2-butyl]imidazole-4-carboxamide 1 (Ki ) 7.7 nM), which we recently reported.
Structure-based drug design (SBDD) utilizing the crystal structure of the 1/ADA complex was
performed in order to obtain structure-activity relationships (SAR) for this type of compound
rationally and effectively. To utilize the newly formed hydrophobic space (F2), replacement of
the benzimidazole ring of 1 with a n-propyl chain (4b) or a simple phenyl ring (4c) was tolerated
in terms of binding activity, and the length of the methylene-spacer was shown to be optimal
at two or three. Replacement of an amide with an ether as a linker was also well tolerated in
terms of binding activity and moreover improved the oral absorption (6a and 6b). Finally,
transformation of indol-1-yl to indol-3-yl resulted in discovery of a novel highly potent and
orally bioavailable ADA inhibitor, 1-[(R)-4-(5-(3-(4-chlorophenyl)propoxy)-1-methylindol-3-yl)-
1-hydroxy-2-butyl]imidazole-4-carboxamide 8c.
In tr od u ction
In recent years, adenosine has come to be considered
as an important factor in the attenuation of inflamma-
tion.1,2 It has been reported that the concentration of
adenosine is increased in inflammatory lesions3 and that
accumulation of adenosine might terminate inflamma-
tion. Several reports have indicated that common an-
tirheumatoid drugs, such as methotrexate, aspirin, and
F igu r e 1. Chemical structures of adenosine and known ADA
inhibitors.
sulfasalazine, might exert antiinflammatory action by
elevating the extracellular adenosine level.4 Moreover,
the pivotal role of adenosine in the physiological reduc-
past decade, ADA, which was considered to be cytosolic,
tion of inflammation was confirmed by using mice
has also been found on the cell surface via binding to
deficient in the adenosine A2a receptor. In A2a-deficient
CD26.7 Since CD26 is strongly up-regulated following
mice, subthreshold doses of an inflammatory stimulus
T-cell activation, it has been demonstrated that ecto-
that causes minimal inflammation in wild-type mice
ADA could perpetuate chronic inflammation by degrad-
were sufficient to induce prolonged severe inflamma-
ing extracellular adenosine or 2′-deoxyadenosine which
tion.2
are toxic for lymphocytes.8 Therefore, an ADA inhibitor
Adenosine deaminase (ADA) (EC 3.5.4.4) is a key
has great potential as an antiinflammatory drug that
enzyme in purine metabolism and catalyzes the ir-
works at inflamed sites selectively.
reversible deamination of both adenosine and 2′-deoxy-
Pentostatin (Figure 1) is well-known as a potent
adenosine to inosine and 2′-deoxyinosine, respectively.5
transition-state ADA inhibitor.9 Pentostatin is the only
ADA is ubiquitous in almost all human tissues, and
ADA inhibitor in clinical use and is currently approved
genetic ADA deficiency results in severe combined
for the treatment of adult patients with hairy cell
immunodeficiency disease by impairment of the dif-
leukemia.10 However, since pentostatin is a nucleoside
ferentiation and maturation of lymphoid cells.6 In the
analogue, it has several toxicities11 and moreover is not
orally bioavailable.12,13 Therefore, pentostatin cannot be
used as an antiinflammatory drug. A number of ground-
* To whom correspondence should be addressed. For T.T.: phone,
+81-6-6390-1286; fax, +81-6-6304-5435; e-mail, tadashi_terasaka@
state ADA inhibitors have been also reported to date
besides pentostatin, for example, (+)-EHNA (Figure 1)14
and related derivatives.15,16 However, none are used
clinically because all are nucleoside analogues or EHNA
derivatives and have poor pharmacokinetics, such as
rapid metabolism.17-19 We speculated that the unfavor-
able properties of the known inhibitors could be im-
po.fujisawa.co.jp. For I.N.: phone, +81-75-753-9273; fax, +81-75-753-
9273; e-mail, isayan@pharm.kyoto-u.ac.jp.
† Medicinal Chemistry Research Laboratories.
‡ Basic Research Laboratories.
§ Medicinal Biology Research Laboratories.
| Biopharmaceutical and Pharmacokinetic Research Laboratories.
# Current address: Department of Theoretical Drug Design, Gradu-
ate School of Pharmaceutical Sciences, Kyoto University, Sakyo-ku,
Kyoto 606-8501, J apan.
10.1021/jm0306374 CCC: $27.50 © 2004 American Chemical Society
Published on Web 06/17/2004

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3731
Ta ble 1. Inhibition Constants for Amide-Linked Derivatives of
1 with ADA
compd
R
n
Ki (nM)a
1
-
H
-
7.7b
1300
24
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
5
1
2
2
3
4
5
2
2
3
2
-
n-C3H7
Ph
Ph
Ph
Ph
4-Me-Ph
4-MeO-Ph
4-Me-Ph
3-Py
-
30
F igu r e 2. Chemical structures of hybrid compound 1 and the
structurally distinct lead compounds 2 and 3.
16
11c
91
38
57
34
proved by changing the nucleoside framework to a non-
nucleoside framework and initiated a search for non-
nucleoside ADA inhibitors.19 Despite the difficulty of
converting a nucleoside to non-nucleoside, a recent
report from our laboratories has described the discovery
of a novel, highly potent non-nucleoside ADA inhibitor
1 (Ki ) 7.7 nM to human ADA) (Figure 2).20 This
compound was discovered by intentional hybridization
of two structurally distinct lead compounds [2 (Ki ) 5.9
µM) and 3 (Ki ) 1.2 µM)] by only two structure-based
drug design (SBDD) iterations (Figure 2), and as a
result, no detailed information on structure-activity
relationships (SAR) for the hybrid compound was avail-
able. Therefore, we next attempted to optimize com-
pound 1 based on the crystal structure of the 1/ADA
complex in order to obtain SAR and to improve activity
and pharmacokinetic properties. In this paper we
disclose these optimization efforts and the discovery of
several analogues with higher inhibitory potency for
ADA.
17
7.5c
a
Ki values were measured with human ADA. Assays were
performed in duplicate. Reference 20. c Tight-binding inhibitory
mode.
b
Ta ble 2. Simulated Interaction Energies for
Methylene-Inserted Analogues of 4c
n ) 2
n ) 3
n ) 4
n ) 5
n ) 6
E (kcal/mol)a -115.2 -115.1 -115.7 -120.9 -110.1
a
Calculated interaction energies.
Resu lts a n d Discu ssion
4a resulted in a substantial deterioration (170-fold) in
the inhibitory potency due to elimination of the van der
Waals and/or hydrophobic interactions with the F2
space (Table 1). However, replacement of the benzimi-
dazole ring of 1 with a n-propyl chain (4b) or a simple
phenyl ring (4c) was a tolerated modification in terms
of binding activity. These modifications resulted in
about a 3-fold loss in ADA inhibitory potency (Ki ) 24
and 30 nM, respectively) (Table 1), which strongly
suggests that extension of the molecule from the indolyl
ring of 4a to the F2 space is important for high potency
due to energy gains from more favorable van der Waals
or hydrophobic interactions. Although compounds 4b
and 4c have similar activity, compound 4c was chosen
for further SAR studies due to synthetic expediency.
Inspection of the docking simulation of 4c with ADA
indicated that the distal hydrophobic space (F2) of the
active site, where the phenyl ring resides, is not fully
occupied by the inhibitor molecule. Hence, in addition
to 4c, we designed compounds possessing methylene
chains of varying lengths between the indole moiety and
the phenyl ring, and docking simulation of the designed
compounds with ADA based on the 1/ADA complex was
performed to prioritize the synthesis. Table 2 shows the
calculated interaction energies of the designed com-
pounds. Since calculated interaction energies consider
only nonbonded energy terms, but not desolvation
energies and other entropic factors, they do not ac-
curately reflect affinities of ligands. Moreover, since the
Compound 1 has a novel structure and highly potent
ADA inhibitory activity, so we opted to perform further
studies based around this structure. The crystal struc-
ture of the 1/ADA complex revealed that ADA performed
an induced fit to bind 1. That is, compound 1 utilized
newly formed binding pockets in the enzyme that are
spatially distinct from the pockets occupied by substrate-
like inhibitors.20 A narrow planar hydrophobic space
(F1) is occupied by the 6-acylaminoindole ring and the
comparatively large hydrophobic space (F2) is occupied
by the benzimidazole ring (Figure 3a). This information
proved very useful for the subsequent design of novel
ADA inhibitors because no comparable data for a semi-
tight-binding inhibitor such as EHNA is available. As
a result, we decided to investigate the SAR for the newly
formed binding pockets (F1 and F2). According to the
crystal structure of the 1/ADA complex, the benz-
imidazole ring binds tightly to the F2 pocket so there is
hardly any space to introduce substituents from the 4-
to 7-positions except for the space to the outside of the
active site(solvent region) (Figure 3a). Hence, we ini-
tially decided to modify the benzimidazole part of 1 in
order to obtain SAR for this part of the structure and
discover structures more amenable to rapid synthetic
modification, while retaining highly potent ADA inhibi-
tory activity.
1. Am id e-Lin k ed Com p ou n d s. Removal of the
benzimidazole moiety from 1 to give the acetyl analogue

3732 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
F igu r e 3. Binding mode of 1 and amide-linked compounds at the ADA active site. Accessible surfaces of the carbon atoms at the
active sites are drawn by mesh. Because the active site is located on the inside of protein, meshed regions indicate regions where
inhibitor molecules can exist without contact to the proten. The upper portion of the figures is the active site entrance (solvent
region). Residues of ADA around the active site are drawn by wire model. Several residues are omitted for clarity. Inhibitor
molecules are rendered as ball-and-stick models colored by atom type: yellow for C, blue for N, red for O, and cyan for H. (a)
Binding orientations of 1 (PDB code: 1NDZ). (b and c) Binding orientations of 4c (PDB code: 1UML) and 4e (PDB code: 1QXL)
(ball-and-stick), respectively, with 1 (stick) superimposed onto the active site surface of the 1/ADA complex. (d) Simulated binding
mole of 4d (ball-and-stick) and 1 (stick) superimposed onto the active site surface of the 4c/ADA complex. The displacements
between 1 and inhibitors are shown with red arrows in c and d.
F2 pocket is a comparatively large hydrophobic space,
it was difficult to predict the precise position of the
phenyl ring in F2 from molecular modeling. However,
we adopted interaction energy to evaluate unfavorable
intermolecular interactions such as large interatomic
clashes or electrostatic repulsions. From this point of
view, methylene chains from three to five carbons in
length could be readily accommodated by the binding
site, because their E values are similar or lower than
that of the two-carbon chain compound (4c). However,
the compound with six methylenes is unsuitable for
binding to the F2 space, because the E value is higher
(∆E > 5 kcal/mol) than the others. Therefore, com-
pounds 4d -f were synthesized and evaluated (Table 1).
Consequently, compounds 4d and 4e exhibited more
potent inhibitory activities compared to 4c; however, the
five methylene chain compound (4f) led to a 3-fold
decrease in enzyme inhibitory potency. Before the
docking simulations were performed, we had considered
from the crystal structure of the 1/ADA complex that
four (4e) and five (4f) methylene chains were too long
to fit the phenyl ring to the F2 space in a similar manner
as the benzimidazole ring of 1, but molecular modeling
predicted that they could fit to the binding site as
described above although these phenyl rings located
almost outside of the F2 pocket. To clarify the binding
mode of 4c and 4e and obtain information for the next
design stage, we carried out determination of the crystal
structures of complexes with ADA. The crystal structure
of the 4c/ADA complex revealed that more space is
available for binding with ADA in the distal F2 space
and the other portions of 4c showed the same binding

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3733
mode as 1 (Figure 3b). On the other hand, the crystal
structure of the 4e/ADA complex revealed that the
indolyl ring had shifted to close proximity to the wall
of the active site compared to 1, and moreover the
phenyl ring shifted to the upper position of the F2
pocket, connected to the outside of the active site
(solvent region), compared to the benzimidazole ring of
1 (Figure 3c). Accordingly, compound 4e could fit to the
active site without clashes with the enzyme wall and
display high inhibitory potency. Compound 4e, however,
shows a different inhibition mode compared to the
others (i.e. tight-binding inhibition like pentostatin).
When pentostatin was analyzed by testing different
concentrations with different enzyme concentration, the
plot of enzyme velocity against enzyme concentration
(Ackermann-Potter plot) was characterized by asymp-
totic concave curves, which revealed the tight-binding
or stoichiometric nature of the inhibitor.21 Compound
4e produces the same pattern as pentostatin in an
Ackermann-Potter plot (data not shown). By inspecting
in detail the crystal structure of the 4e/ADA complex
compared to that of 1 or 4c/ADA, several differences in
the binding mode of 4e were detected; however, we are
unsure as to the exact reason for such an observation.
The inhibitory mode of pentostatin is suspected to be
the reason for chronic inhibition and severe immuno-
deficiency. Therefore, we did not perform further studies
with 4e but rather with the competitive inhibitors (4c
and 4d ) to obtain SAR for binding to the distal hydro-
phobic pocket.
It was considered from the crystal structure of the
4c/ADA complex (Figure 3b) that more space exists to
introduce substituents onto the phenyl ring in F2. On
the other hand, the docking simulation of the 4d /ADA
complex (Figure 3d) shows that the phenyl ring of 4d
shifted to the upper position of the F2 pocket compared
to the benzimidazole ring of 1, hence does not utilize
the F2 pocket substantially. Moreover, the distal phenyl
ring is adjacent to the wall of the active site, in
particular, the para position of the phenyl ring, and
there is no space to introduce a substituent. Therefore,
it was considered that a compound containing a sub-
stituent in the para position of the phenyl ring would
not bind to the active site. On the other hand, crystal
structure analyses of complexes of ADA with our inhibi-
tors (1, 4c, and 4e) revealed that ADA performed
induced fit to bind our inhibitors and that the shape of
the F2 pocket is changeable (Figure 3). Furthermore,
the para-substituted phenyl analogue of 4d could fit to
the active site of ADA by shifting the indolyl ring and
the phenyl ring, as shown by the crystal structure of
the 4e/ADA complex (Figure 3c). Hence, we considered
that it was important for subsequent design to inves-
tigate the limitations of the F2 pocket and obtain the
SAR. From the reasoning described above, we designed
and synthesized the substituted phenyl analogues (4g,
h , i) of compounds 4c and 4d . This resulted in similar
or a small deterioration of inhibitory activity (Table 1).
Occupying the nonutilized spaces on binding of 4c (4g,
h ) was not effective to enhance the inhibitory potency.
That is, a methyl group is tolerated, but a methoxy
group is a little too large, although compound 4h can
still bind to the active site. On the other hand, com-
pound 4i was tolerated in terms of binding activity (2-
fold loss relative to 4d ), despite lack of space to
introduce substituents in the para position of the phenyl
ring of 4d . This observation suggests that either the
active site changes to bind 4i or the indolyl ring and/or
the substituted phenyl ring of 4i shift to appropriate
positions in the active site to fit to ADA.
So far we had replaced the benzimidazole of com-
pound 1 with hydrophobic phenyl or alkyl chains,
because it was considered that the F2 space was highly
hydrophobic. However, the benzimidazole is a basic
heterocyclic ring, so we also attempted to replace the
phenyl ring of 4c with a basic heterocyclic ring, pyridine
(4j). This replacement resulted in a 2-fold increase in
inhibitory activity relative to the phenyl analogue (4c)
(Table 1). Although F2 is a hydrophobic region, the
result for 4j and the potency of 1 suggests tolerance and
a possible preference for occupation of the F2 space by
basic heterocyclic rings and provides useful information
for further optimization.
From the study described above, we obtained com-
parative SAR information for utilizing the F2 space, for
example, the possibility of simple aromatic rings instead
of benzimidazole and the permissible length of spacer
compatible with this site. Taking advantage of this
information, we next investigated the possibility of
transformation of the amide-linker that connected with
the indolyl ring.
2. Tr a n sfor m a tion of th e Am id e-Lin k er Con -
n ectin g th e In d ole to th e F 2 Site Bin d in g Moiety.
2-1. Ur ea -Lin k ed Com p ou n d s. In consideration of the
narrow planar space available at F1 and conformational
restriction of the designed compound, an amide group
was introduced at the 6 position of the indolyl ring to
extend the molecule to the F2 region. It was considered
that this group also plays a role to form an intramo-
lecular hydrogen bond between the NH and the carbonyl
of the imidazolecarboxamide. Further, the crystal struc-
tures of the 1 and 4c/ADA complexes verified formation
of an intramolecular hydrogen bond (distances between
CdO and HN are 3.0 and 3.2 Å, respectively), but they
were not tight bonds. It is well-known that restricting
the flexibility of a molecule can enhance the potency of
an inhibitor.22 To enhance ADA inhibition potency, we
next studied conformational restriction of compound 4c.
In consideration of the formation of a tight intra-
molecular hydrogen bond with the carbonyl group of the
imidazole ring, we replaced the methylene neighboring
the carbonyl group of 4c with NH, which led to urea
type compound 5. This compound was found to be 4-fold
more potent than 4c in the ADA inhibitory assay with
a Ki of 7.5 nM. A crystal structure of the 5/ADA complex
verified that compound 5 bound to the active site of ADA
in a manner similar to 4c, but contrary to our expecta-
tion, the NHs of the urea moiety of 5 did not form
intramolecular hydrogen bonds, as shown in Figure 4.
That is, the distances of 3.3 and 3.5 Å between CdO
and the two HNs are a little too long for a strong
hydrogen bond (typical length ) 2.5-3.0 Å). In search
of a possible explanation for this observation, we care-
fully analyzed the nature of the interaction of 4c and 5
with the active site of ADA (Figure 4b). Both inhibitors
formed hydrogen bonds with ADA through the hydroxyl
and imidazolecarboxamide moieties to H17, D19, G184,
E217, and D296, directly or by bridging through struc-
tural waters. However, the carbamoyl group of the

3734 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
Ta ble 3. Inhibition Constants for Ether-Linked Indol-1-yl
Derivatives with ADA
compd
R
Ki (nM)a
6a
6b
6c
6d
6e
7a
7b
Ph
17
12
55
240
13
1900
6200
PhCH2
n-C3H7
Me
4-Cl-Ph
Ph
n-C3H7
a
See footnote a of Table 1.
compounds 4c and 4d (Table 3), for which Ha/Hd ) 7/3,
to reduce the sum of hydrogen bonds and to discover
improved compounds in terms of oral absorption.
Ether-linked compounds 6a and 6b were found to be
more potent in terms of binding activity than the
corresponding amide-linked compounds 4c and 4d ,
respectively (Table 3), despite not forming an intra-
molecular hydrogen bond. These results support our
hypothesis. Comparison of the ether-linked compounds
6a and 6b with the corresponding amide-linked com-
pounds 4c and 4d indicates that the ether-linked
compounds can form a hydrogen bond with ADA through
the hydroxyl and imidazolecarboxamide moieties in a
manner similar to amide-linked compounds, but could
not form the intramolecular hydrogen bond. Moreover,
dipole repulsion of the oxygen of the ether with the
CdO of carbamoyl of imidazole ring is expected. This
is a potential disadvantage for ether-linked compounds;
however, desolvation energies for the ether-linked com-
pounds are predicted to be lower than for the amide-
linked compounds because the number of heteroatoms
is reduced. Repulsion between oxygen of the ether and
CdO of the carbamoyl is compensated by reduced
desolvation energies. In addition to potent activity,
compounds 6a and 6b showed oral absorption in rats
in preliminary studies of oral administration, in contrast
to no detectable serum levels with amide-linked com-
pounds 1 and 4d (data not shown). These results
motivated us to further SAR studies in this series.
Replacement of the phenyl ring of 6a with a simple
methylene chain, such as n-propyl (6c) was tolerated
in terms of binding activity (3-fold loss), but a methyl
(6d ) led to a 14-fold loss in inhibitory potency (Table
3). These results were similar to the SAR of the amide-
linked series. That is, sufficient occupancy of the F2
pocket with these molecules is crucial for highly potent
ADA inhibition.
F igu r e 4. (a) Binding orientations of 5 (ball and stick)/ADA
complex (PDB code: 1O5R). (b) Hydrogen bond networks
around the carboxamide moieties of 4c (left) and 5 (right).
Hydrogen bonds are drawn by dashed lines with the bonding
distance.
imidazole ring of 5 forms more tight hydrogen bonds
with ADA through three structural water molecules
than that of 4c. Furthermore, repulsion between the
methylene neighboring the carbonyl group of 4c and a
water molecule bridging the carbamoyl group of the
imidazole ring to E217 (d ) 3.0 Å) was observed, but
not for the NH neighboring the carbonyl group of 5 (d
) 3.4 Å). This result suggests that the intramolecular
hydrogen bond is a favorable, but not essential, factor
for highly potent ADA inhibition within this series.
Although compound 5 showed high ADA inhibitory
potency, it displayed a different inhibitory mode com-
pared to the other analogues, similar to 4e. The reason
for the change in inhibition mode is unclear at present.
We did not perform further studies with 5 for the same
reason as with 4e.
2-2. Eth er Lin k ed Com p ou n d s. From the study
described above, we considered that the amide-linker
that connected with the indolyl ring is changeable,
because the intramolecular hydrogen bond might be not
essential for potent ADA inhibition. Moreover, according
to the ‘rule of five’,23 it was predicted that because the
amide- and urea-linked compounds described above
have a high sum of hydrogen bonds acceptors (Ha) and
donors (Hd), they may have poor oral absorption and
permeability (Ha/Hd for compounds 1 and 5 are 10/4
and 9/5, respectively). Hence, we next designed ether-
linked compounds 6a and 6b based on the amide-linked
Since the ether-linked compounds are not able to form
an intramolecular hydrogen bond like the amide com-
pounds, we considered that it is not necessary to connect
at the 6-positon of the indolyl ring. Therefore, in
addition to 6-substituted ether-linked compounds, we
designed 5-substituted ether-linked types 7a and 7b to
determine the best link position. Molecular modeling
studies of 6a and 7a with ADA show that the E value
of 7a is lower (∆E ) ca. 1 kcal/mol) than that of 6a ;
however, the indolyl ring of 7a shifted compared to that

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3735
Ta ble 4. Simulated Interaction Energies for Indol-1-yl and
Indol-3-yl Derivatives
a
Differential of calculated interaction energies between indol-
1-yl and indol-3-yl.
Ta ble 5. Inhibition Constants for Ether-Linked Indol-3-yl
Derivatives with ADA
F igu r e 5. Simulated binding mode of 7a (ball and stick) and
6a (stick) superimposed onto the active site surface of the
1/ADA complex.
of 6a (Figure 5). It is predicted that the structure of 7a
is strained in order to fit to the active site without
clashes with the enzyme wall. Indeed the modeling
showed that the internal energy of 7a is higher (∆E )
ca. 4 kcal/mol) than that of 6a , so that compound 7a is
less stable than 6a . That is, 5-substituted ether-linked
compound (7a ) has a lower interaction energy but
higher internal (strain) energy than 6-substituted ether-
linked compound (6a ) in order to fit to the active site.
Compounds 7a and 7b were actually synthesized and
evaluated, and consequently these compounds showed
a drastic reduction of inhibitory activity (Table 3) as
predicted. Hence, further SAR studies on this type were
not performed.
Inspection of the binding mode of 6a indicates that
there is more space available for binding with ADA in
the F2 region (Figure 5). In the study of amide-linked
compounds, however, introducing methyl or methoxy
groups at the para position of the phenyl ring was not
effective to enhance the inhibitory potency. It was
considered that more planar substituents than methyl
or methoxy groups may be acceptable for binding to the
F2 space. Moreover, an unsubstituted phenyl ring is
relatively susceptible to oxidative metabolism, and it is
well-known that introduction of a halogen atom at the
para position of the phenyl ring can protect from
metabolism.24 Therefore, we designed the p-chloro-
phenyl compound of 6a . The resulting compound (6e)
showed high potency compared to 6a (Table 3). This
approach is effective to enhance the inhibitory potency
and potentially protect the phenyl ring from metabo-
lism. Hence, we also adopted the p-chlorophenyl moiety
to occupy the F2 space in the next design.
compd
R
Ki (nM)a
8a
8b
8c
Ph
26
13
4.9
n-C3H7
4-Cl-Ph
a
See footnote a of Table 1.
stituted indole type, the calculated interaction energies
of indol-3-yl and 1-methylindol-3-yl type compounds
indicate more stability (∆E ) 1.5 and 2.5 kcal/mol,
respectively) than the indol-1-yl type analogues (Table
4). Furthermore, it is well-known in general that the
2-and/or 3-positions of indolyl rings are the main
metabolism sites. Although the indol-1-yl type is un-
protected from metabolism, in the case of the 1-meth-
ylindol-3-yl type, the 3-position of indolyl ring is sub-
stituted in addition to the 1- position. So it may be
possible to protect from metabolism of the 2- and/or
3-positions of the indolyl ring by hindrance effects,
compared to the indol-1-yl type. From this reasoning,
we synthesized 1-methylindol-3-yl types (8a -c) corre-
sponding to the highly potent indol-1-yl types (6a ,c,e)
(Table 5). The resulting compounds 8b and 8c displayed
about a 3-4-fold enhancement in inhibitory activity.
Compound 8c showed the best activity for ADA inhibi-
tion (Ki ) 4.9 nM) among derivatives prepared in these
studies (Table 5). Moreover, compound 8c was also
improved in terms of pharmacokinetics compared to the
lead compound 1. That is, compound 8c showed oral
absorption upon administration to rats at 10 mg/kg
(Cmax ) 0.21 µg/mL. AUC0-8 ) 0.26 µg‚h/mL, oral
bioavailability ) 10%).
h
4. Ch em istr y. The amide- and urea-linked com-
pounds listed in Table 1 were synthesized by the routes
shown in Scheme 1. Alcohol 1025 was converted to
mesylate, followed by displacement with 6-nitroindole
in the presence of NaH in DMF to afford 11. Cleavage
of the ketal of 11 with 1 N hydrochloric acid (HCl) in
THF followed by selective protection of the primary
alcohol afforded tert-butyldimethylsilyl (TBS) derivative
12. The secondary alcohol moiety of 12 was then
3. Tr a n sfor m a tion of th e In d ole Con n ection :
In d ol-3-yl Typ e. All compounds described above were
connected via an indolyl ring at the 1-position with the
imidazole moiety. However, taking into consideration
the interaction energies between the inhibitors and
enzyme, it was considered that a carbon atom was more
advantageous in terms of interaction energy than a
nitrogen atom in this position. That is, with the unsub-

3736 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
Sch em e 1a
a
Reagents and conditions: (a) NaBH4, MeOH, 0 °C-rt, 3 h; (b) MsCl, Et3N, CH2Cl2, 0 °C, 1 h; (c) 6-nitroindole, NaH, DMF, 60 °C, 1.5
h; (d) 1 N HCl, THF, 0 °C-rt, 5 h; (e) TBSCl, imidazole, DMF, 0 °C-rt, overnight; (f) MsCl, Et3N, CH2Cl2, 0 °C, 1 h; (g)
4-imidazolecarboxamide, NaH, DMF, 70 °C, 24 h; (h) Fe, NH4Cl, EtOH-H2O, 90 °C, 40 min; (i) WSCD‚HCl, HOBt, R(CH2)nCOOH, CH2Cl2,
rt, 5 h; (j) benzylisocyanate, THF, rt, 2 h; (k) TBAF, THF, 0 °C, 1 h
Sch em e 2a
a
Reagents and conditions: (a) MsCl, Et3N, CH2Cl2, 0 °C, 1 h; (b) 6-benzyloxyindole, NaH, DMF, 60 °C, 1.5 h; (c) 1 N HCl, THF, 0
°C-rt, 5 h; (d) TBSCl, imidazole, DMF, 0 °C-rt, overnight; (e) MsCl, Et3N, CH2Cl2, 0 °C, 1 h; (f) 4-imidazolecarboxamide, NaH, DMF, 70
°C, 24 h; (g) Pd(OH)2, cyclohexene, EtOH, 90 °C, 30 min; (h) 1. R(CH2)3OH, MsCl, Et3N, CH2Cl2, 0 °C, or R(CH2)3I; 2. K2CO3, DMF, 80-90
°C, 1 day; (i) TBAF, THF, 0 °C, 1 h
converted to mesylate, and displacement by SN2 reaction
with 4-imidazolecarboxamide in the presence of NaH
in DMF afforded 13. Reduction of the nitro group on 13
with iron powder/NH4Cl in ethanol (EtOH)-H2O gave
the key intermediate aniline 14. Amides 15a -j or urea
16 were obtained from condensation of 14 with the
corresponding carboxylic acid or isocyanate, followed by
removal of the TBS group with tetrabutylammonium
fluoride (TBAF) to afford 4a -j and 5 in good yields.
The synthetic routes to the ether-linked compounds
6a -e and 7a ,b are shown in Scheme 2. Compounds 17-
22 were prepared by similar procedures to those of
compounds 11, 12, and 13, respectively. The key inter-
mediate phenols 23 and 24 were obtained by reductive
debenzylation of 21 and 22, respectively, in the presence
of Pd(OH)2 in cyclohexene-EtOH at 90 °C. O-Alkylation
of 23 and 24 with the corresponding mesylate or alkyl
iodide yielded 25a -e and 26a ,b. Deprotection of the
TBS group afforded 6a -e and 7a ,b in good yields.
The ether-linked 1-methylindol-3-yl derivatives 8a -c
were synthesized as shown in Scheme 3. Chain elonga-
tion of 27 under Wittig-Horner conditions with triethyl
phosphonoacetate in the presence of NaH in DMF
yielded 28. Catalytic hydrogenation with palladium on
carbon (Pd-C) in EtOH-THF followed by the reduction
of the ester group with diisobutylaluminum hydride
(DIBAL-H) in CH2Cl2 at -78 °C gave the corresponding
aldehyde 30. Conversion of aldehyde to alkene under
Wittig conditions with Ph3PMeBr in the presence of
n-butyllithium (n-BuLi) in THF yielded 31. Stereose-
lective dihydroxylation of 31 was accomplished by
Sharpless oxidation with AD-mix-R in t-BuOH-H2O at
0 °C,26 followed by selective protection of the primary
alcohol of 32 to afford 33. Compounds 34, 35, 36, and 8
were prepared by similar procedures to those of com-
pound 21, 23, 25, and 6, respectively.
Con clu sion s
We optimized the novel non-nucleoside ADA inhibitor
1 by a structure-based approach using the crystal
structure of the 1/ADA complex and discovered several
new nM potent ADA inhibitors with different structures.

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3737
Sch em e 3a
a
Reagents and conditions: (a) triethyl phosphonacetate, NaH, DMF, 0 °C-rt, overnight; (b) H2/Pd-C, EtOH-THF, rt, 2.5 h; (c) DIBAL-
H, CH2Cl2, rt, 30 min; (d) Ph3PMeBr, n-BuLi, THF, -78 °C-rt, 2 h; (e) AD-mix-R, t-BuOH-H2O, 0 °C, 4 h; (f) TBSCl, imidazone, DMF, 0
°C-rt, overnight; (g) MsCl, CH2Cl2, 0 °C, 1 h; (h); 4-imidazolecarboxamide, NaH, DMF, 70 °C, 24 h; (i) Pd(OH)2, cyclohexene, EtOH, 90
°C, 30 min; (j) 1. R(CH2)3OH, MsCl, Et3N, CH2Cl2, 0 °C or R(CH2)3I; 2. K2CO3, DMF, 80-90 °C, 1 day (k) TBAF, THF, 0 °C, 1 h
1-[2-((4S)-2,2-Dim et h yl-1,3-d ioxola n -4-yl)et h yl]-1H -6-
n itr oin d ole (11). To a stirred mixture of 2-[(4S)-2,2-dimethyl-
1,3-dioxolan-4-yl]ethanol (1.65 g, 11.3 mmol) and methane-
sulfonyl chloride (MsCl) (1.81 g, 15.8 mmol) in dichloromethane
(CH2Cl2) (20 mL) was added dropwise triethylamine (Et3N)
(1.60 g, 15.8 mmol) at ice-bath temperature. After 1 h, the
reaction mixture was partitioned between CH2Cl2 and water.
The organic layer was washed with brine, dried (MgSO4), and
concentrated in vacuo to give the methanesulfonate (2.60 g)
as an oil. This material was used immediately without further
purification.
Furthermore, the SAR of these compounds was revealed
in detail by molecular modeling and crystal structure
analysis of the complexes. Key points are (1) replace-
ment of the benzimidazole of 1 with a simple aromatic
ring and alkyl chains is possible, (2) the permissible
distance from the indole moiety to the distal hydropho-
bic space in the active site was determined, (3) replace-
ment of the amide-linker with urea and ether-linkers
was shown to be viable, (4) introduction of substituents
(especially chlorine) to the distal phenyl ring is feasible,
(5) transformation of indol-1-yl to indol-3-yl improves
potency. Consequently, we discovered new structures
more amenable to synthetic modification, while retain-
ing potent ADA inhibitory activity. There is currently
no other comparable data available for semi-tight-
binding inhibitors. The information from this study
should be useful for the further design of novel ADA
inhibitors. Finally, compound 8c, a potent inhibitor of
ADA, was discovered and found to be orally bioavailable
in rats (BA ) 10%). The discovery of a potent orally
bioavailable ADA inhibitor provides an important tool
to further study the biological function of this enzyme
and to aid in the development of new medications as
antiinflammatory drugs.
To a solution of 6-nitroindole(1.89 g, 11.3 mmol) in DMF
(20 mL) was added NaH (60% in mineral oil, 452 mg, 11.3
mmol) at room temperature. The reaction mixture was stirred
for 30 min. A solution of the methanesulfonate prepared above
in DMF (8 mL) was added, and the resulting mixture was
stirred for 1.5 h at room temperature. The reaction mixture
was cooled to 10 °C in an ice bath, and the insoluble material
was filtered and washed thoroughly with CH2Cl2. The filtrate
and washings were combined and then washed with brine. The
organic layer was dried (MgSO4) and concentrated in vacuo.
The residue was purified by silica gel (40 g) column chroma-
tography eluting with toluene/ethyl acetate (EtOAc) (50:1-
20:1) to give 11 (2.19 g, 66.8%) as a yellow solid; IR (KBr, cm-1
)
1583, 1074; 1H NMR (CDCl3) δ 1.34 (3H, s), 1.48 (3H, s), 1.90-
2.25 (2H, m), 3.45-3.60 (1H, m), 3.85-4.05 (2H, m), 4.41 (2H,
dd, J ) 7.9 Hz, 5.6 Hz), 6.61 (1H, d, J ) 3.1 Hz), 7.42 (1H, d,
J ) 3.1 Hz), 7.65 (1H, d, J ) 8.8 Hz), 8.01 (1H, dd, J ) 8.8 Hz,
2.0 Hz), 8.41 (1H, d, J ) 2.0 Hz); MS (ESI, m/z) 291 (M + H)+;
[R]26 -9.3 ° (c 0.50, EtOH).
Exp er im en ta l Section
D
(S)-[1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-n it r oin d ol-1-
yl)]bu ta n -2-ol (12). To an ice-cooled solution of 11 (1.45 g,
5.00 mmol) in THF (30 mL) was added portionwise 1 N HCl
(20 mL). After the addition was completed, the reaction
mixture was stirred at room temperature for 5 h. An aqueous
solution of NaHCO3 was added, and the resulting mixture was
stirred for several minutes. The mixture was extracted with
EtOAc. The organic layer was washed with brine, dried
(MgSO4), and concentrated in vacuo to give (S)-4-(6-nitroindol-
1-yl)butane-1,2-diol (1.44 g) as an oil. This material was used
immediately without further purification.
Gen er a l Meth od s. Melting points were obtained on a
Buchi 535 apparatus and are uncorrected. Infrared (IR) spectra
1
were measured on a HORIBA FT-710 spectrometer. H NMR
spectra were recorded on a Varian EM-390 NMR spectrometer
or a Bruker AC200P spectrometer. Chemical shifts are re-
ported downfield from tetramethylsilane () 0) for 1H NMR.
Mass spectra (MS) were determined on a Hitachi Model M-80
mass spectrometer (EIMS). Optical rotations were measured
in absolute ethanol on a HORIBA SEPA-300 polarimeter.
Elemental analyses were performed on a Perkin-Elmer 2400
CHN elemental analyzer. Reagents and solvents were used
as obtained from commercial suppliers without further puri-
fication. Chromatographic purification of the compounds was
performed on silica gel 60 (63-200 µm) purchased from Merck
Co.
To an ice-cooled solution of (S)-4-(6-nitroindol-1-yl)butane-
1, 2-diol in DMF (20 mL) was added imidazole (1.02 g, 15.0
mmol) followed by tert-butyldimethylsilyl chloride (TBSCl) (791
mg, 5.25 mmol). After 30 min, the ice-bath was removed, and

3738 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
then the mixture was stirred overnight at room temperature.
The reaction mixture was poured into water (200 mL) and
extracted with EtOAc. The organic layer was washed with
brine, dried (MgSO4), and concentrated in vacuo. The residue
was purified by silica gel (25 g) column chromatography eluting
with hexane/EtOAc (10:1) to give 12 (1.61 g, 88.3%); IR (KBr,
eluting with CHCl3/MeOH(30:1) to give 15c (170.6 mg, 100%);
IR (KBr, cm-1) 3600-3000, 1660; 1H NMR (CDCl3) δ -0.10
(3H, s), -0.08 (3H, s), 0.81 (9H, s), 2.20-2.55 (2H, m), 2.73
(2H, t, J ) 7.6 Hz), 3.10 (2H, t, J ) 7.6 Hz), 3.60-4.25 (5H,
m), 5.42 (1H, brs), 6.45 (1H, d, J ) 3.1 Hz), 6.86 (1H, d, J )
3.1 Hz), 6.90-7.05 (2H, m), 7.15-7.35 (5H, m), 7.40-7.65 (5H,
m); MS (ESI, m/z) 560 (M + H)+.
1
cm-1) 3558, 1511; H NMR (CDCl3) δ 0.04 (6H, s), 0.88 (9H,
s), 1.80-2.00 (2H, m), 2.51 (1H, d, J ) 3.2 Hz), 3.30-3.65 (3H,
m), 4.43 (2H, dd, J ) 7.7 Hz, 6.0 Hz), 6.59 (1H, d, J ) 3.1 Hz),
7.45 (1H, d, J ) 3.1 Hz), 7.65 (1H, d, J ) 8.8 Hz), 8.01 (1H,
dd, J ) 8.8 Hz, 2.0 Hz), 8.39 (1H, d, J ) 2.0 Hz); MS (ESI,
1-[(R)-1-Hyd r oxy-4-(6-(3-p h en ylp r op ion yla m in o)in d ol-
1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (4c). To an ice-
cooled solution of 15c (150 mg, 0.268 mmol) in THF (5 mL)
was added dropwise 1.0 M TBAF in THF (402 µL). After the
addition was completed, the reaction mixture was stirred at
ice-bath temperature for 30 min. 25% AcONH4 (4 mL) was
added, and the resulting mixture was stirred for several
minutes and then extracted with EtOAc. The organic layer
was washed with brine, dried (Na2SO4), and concentrated in
vacuo. The residue was purified by silica gel (6 g) column
chromatography eluting with CHCl3/MeOH (10:1) to give 4c
(96.6 mg, 80.9%) as a white solid; mp 97-100 °C; IR (KBr,
cm-1) 3700-2800, 1656; 1H NMR (DMSO-d6) δ 2.20-2.40 (2H,
m), 2.63 (2H, t, J ) 7.5 Hz), 2.94 (2H, t, J ) 7.5 Hz), 3.60 (2H,
t, J ) 5.2 Hz), 3.80-4.20 (3H, m), 5.07 (1H, t, J ) 5.2 Hz),
6.36 (1H, d, J ) 3.0 Hz), 6.95-7.35 (9H, m), 7.43 (1H, d, J )
8.3 Hz), 7.65-7.90 (3H, m), 9.86 (1H, brs); MS (ESI, m/z) 446
m/z) 365 (M + H)+; [R]26 -62.0° (c 0.50, EtOH).
D
1-[(R)-1-(ter t-Bu tyld im eth ylsilyloxy)-4-(6-n itr oin d ol-1-
yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (13). To a stirred
mixture of 12 (1.53 g, 4.21 mmol) and methanesulfonyl chloride
(675 mg, 5.89 mmol) in CH2Cl2 (30 mL) was added dropwise
Et3N (596 mg, 5.89 mmol) at ice-bath temperature. After 1 h,
the reaction mixture was partitioned between CH2Cl2 and
water. The organic layer was washed with brine, dried
(MgSO4), and concentrated in vacuo to give the methane-
sulfonate (1.84 g) as an oil. This material was used for the
next reaction without further purification.
To a solution of 4-imidazolecarboxamide (467 mg, 4.21
mmol) in DMF (10 mL) was added NaH (60% in mineral oil,
194 mg, 4.84 mmol) at room temperature. The reaction mixture
was stirred for 1 h at 70 °C. The methanesulfonate prepared
above was added, and the resulting mixture was stirred for
38 h at 70 °C. The reaction mixture was cooled to 10 °C in an
ice bath, and the insoluble material was filtered and washed
thoroughly with CH2Cl2. The filtrate and washings were
combined and washed with brine. The organic layer was dried
(Na2SO4) and concentrated in vacuo. The residue was purified
by silica gel (66 g) column chromatography eluting with
chloroform (CHCl3)/methanol (MeOH) (30:1) to give 13 (489
(M + H)+; [R]24 +18.0° (c 0.50, EtOH); Anal. (C25H27N5O3‚
D
0.5H2O) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-a cet yla m i-
n oin d ol-1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (15a ). Re-
placing 3-phenylpropionic acid with acetic acid and following
the same procedure as in the preparation of 15c gave 15a (69.4
mg, 90.3%). IR (KBr, cm-1) 3600-3000, 1664; 1H NMR (CDCl3)
δ -0.10 (3H, s), -0.08 (3H, s), 0.80 (9H, s), 2.23 (3H, s), 2.30-
2.50 (2H, m), 3.60-4.20 (5H, m), 5.51 (1H, brs), 6.46 (1H, d, J
) 3 Hz), 6.88 (1H, d, J ) 3 Hz), 7.00-7.15 (2H, m), 7.40-7.75
(5H, m); MS (ESI, m/z) 470 (M + H)+; [R]26 +16.3° (c 0.50,
1
mg, 25.4%); IR (KBr, cm-1) 3600-3000, 1662, 1504; H NMR
D
EtOH).
(CDCl3) δ -0.08 (3H, s), -0.06 (3H, s), 0.81 (9H, s), 2.25-2.65
(2H, m), 3.65-4.20 (5H, m), 5.44 (1H, brs), 6.62 (1H, d, J )
3.1 Hz), 6.97 (1H, brs), 7.16 (1H, d, J ) 3.1 Hz), 7.40 (1H, s),
7.66 (1H, s), 7.68 (1H, d, J ) 8.8 Hz), 8.04 (1H, dd, J ) 8.8
Hz, 2.0 Hz), 8.18 (1H, d, J ) 2.0 Hz); MS (ESI, m/z) 458 (M +
1-[(R)-4-(6-Acetyla m in oin d ol-1-yl)-1-h yd r oxy-2-bu tyl]-
im id a zole-4-ca r boxa m id e (4a ). To an ice-cooled solution of
15a (62 mg, 0.133 mmol) in THF (5 mL) was added dropwise
1.0 M TBAF in THF (199 µL). After addition, the reaction
mixture was stirred at ice-bath temperature for 1 h. 25%
AcONH4 (5 mL) was added, and the resulting mixture was
stirred for several minutes and then extracted with water.
Because of high water solubility of 4a , the aqueous layer was
washed with EtOAc and was purified by HP-20 (15 cm3)
column chromatography eluting with water/2-propanol (IPA)
(9:1) and lyophilized to give 4a (41 mg, 86.8%) as an amor-
phous solid. IR (KBr, cm-1) 3600-3000, 1662; 1H NMR
(DMSO-d6) δ 2.05 (3H, s), 2.20-2.40 (2H, m), 3.50-4.20 (5H,
m), 5.11 (1H, brs), 6.36 (1H, d, J ) 3 Hz), 7.00-7.15 (2H, m),
7.17 (1H, d, J ) 3 Hz), 7.30 (1H, brs) 7.43 (1H, d, J ) 9 Hz),
7.70-7.85 (3H, m), 9.89 (1H, brs); MS (ESI, m/z) 356 (M +
H)+; [R]27 -3.3° (c 0.50, EtOH).
D
1-[(R)-4-(6-Am in oin d ol-1-yl)-1-(ter t-b u t yld im et h ylsil-
yloxy)-2-bu tyl]im id a zole-4-ca r boxa m id e (14). To a me-
chanically stirred mixture of 13 (450 mg, 0.98 mmol) and
ammonium chloride (45 mg) in EtOH (10 mL) and water (5
mL) was added portionwise iron powder (450 mg) at 90 °C.
The resulting mixture was stirred for 40 min at 90 °C. After
cooling, the insoluble material was filtered through Celite and
washed with ethanol and ethyl acetate. The filtrate and
washings were combined and concentrated in vacuo. The
resulting residue was extracted with EtOAc and washed with
brine. The organic layer was dried (Na2SO4) and concentrated
in vacuo to give 14 (397 mg, 94.7%). This material was used
for the next reaction without further purification. IR (KBr,
cm-1) 3700-3000, 1662, 1627; 1H NMR (CDCl3) δ -0.10 (3H,
s), -0.08 (3H, s), 0.81 (9H, s), 2.00-2.70 (4H, br), 3.60-4.15
(5H, m), 5.44 (1H, brs), 6.36 (1H, d, J ) 1.9 Hz), 6.38 (1H, d,
J ) 3.2 Hz), 6.57 (1H, dd, J ) 8.3 Hz, 1.9 Hz), 6.71 (1H, d, J
) 3.2 Hz), 7.00 (1H, brs), 7.39 (1H, d, J ) 8.3 Hz), 7.41 (1H,
s), 7.68 (1H, s); MS (ESI, m/z) 428 (M + H)+.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(3-p h en yl-
p r op ion yla m in o)in d ol-1-yl)-2-bu tyl]-im id a zole-4-ca r box-
a m id e (15c). To a stirred mixture of 3-phenylpropionic acid
(46 mg, 0.304 mmol) and 1-hydroxybenzotriazole (45 mg, 0.334
mmol) in CH2Cl2 (5 mL) was added 1-(3-dimethylaminopropyl)-
3-ethylcarbodiimide (52 mg, 0.334 mmol) at ice-bath temper-
ature. The mixture was stirred for 40 min at room tempera-
ture, and then 14 (130 mg, 0.304 mmol) was added to the
mixture at ice-bath temperature. After addition, the resulting
mixture was stirred for 4 h at room temperature. The solvent
was concentrated in vacuo, and the residue extracted with
ethyl acetate and washed with water and brine. The organic
layer was dried (Na2SO4) and concentrated in vacuo. The
residue was purified by silica gel (6 g) column chromatography
H)+; [R]27 +13.6° (c 0.50, MeOH); Anal. (C18H21N5O3‚0.5IPA‚
D
1.0H2O) C, H, N.
1-[(R)-1-(ter t-Bu tyldim eth ylsilyloxy)-4-(6-h exan oylam i-
n oin d ol-1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (15b). Re-
placing 3-phenylpropionic acid with hexanoic acid and follow-
ing the same procedure as in the preparation of 15c gave 15b
1
(149 mg, 93.2%). IR (KBr, cm-1) 3600-3000, 1660; H NMR
(CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.80 (9H, s), 0.93 (3H,
t, J ) 6.6 Hz), 1.30-1.50 (4H, m), 1.70-1.90 (2H, m), 2.20-
2.50 (4H, m), 3.60-4.25 (5H, m), 5.43 (1H, brs), 6.45 (1H, d, J
) 3.1 Hz), 6.84 (1H, d, J ) 3.1 Hz), 6.95-7.10 (2H, m), 7.45-
7.60 (3H, m), 7.63 (1H, s), 7.75 (1H, s); MS (ESI, m/z) 526 (M
+ H)+; [R]24 +19.9° (c 0.50, EtOH).
D
1-[(R )-4-(6-H e xa n oyla m in oin d ol-1-yl)-1-h yd r oxy-2-
bu tyl]im id a zole-4-ca r boxa m id e (4b). Replacing 15c with
15b and following the same procedure as in the preparation
of 4c gave 4b (89.9 mg, 79.7%) as an amorphous solid. IR (KBr,
1
cm-1) 3600-3000, 1656; H NMR (CDCl3) δ 0.89 (3H, t, J )
6.5 Hz), 1.20-1.45 (4H, m), 1.61 (2H, qui, J ) 7.0 Hz), 2.20-
2.40 (4H, m), 3.60 (2H, t, J ) 5.2 Hz), 3.85-4.25 (3H, m), 5.06
(1H, t, J ) 5.2 Hz), 6.35 (1H, d, J ) 3.1 Hz), 7.00-7.15 (2H,
m), 7.16 (1H, d, J ) 3.1 Hz), 7.29 (1H, brs), 7.42 (1H, d, J )

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3739
1
8.5 Hz), 7.74 (1H, s), 7.78 (1H, s), 7.84 (1H, s), 9.80 (1H, brs);
MS (ESI, m/z) 412 (M + H)+; [R]26D +25.8° (c 0.50, EtOH); Anal.
(C22H29N5O3‚0.25IPE‚0.5H2O) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(4-p h en yl-
b u t yr yla m in o)in d ol-1-yl)-2-b u t yl]im id a zole-4-ca r b ox-
a m id e (15d ). Replacing 3-phenylpropionic acid with 4-phen-
ylbutyric acid and following the same procedure as in the
80.7%). IR (KBr, cm-1) 3600-3000, 1664; H NMR (CDCl3) δ
-0.10 (3H, s), -0.08 (3H, s), 0.81 (9H, s), 2.20-2.55 (5H, m),
2.71 (2H, t, J ) 7.6 Hz), 3.05 (2H, t, J ) 7.6 Hz), 3.60-4.20
(5H, m), 5.42 (1H, brs), 6.45 (1H, d, J ) 3.0 Hz), 6.85 (1H, d,
J ) 3.0 Hz), 6.90-7.25 (6H, m), 7.40-7.70 (5H, m); MS (ESI,
m/z) 574 (M + H)+; [R]26 +16.4° (c 0.50, EtOH).
D
1-[(R )-1-H yd r oxy-4-(6-(3-(4-m e t h ylp h e n yl)p r op ion -
ylam in o)in dol-1-yl)-2-bu tyl]im idazole-4-car boxam ide (4g).
Replacing 15c with 15g and following the same procedure as
in the preparation of 4c gave 4g (71 mg, 92.5%) as an
preparation of 15c gave 15d (170.6 mg, 100%). IR (KBr, cm-1
)
3600-3000, 1662; 1H NMR (CDCl3) δ -0.10 (3H, s), -0.08 (3H,
s), 0.80 (9H, s), 2.00-2.20 (2H, m), 2.20-2.50 (4H, m), 2.75
(2H, t, J ) 7.4 Hz), 3.60-4.25 (5H, m), 5.34 (1H, brs), 6.45
(1H, d, J ) 3.2 Hz), 6.86 (1H, d, J ) 3.2 Hz), 6.90-7.10 (2H,
m), 7.15-7.40 (5H, m), 7.40-7.70 (5H, m); MS (ESI, m/z) 574
amorphous solid. IR (KBr, cm-1) 3600-3000, 1656; H NMR
1
(DMSO-d6) δ 2.10-2.40 (5H, m), 2.60 (2H, t, J ) 7.5 Hz), 2.89
(2H, t, J ) 7.5 Hz), 3.61 (2H, t, J ) 5.2 Hz), 3.85-4.25 (3H,
m), 5.08 (1H, brs), 6.36 (1H, d, J ) 3.0 Hz), 6.95-7.25 (7H,
m), 7.41 (1H, brs), 7.43 (1H, d, J ) 8.5 Hz), 7.70-8.00 (3H,
(M + H)+; [R]23 +15.1° (c 0.50, EtOH).
D
1-[(R)-1-Hyd r oxy-4-(6-(4-p h en ylbu tyr yla m in o)in d ol-1-
yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (4d ). Replacing 15c
with 15d and following the same procedure as in the prepara-
tion of 4c gave 4d (73.4 mg, 68.4%) as an amorphous solid. IR
(KBr, cm-1) 3600-3000, 1656; 1H NMR (DMSO-d6) δ 1.80-
2.05 (2H, m), 2.15-2.40 (4H, m), 2.64 (2H, t, J ) 7.5 Hz), 3.59
(2H, t, J ) 5.2 Hz), 3.80-4.20 (3H, m), 5.06 (1H, t, J ) 5.2
Hz), 6.36 (1H, d, J ) 3.0 Hz), 6.95-7.40 (9H, m), 7.43 (1H, d,
J ) 8.5 Hz), 7.70-7.90 (3H, m), 9.82 (1H, brs); MS (ESI, m/z)
460 (M + H)+; [R]24D +19.0° (c 0.50, EtOH); Anal. (C26H29N5O3‚
0.5H2O) C, H, N.
m), 9.84 (1H, brs); MS (ESI, m/z) 460(M + H)+; [R]27 +11.9°
D
(c 0.50, EtOH); Anal. (C26H29N5O3‚0.8H2O) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(3-(4-m et h -
oxylph en yl)pr opion ylam in o)in dol-1-yl)-2-bu tyl]im idazole-
4-ca r boxa m id e (15h ). Replacing 3-phenylpropionic acid with
3-(4-methoxylphenyl)propionic acid and following the same
procedure as in the preparation of 15c gave 15h (105 mg,
80.2%). IR (KBr, cm-1) 3600-3000, 1664, 1247; 1H NMR
(CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.81 (9H, s), 2.20-2.55
(2H, m), 2.69 (2H, t, J ) 7.6 Hz), 3.04 (2H, t, J ) 7.6 Hz),
3.55-4.25 (10H, m), 5.44 (1H, brs), 6.45 (1H, d, J ) 3.0 Hz),
6.75-7.05 (5H, m), 7.15-7.25 (2H, m), 7.40-7.65 (5H, m); MS
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(5-p h en yl-
va le r yla m in o)in d ol-1-yl)-2-b u t yl]im id a zole -4-ca r b ox-
a m id e (15e). Replacing 3-phenylpropionic acid with 5-phen-
ylvaleric acid and following the same procedure as in the
(ESI, m/z) 590 (M + H)+; [R]27 +16.8° (c 0.50, EtOH).
D
1-[(R)-1-Hyd r oxy-4-(6-(3-(4-m et h oxylp h en yl)p r op ion -
ylam in o)in dol-1-yl)-2-bu tyl]im idazole-4-car boxam ide (4h ).
Replacing 15c with 15h and following the same procedure as
in the preparation of 4c gave 4h (67 mg, 85.6%) as an
preparation of 15c gave 15e (82.4 mg, 100%). IR (KBr, cm-1
)
3600-3000, 1664; 1H NMR (CDCl3) δ -0.10 (3H, s), -0.08 (3H,
s), 0.80 (9H, s), 1.60-1.95 (4H, m), 2.20-2.55 (4H, m), 2.69
(2H, t, J ) 7.0 Hz), 3.60-4.25 (5H, m), 5.37 (1H, brs), 6.45
(1H, d, J ) 3.2 Hz), 6.85 (1H, d, J ) 3.2 Hz), 6.90-7.10 (2H,
m), 7.10-7.35 (5H, m), 7.40-7.60 (3H,m), 7.70 (1H, s), 7.78
amorphous solid. IR (KBr, cm-1) 3700-3000, 1656, 1241; H
1
NMR(DMSO-d6) δ 2.20-2.40 (2H, m), 2.59 (2H, t, J ) 7.5 Hz),
2.87 (2H, t, J ) 7.5 Hz), 3.60 (2H, t, J ) 5.2 Hz), 3.71 (3H, S),
3.90-4.20 (3H, m), 5.06 (1H, t, J ) 5.2 Hz), 6.36 (1H, d, J )
3.0 Hz), 6.85 (2H, d, J ) 8.6 Hz), 7.00-7.25 (5H, m), 7.28 (1H,
brs), 7.43 (1H, d, J ) 8.5 Hz), 7.70-7.90 (3H, m), 9.83 (1H,
brs); MS (ESI, m/z) 476 (M + H)+; [R]28D +17.1° (c 0.50, EtOH);
Anal. (C26H29N5O4‚0.6H2O) C, H, N.
(1H, s); MS (ESI, m/z) 588 (M + H)+; [R]21 +16.6° (c 0.50,
D
EtOH).
1-[(R)-1-Hyd r oxy-4-(6-(5-p h en ylva ler yla m in o)in d ol-1-
yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (4e). Replacing 15c
with 15e and following the same procedure as in the prepara-
tion of 4c gave 4e (54.1 mg, 70.3%) as an amorphous solid. IR
(KBr, cm-1) 3600-3000, 1664, 1656; 1H NMR (DMSO-d6) δ
1.50-1.75 (4H, m), 2.10-2.45 (4H, m), 2.50-2.75 (2H, m), 3.59
(2H, t, J ) 5.2 Hz), 3.80-4.20 (3H, m), 5.06 (1H, t, J ) 5.2
Hz), 6.36 (1H, d, J ) 2.8 Hz), 6.95-7.35 (9H, m), 7.43 (1H, d,
J ) 8.5 Hz), 7.70-7.90 (3H, m), 9.82 (1H, brs); MS (ESI, m/z)
474 (M + H)+; [R]27D +19.0° (c 0.50, EtOH); Anal. (C27H31N5O3‚
0.5H2O) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(4-(4-m et h -
ylp h en yl)bu tyr yla m in o)in d ol-1-yl)-2-bu tyl]im id a zole-4-
ca r boxa m id e (15i). Replacing 3-phenylpropionic acid with
4-(4-methylphenyl)butyric acid and following the same proce-
dure as in the preparation of 15c gave 15i (103 mg, 78.9%).
IR (KBr, cm-1) 3600-3000, 1664; 1H NMR (CDCl3) δ -0.10
(3H, s), -0.08 (3H, s), 0.80 (9H, s), 1.95-2.20 (2H, m), 2.20-
2.50 (7H, m), 2.71 (2H, t, J ) 7.4 Hz), 3.50-4.25 (5H, m), 5.37-
(1H, brs), 6.45 (1H, d, J ) 3.0 Hz), 6.85 (1H, d, J ) 3.0 Hz),
6.90-7.20 (6H, m), 7.40-7.70 (5H, m); MS (ESI, m/z) 588 (M
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(6-p h en yl-
h exa n oyla m in o)in d ol-1-yl)-2-b u t yl]im id a zol-4-ca r b ox-
a m id e (15f). Replacing 3-phenylpropionic acid with 6-phen-
ylhexanic acid and following the same procedure as in the
+ H)+; [R]25 +14.7° (c 0.50, EtOH).
D
1-[(R)-1-Hyd r oxy-4-(6-(4-(4-m eth ylp h en yl)bu tyr yla m i-
n o)in d ol-1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (4i). Re-
placing 15c with 15i and following the same procedure as in
the preparation of 4c gave 4i (63.8 mg, 82.5%) as an amor-
phous solid. IR (KBr, cm-1) 3700-3000, 1656; 1H NMR
(DMSO-d6) δ 1.80-2.00 (2H, m), 2.20-2.40 (7H, m), 2.60 (2H,
t, J ) 7.5 Hz), 3.60 (2H, d, J ) 5.2 Hz), 3.90-4.25 (3H, m),
5.08 (1H, brs), 6.36 (1H, d, J ) 3.0 Hz), 7.00-7.25 (7H, m),
7.40 (1H, brs), 7.42 (1H, d, J ) 8.5 Hz), 7.75-8.00 (3H, m),
preparation of 15c gave 15f (73.7 mg, 74.8%). IR (KBr, cm-1
)
3600-3000, 1658; 1H NMR(CDCl3) δ -0.10 (3H, s), -0.08 (3H,
s), 0.80 (9H, s), 1.40-1.95 (6H, m), 2.20-2.50 (4H, m), 2.64
(2H, t, J ) 8 Hz), 3.55-4.25 (5H, m), 5.42 (1H, brs), 6.45 (1H,
d, J ) 3 Hz), 6.85 (1H, d, J ) 3 Hz), 6.90-7.35 (7H, m), 7.45-
7.75 (5H, m); MS (ESI, m/z) 602 (M + H)+; [R]23 +15.9° (c
D
0.50, EtOH).
1-[(R)-1-Hyd r oxy-4-(6-(6-p h en ylh exa n oyla m in o)in d ol-
1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (4f). Replacing 15c
with 15f and following the same procedure as in the prepara-
tion of 4c gave 4f (47 mg, 86.6%) as an amorphous solid. IR
(KBr, cm-1) 3700-3000, 1657; 1H NMR (DMSO-d6) δ 1.20-
1.75 (6H, m), 2.20-2.70 (6H, m), 3.50-4.20 (5H, m), 5.06 (1H,
brs), 6.36 (1H, d, J ) 3 Hz), 7.00-7.35 (9H, m), 7.43 (1H, d, J
) 9 Hz), 7.70-7.90 (3H, m), 9.80 (1H, brs); MS (ESI, m/z) 488
9.81 (1H, brs); MS (ESI, m/z) 474 (M + H)+; [R]27 +13.0° (c
0.50, EtOH); Anal. (C27H31N5O3‚1.0H2O) C, H, N.
D
1-{(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-[6-(3-(3-p yr id -
yl)pr opion ylam in o)in dol-1-yl]-2-bu tyl}im idazole-4-car box-
a m id e (15j). Replacing 3-phenylpropionic acid with 3-(3-
pyridyl)propionic acid and following the same procedure as in
the preparation of 15c gave 15j (103 mg, 82.8%). IR (KBr,
cm-1) 3600-3000, 1664; 1H NMR (CDCl3) δ -0.09 (3H, s),
-0.07 (3H, s), 0.81 (9H, s), 2.20-2.50 (2H, m), 2.76 (2H, t, J
) 8 Hz), 3.11 (2H, t, J ) 8 Hz), 3.55-4.20 (5H, m), 5.46 (1H,
brs), 6.46 (1H, d, J ) 3 Hz), 6.89 (1H, d, J ) 3 Hz), 7.00 (1H,
brs), 7.10-7.30 (2H, m), 7.40-7.70 (5H, m), 7.89 (1H, brs),
8.40-8.60 (2H, m); MS (ESI, m/z) 561 (M + H)+; [R]26D +13.3°
(c 0.50, EtOH).
(M + H)+; [R]27 +20.5° (c 0.50, EtOH); Anal. (C28H33N5O3‚
D
0.25H2O) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(3-(4-m et h -
ylp h en yl)p r op ion yla m in o)in d ol-1-yl)-2-bu tyl]im id a zole-
4-ca r boxa m id e (15g). Replacing 3-phenylpropionic acid with
3-(4-methylphenyl)propionic acid and following the same
procedure as in the preparation of 15c gave 15g (103 mg,

3740 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
1-{(R)-1-Hyd r oxy-4-[6-(3-(3-p yr id yl)p r op ion yla m in o)-
in d ol-1-yl]-2-bu tyl}im id a zole-4-ca r boxa m id e (4j). Replac-
ing 15a with 15j and following the same procedure as in the
preparation of 4a gave 4j (73.4 mg, 97.2%) as an amorphous
2.60 (4H, m), 2.86 (2H, t, J ) 8 Hz), 3.60-4.20 (7H, m), 5.40
(1H, brs), 6.44 (1H, d, J ) 3 Hz), 6.56 (1H, d, J ) 2 Hz), 6.78
(1H, d, J ) 3 Hz), 6.80 (1H, dd, J ) 9 Hz, 2 Hz), 6.95 (1H,
brs), 7.10-7.35 (5H, m), 7.38 (1H, s), 7.49 (1H, d, J ) 9 Hz),
1
solid. IR (KBr, cm-1) 3600-2800, 1664; H NMR (DMSO-d6)
7.67 (1H, s); MS (ESI, m/z) 547 (M + H)+; [R]28 +8.6° (c 0.50,
D
δ 2.20-2.40 (2H, m), 2.67 (2H, t, J ) 8 Hz), 2.95 (2H, t, J ) 8
Hz), 3.50-4.25 (5H, m), 5.09 (1H, brs), 6.36 (1H, d, J ) 3 Hz),
7.00-7.16 (2H, m), 7.18 (1H, d, J ) 3 Hz), 7.20-7.40 (2H, m),
7.43 (1H, d, J ) 9 Hz), 7.65-7.90 (4H, m), 8.30-8.55 (2H, m),
EtOH).
1-[(R)-1-Hyd r oxy-4-(6-(3-p h en ylp r op oxy)in d ol-1-yl)-2-
bu tyl]im id a zole-4-ca r boxa m id e (6a ). Replacing 15c with
25a and following the same procedure as in the preparation
of 4c gave 6a (55.6 mg, 78.1%) as a white solid; mp 135-136
9.87 (1H, brs); MS (ESI, m/z) 447 (M + H)+; [R]27 +12.4 ° (c
D
1
°C; IR (KBr, cm-1) 3319, 3186, 1680, 1238; H NMR (DMSO-
0.40, EtOH); Anal. (C24H26N6O3‚0.5IPA‚1.0H2O) C, H, N.
d6) δ 1.90-2.40 (4H, m), 2.78 (2H, t, J ) 8 Hz), 3.60 (2H, t, J
) 5 Hz), 3.80-4.20 (5H, m), 5.04 (1H, t, J ) 5 Hz), 6.34 (1H,
d, J ) 3 Hz), 6.68 (1H, dd, J ) 9 Hz, 2 Hz), 6.73 (1H, d, J )
2 Hz), 7.00-7.45 (9H, m), 7.70 (1H, s), 7.80(1H, s); MS (ESI,
1-[(R)-4-(6-(3-Ben zylu r eid o)in d ol-1-yl)-1-(ter t-bu tyld i-
m eth ylsilyloxy)-2-bu tyl]im id a zole-4-ca r boxa m id e (16).
To a solution of 14 (100 mg, 0.234 mmol) in THF (5 mL) was
added dropwise benzyl isocyanate (32 µL, 0.257 mmol). The
reaction mixture was stirred at room temperature for 2 h and
the solvent was concentrated in vacuo. The residue was
purified by silica gel (4.5 g) column chromatography eluting
with CHCl3/MeOH (30:1) to give 16 (101 mg, 77%) as a pale
yellow solid. IR (KBr, cm-1) 3600-3000, 1654; 1H NMR
(CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.81 (9H, s), 2.20-2.40
(2H, m), 3.55-4.10 (5H, m), 4.47 (2H, d, J ) 6 Hz), 5.39 (1H,
brs), 5.66 (1H, t, J ) 6 Hz), 6.44 (1H, d, J ) 3 Hz), 6.85-6.92
(2H, m), 7.01 (1H, brs), 7.15-7.40 (7H, m), 7.40-7.65 (3H, m);
m/z) 433(M + H)+; [R]27 +16.3° (c 0.50, MeOH); Anal.
D
(C25H28N4O3) C, H, N.
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(4-p h en yl-
bu toxy)in dol-1-yl)-2-bu tyl]im idazole-4-car boxam ide (25b).
Replacing 3-phenylpropanol with 4-phenylbutanol and follow-
ing the same procedure as in the preparation of 25a gave 25b
(92 mg, 78.1%). IR (neat, cm-1) 3700-3000, 1668, 1250; 1H
NMR (CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.80 (9H, s), 1.75-
2.00 (4H, m), 2.20-2.80 (4H, m), 3.60-4.25 (7H, m), 5.32 (1H,
brs), 6.43 (1H, d, J ) 3 Hz), 6.55 (1H, d, J ) 2 Hz), 6.70-6.85
(2H, m), 6.95 (1H, brs), 7.15-7.35 (6H, m), 7.37 (1H, s), 7.48
(1H, d, J ) 9 Hz), 7.80 (1H, s); MS (ESI, m/z) 561 (M + H)+;
MS (ESI, m/z) 561 (M + H)+; [R]26 +17.9° (c 0.50, EtOH).
D
1-[(R)-4-(6-(3-Ben zylu r eid o)in d ol-1-yl)-1-h yd r oxy-2-
bu tyl]im id a zole-4-ca r boxa m id e (5). Replacing 15c with 16
and following the same procedure as in the preparation of 4c
[R]27 +8.2° (c 0.50, EtOH).
D
gave 5 (75 mg, 98.2%) as an amorphous solid. IR (KBr, cm-1
)
1-[(R)-1-H yd r oxy-4-(6-(4-p h en ylb u t oxy)in d ol-1-yl)-2-
bu tyl]im id a zole-4-ca r boxa m id e (6b). Replacing 15c with
25b and following the same procedure as in the preparation
of 4c and washing with IPE gave 6b (72.7 mg, 100%) as an
1
3600-2800, 1654; H NMR (DMSO-d6) δ 2.15-2.40 (2H, m),
3.59 (2H, t, J ) 5 Hz), 3.85-4.20 (3H, m), 4.32 (2H, d, J ) 6
Hz), 5.05 (1H, t, J ) 5 Hz), 6.32 (1H, d, J ) 3 Hz), 6.56 (1H,
t, J ) 6 Hz), 6.90 (1H, dd, J ) 8 Hz, 2 Hz), 6.95-7.45 (9H, m),
7.62 (1H, d, J ) 2 Hz), 7.73 (1H, s), 7.79 (1H, s), 8.46 (1H, s);
1
amorphous solid. IR (KBr, cm-1) 3357, 3182, 1664, 1252; H
NMR (DMSO-d6) δ 1.75-1.90 (4H, m), 2.10-2.40 (2H, m),
2.55-2.75 (2H, br), 3.60 (2H, t, J ) 5 Hz), 3.80-4.20 (5H, m),
5.04 (1H, t, J ) 5 Hz), 6.33 (1H, d, J ) 3 Hz), 6.64 (1H, dd, J
) 9 Hz, 2 Hz), 6.73 (1H, d, J ) 2 Hz), 7.00-7.45 (9H, m), 7.70
(1H, s), 7.80 (1H, s); MS (ESI, m/z) 447 (M + H)+; [R]27D +15.8°
(c 0.50, MeOH); Anal. (C26H30N4O3‚0.35H2O) C, H, N.
MS (ESI, m/z) 447 (M + H)+; [R]24 +10.0° (c 0.20, MeOH);
D
Anal. (C24H26N6O3‚1.0H2O) C, H, N.
1-[(R)-1-(ter t-Bu tyldim eth ylsilyloxy)-4-(6-h ydr oxyin dol-
1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (23). To a solution
of 21 (975 mg, 1.88 mmol) in cyclohexene (10 mL) and EtOH
(20 mL) was added 20% palladium hydroxide on carbon (488
mg). The resulting mixture was stirred at reflux for 1 h. After
cooling to room temperature, the mixture was filtered through
Celite and washed with EtOH. The filtrate was concentrated
in vacuo and the residue purified by silica gel (20 g) chroma-
tography eluting with CHCl3/MeOH (100:1 to 50:1) to give 23
(772 mg, 95.8%) as a colorless solid. IR (KBr, cm-1) 3700-
3000, 1658, 1255; 1H NMR (CDCl3) δ -0.09 (3H, s), -0.07 (3H,
s), 0.81 (9H, s), 2.20-2.55 (2H, m), 3.55-4.05 (5H, m), 5.64
(1H, brs), 6.43 (1H, d, J ) 3 Hz), 6.56 (1H, d, J ) 2 Hz), 6.74
(1H, dd, J ) 8 Hz, 2 Hz), 6.81 (1H, d, J ) 3 Hz), 7.06 (1H,
brs), 7.44 (1H, d, J ) 8 Hz), 7.46 (1H, s), 7.73 (1H, s); MS (ESI,
1-[(R)-4-(6-Bu toxyin d ol-1-yl)-1-(ter t-bu tyld im eth ylsil-
yloxy)-2-bu tyl]im id a zole-4-ca r boxa m id e (25d ). To a solu-
tion of 23 (80 mg, 0.187 mmol) in DMF (5 mL) was added
potassium carbonate (38.7 mg, 0.280 mmol) at room temper-
ature. The reaction mixture was stirred for 30 min, 1-iodo-
butane (51.5 mg, 0.280 mmol) was added, and the resulting
mixture was stirred for 24 h at 60-80 °C. The reaction mixture
was poured into water (50 mL) and extracted with ethyl
acetate. The organic layer was washed with brine, dried (Na2-
SO4), and evaporated in vacuo. The residue was purified by
silica gel (3 g) chromatography eluting with CHCl3/MeOH (50:
1) to give 25d (66.3 mg, 73.3%). IR (KBr, cm-1) 3600-3000,
m/z) 429 (M + H)+; [R]26 +25.7° (c 0.50, EtOH).
1
1666, 1255; H NMR (CDCl3) δ -0.10 (3H, s), -0.08 (3H, s),
D
0.81 (9H, s), 1.00 (3H, t, J ) 7 Hz), 1.40-1.95 (4H, m), 2.20-
2.60 (2H, m), 3.55-4.25 (7H, m), 5.39 (1H, brs), 6.43 (1H, d, J
) 3 Hz), 6.58 (1H, d, J ) 2 Hz), 6.78 (1H, d, J ) 3 Hz), 6.80
(1H, dd, J ) 9 Hz, 2 Hz), 6.97 (1H, brs), 7.38 (1H, s), 7.49 (1H,
1-[(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-(6-(3-p h en yl-
p r op oxy)in d ol-1-yl)-2-b u t yl]im id a zole-4-ca r b oxa m id e
(25a ). To a stirred mixture of 3-phenylpropanol (40 mg, 0.294
mmol) and MsCl (40 mg, 0.353 mmol) in CH2Cl2 (5 mL) was
added dropwise Et3N (36 mg, 0.353 mmol) at ice-bath tem-
perature. After 1 h, the reaction mixture was partitioned
between CH2Cl2 and water. The organic layer was washed with
brine, dried (MgSO4), and concentrated in vacuo to give the
methanesulfonate (62.7 mg, 100%) as an oil. This material was
used immediately without further purification.
d, J ) 9 Hz), 7.68 (1H, s); MS (ESI, m/z) 485 (M + H)+; [R]28
D
+12.7° (c 0.50, EtOH).
1-[(R)-4-(6-Bu toxyin d ol-1-yl)-1-h yd r oxy-2-bu tyl]im id a -
zole-4-ca r boxa m id e (6d ). Replacing 15c with 25d and
following the same procedure as in the preparation of 4c gave
6d (40.8 mg, 92.6%) as a white solid; mp 176-178 °C; IR (KBr,
cm-1) 3317, 3236, 3163, 1676, 1238; 1H NMR (DMSO-d6) δ 0.95
(3H, t, J ) 7 Hz), 1.35-1.85 (4H, m), 2.10-2.40 (2H, m), 3.50-
4.20 (7H, m), 5.05 (1H, brs), 6.33 (1H, d, J ) 3 Hz), 6.65 (1H,
dd, J ) 9 Hz, 2 Hz), 6.73 (1H, d, J ) 2 Hz), 7.07 (1H, brs),
7.11 (1H, d, J ) 3 Hz), 7.27 (1H, brs), 7.38 (1H, d, J ) 9 Hz),
Under N2 atmosphere, to a solution of 23 (80 mg, 0.187
mmol) in DMF (4 mL) was added potassium carbonate (38.7
mg, 0.280 mmol) at room temperature. The reaction mixture
was stirred for 20 min, the methanesulfonate prepared above
in DMF (2 mL) was then added, and the resulting mixture
was stirred for 24 h at 80 to 90 °C. The reaction mixture was
poured into water (20 mL) and extracted with EtOAc. The
extract was washed with brine, dried (Na2SO4), and evaporated
in vacuo. The residue was purified by silica gel (4 g) chroma-
tography eluting with CHCl3/MeOH (100:1 to 10:1) to give 25a
7.70 (1H, s), 7.80 (1H, s); MS (ESI, m/z) 371 (M + H)+; [R]27
+21.5° (c 0.50, MeOH); Anal. (C20H26N4O3) C, H, N.
D
1-[(R)-1-(ter t-Bu tyld im eth ylsilyloxy)-4-(6-h exyloxyin -
d ol-1-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (25c). Replac-
ing 1-iodobutane with 1-iodohexane and following the same
procedure as in the preparation of 25d gave 25c (72.9 mg,
76.1%). IR (neat, cm-1) 3700-3000, 1666, 1254; 1H NMR
1
(93.6 mg, 81.5%). IR (neat, cm-1) 3600-3000, 1666, 1250; H
NMR (CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.80 (9H, s), 2.00-

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3741
(CDCl3) δ -0.10 (3H, s), -0.08 (3H, s), 0.81 (9H, s), 0.92 (3H,
t, J ) 7 Hz), 1.20-1.95 (8H, m), 2.20-2.60 (2H, m), 3.55-
4.25 (7H, m), 5.40 (1H, brs), 6.43 (1H, d, J ) 3 Hz), 6.59 (1H,
d, J ) 2 Hz), 6.78 (1H, d, J ) 3 Hz), 6.80 (1H, dd, J ) 9 Hz,
2 Hz), 6.97 (1H, brs), 7.38 (1H, s), 7.49 (1H, d, J ) 9 Hz), 7.68
9 Hz, 2 Hz), 7.15-7.60 (8H, m), 7.86 (1H, d, J ) 16 Hz); MS
(ESI, m/z) 336 (M + H)+.
E t h yl 3-(5-b en zyloxy-1-m et h ylin d ol-3-yl)p r op ion a t e
(29). To a solution of 28 (11.7 g, 34.9 mmol) in EtOH‚THF
(300 mL, 2:1) was added 10% palladium on carbon (2.34 g),
and the mixture was stirred under hydrogen (2 atm) for 1.5 h
at room temperature. The mixture was filtered through Celite
and washed with THF. The filtrate was concentrated in vacuo
(1H, s); MS (ESI, m/z) 513 (M + H)+; [R]28 +10.4° (c 0.50,
D
EtOH).
1-[(R)-4-(6-Hexyloxyin d ol-1-yl)-1-h yd r oxy-2-bu tyl]im i-
d a zole-4-ca r boxa m id e (6c). Replacing 15c with 25c and
following the same procedure as in the preparation of 4c gave
6c (39.3 mg, 78.5%) as a white solid; mp 120-123 °C; IR (KBr,
cm-1) 3500-3000, 1662, 1252; 1H NMR (DMSO-d6) δ 0.89 (3H,
t, J ) 7 Hz), 1.15-1.55 (6H, m), 1.60-1.85 (2H, m), 2.10-
2.40 (2H, m), 3.50-4.20 (7H, m), 5.04 (1H, brs), 6.33 (1H, d, J
) 3 Hz), 6.64 (1H, dd, J ) 9 Hz, 2 Hz), 6.74 (1H, d, J ) 2 Hz),
7.06 (1H, brs), 7.11 (1H, d, J ) 3 Hz), 7.26 (1H, brs), 7.38 (1H,
d, J ) 9 Hz), 7.70 (1H, s), 7.80 (1H, s); MS (ESI, m/z) 399 (M
to give 29 (10.6 g, 90.0%) as a pale-yellow oil. IR (neat, cm-1
)
1730, 1236; 1H NMR (CDCl3) δ 1.25 (3H, t, J ) 7 Hz), 2.66
(2H, t, J ) 8 Hz), 3.04 (2H, t, J ) 8 Hz), 3.71 (3H, s), 4.14 (2H,
q, J ) 7 Hz), 5.11 (2H, s), 6.84 (1H, s), 6.96 (1H, dd, J ) 9 Hz,
2 Hz), 7.12 (1H, d, J ) 2 Hz), 7.18 (2H, d, J ) 9 Hz), 7.30-
7.55 (5H, m); MS (ESI, m/z) 338 (M + H)+.
3-(5-Ben zyloxy-1-m et h ylin d ol-3-yl)p r op ion a ld eh yd e
(30). Under nitrogen, to a stirred solution of 29 (12.0 g, 35.6
mmol) in CH2Cl2 (120 mL) was added dropwise 1.0 M DIBAL-H
in hexane (37.4 mL, 35.6 mmol) at -78 °C (dry ice/acetone)
for 25 min. After 30 min, MeOH (5.7 mL) was added dropwise
at -78 °C, and the mixture was stirred at room temperature
for 30 min. The mixture was filtered through Celite and
washed with CH2Cl2. The filtrate was concentrated in vacuo,
and then the residue was purified by silica gel (90 g) chroma-
tography eluted with toluene/EtOAc (10:1 to 5:1) to give 30
+ H)+; [R]28 +18.9° (c 0.50, EtOH); Anal. (C22H30N4O3) C, H,
D
N.
1-{(R)-1-(ter t-Bu t yld im et h ylsilyloxy)-4-[6-(3-(4-ch lo-
r op h en yl)p r op oxy)in d ol-1-yl]-2-b u t yl}im id a zole-4-ca r -
b oxa m id e (25e). Replacing 3-phenylpropanol with 3-(4-
chlorophenyl) propanol and following the same procedure as
in the preparation of 25a gave 25e (90.5 mg, 83.4%). IR (neat,
cm-1) 3700-3000, 1668, 1254, 1093; 1H NMR (CDCl3) δ -0.10
(3H, s), -0.08 (3H, s), 0.81 (9H, s), 2.00-2.55 (4H, m), 2.83
(2H, t, J ) 8 Hz), 3.60-4.25 (7H, m), 5.38 (1H, brs), 6.44 (1H,
d, J ) 3 Hz), 6.54 (1H, d, J ) 2 Hz), 6.75-6.90 (2H, m), 6.96
(1H, brs), 7.10-7.35 (5H, m), 7.38 (1H, s), 7.49 (1H, d, J ) 9
1
(8.03 g, 77.0%). IR (KBr, cm-1) 1712, 1223; H NMR (CDCl3)
δ 2.80 (2H, t, J ) 7 Hz), 3.05 (2H, t, J ) 7 Hz), 3.71 (3H, s),
5.12 (2H, s), 6.82 (1H, s), 6.97 (1H, dd, J ) 9 Hz, 2 Hz), 7.09
(1H, d, J ) 2 Hz), 7.15-7.55 (6H, m), 9.83 (1H, s); MS (ESI,
m/z) 294 (M + H)+.
Hz), 7.67 (1H, s); MS (ESI, m/z) 581 (M + H)+; [R]28 +7.7° (c
D
5-Ben zyloxy-3-(3-bu ten yl)-1-m eth ylin d ole (31). Meth-
yltriphenylphosphonium bromide (Ph3PMeBr) (13.6 g, 37.9
mmol) was dried in vacuo for 2 h and then suspended with
stirring in THF (150 mL) under nitrogen. The mixture was
cooled to -78 °C, and 1.5 M n-BuLi in THF (25.3 mL, 37.9
mmol) was added dropwise. After stirring for 2 h while
warming to room temperature, the mixture was again cooled
to -78 °C, and 30 (7.95 g, 27.1 mmol) in THF (50 mL) was
added dropwise. After 2 h at room temperature, the reaction
was quenched by addition of aqueous ammonium chloride. This
mixture was then extracted with EtOAc, and the extracts were
washed with brine, dried (MgSO4), and evaporated in vacuo.
The residue was purified by silica gel (120 g) chromatography
eluting with hexane/EtOAc (10:1) to give 31 (6.92 g, 87.7%)
as a white solid. IR (KBr, cm-1) 1223; 1H NMR (CDCl3) δ 2.43
(2H, q, J ) 7 Hz), 2.79 (2H, t, J ) 7 Hz), 3.71 (3H, s), 4.90-
5.20 (4H, m), 5.80-6.05 (1H, m), 6.82 (1H, s), 6.95 (1H, dd, J
) 9 Hz, 2 Hz), 7.12 (1H, d, J ) 2 Hz), 7.18 (1H, d, J ) 9 Hz),
7.30-7.55 (5H, m); MS (ESI, m/z) 292 (M + H)+.
0.50, EtOH).
1-{(R)-4-[6-(3-(4-Ch lor op h en yl)p r op oxy)in d ol-1-yl]-1-
h yd r oxy-2-bu tyl}im id a zole-4-ca r boxa m id e (6e). Replacing
15c with 25e and following the same procedure as in the
preparation of 4c gave 6e (61.6 mg, 93.6%) as an amorphous
1
solid. IR (KBr, cm-1) 3600-3000, 1658, 1250, 1088; H NMR
(DMSO-d6) δ 1.90-2.40 (4H, m), 2.77 (2H, t, J ) 8 Hz), 3.50-
4.20 (7H, m), 5.05 (1H, brs), 6.34 (1H, d, J ) 3 Hz), 6.67 (1H,
dd, J ) 9 Hz, 2 Hz), 6.70 (1H, d, J ) 2 Hz), 7.08 (1H, brs),
7.12 (1H, d, J ) 3 Hz), 7.15-7.45 (6H, m), 7.70 (1H, s), 7.80
(1H, s); MS (ESI, m/z) 467 (M + H)+; [R]28 +16.6° (c 0.50,
D
EtOH); Anal. (C25H27ClN4O3) C, H, N.
1-[(R)-1-Hyd r oxy-4-(5-(3-p h en ylp r op oxy)in d ol-1-yl)-2-
bu tyl]im id a zole-4-ca r boxa m id e (7a ). Replacing 15c with
26a and following the same procedure as in the preparation
of 4c gave 7a (51.3 mg, 76.3%) as a white solid; mp 117-119
°C; IR (KBr, cm-1) 3600-3000, 1658, 1238; 1H NMR (DMSO-
d6) δ 1.90-2.40 (4H, m), 2.76 (2H, t, J ) 8 Hz), 3.59 (2H, t, J
) 5 Hz), 3.80-4.20 (5H, m), 5.03 (1H, t, J ) 5 Hz), 6.32 (1H,
d, J ) 3 Hz), 6.77 (1H, dd, J ) 9 Hz, 2 Hz), 6.95-7.40 (10H,
m), 7.69 (1H, s), 7.78 (1H, s); MS (ESI, m/z) 433 (M + H)+;
(S)-4-(5-Ben zyloxy-1-m et h ylin d ol-3-yl)b u t a n e-1,2-d i-
ol (32). To a mixture of tert-butyl alcohol (124 mL) and water
(124 mL) was added AD-mix-R (34.6 g). The mixture was
stirred at room temperature until both phases were clear and
then cooled to 0 °C. Compound 31 (7.20 g, 24.7 mmol) was
added, and the heterogeneous slurry was stirred vigorously
at 0 °C for 4 h. With stirring at 0 °C, solid sodium sulfite (37.1
g) was then added and the mixture allowed to warm to room
temperature and stirred for 1 h. This mixture was extracted
with EtOAc, and the extracts were washed with brine, dried
(MgSO4), and evaporated in vacuo. The residue was purified
by silica gel (215 g) chromatography eluting with CHCl3/MeOH
(100:1 to 20:1) to give 32 (4.30 g, 53.5%). IR (KBr, cm-1) 3365,
[R]27 +37.0° (c 0.50, MeOH); Anal. (C25H28N4O3) C, H, N.
D
1-[(R)-4-(5-Hexyloxyin d ol-1-yl)-1-h yd r oxy-2-bu tyl]im i-
d a zole-4-ca r boxa m id e (7b). Replacing 15c with 26b and
following the same procedure as in the preparation of 4c gave
7b (38.2 mg, 92.7%) as a white solid; mp 103-106 °C; IR (KBr,
cm-1) 3600-3000, 1658, 1238; 1H NMR (DMSO-d6) δ 0.88(3H,
t, J ) 7 Hz), 1.10-1.85 (8H, m), 2.10-2.40 (2H, m), 3.59 (2H,
t, J ) 5 Hz), 3.80-4.20 (5H, m), 5.04 (1H, t, J ) 5 Hz), 6.32
(1H, d, J ) 3 Hz), 6.74 (1H, dd, J ) 9 Hz, 2 Hz), 6.90-7.40
(5H, m), 7.69 (1H, s), 7.78 (1H, s); MS (ESI, m/z) 399 (M +
1
H)+; [R]28 +35.9° (c 0.50, EtOH); Anal. (C22H30N4O3) C, H, N.
1227, 1055; H NMR (CDCl3) δ 1.65-1.95 (3H, m), 2.03 (1H,
D
d, J ) 4 Hz), 2.70-3.00 (2H, m), 3.35-3.90 (6H, m), 5.12 (2H,
s), 6.83 (1H, s), 6.96 (1H, dd, J ) 9 Hz, 2 Hz), 7.11 (1H, d, J )
2 Hz), 7.18 (1H, d, J ) 9 Hz), 7.30-7.55 (5H, m); MS (ESI,
Eth yl 3-(5-Ben zyloxy-1-m eth ylin d ol-3-yl)a cr yla te (28).
To an ice-cooled solution of 5-benzyloxy-1-methylindole-3-
carboxaldehyde (13.0 g, 49.0 mmol) and triethyl phosphono-
acetate (13.2 g, 58.8 mmol) in DMF (100 mL) was added NaH
(60% in mineral oil, 2.74 g, 68.6 mmol). The mixture was
stirred overnight at room temperature. The reaction mixture
was poured into ice-water (700 mL) and stirred 30 min. The
resulting precipitates were collected by filtration, washed with
water, and dried in vacuo to give 28 (14.2 g, 86.3%) as a pale
yellow solid. IR (KBr, cm-1) 1703, 1280, 1228, 1176; 1H NMR
(CDCl3) δ 1.36 (3H, t, J ) 7 Hz), 3.78 (3H, s), 4.27 (2H, q, J )
7 Hz), 5.15 (2H, s), 6.30 (1H, d, J ) 16 Hz), 7.03 (1H, dd, J )
m/z) 326 (M + H)+; [R]24 -21.6° (c 0.50, EtOH).
D
(S)-4-(5-Ben zyloxy-1-m eth ylin d ol-3-yl)-1-(ter t-bu tyld i-
m eth ylsilyloxy)bu ta n -2-ol (33). To an ice-cooled solution of
32 (4.25 g, 13.1 mmol) in DMF (40 mL) was added imidazole
(2.67 g, 39.2 mmol) followed by TBSCl (2.07 g, 13.7 mmol).
After 30 min, the ice-bath was removed, and then the mixture
was stirred overnight at room temperature. The reaction
mixture was poured into water (100 mL) and extracted with
EtOAc. The organic layer was washed with brine, dried

3742 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15
Terasaka et al.
(MgSO4), and concentrated in vacuo. The residue was purified
by silica gel (105 g) column chromatography eluting with
hexane/EtOAc (20:1) to give 33 (4.85 g, 84.5%) as a colorless
was 10.3%, the ratio I/σ(I) was 4.5, and the data 99.7%
complete. Corresponding values for the high-resolution data
shell (2.66-2.50 Å) are 28.9%, 2.6, and 98.5%, respectively.
The structure of the 4c/ADA complex was solved and refined
using these data, program AMoRe28 and X-PLOR (Accelrys),
and the protein model from the 4e/ADA complex. The conven-
tional and free R-factors after refinement are 22.6% and 28.6%.
The rms deviations between model and ideal bond distances,
bond angles, dihedral angles, and improper angles are 0.036
Å, 5.3°, 27.8°, and 3.98°. The model coordinates have been
deposited in the Protein Data Bank with the code 1UML.
1
oil. IR (neat, cm-1) 3458, 1252; H NMR (CDCl3) δ 0.07 (6H,
s), 0.90 (9H, s), 1.65-1.90 (2H, m), 2.44 (1H, d, J ) 4 Hz),
2.65-3.05 (2H, m), 3.30-3.85 (6H, m), 5.11 (2H, s), 6.83 (1H,
s), 6.95 (1H, dd, J ) 9 Hz, 2 Hz), 7.13 (1H, d, J ) 2 Hz), 7.18
(1H, d, J ) 9 Hz), 7.30-7.55 (5H, m); MS (ESI, m/z) 440 (M +
H) +; [R]23 -25.5° (c 0.50, EtOH).
D
1-[(R)-1-Hydr oxy-4-(1-m eth yl-5-(3-ph en ylpr opoxy)in dol-
3-yl)-2-bu tyl]im id a zole-4-ca r boxa m id e (8a ). Replacing 15c
with 36a and following the same procedure as in the prepara-
tion of 4c gave 8a (26.2 mg, 99.8%) as a white solid; mp 116-
119 °C; IR (KBr, cm-1) 3700-3000, 1653, 1236; 1H NMR
(DMSO-d6) δ 1.90-2.20 (4H, m), 2.30-2.60 (2H, m), 2.77 (2H,
t, J ) 8 Hz), 3.50-3.80 (5H, m), 3.95 (2H, t, J ) 6 Hz), 4.15
(1H, m), 4.99 (1H, t, J ) 5 Hz), 6.78 (1H, dd, J ) 9 Hz, 2 Hz),
6.86 (1H, d, J ) 2 Hz), 6.90-7.40 (9H, m), 7.73 (1H, s), 7.77
Cr ysta llogr a p h y of 5/ADA. Cocrystals of the bovine
intestine ADA were obtained using procedures similar to the
4e/ADA complex. X-ray diffraction data were collected from
this P43212 form: a ) 78.44 Å, c ) 137.71 Å, at SP-ring8
beamline 24XU. Data resolution was from 51.75 to 2.35 Å;
238710 observations were scaled and merged into 19201
unique reflections using Crystal Clear (RIGAKU). The overall
R-merge was 7.4%, the ratio I/σ(I) was 2.0, and the data 98.9%
complete. Corresponding values for the high-resolution data
shell (2.50-2.35 Å) are 35.2%, 6.3, and 99.4%, respectively.
The structure of the 5/ADA complex was solved and refined
using these data, program AMoRe28 and X-PLOR (Accelrys),
and the protein model from the 4e/ADA complex. The conven-
tional and free R-factors after refinement are 22.8% and 24.2%.
The rms deviations between model and ideal bond distances,
bond angles, dihedral angles, and improper angles are 0.042
Å, 5.0°, 26.7°, and 4.90°. The model coordinates have been
deposited in the Protein Data Bank with the code 1O5R.
(1H, s); MS (ESI, m/z) 447 (M + H) +; [R]27 +24.6° (c 0.50,
D
EtOH); Anal. (C26H30N4O3?0.3H2O) C, H, N.
1-[(R)-4-(5-Hexyloxy-1-m eth ylin d ol-3-yl)-1-h yd r oxy-2-
bu tyl]im id a zole-4-ca r boxa m id e (8b). Replacing 15c with
36b and following the same procedure as in the preparation
of 4c gave 8b (38.8 mg, 93.5%) as a white solid; mp 136-139
°C; IR (KBr, cm-1) 3600-3000, 1676, 1227; 1H NMR (DMSO-
d6) δ 0.88 (3H, t, J ) 7 Hz), 1.10-1.55 (6H, m), 1.60-1.80 (2H,
m), 2.00-2.20 (2H, m), 2.35-2.60 (2H, m), 3.50-3.80 (5H, m),
3.93 (2H, t, J ) 6 Hz), 4.15 (1H, m), 4.99 (1H, t, J ) 5 Hz),
6.75 (1H, dd, J ) 9 Hz, 2 Hz), 6.86 (1H, d, J ) 2 Hz), 6.90-
7.35 (4H, m), 7.72 (1H, s), 7.77 (1H, s); MS (ESI, m/z) 413 (M
Dock in g Sim u la tion . Inhibitor molecules were modeled
based on the active conformation of 1, taken from the crystal
structure of 1/ADA complex, using InsightII(Accelrys). The
modeled structures were energy minimized by Discover-
(Accelrys) adopting the cvff force field parameters. The initial
binding status of each inhibitor molecule was simulated using
GREEN(IMMD) in the following way: the modeled compound
was placed at the active site manually referring to the crystal
structure of 1/ADA complex, and energetically stable confor-
mations were explored using the Monte Carlo simulation mode
considering the flexibility of the rotatable bonds. The binding
mode which gave the most stable interaction energy was
adopted as initial binding status. This binding status was
further minimized by Discover(Accelrys) using the following
protocol: main chain atoms of ADA were set fixed to the
original position during the minimization, positions of side
chain atoms of the residues coordinating to the zinc atom, that
is His15, His17, His214 and Asp295, were also fixed. In
addition, one water molecule that coordinates to Zn and two
water molecules that form hydrogen bonds with the carboxa-
mide moiety of 1, were fixed. Other non-hydrogen atoms were
tethered at the original position with a force constant of 25
kcal/Å. The minimization of the whole complex structure was
carried out by 50 steps of the steepest descent method followed
by 10000 steps of the conjugate gradient method until the
maximum derivative is less than 0.1 kcal/Å. The interaction
energies between ADA and inhibitor molecules are obtained
by InsightII(Accelrys) based on the minimized complex struc-
tures.
Ra t P h a r m a cok in etic Stu d y. Male Sprague-Dawley
rats, weighing between 300 and 350 g, were used. The animals
were housed in a controlled room at 20-26 °C with a 12 h
light/dark cycle. Food and water were provided ad libitum.
After overnight fasting, rats were used for pharmacokinetic
experiments. For intravenous dosing, compound 8c solution
was injected into the femoral vein at a dose of 10 mg/kg to
rats. For oral dosing, the solution was given by gavage at a
dose of 10 mg/kg to rats. Blood samples were collected at 0.083,
0.25, 0.5, 1, 2, 4, 6, and 8 h after intravenous administration
and 0.25, 0.5, 1, 2, 4, 6, and 8 h after oral administration. Blood
samples were collected in heparinized tubes and then centri-
fuged to separate plasma samples. All samples were stored
on ice until analyzed. Plasma samples were added into
methanol to precipitate protein and centrifuged. Aliquots of
the supernatant were injected into HPLC for assay.
+ H) +; [R]27 +30.3° (c 0.50, MeOH); Anal. (C23H32N4O3) C,
D
H, N.
1-{(R)-4-[5-(3-(4-Ch lor oph en yl)pr opoxy)-1-m eth ylin dol-
3-yl]-1-h yd r oxy-2-bu t yl}im id a zole-4-ca r boxa m id e (8c).
Replacing 15c with 36c and following the same procedure as
in the preparation of 4c gave 8c (228 mg, 70.5%) as an
amorphous solid. IR (KBr, cm-1) 3600-3000, 1658, 1232, 1082;
1H NMR (DMSO-d6) δ 1.90-2.25 (4H, m), 2.30-2.60 (2H, m),
2.77 (2H, t, J ) 8 Hz), 3.55-3.80 (5H, m), 3.94 (2H, t, J ) 6
Hz), 4.15 (1H, m), 4.99 (1H, t, J ) 5 Hz), 6.77 (1H, dd, J ) 9
Hz, 2 Hz), 6.85 (1H, d, J ) 2 Hz), 6.95-7.45 (8H, m), 7.72
(1H, s), 7.77 (1H, s); MS (ESI, m/z) 481 (M + H)+ (35Cl), 483
(M + H)+ (37Cl); [R]27 +21.3° (c 0.50, EtOH); Anal. (C26H29
-
D
ClN4O3‚0.3H2O) C, H, N.
Biologica l Meth od s. Expression and purification of human
recombinant ADA as well as in vitro ADA enzyme assay have
been described previously.19
Cr ysta llogr a p h y of 4e/ADA. Cocrystals of the bovine
intestine ADA were obtained using procedures similar to those
previously described.27 In this study, 100 mM MES, pH 6.0,
2.1 M ammonium sulfate, and 2.5% (v/v) PEG-400 were used
as precipitant. X-ray diffraction data were collected from this
P43212 form: a ) 78.37 Å, c ) 137.82 Å, at SP-ring8 beamline
24XU. Data resolution was from 51.75 to 2.25 Å; 137589
observations were scaled and merged into 20752 unique
reflections using Crystal Clear (RIGAKU). The overall R-merge
was 9.0%, the ratio I/σ(I) was 8.7, and the data 98.1% complete.
Corresponding values for the high-resolution data shell (2.35-
2.25 Å) are 33.8%, 3.0, and 96.6%, respectively. The structure
of the 4e/ADA complex was solved and refined using these
data, program AMoRe28 and X-PLOR (Accelrys), and the
protein model 1KRM from the Protein Data Bank.29 The
conventional and free R-factors after refinement are 23.2% and
25.3%. The rms deviations between model and ideal bond
distances, bond angles, dihedral angles, and improper angles
are 0.021 Å, 3.9°, 26.7°, and 2.98°. The model coordinates have
been deposited in the Protein Data Bank with the code 1QXL.
Cr ysta llogr a p h y of 4c/ADA. Cocrystals of the bovine
intestine ADA were obtained using procedures similar to the
4e/ADA complex. X-ray diffraction data were collected from
this P43212 form: a ) 78.32 Å, c ) 138.32 Å, at SP-ring8
beamline 24XU. Data resolution was from 37.68 to 2.5 Å;
89568 observations were scaled and merged into 15505 unique
reflections using Crystal Clear (RIGAKU). The overall R-merge

Non-nucleoside Adenosine Deaminase Inhibitors
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 15 3743
(13) Lathia, C.; Fleming, G. F.; Meyer, M.; Ratain, M. J .; Whitfield,
L. Pentostatin pharmacokinetics and dosing recommendations
in patients with mild renal impairment. Cancer Chemother.
Pharmacol. 2002, 50, 121-126.
Ack n ow led gm en t. We thank Dr. David Barrett,
Fujisawa Pharmaceutical Co. Ltd., for valuable com-
ments and help in the preparation of the manuscript.
(14) (a) Baker, D. C.; Hanvey, J . C.; Hawkins, L. D.; Murphy, J .
Identification of the bioactive enantiomer of erythro-3-(adenine-
9-yl)-2-nonanol (EHNA), a semi-tight binding inhibitor of ad-
enosine deaminase. Biochem. Pharmacol. 1981, 30, 1159-1160.
(b) Bessodes, M.; Bastian, G.; Abushanab, E.; Panzica, R. P.;
Berman, S. F.; Marcaccio, E. J .; Chen, S. F.; Stoeckler, J . D.;
Parks, R. J . Effect of chirality in erythro-9-(2-hydroxy-3-nonyl)
adenine (EHNA) on adenosine deaminase inhibition. Biochem.
Pharmacol. 1982, 31, 879-882.
(15) Cristalli, G.; Eleuteri, A.; Franchetti, P.; Grifantini, M.; Vittori.
S.; Lupidi, G. Adenosine deaminase inhibitors: Synthesis and
structure-activity relationships of imidazole analogue of erythro-
9-(2-hydroxy-3-nonyl) adenine. J . Med. Chem. 1991, 34, 1187-
1192.
(16) Pragnacharyulu, P. V. P.; Varkhedkar, V.; Curtis, M. A.; Chang,
I. F.; Abushanab, E. Adenosine deaminase inhibitors: Synthesis
and biological evaluation of unsaturated, aromatic, and oxo
derivatives of (+)-erythro-9-(2′S-hydroxy-3′R-nonyl) adenine [(+)-
EHNA]. J . Med. Chem. 2000, 43, 4694-4700.
(17) McConnell, W. R.; El-Dareer, S. M.; Hill, D. L. Metabolism and
disposition of erythro-9-(2-hydroxy-3-nonyl)[14C] adenine in the
rhesus monkey. Drug Metab. Dispos. 1980, 8, 5-7.
Su p p or tin g In for m a tion Ava ila ble: Description of the
synthesis and experimental data for the preparation of com-
pounds 17-22, 24, 26, and 34-36. This material is available
free of charge via the Internet at http://pubs.acs.org.
Refer en ces
(1) Cronstein. B. N. Adenosine, An endogenous antiinflammatory
agent. J . Appl. Physiol. 1994, 76, 5-13.
(2) Ohta, A.; Sitkovsky, M. Role of G-protein-coupled adenosine
receptors in downregulation of inflammation and protection from
tissue damage. Nature 2001, 414, 916-920.
(3) (a) Rudolphi, K. A.; Schubert, P.; Parkinson, F. E.; Fredholm,
B. B. Neuroprotective role of adenosine in cerebral ischemia.
Trends. Pharmacol. Sci. 1992, 13, 439-445. (b) Marquardt, D.
L.; Gruber, H. E.; Wasserman, S. I. Adenosine release from
stimulated mast-cells. Proc. Natl. Acad. Sci. U.S.A. 1984, 81,
6192-6196. (c) Filippini, A.; Taffs, R. E.; Sitkovsky, M. V.
Extracellular ATP in T-lymphocyte activation: Possible role in
effector functions. Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 8267-
8271.
(4) (a) Cronstein, B. N.; Naime, D.; Ostad, E. The antiinflammatory
mechanism of Methotrexate. Increased adenosine release at
inflamed sites diminishes leukocyte accumulation in an in vivo
model of inflammation. J . Clin. Invest. 1993, 92, 2675-2682.
(b) Cronstein, B. N.; Montesinos, M. C.; Weissmann, G. Sites of
action for future therapy: an adenosine-dependent mechanism
by which aspirin retains its antiinflammatory activity in cy-
clooxygenase-2 and NFκB knockout mice. Osteoarthritis Carti-
lage 1999, 7, 361-363. (c) Cronstein, B. N.; Montesinos, M. C.;
Weissmann, G. Salycylates and sulfasalazine but not glucocor-
ticoids, inhibit leukocyte accumulation by an adenosine-depend-
ent mechanism that is independent of inhibition of prostaglandin
synthesis and p105 of NFkB. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 6377-6381.
(18) Lambe, C. U.; Nelson, D. J . Pharmacokinetics of inhibition of
adenosine deaminase by erythro-9-(2-hydroxy-3-nonyl) adenine
in CBA mice. Biochem. Pharmacol. 1982, 31, 535-539.
(19) Terasaka, T.; Nakanishi, I.; Nakamura, K.; Eikyu, Y.; Kinoshita,
T.; Nishio, N.; Sato, A.; Kuno, M.; Seki, N.; Sakane, K. Structure-
based de novo design of non-nucleoside adenosine deaminase
inhibitors. Bioorg. Med. Chem. Lett. 2003, 13, 1115-1118.
(20) Terasaka, T.; Kinoshita,T.; Kuno, M.; Nakanishi, I. A highly
potent non-nucleoside adenosine deaminase inhibitor: Efficient
drug discovery by intentional lead hybridization. J . Am. Chem.
Soc. 2004, 126, 34-35.
(21) Agerwal, R. P. Inhibitors of adenosine deaminase. Pharmacol.
Ther. 1982, 17, 399-429.
(22) Alberg, D. G.; Schreiber, S. L. Structure-based design of a
cyclophilin-calcineurin bridging ligand. Science 1993, 262, 248-
250.
(23) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J .
Experimental and computational approaches to estimate solubil-
ity and permeability in drug discovery and development settings.
Adv. Drug Delivery Rev. 1997, 23, 3-25.
(24) Smith, D. A.; J ones, B. C.; Walker, D. K. Design of drugs
involving the concepts and theories of drug metabolism and
pharmacokinetics. Med. Res. Rev. 1996, 16, 243-266.
(25) Terasaka, T.; Seki, N.; Tsuji, K.; Nakanishi, I.; Kinoshita,T.;
Nakamura, K. WO 00/55155, 2000.
(26) Sharpless, K. B.; Amberg, W.; Bennani, Y. L.; Crispino, G. A.;
Hartung, J .; J eong, K. S.; Kwong, H. L.; Morikawa, K.; Wang,
Z. M.; Xu, D. Q.; Zhang, X. L. The osmium-catalyzed asymmetric
dihydroxylation: A new ligand class and a process improvement.
J . Org. Chem. 1992, 57, 2768-2771.
(27) Kinoshita, T.; Nishio, N.; Sato, A.; Murata, M. Crystallization
and preliminary analysis of bovine adenosine deaminase. Acta
Crystallogr. 1999, D55, 2031-2032.
(28) Navaza, J . On the computation of the fast rotation function. Acta
Crystallogr. 1993, D49, 588-591.
(29) Kinoshita, T.; Nishio, N.; Nakanishi, I.; Sato, A.; Fujii, T. Crystal
structure of bovine adenosine deaminase complexed with 6-hy-
droxyl-1,6-dihydropurine riboside. Acta Crystallogr. 2003, D59,
299-303.
(5) Cristalli, G.; Costanzi, S.; Lambertucci, C.; Lupidi, G.; Vittori,
S.; Volpini, R.; Camaioni, E. Adenosine deaminase: Functional
implications and different classes of inhibitors. Med. Res. Rev.
2001, 21, 105-128.
(6) Resta, R.; Thompson, L. F. SCID: the role of adenosine deami-
nase deficiency. Immunol. Today 1997, 18, 371-374.
(7) Kameoka, J .; Tanaka, T.; Nojima, Y.; Schlossman, S. F.; Morim-
oto, C. Direct association of adenosine deaminase with T cell
activation antigen, CD26. Science 1993, 261, 466-469.
(8) (a) Franco, R.; Valenzuela, A.; Lluis, C.; Blanco, J . Enzymatic
and extraenzymatic role of ecto-adenosine deaminase in lym-
phocytes. Immunol. Rev. 1998, 161, 27-42. (b) Morimoto, C.;
Schlossman, S. F. The structure and function of CD26 in the
T-cell immune response. Immunol. Rev. 1998, 161, 55-70.
(9) Agerwal, R. P.; Spector, T.; Parks, R. E. Tight-binding inhibitors
IV. Inhibition of adenosine deaminase by various inhibitors.
Biochem. Pharmacol. 1977, 26, 359-367.
(10) Rafel, M.; Cervantes, F.; Beltran, J . M.; Zuazu, J .; Nieto, L. H.;
Rayon, C.; Talavera, J . G.; Montserrat, E. Deoxycoformycin in
the treatment of patients with hairy cell leukemia. Cancer 2000,
88, 352-357.
(11) Brogden, R. N.; Sorkin, E. M. Pentostatin. A review of its
pharmacodynamic and pharmacokinetic properties, and thera-
peutic potential in lymphoproliferative disorders. Drugs 1993,
46, 652-677.
(12) McConnell, W. R.; Furner, R. L.; Hill, D. L. Pharmacokinetics
of 2′-deoxycoformycin in normal and L1210 leukemic mice. Drug
Metab. Dispos. 1979, 7, 11-13.
J M0306374