Inhibition of Adenosine Deaminase by Novel 5:7 Fused Heterocycles Containing the Imidazo[4,5-e][1,2,4]triazepine Ring System: A Structure-Activity Relationship Study

Article abstract of DOI:10.1021/jm0304257

As part of a program to explore structure-activity relationships for the extremely tight binding inhibition characteristics of coformycins to adenosine deaminase, a series of analogues (1a-1h) containing the imidazo[4,5-e][1,2,4]triazepine ring system has

Full text of DOI:10.1021/jm0304257

1044
J . Med. Chem. 2004, 47, 1044-1050
In h ibition of Ad en osin e Dea m in a se by Novel 5:7 F u sed Heter ocycles Con ta in in g
th e Im id a zo[4,5-e][1,2,4]tr ia zep in e Rin g System : A Str u ctu r e-Activity
Rela tion sh ip Stu d y
Ayub Reayi and Ramachandra S. Hosmane*
Laboratory for Drug Design and Synthesis, Department of Chemistry and Biochemistry,
University of Maryland, Baltimore County, 1000 Hilltop Circle, Baltimore, Maryland 21250
Received September 2, 2003
As part of a program to explore structure-activity relationships for the extremely tight binding
inhibition characteristics of coformycins to adenosine deaminase, a series of analogues (1a -
1h ) containing the imidazo[4,5-e][1,2,4]triazepine ring system has been synthesized and
screened in vitro against a mammalian adenosine deaminase for inhibitory activity. While
compounds 1a and 1b were synthesized in five steps starting from 4-nitroimidazole, others
were derived from 1a through simple exchange reactions with the appropriate alcohols. The
observed kinetics profiles and Ki values suggest that the target compounds are competitive
inhibitors that bind 6-9 orders of magnitude less tightly to the enzyme. Compounds 1c and
1d were the most active in the series with Ki’s ranging from 12 to 15 µM.
In tr od u ction
cordycepin,26 which would otherwise be hydrolyzed by
ADA into their inactive inosine counterparts.
Adenosine deaminase (ADA, EC 3.5.4.4) catalyzes the
hydrolytic deamination of adenosine and 2′-deoxyad-
enosine to the corresponding inosines.1 ADA is a pivotal
enzyme of purine metabolism, whose genetic deficiency
is associated with a phenotype of severe combined
immunodeficiency disorder (SCID).2 As SCID is caused
by high elevations in the concentration of 2′-deoxyad-
enosine-5′-triphosphate (dATP), which is lymphotoxic,
inhibitors of ADA became the target of many investiga-
tions for the selective treatment of lymphoproliferative
malignancies.3,4 Indeed, the potent antitumor properties
of two natural products, coformycin (1)5-8 and 2′-
The high promise of coformycins in cancer treatment,
however, met severe setbacks in the clinic due to high
incidences of unacceptable toxicities involving liver,
kidney, and the central nervous system.27-29 The long
and nearly irreversible, extremely tight binding inhibi-
tion of intracellular ADA has been given as the logical
explanation for the observed toxicities with the cofor-
mycin therapy.30 The synthesis of a new enzyme mol-
ecule was believed necessary for recovering each time
from the toxic effects of coformycin inhibition.30 There-
fore, an analogue of coformycin that is somewhat less
tight binding to ADA, is readily reversible, and has a
shorter duration of action to allow faster enzyme
recovery will be needed for overcoming the undesirable
toxic effects. While a few synthetic analogues of cofor-
mycin have been reported,31-33 an ideal clinical candi-
date with the desired characteristics is yet to be found.
To this end, our lab has been involved in performing
systematic structure-activity relationship (SAR) stud-
ies in this area and is currently focusing on probing the
molecular basis of tight-binding interactions of cofor-
mycin with ADA.34,35 The outcome of such studies is
believed to provide the much needed guidance for
eventually rational structural modifications of coformy-
cin that would afford to somewhat loosen its extremely
tight protein binding without significantly compromis-
ing its efficacy.
deoxycoformycin (pentostatin) (2),9-14 the strongest
inhibitors of ADA known to-date, with Ki ranging from
10-11 to 10-12 M,15-20 were found to bear a relationship
with their ability to mimic ADA deficiency and act as
immunosuppressants to control cancers associated with
the hyperimmune system, such as leukemia and
lymphoma.21-23 Furthermore, the coformycins were
discovered to cause synergistic effects upon coadminis-
tration with other antitumor compounds that are ana-
logues of adenosine, such as Ara-A,24 formycin,25 or
The single-crystal X-ray structure of a complex of
ADA with 2′-deoxycoformycin (dCF) bound at the active
site,36,37 schematically shown in Figure 1, surprisingly
revealed the absence of any H-bonding interactions of
the heterocyclic ring nitrogen atoms of the ligand with
the protein. As a matter of fact, the only conspicuous
interaction of the heterocycle is a coordination bond
between the 8-OH group of dCF and the active site zinc
of ADA.36,37 However, the crystal structure of the
complex did show the existence of several hydrogen
* To whom the correspondence should be addressed. Tel: (410) 455-
2520. Fax: (410) 455-1148. E-mail: hosmane@umbc.edu.
10.1021/jm0304257 CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/13/2004

Inhibition of Adenosine Deaminase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 1045
Sch em e 1
structure-activity relationship (SAR) studies,34,35 which
indicated that the complete removal of the sugar moiety
of coformycins leads to total inactivity against ADA,
while the substitution of a benzyl group in place of the
sugar retains a good part of the original activity. We
also speculated that an aralkyl group at position-1 may
be preferred to position-3, as it would avoid the unfavor-
able hydrophobic interactions of the aromatic ring in a
presumably hydrophilic sink, where the original sugar
hydroxyls of coformycins lay in the active site of ADA.
Furthermore, there exist several hydrophobic residues
in the active site of ADA where the departing molecule
of ammonia exits,36 so the hydrophobic interactions of
those residues with an aralkyl group of 1 should be
favorable. (c) The somewhat weakened nucleophilicity
of the respective 8-OH group by the electron-withdraw-
ing inductive effect of an additional ring nitrogen atom
inserted at position-7 is important. (d) The attachment
of a methyl group at each of the positions 6 and 7 of the
heterocyclic ring is consequential. This is because the
5:7-fused imidazotriazepine ring system of 1 is prone
to facile opportunistic rearrangements to form a 5:6-
fused system,43-46 unless substituted at one or both of
6- and 7-positions of the triazepine ring. For the same
reason, an analogue containing the parent 8-OH func-
tion (1, R ) H), as in coformycins, was excluded from
the target list, as such a compound would constitute the
characteristics of a carbinolamine, which is prone to
facile ring-opening to form an aldehyde derivative of an
imidazole. We present here our SAR studies with eight
analogues of coformycins, a -h , bearing the general
structural features of 1.
F igu r e 1. Schematic representation of important protein-
ligand interactions in the reported36,37 crystal structure of
ADA-dCF complex.
bonds between the sugar hydroxyls of dCF and the
active site amino acid residues of ADA, in particular
His 117 and Asp 19, in addition to H-bonding with an
adventitious molecule of water.
A close inspection of the X-ray structure suggested
that the elimination of H-bonding interactions of the
sugar hydroxyls, coupled with the weakening of the
coordination bond between the 8-OH and zinc, would
considerably loosen the tight-binding interactions of
coformycins with ADA. In this regard, compounds
bearing the general structure 1 were envisioned to fulfill
both of these demands, and are thus the targets of the
present investigation.
Resu lts a n d Discu ssion
Ch em istr y. The common starting material for the
target compounds is 4-amino-1-p-methoxybenzylimida-
zole-5-carbaldehyde (6) (Scheme 1), which was synthe-
sized in four steps starting from the commercially
available 4-nitro-1H-imidazole (2). The latter was con-
densed with p-methoxybenzyl chloride, catalyzed by
potassium carbonate in dimethylformamide (DMF), to
obtain 4-nitro-1-p-methoxybenzylimidazole (3) in >90%
yield. The position of attachment of the p-methoxyben-
zyl group was confirmed by two-dimensional nuclear
Overhauser and exchange spectroscopy (NOESY), which
showed correlation between the methylene protons of
the p-methoxybenzyl substituent and the proton at
position-5. A dichloromethyl group was then introduced
at position-5, employing the vicarious nucleophilic
substitution procedure of Ostrowski.47 Thus, the reac-
tion of 3 with chloroform in the presence of potassium
The important structural features of compounds 1a -
1h include (a) the total absence of the sugar moiety, and
hence, the inability to form the crucial H-bonds associ-
ated with the sugar hydroxyls of coformycins. The sugar
moiety is not absolutely essential for activity, as we had
already discovered from our earlier SAR studies.34,35
While ADA inhibitory activity has also been reported
with many other different classes of compounds besides
nucleosides, including flavanoids,38,39 steroids,40 terpen-
oids,40 and other natural products,41,42 it is not clear if
these compounds bind to the active or allosteric site of
the enzyme. (b) The presence of an aralkyl substituent
at position-1 is significant. This is based on our earlier

1046 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Reayi and Hosmane
Sch em e 2
Sch em e 3
tert-butoxide in DMF at < -60 °C afforded 4 in 53%
yield. Compound 4 was hydrolyzed to the aldehyde 5
by reaction with aqueous calcium carbonate at 65-75
°C in 61% yield. The reduction of 5 to the desired 6 was
accomplished by catalytic hydrogenation over 5% Pd on
carbon in methanol in 34% yield. Although the use of
ethyl acetate as the solvent medium for this reaction
gave even lower yield than the one with methanol, the
product was cleaner and did not require any chromato-
graphic purification.
The target compounds 1a and 1b were synthesized
in 62% and 30% yields, respectively, in a single-pot
reaction (Scheme 2) of 6 with 1,2-dimethylhydrazine
dihydrochloride and trimethyl or triethyl orthoformate
in the corresponding alcohol used as a solvent. The
reaction pathway to the products is believed to involve
the initial formation of the imidate 7, which upon ring-
closure would form an equilibrium mixture containing
the aminol-iminium species (8). The latter upon reac-
tion with the alcohol will form the target imidazotriaz-
epinols 1a and 1b. In corroboration of this mechanism
is the isolation of 7 in a separate reaction of 6 with
triethyl orthoformate at reflux, whether in the presence
or absence of an acid catalyst.
benzyl alcohol, and 3-methylbenzyl alcohol, respectively,
under catalysis by trifluoroacetic acid at room temper-
ature, in yields of 32%-77%.
The above exchange reactions could also be carried
out without the acid catalysis by simply heating the
mixture of 1a in the appropriate alcohol at reflux
temperature. This observation, however, raised some
concern about the stability of the final products in an
aqueous medium that is to be employed for the intended
ADA inhibition studies. The iminium ion intermediate,
if formed at room temperature even without the cataly-
sis, could produce an aminol (9) (Scheme 4) upon
reaction with water and would subsequently ring-open
to produce an aldehyde (10). We were relieved to find
out that such a transformation does not occur without
the acid catalysis, as shown in Scheme 3, or without
1
heating at high temperature. The H NMR spectrum of
1a in a 4:1 mixture of D2O-DMSO-d6 exhibited no
change, even after standing at room temperature for
more than 20 h.
The target compounds 1a -1h are somewhat unstable,
especially when the trace solvent molecules that tend
to adhere to the compounds are completely stripped off
by elaborate drying. However, the compounds are
reasonably stable as a solution in dimethyl sulfoxide
and/or ethanol or when allowed to retain residual
amounts of these solvents during the purification pro-
cess. Therefore, satisfactory microanalytical data for
these compounds could only be obtained by accounting
for the adventitious solvent molecules. This observation
is also consistent with the reported instability and
elemental microanalyticatical data of the aglycon of
The speculated intermediacy of aminol-iminium spe-
cies 8 in the formation of the target 1a and 1b further
suggested that the remaining targets 1c-1h could
simply be prepared through exchange reactions of 1a
or 1b with the appropriate alcohols. This was indeed
proven to be correct as compounds 1c-1h were all
obtained (Scheme 3) by reactions of 1a with n-propanol,
n-butanol, 2-propanol, 2-methoxyethanol, 4-methoxy-

Inhibition of Adenosine Deaminase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 1047
Sch em e 4
Ta ble 1. Inhibitory Activities of Target Compounds 1a -1h
tent). The increasing inhibitory activity in going from
1a to 1d may well be due to the parallel increase in
electron-donating inductive effect of the alkyl group,
which can render the 8-OR moiety increasingly electron-
rich and, in turn, strengthen the 8-OR-mediated coor-
dination bond between the heterocycle and the active
site zinc of the protein. However, it is also possible that
the increased steric bulk associated with the increasing
inductive effect of the alkyl homologues will offset some
of the gain made in bond strength by electronic factors.
This is evident in the increased Ki of compounds 1e, 1g,
and 1h , where steric interference appears to suppress
the electronic effect. In the case of 1g and 1h , the bad
steric interactions with the p-methoxybenzyl group at
the peri position may further distort and weaken the
zinc coordination. Finally, the replacement of a terminal
methylene group of 1d with an electron-withdrawing
oxygen atom, as in 1f, results in a slight increase of Ki,
as anticipated.
Compounds 1a -1h contain a chiral center at position
8, so each can exist in two stereoisomeric forms, R and
S. In view of the fact that the R isomer of coformycin is
considerably more potent than the S,15 it is similarly
likely that one of the two isomers in each of the racemic
final products reported herein would be more potent
than the other. Nevertheless, no further attempts were
made to separate the two enantiomers from the racemic
mixture, as the observed inhibition was only modest,
coupled with the fact that the thrust of the present
study was rather to explore the molecular basis of tight-
binding interactions of coformycin with ADA than to
discover an ideal inhibitor of the enzyme, per se.
against Calf Spleen Adenosine Deaminasea
compd
Ki (µM)
compd
Ki (µM)
1a
1b
1c
1d
24 ( 0.5
20 ( 0.6
12 ( 5.8
15 ( 1.0
1e
1f
1g
1h
30 ( 5.5
23 ( 5.4
77 ( 24.3
93 ( 30.6
a
Ki values were computed from Lineweaver-Burk plots.
pentostatin, which carried an adventitious molecule of
1
DMSO. In any case, the H NMR spectra, coupled with
the high-resolution mass spectral data, leave little doubt
about the accuracy of the molecular structures of 1a -
1h . An interesting feature of the 1H NMR spectra of the
target compounds is the resolution of the two methylene
protons of the p-methoxybenzyl groups into two distinct
doublets, separated by as much as 0.14 ppm, with a
large (∼15 Hz) coupling constant. This appears to due
to the restricted rotation of the methylene protons,
imposed by steric interactions with the peri-alkoxy
group at position-8. Our molecular modeling studies
(Insight/Discover)48 revealed that the alkoxy oxygen
atom of 1 lies within 2.5 Å of the methylene protons of
the aralkyl group. The restricted rotation could also be
discerned in the two methylene protons of the 8-alkoxy
group of compounds 1b, 1c, and 1f-1h , as they ap-
peared as two distinct, broad multiplets.
Bioch em istr y. Compounds 1a -1h were screened for
inhibition of ADA from calf spleen (Sigma) in a 50 mM
phosphate buffer (pH 7.0) at 25 °C by spectrophoto-
metrically monitoring the rate of hydrolysis of the
substrate adenosine into product inosine at 265 nm. The
Ki’s were computed from Lineweaver-Burk plots (see
Figure 2 of the Supporting Information) and are listed
in Table 1. All of the target compounds were determined
to be competitive inhibitors of ADA with Ki’s in the
range ∼10-100 µM. The n-propyl (1c) and n-butyl (1d )
analogues were found to be the most potent compounds
among the group.
The target compounds of this study contained a
hydrophobic substituent at the N-1 position of 1. It
would be interesting to explore how the hydrophilic
groups attached to either the N-1 or the N-3 atom of 1
would affect the overall enzyme binding and inhibition.
The synthesis and screening of a variety of additional
analogues of 1 are anticipated to shed further light on
the SAR of coformycin, and such an endeavor is cur-
rently in progress in our laboratory.
Con clu sion s
The observed Ki’s of 1a -1h suggest that they bind
about 6- 9 orders of magnitude less tightly to ADA than
Exp er im en ta l Section
do coformycin (Ki ) 10-11 15
and deoxycoformycin (Ki )
)
(A) Or ga n ic Syn th esis. Melting points were taken on a
Thomas-Hoover capillary melting point apparatus and are
uncorrected. 1H and 13C NMR were obtained on a 300-MHz
instrument equipped with the NMR software Aquerius by
Techmag. Chemical shifts are reported downfield from DMSO-
d6. Spectral data are reported as follows: s ) singlet, d )
doublet, dd ) doublet of doublets, t ) triplet, dt ) doublet of
triplets, q ) quartet, m ) multiplet, and b ) broad. Thin-layer
chromatography (TLC) was carried out on Kieselgel silica gel
60 F254 plates. Column chromatography was performed using
32-63 mesh silica gel. Elemental microanalyses were per-
formed by Atlantic Microlab, Inc., Norcross, Georgia. Mass
10-12).15 The considerably weaker inhibition of 1a -1h
compared to coformycins is consistent with the loss of
the sugar hydroxyl-protein hydrogen bonds, which are
partly responsible for the extremely tight enzyme bind-
ing of compounds of general formula 1. In the homolo-
gous series 1a -1d , the inhibition seems to correlate
with the hydrophobicity of the alkyl group (note: al-
though compound 1c appears to deviate slightly from
the trend, the larger error bar associated with its Ki as
compared to 1d will make this correlation still consis-

1048 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Reayi and Hosmane
spectra were recorded at the Mass Spectral Facility, Depart-
ment of Biochemistry, Michigan State University or the
University of Maryland, College Park. Anhydrous solvents
DMF and acetonitrile were purchased and used as is. All
alcohols and THF were dried over sodium metal.
(d, 2H, J ) 8.7 Hz, Ph-H), 5.50 (s, 2H, CH2), 3.71 (s, 3H, OCH3);
13C NMR (DMSO-d6) δ 181.4 (CHO), 159.1 (C-4), 150.4 (C-5),
140.1 (C-2), 129.0 (Ph-C), 127.5 (Ph-C), 124.6 (Ph-C), 114.0 (Ph-
C), 55.1 (CH2 of benzyl), 49.9 (OCH3); MS (EI) m/z 261 (M+).
Anal. C, H, N.
4-Am in o-1-p -m e t h oxyb e n zylim id a zole -5-ca r b a ld e -
h yd e (6). Meth od A. In a Parr hydrogenation bottle, 5 (1 g,
3.8 mmol) was dissolved in 50-75 mL of methanol. Raney
nickel (1 g, wet weight) that was washed several times with
methanol was added. The bottle was sealed and hydrogenated
at 45 psi at room temperature for 3 h. The catalyst was filtered
using Celite and washed with methanol. The filtrate was
evaporated, and the residue was purified by silica gel flash
chromatography, using a mixture of chloroform-acetone (10:
1) as an eluting solvent system. The appropriate fractions were
pooled and evaporated to obtain 6 as a white solid (0.3 g,
34%): mp 153-161 °C (dec); 1H NMR (DMSO-d6) δ 9.42 (s,
1H, CHO), 7.70 (s, 1H, H-2), 7.17 (b, 2H, Ph-H), 6.88 (b, 2H,
Ph-H), 6.37 (s, 1H, NH), 5.21 (s, 2H, CH2), 3.71 (s, 3H, OCH3);
13C NMR (DMSO-d6) δ 174.5 (CHO), 158.6 and 141.5 (C-2 +
C-4 + C-5), 128.3 (Ph-C), 113.8 (Ph-C), 54.9 (CH2 of benzyl),
47.8 (OCH3). Anal. C, H, N.
1-p-Meth oxyben zyl-4-n itr oim id a zole (3). A suspension
of 4-nitroimidazole (14.9 g, 131.8 mmol) and potassium
carbonate (21.9 g, 158.4 mmol) in 80 mL of anhydrous DMF
was stirred for 5 min, and it became creamy yellow. To the
reaction mixture was added p-methoxybenzyl chloride (21.6
mL, 159.3 mmol), followed by 20 mL of anhydrous DMF. The
mixture was stirred at room temperature overnight. The
creamy yellow reaction mixture was rotary evaporated under
high vacuum at 50 °C to remove DMF. Ice water was then
added to the reaction mixture and left at room temperature
for 1 h to dissolve the potassium carbonate. The crude product
was suction filtered and washed successively with water and
methanol to obtain 3 as a white crystalline solid (28.9 g,
94%): mp 132-135 °C; 1H NMR (DMSO-d6) δ 8.45 (s, 1H, H-5),
7.98 (s, 1H, H-2), 7.37 (d, 2H, J ) 9.0 Hz, Ph-H), 6.92 (d, 2H,
J ) 8.4 Hz, Ph-H), 5.22 (s, 2H, CH2), 3.73 (s, 3H, OCH3); 13C
NMR (DMSO-d6) δ 159.2 (C-4), 147.0 (C-5), 136.8 (C-2), 129.4
(Ph), 127.8 (Ph), 120.8 (Ph), 114.2 (Ph), 54.9 (CH2 of benzyl),
50.1 (OCH3); MS (FAB) m/z 234 (MH+). Anal. C, H, N.
Meth od B. In a 250-mL round-bottom flask, 5 (1 g, 3.8
mmol) was dissolved in 50-75 mL of ethyl acetate, and 5%
Pd/C (0.45 g) was added to the flask. The flask was sealed with
a rubber septum and evacuated using a needle attached to a
hose and connected to the water aspirator. Then a balloon filled
with hydrogen gas was attached to the septum using a needle.
The reaction was hydrogenated at 1 atm at room temperature
while being stirred for 6 h. The reaction mixture was then
filtered using Celite and the catalyst was washed with ethyl
acetate several times. Rotary evaporation of the filtrate
5-Dich lor om et h yl-1-p -m et h oxyb en zyl-4-n it r oim id a -
zole (4). To a 250-mL round-bottom flask was added t-BuOK
(6 g, 53.5 mmol) and a mixture of dry THF-DMF (2:1, 45 mL).
The reaction mixture was stirred vigorously under N2 at -100
°C (dry ice/ether). In a separate flask, 1-p-methoxybenzyl-4-
nitroimidazole (3) (3 g, 12.9 mmol) was dissolved in dry
chloroform (1.2 mL, 14.9 mmol), and the solution was added
to 10 mL of anhydrous DMF. This solution was then added to
the 250-mL round-bottom flask during a period of 10 min. The
reaction mixture was stirred for 10 min and then an additional
0.5 mL (6.3 mmol) of dry chloroform was added. The reaction
mixture continued to stir for an additional 10 min at -100 °C
and was quenched with a solution of glacial acetic acid (5 mL)
in 5 mL of methanol, when it turned reddish brown. To the
acidified mixture, 100 mL of ice water was added. The mixture
was extracted with dichloromethane (25 mL × 3), the organic
layers were combined, washed twice with water, and dried over
anhydrous magnesium sulfate. After filtration, the dichlo-
romethane was removed in low vacuum at 40 °C and the DMF
was removed under high vacuum at 75 °C. The crude product
was purified using silica gel column chromatography [eluent:
n-heptane to a mixture of n-heptane-chloroform (1:1) to
chloroform]. The desired fractions were pooled and the solvent
was removed. Addition of a minimal amount of ethyl acetate
followed by the addition of hexane precipatated 4 as white
crystals (2.15 g, 53%): mp 97-100 °C; 1H NMR (DMSO-d6) δ
8.01 (s, 1H, CHCl2), δ 7.96 (s, 1H, H-2), 7.27 (d, 2H, J ) 8.4
Hz, Ph-H), 6.95 (d, 2H, J ) 8.4 Hz, Ph-H), 5.56 (s, 2H, CH2),
3.73 (s, 3H, OCH3); 13C NMR (DMSO-d6) δ 159.1 (C-4), 142.0
(C-5), 138.3 (C-2), 129.0 (Ph-C), 126.7 (Ph-C), 126.4 (Ph-C),
114.0 (Ph-C), 58.5 (CHCl2), 54.9 (CH2 of benzyl), 49.3 (OCH3);
MS (EI) m/z 315, 317 (M+). Anal. C, H, N, Cl.
1
produced a pure light-yellow compound (0.19 g, 22%). The H
NMR of this compound was identical to that of 6 obtained by
method A above.
7,8-Dih yd r o-8-m eth oxy-1-p-m eth oxyben zyl-6H-6,7-d i-
m eth ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1a ). In a dry round-
bottom flask were added consecutively compound 6 (0.08 g,
0.35 mmol), 1,2-dimethylhydrazine dihydrochloride (0.08 g,
0.60 mmol), dry methanol (10-15 mL), and anhydrous tri-
methyl orthoformate (0.5 mL) (Aldrich). The flask was stirred
at room temperature for 1.5 h and then refluxed for 7-8 h
until all the starting material disappeared, as observed by TLC
[chloroform/methanol (10:1)]. The reaction flask was allowed
to cool, then triethylamine (0.3 mL) was added, and the
mixture was allowed to stir for 5 min at room temperature. A
new spot moving slower than the starting material was
observed by TLC [chloroform/methanol (10:1)]. The contents
of the flask were rotary evaporated (not to complete dryness)
and a minimal amount of chloroform was added. The chloro-
form mixture was loaded on a silica gel column [eluent:
chloroform f chloroform/methanol (10:1)]. The desired frac-
tions were pooled and evaporated to obtain 1a as a thick brown
oil (0.068 g, 62%): 1H NMR (DMSO-d6) δ 7.53 (s, 1H, H-2),
7.12 (d, 2H, J ) 7.8 Hz, Ph-H), 6.98 (s, 1H, H-5), 6.89 (d, 2H,
J ) 8.1 Hz, Ph-H), 4.98 (m, 2H, CH2), 4.88 (s, 1H, H-8), 3.71
(s, 6H, OCH3), 3.30 (H2O), 3.05 (s, 3H, N6-CH3), 2.04 (s, 3H,
N7-CH3); HRMS (DEI) calcd for C16H21N5O2 (M+) m/z 315.1695,
found 315.1692. Anal. C, H, N.
1-p-Met h oxyb en zyl-4-n it r oim id a zole-5-ca r ba ld eh yd e
(5). In a three-neck round-bottom flask, crushed calcium
carbonate (6 g, 60 mmol) was stirred in 80-100 mL of water
for 30 min at 65-75 °C. To the flask was added 4 (6 g, 18.9
mmol), and the heterogeneous reaction mixture was stirred
for 40-48 h at 65-75 °C until no further starting material
was present as shown by TLC [chloroform/methanol (50:1)].
After cooling, the calcium carbonate was filtered and washed
with a minimum amount of acetone, the filtrate was extracted
with dichloromethane (3 × 25 mL), and the combined organic
extracts were washed with 25 mL of water. After silica gel
chromatography (eluent 1:1 f 2:1 chloroform/heptane), the
desired fractions were pooled and rotary evaporated. Addition
of ethyl acetate to the residue, followed by hexane, yielded 5
as yellow crystals (3 g, 61%). Compound 5 is light sensitive
and discolors in daylight; therefore, it needs to be stored in
7,8-Dih ydr o-8-eth oxy-1-p-m eth oxyben zyl-6H-6,7-dim eth -
ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1b). In
a dry round-
bottom flask were added consecutively compound 6 (0.08 g,
0.35 mmol), 1,2-dimethylhydrazine dihydrochloride (0.08 g,
0.60 mmol), dry ethanol (10-15 mL), and anhydrous triethyl
orthoformate (0.5-1.0 mL) (Lancaster). The rest of the pro-
cedure is the same as the one described above for 1a . The
desired fractions were pooled and evaporated to obtain 1b as
a thick brown oil (0.034 g, 30%): 1H NMR (DMSO-d6) δ 7.53
(s, 1H, H-2), 7.10 (d, 2H, J ) 7.9 Hz, Ph-H), 6.98 (s, 1H, H-5),
6.88 (d, 2H, J ) 8.1 Hz, Ph-H), 5.03 (m, 2H, CH2), 4.96 (s, 1H,
H-8), 3.71 (s, 3H, OCH3), 3.80 (m, 1H, OCH2), 3.48 (m, 1H,
OCH2), 3.03 (s, 3H, N6-CH3), 2.03 (s, 3H, N7-CH3), 1.04 (t, 3H,
CH3, J ) 8 Hz, ether CH3); HRMS (DEI) calcd for C17H23N5O2
(M+) m/z 329.1852, found 329.1844. Anal. C, H, N.
1
the dark: mp 87-90 °C; H NMR (DMSO-d6) δ 10.22 (s, 1H,
CHO), 8.34 (s, 1H, H-2), 7.23 (d, 2H, J ) 8.4 Hz, Ph-H), 6.89

Inhibition of Adenosine Deaminase
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4 1049
7,8-Dih ydr o-1-p-m eth oxyben zyl-6H-6,7-dim eth yl-8-pr o-
p oxyim id a zo[4,5-e][1,2,4]tr ia zep in e (1c). Compound 1a
(0.045 g, 0.14 mmol) was dissolved in excess dry n-propanol
(10-20 mL) in a dry round-bottom flask, and a catalytic
amount of dry trifluoroacetic acid (5 µL) was added to the
solution. The reaction mixture was stirred at room tempera-
ture for 5 days. It was then rotary evaporated (not to complete
dryness), and the residue was purified by flash silica gel
chromatography [eluent: chloroform f chloroform-methanol
(10:1)]. The desired fractions were pooled and evaporated to
obtain 1c as a thick brown oil (0.018 g, 37%): 1H NMR (DMSO-
d6) δ 7.51 (s, 1H, H-2), 7.08 (d, 2H, J ) 9 Hz, Ph-H), 6.98 (s,
1H, H-5), 6.89 (d, 2H, J ) 9 Hz, Ph-H), 5.03 (m, 2H, CH2),
4.97 (s, 1H, H-8), 3.76 (m, 1H, OCH2), 3.62 (m, 1H, OCH2),
3.71 (s, 3H, OCH3), 3.04 (s, 3H, N6-CH3), 2.04 (s, 3H, N7-CH3),
1.44 (q, 2H, J ) 9 Hz, CH2), 0.82 (t, 3H, J ) 8 Hz, CH3); HRMS
calcd for C23H28N5O3 (MH+) m/z 422.2192, found 422.2174.
Anal. C, H, N.
7,8-Dih ydr o-1-p-m eth oxyben zyl-8-(3-m eth ylben zyloxy)-
6H-6,7-d im eth ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1h ). Com-
pound 1a (0.075 g, 0.24 mmol) was dissolved in excess
3-methylbenzyl alcohol (10 mL), and a catalytic amount of dry
trifluoroacetic acid (3 µL) was added to the solution. The
reaction mixture was stirred at room temperature for 3-5 days
until TLC showed the complete disappearance of the starting
material. The rest of the procedure is the same as the one
described above for 1c. The purification needed repeated
column chromatography. The desired fractions were pooled
and evaporated to obtain 1h as a thick brown oil (0.031 g,
32%): 1H NMR (DMSO-d6) δ 7.54 (s, 1H, H-2), 6.78-7.21 (m,
H-5 and Ph-H), 5.07 (m, 2H from CH2, 1H from H-8), 4.49 (m,
1H, OCH2), 4.79 (m, 1H, OCH2), 3.68 (s, 3H, OCH3), 3.3 (H2O),
3.08 (s, 3H, N6-CH3), 2.26 (d, 3H, J ) 9 Hz, m-CH3), 2.07 (s,
3H, N7-CH3); HRMS (FAB) calcd for C23H28N5O2 (MH+) m/z
406.2243, found 406.2250. Anal. C, H, N.
(FAB) calcd for
C
18H26N5O2 (MH+) m/z 344.2087, found
344.2089. Anal. C, H, N.
8-Bu toxy-7,8-dih ydr o-1-p-m eth oxyben zyl-6H-6,7-dim eth -
ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1d ). Compound 1a (0.045
g, 0.14 mmol) was dissolved in excess dry butanol (10-20 mL),
and a catalytic amount of dry trifluoroacetic acid (5 µL) was
added to the solution. The rest of the procedure is the same
as the one described above for 1c. The desired fractions were
pooled and evaporated to obtain 1d as a thick brown oil (0.022
g, 43%): 1H NMR (DMSO-d6) δ 7.51 (s, 1H, H-2), 7.06 (d, 2H,
J ) 9 Hz, Ph-H), 6.98 (s, 1H, H-5), 6.88 (d, 2H, J ) 9 Hz, Ph-
H), 5.02 (m, 2H, CH2), 4.97 (s, 1H, H-8), 3.78 (m, 1H, OCH2),
3.64 (m, 1H, OCH2), 3.71 (s, 3H, OCH3), 3.37 (H2O), 3.03 (s,
3H, N6-CH3), 2.04 (s, 3H, N7-CH3), 1.33 (m, 4H, CH2), 0.85 (t,
3H, J ) 8 Hz, CH3); HRMS (DEI) calcd for C19H27N5O2 (M+)
m/z 357.2165, found m/z 357.2174. Anal. C, H, N.
4-(Eth oxym eth ylen e)im in o-1-p-m eth oxyben zylim id a -
zole-5-ca r ba ld eh yd e (7). To a solution of compound 6 (0.1
g, 0.43 mmol) in 10 mL of dry THF was added anhydrous
triethyl orthoformate (2 mL), and the reaction mixture was
refluxed until all the starting material was consumed (usually
overnight). A new spot moving faster than the starting
material was formed, as observed by TLC. The reaction
mixture was rotary evaporated in high vacuum to remove
excess triethyl orthoformate. The solid residue obtained was
redissolved in dry THF and rotary evaporated again to obtain
7 as a crystalline solid (0.11 g, 89%): mp 105-106 °C; 1H NMR
(DMSO-d6) δ 9.73 (s, 1H, CHO), 8.53 (s, 1H, H-5), 8.05 (s, 1H,
H-2), 7.21 (d, 2H, J ) 9 Hz, Ph), 6.87 (d, 2H, J ) 9 Hz, Ph),
5.36 (s, 2H, CH2), 4.28 (q, 2H, OCH2), 3.69 (s, 3H, OCH3), 1.27
(t, 3H, J ) 9 Hz, ether CH3); MS (EI) m/z 287 (M+). Anal. C,
H, N.
7,8-Dih yd r o-8-isop r op oxy-1-p-m eth oxyben zyl-6H-6,7-
d im eth ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1e). Compound
1a (0.045 g, 0.14 mmol) was dissolved in excess dry 2-propanol
(10-20 mL), and a catalytic amount of dry trifluoroacetic acid
(5 µL) was added to the solution. The rest of the procedure is
the same as the one described above for 1c. The desired
fractions were pooled and evaporated to obtain 1e as a thick
brown oil (0.038 g, 77%): 1H NMR (DMSO-d6) δ 7.46 (s, 1H,
H-2), 7.01 (d, 2H, J ) 9 Hz, Ph-H), 6.99 (s, 1H, H-5), 6.90 (d,
2H, J ) 9 Hz, Ph-H), 5.02 (m, 2H, CH2), 5.06 (m, 1H from
H-8, 1H from CH), 3.71 (s, 3H, OCH3), 3.3 (H2O), 3.04 (s, 3H,
N6-CH3), 2.05 (s, 3H, N7-CH3), 1.08 (t, 6H, J ) 8 Hz, CH3);
HRMS (FAB) calcd for C18H26N5O2 (MH+) m/z 344.2087, found
344.2093. Anal. C, H, N.
(B) In h ibition Stu d ies. The target compounds (1a -1h )
were screened against calf spleen ADA (Sigma) in vitro in a
50 mM phosphate buffer (pH 7.0) at 25 °C. The rate of
hydrolysis of adenosine into inosine was monitored spectro-
photometrically by following the absorbance decrease over 2
min at 265 nm using a Varian Cary UV-visible spectropho-
tometer (50-BIO). All absorbances were maintained below 1
absorbance unit in a 1 cm path length cuvette. The enzyme
and substrate stock solutions were prepared using a 50 mM
phosphate buffer (pH 7.0). While holding the enzyme and
inhibitor concentration constant, the substrate concentration
was varied to obtain the kinetic data. The substrate concentra-
tion in each assay was measured using up to five different
concentrations, ranging from 20 to 60 µM. This was repeated
with a different inhibitor concentration. The inhibitor concen-
tration ranged from 8 to 15 µM. The enzyme concentration
(0.022 units/mL) was kept constant in all assays. Lineweaver-
Burk (1/V vs 1/[S]) plots were constructed for each assay and
the Ki was calculated using Graph Pad Prism Software. In each
assay, the inhibitors were used without preincubation with
enzyme. The substrate and inhibitor were added to the
phosphate buffer (50 mM, pH 7.0), followed by the addition of
enzyme. The enzyme solution was kept in an ice bath prior to
performing each assay. Assays were performed in duplicate
and in some cases triplicate.
7,8-Dih yd r o-1-p-m eth oxyben zyl-8-(2-m eth oxyeth oxy)-
6H-6,7-d im eth ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1f). Com-
pound 1a (0.045 g, 0.14 mmol) was dissolved in excess dry
2-methoxyethanol (10-20 mL), and a catalytic amount of dry
trifluoroacetic acid (5 µL) was added to the solution. The rest
of the procedure is the same as the one described above for
1c. The desired fractions were pooled and evaporated to obtain
1f as a thick brown oil (0.024 g, 47%): 1H NMR (DMSO-d6) δ
7.52 (s, 1H, H-2),7.10 (d, 2H, J ) 9 Hz, Ph-H), 6.98 (s, 1H,
H-5), 6.88 (d, 2H, J ) 9 Hz, Ph-H), 5.05 (m, 2H, CH2), 5.03 (s,
1H, H-8), 3.83 (m, 2H, OCH2), 3.65 (m, 2H, OCH2), 3.71 (s,
3H, OCH3), 3.20 (s, 3H, OCH3), 3.04 (s, 3H, N6-CH3), 2.04 (s,
3H, N7-CH3); HRMS (FAB) calcd for C18H26N5O3 (MH+) m/z
360.2036, found 360.2039. Anal. C, H, N.
7,8-Dih yd r o-1-p -m et h oxyb en zyl-8-(4-m et h oxyb en zyl-
oxy)-6H-6,7-d im eth ylim id a zo[4,5-e][1,2,4]tr ia zep in e (1g).
Compound 1a (0.08 g, 0.25 mmol) was dissolved in excess
4-methoxybenzyl alcohol (10 mL), and a catalytic amount of
dry trifluoroacetic acid (3 µL) was added to the solution. The
rest of the procedure is the same as the one described above
for 1c. The purification needed repeated column chromatog-
raphy. The desired fractions were pooled and evaporated to
obtain 1g as a thick brown oil (0.044 g, 41%): 1H NMR (DMSO-
d6) δ 7.53 (s, 1H, H-2), 7.18 (m, 4H, Ph-H), 7.00 (s, 1H, H-5),
6.83 (m, 4H, Ph-H), 5.04 (m, 2H from CH2, 1H from H-8), 4.74
(m, 1H, OCH2), 4.46 (m, 1H, OCH2), 3.71 (b, 6H, OCH3), 3.3
(H2O), 3.09 (s, 3H, N6-CH3), 2.05 (s, 3H, N7-CH3); HRMS (FAB)
Ack n ow led gm en t. This research was supported by
a grant (#9 RO1 AI55452) from the National Institute
of Allergy and Infectious Diseases of the National
Institutes of Health. The authors thank Mrs. Dalia
Tadros for assistance in biochemical studies and Dr.
Diana S. Hamilton for advice on enzyme kinetics.
Su p p or tin g In for m a tion Ava ila ble: Lineweaver-Burk
plots used for computation of Ki’s of target compounds 1a -1h
(Figure 2).

1050 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 4
Reayi and Hosmane
(25) Umezawa, H.; Tsutomu, S.; Yasuo, F.; Ikuyo, H.; Masaaki, I.;
Tomio, T. Formycin and formycin B in cells of Ehrlich carcinoma
and Escherichia coli. J . Antibiot. (Tokyo), Ser. A. 1967, 20, 308-
316.
(26) J ohns, D. G.; Adamson, R. H. Enhancement of the biological
activity of cordycepin (3′-deoxyadenosine) by the adenosine
deaminase inhibitor 2′-deoxycoformycin. Biochem. Pharmacol.
1976, 25, 1441-1444.
(27) Major, P. P.; Agarwal, R. P.; Kufe, D. W. Deoxycoformycin:
Neurological toxicity. Cancer Chemother. Pharmacol. 1981, 5,
193-196.
(28) Major, P. P.; Agarwal, R. P.; Kufe, D. W. Clinical pharmacology
of deoxycoformycin. Blood 1981, 58, 91-96.
(29) Grever, M. R.; Siaw, M. F.; J acob, W. F.; Niedhart, J . A.; Miser,
J . S.; Coleman, M. S.; Hutton, J . J .; Balcerzak, S. P. The
biochemical and clinical consequences of 2′-deoxycoformycin in
refractory lymphoproliferative malignancy. Blood 1981, 57, 406-
417.
(30) Lambe, C.; Bugge, C. J . L.; Lafon, S. W.; Nelson, D. J .; Elion, G.
B. Fed. Proc. Am. Soc. Exp. Biol. 1979, 38, 670.
(31) Acevedo, O. L.; Krawczyk, S. H.; Townsend, L. B. A novel
approach towards the synthesis of the 3,4,5-trihydro-1,3-diaz-
epin-5-ol ring structure. Tetrahederon Lett. 1983, 24, 4789-4792.
(32) Acevedo, O. L.; Krawczyk, S. H.; Townsend, L. B. Synthesis of
the new ring system, pyrazolo[3,4-d][1,3]diazepines. J . Hetero-
cycl. Chem. 1985, 22, 349-352.
(33) Acevedo, O. L.; Krawczyk, S. H.; Townsend, L. B. Total synthesis
of (4R)- and (4S)-5,6-dihydro-1-â-D-ribofuranosyl-4H-pyrazolo-
[3,4-d][1,3]diazepin-4-ol and (8R)- and (8S)-7,8-dihydro-3-â-D-
ribofuranosyl-6H-v-triazolo[4,5-d][1,3]diazepin-8-ol: Two het-
erocyclic analogues of the nucleoside antibiotic coformycin. J .
Org. Chem. 1986, 51, 1050-1058.
(34) Hong, M.; Hosmane, R. S. Irreversible, tight-binding inhibition
of adenosine deaminase by coformycins: Inhibitor structural
features that contribute to the mode of inhibition. Nucleosides
Nucleotides 1997, 16, 1053-1057.
(35) Hosmane, R. S.; Hong, M. How important is the N-3 sugar
moiety in the tight-binding interaction of coformycin with
adenosine deaminase? Biochem. Biophys. Res. Commun. 1997,
236, 88-93.
(36) Wang, Z.; Quiocho, F. A. Complexes of adenosine deaminase with
two potent inhibitors: X-ray structures in four Independent
molecules at pH of maximum activity. Biochemistry 1998, 37,
8314-8324.
(37) Marrone, T. J .; Straatsma, T. P.; Briggs, J . M.; Wilson, D. K.;
Quiocho, F. A.; McCammon, J . A. Theoretical study of inhibition
of adenosine deaminase by (8R)-coformycin and (8R)-deoxyco-
formycin. J . Med. Chem. 1996, 39, 277-284.
Refer en ces
(1) 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.
(2) Aiuti, A. Advances in gene therapy for ADA-deficient SCID.
Curr. Opin. Mol. Therapeut. 2002, 4, 515-522.
(3) Lum, C. T.; Sutherland, D. E.; Najarian, J . R. Inhibition for
immunosuppression. New Eng. J . Med. 1977, 296, 819.
(4) Chassin, M. M.; Chirigos, M. A.; J ohns, D. G.; Adamson, R. H.
Adenosine deaminase inhibition for immunosuppression. New
Eng. J . Med. 1977, 296, 1232.
(5) Ohno, M.; Yagisawa, N.; Shibahara, S.; Kondo, S.; Maeda, K.;
Umezawa, H. Synthesis of coformycin. J . Am. Chem. Soc. 1974,
96, 4326-4327.
(6) Nakamura, H.; Koyama, G.; Iitaka, Y.; Ohno, M.; Yagisawa, N.;
Kondo, S.; Maeda, K.; Umezawa, H. Structure of coformycin, an
unusual nucleoside of microbial origin. J . Am. Chem. Soc. 1974,
96, 4327-4328.
(7) Nakamura, H.; Koyama, G.; Umezawa, H.; Iitaka, Y. The crystal
and molecular structure of coformycin. Acta Crystallogr. 1976,
B32, 1206-1212.
(8) Hawkins, L. D.; Hanvey, J . C.; Boyd, F. L., J r.; Baker, D. C.;
Showalter, H. D. H. Inhibitors of adenosine deaminase. Syn-
thesis of coformycin and 3′-deoxycoformycin. Nucleosides Nucleo-
tides 1983, 2, 479-494.
(9) Woo, P. W. K.; Dion, H. W.; Lange, S. M.; Dahl, L. F.; Durham,
L. J . A novel adenosine and ara-A deaminase Inhibitor, (R)-3-
(2-deoxy-â-D-erythropentofuranosyl)-3,6,7,8-tetrahydroimidazo-
[4,5-d]diazepin-8-ol. J . Heterocycl. Chem. 1974, 11, 641-643.
(10) Baker, D. C.; Putt, S. R. A total synthesis of pentostatin, the
potent inhibitor of adenosine deaminase. J . Am. Chem. Soc.
1979, 101, 6127-6128.
(11) Baker, D. C.; Putt, S. R.; Showalter, H. D. H. Studies related to
the total synthesis of pentostatin. Approaches to the synthesis
of (8R)-3,6,7,8-tetrahydroimidazo[4,5-d][1,3]diazepin-8-ol and
N-3 alkyl congeners. J . Heterocycl. Chem. 1983, 20, 629-634.
(12) Chan, E.; Putt, S. R.; Showalter, H. D. H.; Baker, D. C. Total
synthesis of (8R)-3-(2-deoxy-â-D-erythropentofuranosyl-3,6,7,8-
tetrahydroimidazo[4,5-d][1,3]diazepin-8-ol (pentostatin), the po-
tent inhibitor of adenosine deaminase. J . Org. Chem. 1982, 47,
3457-3464.
(13) Hanvey, J . C.; Hardman, J . K.; Suhadolnik, R. J .; Baker, D. C.
Evidence for the conversion of adenosine to 2′-deoxycoformycin
by streptomyces antibiotics. Biochemistry 1984, 23, 904-907.
(14) Hanvey, J . C.; Hawkins, E. S.; Tunac, J . B.; Dechter, J . J .; Baker,
D. C.; Suhadolnik, R. J . Biosynthesis of 2′-deoxycoformycin:
Evidence for ring-expansion of the adenine moiety of adenosine
to a tetrahydroimidazo[4,5-d][1,3]diazepine system. Biochemistry
1987, 26, 5636-5641.
(38) Melzig, M. F. Inhibition of adenosine deaminase activity of aortic
endothelial cells by selected flavonoids. Planta Medica 1996, 62,
20-21.
(15) Agarwal, R. P.; Cha, S.; Crabtree, G. W.; Parks, R. E. Coformycin
and deoxycoformycin: Tight-binding inhibitors of adenosine
deaminase. Chemistry and Biology of Nucleosides and Nucleo-
tides; Academic Press: New York, 1978; pp 159-197.
(16) Wolfenden, R.; Wentworth, D. F.; Mitchell, G. N. Influence of
substituent ribose on transition state affinity in reactions
catalyzed by adenosine deaminase. Biochemistry 1977, 16,
5071-5077.
(17) Frick, L.; Wolfenden, R.; Smal, E.; Baker, D. C. Transition-state
stabilization by adenosine deaminase: Structural studies of its
inhibitory complex with deoxycoformycin. Biochemistry 1986, 25,
1616-1621.
(18) Kati, W. M.; Wolfenden, R. Contribution of a single hydroxyl
group to transition-state discrimination by adenosine deami-
nase: Evidence for an “entropy trap” mechanism. Biochemistry
1989, 28, 7919-7927.
(19) Kurz, L. C.; Moix, L.; Riley, M. C.; Frieden, C. The rate of
formation of transition-state analogues in the active site of
adenosine deaminase is encounter-controlled: Implications for
the mechanism. Biochemistry 1992, 31, 39-48.
(20) Agarwal, R. P.; Spector, T.; Parks, R. E., J r. Tight-binding
inhibitors. IV. Inhibition of adenosine deaminases by various
inhibitors. Biochem. Pharmacol. 1977, 26, 359-367.
(21) Smyth, J . F.; Harrap, K. R. Adenosine deaminase activity in
leukemia. Br. J . Cancer 1975, 31, 544-549.
(22) Smyth, J . F.; Poplack, D. G.; Holiman, B. J .; Leventhal, B. G.;
Yarbro, G. Correlation of adenosine deaminase activity with cell
surface markers in acute lymphoblastic leukemia. J . Clin. Invest.
1978, 62, 710-712.
(23) Smyth, J . F.; Prentice, H. G.; Proctor, S.; Hoffbrand, A. V.
Deoxycoformycin in the treatment of leukemias and lymphomas.
Ann. N. Y. Acad. Sci. 1985, 451, 123-128.
(24) LePage, G. A.; Worth, L. S.; Kimball, A. P. Enhancement of the
antitumor activity of arabinofuranosyladenine by 2′-deoxycofor-
mycin. Cancer Res. 1976, 36, 1481-1485.
(39) Koch, H. P.; J ager, W.; Groh, U.; Plank, G. In vitro inhibition of
adenosine deaminase by flavonoids and related compounds. New
insight into the mechanism of action of plant phenolics. Methods
Find. Exp. Clin. Pharmacol. 1992, 14, 413-417.
(40) Koch, H. P.; Aichinger, A.; Bohne, B.; Plank, G. In vitro inhibition
of adenosine deaminase by a group of steroid and triterpenoid
compounds. Phytother. Res. 1994, 8, 109-111.
(41) Koch, H. P.; J ager, W.; Hysek, J .; Korpert, B. Garlic and onion
extracts. In vitro inhibition of adenosine deaminase. Phytother.
Res. 1992, 6, 50-52.
(42) Sheid, B. Trazodone, a nontricyclic antidepressant, is an inhibi-
tor of adenosine deaminase. Res. Commun. Chem. Pathol.
Pharmacol. 1985, 47, 149-152.
(43) Hosmane, R. S.; Lim, B. B. Chemistry of the 5:7-fused heteroaro-
matic systems: A novel rearrangement involving the imidazo-
[4,5-e][1,3,4]triazepine and pyrazolo[3,4-e][1,3,4]triazepine sys-
tems. Heterocycles 1988, 27, 31-34.
(44) Hosmane, R. S.; Lim, B. B. A novel method for the synthesis of
6-(â-methyl)hydrazinopurine and 4-(â-methyl)hydrazinopyra-
zolo[3,4-d]pyrimidine. Synthesis 1988, 242-244.
(45) Hosmane, R. S.; Lim, B. B.; Burnett, F. N. Rearrangements in
heterocyclic synthesis: A novel translocation of an (N-amino-
N-methyl)formamidine side-chain to the vinylogous nitrile func-
tion. J . Org. Chem. 1988, 53, 382-386.
(46) Hosmane, R. S.; Lim, B. B.; Summers, M. F.; Siriwardane, U.;
Hosmane, N. S.; Chu, S. C. Rearrangements in heterocyclic
synthesis. 2. Novel transformations of 2-aminonicotino- and
anthranilonitriles. J . Org. Chem. 1988, 53, 5309-5315.
(47) Ostrowski, S. Synthesis of some new imidazole derivatives.
Polish J . Chem. 1994, 68, 2237-2247.
(48) Molecular modeling studies were performed using Insight/
DiscoverTM software package, available from Molecular Simula-
tions, Inc., San Diego, California.
J M0304257