Synthesis, mechanism of action, and antiviral activity of a new series of covalent mechanism-based inhibitors of S-adenosyl-L-homocysteine hydrolase

Article abstract of DOI:10.1021/jm0108350

A direct method for the preparation of 5′-S-alkynyl-5′-thioadenosine and 5′-S-allenyl-5′thioadenosine has been developed. Treatment of a protected 5′-acetylthio-5′-deoxyadenosine with sodium methoxide and propargyl bromide followed by deprotection gave th

Full text of DOI:10.1021/jm0108350

J . Med. Chem. 2001, 44, 2743-2752
2743
Syn th esis, Mech a n ism of Action , a n d An tivir a l Activity of a New Ser ies of
Cova len t Mech a n ism -Ba sed In h ibitor s of S-Ad en osyl-L-Hom ocystein e Hyd r ola se
,
†
†
†
§
§
Georges Guillerm,* Danielle Guillerm, Corinne Vandenplas-Witkowki, H e´ l e` ne Rogniaux, Nathalie Carte,
§
§
⊥
|
Emmanuelle Leize, Alain Van Dorsselaer, Erik De Clercq, and Christine Lambert
Universit e´ de Reims Champagne-Ardenne, UMR 6519, UFR Sciences, BP 1039, 51687 Reims Cedex, France,
Universit e´ Louis Pasteur ECPM UMR 7507, 25, rue Becquerel F, 67087 Strasbourg Cedex 2, France, Rega Institute for
Medical Research, Katholieke Universiteit of Leuven, Leuven, Belgium, and AstraZeneca - Reims, Centre de Recherches,
BP 1050, Chemin de Vrilly, 51689 Reims Cedex 2, France
Received February 5, 2001
A direct method for the preparation of 5′-S-alkynyl-5′-thioadenosine and 5′-S-allenyl-5′-
thioadenosine has been developed. Treatment of a protected 5′-acetylthio-5′-deoxyadenosine
with sodium methoxide and propargyl bromide followed by deprotection gave the 5′-S-propargyl-
5
′-thioadenosine 4. Under controlled base-catalysis with sodium tert-butoxide in tert-butyl
alcohol 4 was quantitatively converted into 5′-S-allenyl-5′-thioadenosine 5 or 5′-S-propynyl-
′-thioadenosine 6. Incubation of recombinant human placental AdoHcy hydrolase with 4, 5,
or 6 resulted in time- and concentration-dependent inactivation of the enzyme (K : 45 ( 0.5,
5
i
1
6 ( 1, and 15 ( 1 µM, respectively). Compound 4 caused complete conversion of the enzyme
+
from its E-NAD to E-NADH form during the inactivation process. This indicates that 4 is a
substrate for the 3′-oxidative activity of AdoHcy hydrolase (type I inhibitor). In contrast, the
NAD /NADH content of the enzyme was not affected during the inactivation process with 5
+
and 6, and their mechanism of inactivation was further investigated. Addition of enzyme-
sequestered water on the S-allenylthio group of 5 or S-propynylthio group of 6 within the active
site should lead to the formation of the corresponding thioester 7. This acylating-intermediate
agent could then undergo nucleophilic attack by a protein residue, leading to a type II
mechanism-based inactivation. ElectroSpray mass spectra analysis of the inactivated protein
n
by 5 supports this mechanistic proposal. Further studies (MALDI-TOF and ESI/MS experi-
ments) of the trypsin and endo-Lys-C proteolytic cleavage of the fragments of inactivated
AdoHcy hydrolase by 5 were carried out for localization of the labeling. The antiviral activity
of 4, 5, and 6 against a large variety of viruses was determined. Significant activity (EC50: 1.9
µM) was noted with 5 against vaccinia virus.
In tr od u ction
tive action of the enzyme and irreversibly keep the
AdoHcy hydrolase in its NADH form, thus disabling the
S-Adenosylhomocysteine (AdoHcy) is the product of
all biological methylation in which S-Adenosylmethion-
ine (AdoMet) is utilized as a methyl donor. This impor-
tant metabolite is reversibly hydrolyzed to L-homocys-
teine and adenosine (Ado) by S-adenosylhomocysteine
hydrolase (E.C. 3.3.1.1).1 Inhibition of this cellular
enzyme results in intracellular accumulation of AdoHcy
leading to a feedback inhibition of AdoMet-dependent
3
a,5
cycle of the overall enzyme reaction.
Type II mech-
anism-based inhibitors utilize the same “oxidative ac-
tion” of the enzyme or its “hydrolytic activity” to
generate an electrophilic site on the inhibitor which can
6
,7
then bind an active site nucleophile.
We recently found that fluorinated analogues of 5′-
methylthioadenosine are potent inactivators of AdoHcy
8
hydrolase. The interaction of these derivatives with the
methylation reactions (i.e., viral mRNA methylation)
which are essential for viral replication.2 Therefore,
different catalytic steps of the enzyme was presumed
to generate in the enzyme active cavity highly reactive
electrophilic species that could bind covalently with
AdoHcy hydrolase has become an attractive target for
the molecular design of antiviral agents. During the
3
9
active site residue. This result led us to consider that
past decade, the design and synthesis of mechanism-
based inhibitors have received considerable attention,
particularly since the mechanism of the catalysis of
AdoHcy hydrolase was elucidated by Palmer and Abe-
other thionucleosides such as 5′-S-allenyl-5′-thioadenos-
ine 5 or 5′-S-propynyl-5′-thioadenosine 6 (Figure 1)
could also function as alternative substrates for AdoHcy
hydrolase and be good candidates as new covalent
mechanism-based inhibitors.
4
les. Two classes of potent irreversible inhibitors have
been identified for AdoHcy hydrolase. Type I mecha-
Two plausible mechanisms by which 5 and 6 should
inactivate AdoHcy hydrolase can be envisioned. We
reasoned that enzymatic oxidation of 5 and 6 into their
corresponding 3′-ketoderivatives could be accompanied
nism-based inhibitors are substrates for the C-3′ oxida-
*
To whom correspondence should be addressed. Tel: +33 3 26 91
3
3 08. Fax: +33 3 26 91 31 66. E-mail: georges.guillerm@univ-reims.fr.
†
10
Universit e´ de Reims Champagne-Ardenne.
Universit e´ Louis Pasteur - Strasbourg.
Katholieke Universiteit of Leuven.
by a â-elimination of an allenyl or propynyl mercaptan.
§
⊥
Under their presumed tautomeric thioaldehyde forms
|
11
AstraZeneca - Reims.
these intermediates are quite reactive and might
1
0.1021/jm0108350 CCC: $20.00 © 2001 American Chemical Society
Published on Web 07/18/2001

2
744 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17
Guillerm et al.
In a ctiva tion of S-Ad en osyl-L-h om ocystein e
Hyd r ola se
5
′-S-Allenyl-5′-thioadenosine 5 and 5′-S-propynyl-5′-
thioadenosine 6 and their precursor the 5′-S-propargyl-
5
′-thioadenosine 4 were evaluated for inhibition of
purified recombinant human placental AdoHcy hydro-
1
4
lase. Incubation of the enzyme with 4, 5, and 6
resulted in time- and concentration-dependent inactiva-
tion of the enzyme. As shown in Figure 2 (panels B and
C), the inactivation by 5 and 6 appeared to be biphasic
showing pseudo-first-order kinetics only in the first
period of inactivation.
Using the Kitz and Wilson method,15 a double recip-
rocal plot of the initial pseudo-first-order inactivation
rate constant (1/Kapp) versus 1/[I] gave the Ki and Kinact
values listed Table 1. Ki values for 4, 5, and 6 are in
the same range (16-40 µM), but 5 and 6 appeared to
be better inactivators in terms of their Kinact values.
When AdoHcy hydrolase (1 unit) was incubated in
assay buffer with 4, 5, or 6 (200 µM), total inactivation
of the enzyme occurred after 15 min. In each case, the
inactivation was irreversible since the enzyme activity
lost could not be restored after dialysis (24 h) against
assay buffer. We confirmed that the action of each of
the nucleosides 4, 5, and 6 was active site-directed by
F igu r e 1.
irreversibly acylate nucleophilic residues involved in the
catalytic process within the enzyme cavity (Scheme 1,
pathway A). An alternative process involving the hy-
drolytic activity of AdoHcy hydrolase should not be
excluded. Addition of the enzyme’s sequestered water
to the S-allenylthio group of 5 or the S-propynylthio
group of 6 might generate the formation of the corre-
sponding reactive thioester 7 (Figure 1). Attack of such
an intermediate by amino acid functionalities of the
active site might also cause type II (covalent binding)
inhibition of the enzyme (Scheme 1, pathway B).
1
6
means of protection experiments with adenosine.
+
The effects of 4, 5, and 6 on the NAD /NADH content
of AdoHcy hydrolase were also determined after com-
plete inactivation of the enzyme. The freshly prepared
recombinant placental AdoHcy hydrolase was deter-
+
mined to contain 0.75 mol of NAD and 0.25 mol of
We now describe the syntheses of the new 5′-S-
propargyl-5′-thioadenosine 4, 5′-S-allenyl-5′-thioadeno-
sine 5, and 5′-S-propynyl-5′-thioadenosine 6, their in-
hibitory effects on AdoHcy hydrolase, and their antiviral
activities. Compounds 4, 5, and 6 are irreversible
inactivators of the enzyme, and we undertook a mass
spectral analysis of the inactivated protein to provide
evidence for the formation a covalent adduct induced
by mechanism-based inhibitors.
NADH per enzyme subunit, in accord with reported
6
b
data.
5′-S-Propargyl-5′-thioadenosine 4 (600 µM) caused
complete conversion of the enzyme (20 µM) from its
+
E-NAD to E-NADH form during the inactivation
process. This is indicative that 4 is substrate for the 3′-
oxidative activity of AdoHcy hydrolase. Surprisingly,
inactivation of AdoHcy hydrolase, carried out under the
same conditions, with 5 and 6 resulted in only minor
+
changes in the initial NAD /NADH ratio (Table 2).
Ch em istr y
Upon complete inactivation of the enzyme (25 µM)
with 200 µM of 4, 5, and 6, the reaction products were
analyzed by liquid chromatography/electrospray ioniza-
tion-mass spectrometry (LC/ESI-MS). With 4, no trace
of product other than unreacted nucleoside 4 was
detected.
The general synthetic procedure used for the prepara-
tion of 4, 5, and 6 was as follows (Scheme 2). Tosylation
6
of N -benzoyl-2′,3′-isopropylidene adenosine 1, readily
available from adenosine,12 followed by reaction with
potassium thioacetate gave rise to thioacetate 2. Treat-
ment of 2 with 1 equiv of sodium methoxide in methanol
followed by alkylation with propargyl bromide gave the
corresponding 5′-S-propargyl-5′-thioadenosine deriva-
tive. In situ removal of the N-benzoyl protecting group
of this intermediate can be achieved readily by stirring
the resulting solution with a second equivalent of
MeONa at 40 °C to afford 3 in a total yield of 72% from
In contrast, in experiments carried out with 5 and 6,
among few other detected compounds, adenosine was
identified by its retention time on the column (compared
with Ado used as authentic) and its molecular peak (M
+
+ H) m/z ) 268. This result indicated that the
inhibition process with 5 and 6 is competing with
substrate activity. The partition ratio was estimated to
be 90/10 in favor of inactivation. This result was
confirmed by the inactivation assays of AdoHcy hydro-
lase carried out with 5 and 6 in the presence of a large
excess (2.5 mM) of D,L-homocysteine. In these experi-
ments, AdoHcy was detected and identified by its
2
. Because of their acid instability, the desired com-
pounds 5 and 6 had to be prepared from the deprotected
S-propargylthioadenosine 4. Removal of the isopropy-
lidene protecting group of 3 with aqueous 80% HCOOH
resulted in the formation of 5′-S-propargyl-5′-thioad-
1
3
+
enosine 4 in 72% yield. Base-catalyzed isomerization
molecular peak (M + H) m/z ) 385 in the reaction
of 4, using 1 or 3.5 equiv of potassium tert-butoxide in
tert-butyl alcohol at 50 °C afforded the desired target
nucleosides 5′-S-allenyl-5′-thioadenosine 5 or 5′-S-pro-
pargyl-5′-thioadenosine 6, respectively.
product.
This result is important to explain the biphasic
inactivation-time profiles observed for inactivation of
AdoHcy hydrolase with 5 and 6. A progressive formation

Synthesis of AdoHcy Hydrolase Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17 2745
Sch em e 1. Possible Generation of Active Intermediates from 5 and 6 after Interaction with AdoHcy Hydrolase
Sch em e 2a
NADH content of the enzyme suggests that the main
pathway by which 5 and 6 proceed to inactivate AdoHcy
hydrolase do not involve its 3′-oxidative activity.
The tertiary structure of human AdoHcy hydrolase
+
complexed with NAD and the type I mechanism-based
inhibitor (1′R,2′S,3′R)-9-(2′,3′-dihydroxycyclopentan-1′-
1
7
yl)-adenine is presently known. Taking in account the
information on the position and interactions of this
inhibitor cocrystallized in the AdoHcy hydrolase struc-
ture, a computer model of 5, docked at the active site of
the enzyme, was established (Figure 3). This model
revealed the proximity of the sequestered water mol-
ecule involved in the hydrolytic activity of the enzyme
to the C-6′ of the allenylthio group of 5. This observa-
tion, and the fact that 5 and 6 can be easily transformed
by acid treatment (aqueous formic acid) into 5′-S-
propionyl-5′-thioadenosine 7, strongly suggests the pos-
sible reaction of sequestered water with 5 to generate
a thioester intermediate. Attack by amino acid func-
tionalities of the enzyme’s active site might cause
irreversible covalent inactivation accompanied by a
covalent linkage of 54 Da on each enzyme subunit.
In the second proposed mechanism of inactivation
(Scheme 1, pathway A), each catalytic event with 5 and
6
also produces 3′-keto-4′,5′-dehydroadenosine, an in-
termediate which is converted by normal enzymatic
reaction into adenosine, in the absence of homocysteine.
If reactive mercaptans produced in this process are not
immediatly quenched by reaction with a nucleophile,
adenosine might be formed. Since the formation of
adenosine was observed upon incubation of 5 and 6 with
the enzyme, the partial involvement of allenyl or pro-
pynyl mercaptan groups in the covalent irreversible
inactivation of AdoHcy hydrolase cannot be excluded.
If occurring, such an inactivation process might be
accompanied this time by a covalent linkage of 72 Da.
a
Reagents and conditions: (a)TsCl, pyridine, 95%; (b) CH3COSK,
DMF, 60 °C, 70%; (c) MeONa, Me0H; HC≡CCH2Br (72%); (d) 80%
HCOOH, 40 °C, 72%; (e) 1 equiv tBuOK , tBuOH, 50 °C, 1H (55%);
(
f) 3.5 equiv tBuOK, tBuOH, 50 °C, 3H (60%).
of Ado during the interaction of 5 and 6 with the enzyme
could protect AdoHcy hydrolase against its irreversible
inactivation by 5 and 6. Similar biphasic kinetics have
already been observed when AdoHcy hydrolase was
To discriminate between these two possibilities, the
mechanism of inactivation was further investigated
using ESI-MS analysis of AdoHcy hydrolase inactivated
by 5.
inactivated by 5′,6′-didehydro-6′-deoxy-6′-fluorohomo-
adenosine.6b
From these initial results it appeared that inactiva-
tion of AdoHcy hydrolase by 4 involves a type I mech-
anism (cofactor depletion).5
Accurate molecular weight of AdoHcy hydrolase sub-
unit was determined from the wide distribution of
highly charged states observed in ESI-MS analysis
under the acidic conditions described in Experimental
Section. As shown in Figure 5 (panel A), the freshly
In contrast, the fact that the interacion of 5 and 6 with
+
AdoHcy hydrolase causes minor changes in the NAD /

2
746 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17
Guillerm et al.
Ta ble 1. Inhibition Constants for 4, 5, and 6 with AdoHcy
Hydrolase
Ki
kinact
t1/2
(µM) (min-1) (min)
compd
type of inhibition
4
5
6
irreversible, time-dependent
irreversible, time-dependent
irreversible, time-dependent
45
16
16
0.0074
0.12
0.062
19a
b
7
c
7
a
b
The half-time (t1/2) of enzyme inactivation at 15 µM of 4. The
half-time (t1/2) of enzyme inactivation at 3 µM of 5. c The half-time
t1/2) of enzyme inactivation at 40 µM of 6.
(
+
Ta ble 2. Effect of 4, 5, and 6 on NAD /NADH Content of
AdoHcy Hydrolasea
NAD+ content
mol/mol of
NADH content
(mol/mol of
(
sample
enzyme subunit)
enzyme subunit)
purified enzyme
enzyme + 4
enzyme + 5
enzyme + 6
0.75 ( 0.02
0
0.70 ( 0.02
0.68 ( 0.02
0.25 ( 0.04
1 ( 0.04
0.30 ( 0.04
0.32 ( 0.04
a
AdoHcy hydrolase (20 µM) was incubated with 600 µM 4, 5,
+
and 6 in assay buffer at 37 °C until total inactivation. The NAD /
NADH content was assayed as described in the Experimental
Section.
F igu r e 3. Compound 5 (colored by atom type) docked into
the active site of AdoHcy hydrolase. For clarity, only the
residues providing the main hydrogen bond interactions and
the NAD+ molecule are shown. Hydrogen bonds are repre-
sented as dotted lines. The proximity of a water molecule (red
sphere) to the allenyl moiety of the inhibitor strongly suggests
the possibility of a reaction with water to give the thioester
intermediate 7.
charged species was detected at 664.2 m/z which was
+
attributed to NAD and/or NADH, the cofactors present
in AdoHcy hydrolase. After incubation with 5, the
inactivated enzyme was detected at a molecular weight
of 47659.5 ( 3.0 Da (Figure 4, panel B) showing that
inhibition of the enzyme was accompanied by a covalent
linkage of 54 ( 4.5 Da.
This result supports the mechanistic proposal de-
scribed in Scheme 1 (pathway B) via the putative
formation of a thioester intermediate which could ir-
reversibly acylate a nucleophilic residue on the enzyme.
Further evidence in support of this mechanism comes
from inactivation experiments carried out with 5′-S-
propionyl-5′-thioadenosine 7 prepared in our laboratory.
F igu r e 2. Time-dependent inactivation of AdoHcy hydrolase
with 4 (panel A), 5 (panel B), and 6 (panel C). AdoHcy
hydrolase was incubated with inhibitors at concentrations of
5
-40 µM in assay buffer at 37 °C. At the indicated time points,
residual enzyme activity was determined in the synthetic
direction as described in the Experimental Section.
purified enzyme was present with a major molecular
species being detected at 47604.8 ( 1.5 Da. A singly

Synthesis of AdoHcy Hydrolase Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17 2747
2 3
F igu r e 4. Panel A: Positive ion-ESI mass spectrum recorded for purified AdoHcy hydrolase (5 µM in 49.5:49.5:1 (v/v) H O-CH -
CN-HCOOH). The enzyme displays a wide distribution of highly protonated states, some of which are indicated. The peak labeled
with a solid circle (b) corresponds to the singly charged ion produced by NADH (664 Da), the known cofactor of AdoHcy hydrolase.
The peaks labeled with a star (*) in which mass exceeds by 665 ( 5 Da may possibly come from residual cofactor associated to
the enzyme. Panel B: ESI mass spectrum of Ado Hcy hydrolase inactivated with 5.
Inactivation of AdoHcy hydrolase with 7 was shown to
be both concentration- and time-dependent as predicted
for the chemical modification observed on the inacti-
vated protein (∆M ) 54.8 ( 4.5 Da). These studies
(described in Experimental Section) allowed the deter-
mination of the whole primary sequence of the peptide
detected at 2414.0 m/z, which was identified as the
Endo-LysC cleaved C-terminal peptide of AdoHcy hy-
drolase QAQYLGMSCDGP F KP DHYRY containing
an unlocalized adduct of 38 Da. The excess mass (38
Da) recovered by analysis of the cleavage products of
inactivated AdoHcy hydrolase is 17 Da lower than
expected from the protein subunit before proteolysis.
The loss of OH group (-17 Da) from the C-terminal end
peptide of the protein, possibly induced by the MALDI
process itself, might explain this difference.
(Figure 6), and kinetic analysis of the inactivation
process by the Kitz and Wilson method gave kinetic
-1
constants (Ki ) 78 µM; Kinact ) 0.155 min ) in the range
of those obtained with 5 and 6.
To localize the amino acid(s) involved in the binding
of ∆M ) 54.8 ( 4.5 Da, AdoHcy hydrolase inhibited by
S-allenylthioadenosine 5 was submitted to proteolytic
cleavage by Endo-LysC. The cleavage products (i.e., a
mixture of peptides C-terminally cleaved at Lys resi-
dues) were analyzed without further separation by
MALDI-MS, and compared to those obtained from
native AdoHcy hydrolase. This comparison revealed a
singly charged ion at 2414.0 m/z in the MALDI mass
spectrum of the inactivated protein, which was absent
in the mass spectrum of the native protein (Figure 5).
A separation of the proteolysis cleavage peptides was
achieved by LC/ ESI-MS for both the native and the
inhibited AdoHcy hydrolase. The fraction containing the
peptide at 2414.0 m/z was isolated. This fraction was
An tivir a l Activity
Compounds 4, 5, and 6 were evaluated for antiviral
1
8
activity in cell cultures (Table 3). No significant
antiviral activity was observed against Parainfluenza-
3, Sindbis, Punta Toro, or Coxsackie virus B4 in Vero
cells at concentrations that were significantly (>5-fold)
lower than the cytotoxicity threshold. No specific anti-
viral activity was noted against any other viruses, not
even vesicular stomatisis virus or reovirus which nor-
n
submitted to further nanoESI-MS experiments on an
Ion-Trap instrument in order to characterize its primary
structure and to determine whether it was responsible

2
748 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17
Guillerm et al.
F igu r e 5. MALDI mass spectra recorded in the reflector positive mode (range 2345-2435 m/z) for the peptides mixture obtained
by enzymatic (Endo-LysC) proteolysis of AdoHcy hydrolase (A) and inhibited AdoHcy hydrolase (B). The singly charged ion detected
at 2414.0 m/z (5B) is not present on the mass spectrum of the purified protein (5A), which may suggest that this peptide carries
the modification observed on the entire (inhibited) protein. The ion detected at 2376.0 m/z for both the purified and the inactivated
protein is attributed to the singly charged ion produced by the Endo-LysC cleaved C-terminal peptide of the protein (i.e., 20 last
amino acids of the protein: QAQYLGMSCDGPFKPDHYRY). While the singly charged ion detected at 2359.0 m/z does not
correspond to any expected cleavage product and is rather imputed to the loss of an OH group (-17 Da) from the C-terminal-end
peptide of the protein. This loss may possibly be induced by the MALDI process itself.
time- and concentration-dependent inactivation of
AdoHcy hydrolase.
Our results reveal that inactivation of AdoHcy hy-
drolase by 4 involves a type I mechanism. Compounds
5
and 6 inactivate the enzyme in a different manner
+
since these two inactivators do not induce NAD deple-
tion. The main pathways of the inactivation process
have been determined with mass spectral analysis of
the inactivated protein. ESI-MS analysis of the enzyme
inactivated with 5 demonstrates that inhibition of
AdoHcy hydrolase is accompanied by covalent modifica-
tion of each subunit of the enzyme.
Moreover, 5 and 6 are easily transformed by acid
hydrolysis into 5′-S-propionyl-5′-thioadenosine, a reac-
tive thioester which can inactivate AdoHcy hydrolase
in a time- and concentration-dependent manner. All
these results are consistent with the mechanistic pro-
posal depicted in Scheme 1 (pathway B) in which 5 or 6
could be activated by the “hydrolytic activity” of AdoHcy
hydrolase. Thus, they represent a new class of type II
covalent inhibitors of AdoHcy hydrolase. In addition, we
demonstrated that ESI-MS and Nano ESI-MSn are
valuable tools for studying the nature and localization
of the covalent adduct formed during covalent mecha-
nism-based inhibition.
F igu r e 6. Time-dependent inactivation of AdoHcy hydrolase
with 5′-S-propionyl-5′-thioadenosine 7. AdoHcy hydrolase was
incubated with this inhibitor at concentrations of 5, 10, 15,
and 20 µM in assay buffer at 37° for various times. At the
indicated time, residual enzyme activity was determined in
the synthetic direction as described in the Experimental
Section.
Dihalovinyl and dihaloacetylene analogues derived
from adenosine have been characterized as first type II
mechanism-based inhibitors of human placental AdoHcy
hydrolase that relies only on the enzyme’s hydrolytic
7
b-d
activity”.
Arg 196 and Lys 318 were identified as
mally fall within the activity spectrum of the AdoHcy
hydrolase inhibitors. However, 6 and particularly 5
proved to be specifically active against vaccinia virus
in E6SM cells, at a concentration that was >10-fold
lower than the cytotoxic concentration. For 5, the EC50
amino acid residues that attack an activated intermedi-
ate catalytically generated from 6′-bromo-5′,6′-didehy-
7
c
dro-6′-deoxy-6′-fluorohomoadenosine and 5′,5′,6′,6′-
tetrahydro-6′-deoxy-6′-chloro- and 6′-iodoadenosine,7d
respectively, whereas, from our results it appeared that
5 covalently modifies each enzyme’s subunit at the
C-terminal peptide of the protein (i.e., 20 last amino
acids)sa peptide sequence which contains Lys 426 and
Tyr 430, two essential amino acids for binding of the
(50% effective concentration) in inhibiting vaccinia virus
replication was 1.9 µM.
Con clu sion
In summary, we have synthesized a new series of
nucleosides containing a 5′-sulfur atom that produces
1
7,19
NAD cofactor and enzyme activity.

Synthesis of AdoHcy Hydrolase Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17 2749
Ta ble 3. Antiviral Evaluation of 4-6
b
EC50 (µM)
HSV-1/TK-
HSV-1 HSV-2 B2006 or vaccinia
KOS) (G) VMW1837 virus
SM HeLa Vero (E SM) (E SM) (E SM) (E SM) (E
Punta
MICa (µM)
Coxsackie
parainfluenza-3
virus
Toro
reovirus virus
(Vero) (Vero)
(
VSV
6
VSV virus B4 RSV
RSV
compd
E
6
6
6
6
6
SM) (HeLa) (HeLa) (Vero) (HeLa)
(Vero)
4
5
6
c
d
e
1240 1240 1240 >250 >250
>250
150
>250
>250
1.9
49
>250
>250
>250
>250
>120
>600
>120
230
>250 >250
>250 >250
>1240 >250
>2000 2000
>1600 1.5
>120
>620
>120
230
>120
>380
>120
45 >2000
40 40
>120
>620
>120
1240 1240 1240 250
1240 >1240 1240 250
>2000 >2000
>250
620
>250 >1240
230
1000
>1600 >1600 >1600 320
320
0.005
65
>100
40
>100
>320
196
>1600
0.0012 0.0012
b
EC50 (µM)
+
-
TK VZV
TK VZV
CMV
MICa (µM)
YS
strain
(HEL)
OKA
strain
(HEL)
07/01
strain
(HEL)
YS/R
strain
(HEL)
AD-169
strain
(HEL)
Davis
strain
(HEL)
Sindbis
J unin
virus
(Vero)
Tacaribe
virus
virus
compd
HEL
Vero
(Vero)
(Vero)
4
5
6
f
d
e
>200
>200
>200
>150
>200
>200
>200
>200
>200
>200
4.09
>200
>200
>200
4.53
154
87
200
41
126
>200
200
>200
>200
>200
>200
>200
>200
>40
>200
>40
>200
>200
>200
>200
>200
>200
63
>500
240
3.2
3.2
>150
7.8
11.8
a
b
Minimal inhibitory concentration required to elicit a microscopically visible alteration of cell morphology. Effective concentration
required to inhibit virus-induced cytopathicity by 50%. (S)-2,3-Dihydroxypropyladenine [(S)-DHPA]. d Ribavirin. e 9-(1,3-Dihydroxy-2-
propoxymethyl)guanine (DHPG, gancyclovir). f 9-(2-Hydroxyethoxymethyl) guanine (ACV, acyclovir). Abbreviations: TK , thymine kinase
deficient; HSV, Herpes simplex virus; VSV, vesicular stomatitis virus; RSV, respiratory synaptial virus; E6SM, embryonic skim-muscle
cells; VZV, variecella-zoster virus; CMV cytomegalovirus.
c
-
1H, H-1′, J ) 2.29); 7.93 and 8.35 (2 s, 2H, H-2 and H-8). 13C
NMR, CDCl , δ (ppm): 19.6 (CH CC); 25.3 and 27.0 (CH ) C);
Additional studies are in progress using nano ESI-
MS techniques to elucidate the exact localization of the
n
3
2
3 2
3
3.5 (C-5′); 71.5(CCH); 79.4 (CCH); 83.1 (C-4′); 83.8 (C-3′); 86.7
C-2′); 90.7 (C-1′); 114.4 ((CH C); 120.2 (C-6); 139.9 and 153.0
C-2 and C-8); 149.1 and 155.8 (C-4 and C-5).
′-S-P r op a r gyl-5′-th ioa d en osin e 4. A total of 315 mg
0.87 mmol) of 3 in 3 mL of formic acid/water solution (8/2)
covalent linkage induced by AdoHcy with 5, 6, and 7 as
well as to clarify the mechanism of inactivation of other
analogues such as 5′-S-ethynyl-5′-thioadenosine and 5′-
S-vinyl-5′-thioadenosine identified recently as new mech-
anism-based inhibitors of the enzyme.
(
(
3 2
)
5
(
was heated at 40 °C for 1.5 h. After cooling, formic acid was
evaporated under reduced pressure, and traces of acid were
removed by coevaporation with water (3 times). The crude
material was dissolved in 3 mL of water, and the pH of the
Exp er im en ta l Section
Gen er a l. NMR spectra ( H, 250 MHz; 13C, 62.5 MHz) were
recorded on a Brucker spectrometer. IR spectra were recorded
on a FTIR Midac. Optical rotations were measured on a
Perkin-Elmer 241 polarimeter. Mass spectra were obtained
using a VG Instruments AUTOSPEC spectrometer. Protein
electrospray ionization were performed on a VQ Bio Q triple
quadrupole spectrometer. Flash column chromatography was
performed using 9385 Merck Kieselgel 60 and TLC with silica
gel plates Merck 60 F254. Purification on hydrophobic resin was
performed using a HP20SS Resin from Mitsubishi Fine
Chemicals. Analytical reversed-phase HPLC analyses were
performed with a C-18 column (ODS2 spherisorb 5 µm; 250 ×
1
4
solution was adjusted to pH 7 with a few drops of 2 N NH -
OH. Crude 4 was purified on a column of hydrophobic resin
HP20SS (elution: methanol/water 3/7). Evaporation of metha-
nol and lyophilization gave 4 (200 mg, 72%). mp: 87-89 °C.
2
0
[
R] -20.5° (c 0.23; CHCl
3
/MeOH: 4/1). IR (KBr) 2170, 3281
, ꢀ CD OD δ (ppm), J (Hz): 2.27 (t, 1H,
CCH, J 2.67); 3.01 (dd, 1H, H5′R, J ) 14.11, 6.1); 3.12 (dd,
H, H5′â, J ) 14 0.11, 4.57); 3.28 (d, 2H, CH CC, 2.67); 4.21
m, 1H, H-4′); 4.29 (dd, 1H, H-3′, J ) 5.34, 5,72); 4.46 (dd, 1H,
H-2′, J 5.34, 3.81); 5.81 (d, 1H, H-1′, J ) 3.81); 8.09 and 8.21
2 s, 2H, H-2 and H-8). 13C NMR, CDCl
, ꢀ CD OD, δ (ppm):
CC); 33.3 (C-5′); 71.6(CCH); 72.3 (C-4′); 74.2 (C-3′);
0.0 (CCH); 83.9 (C-2′); 89.6 (C-1′); 120.0 (C-6); 139.3 and 152.8
D
-
1 1
cm . H NMR, CDCl
3
3
1
(
2
(
2
8
3
3
0.0 (CH
2
4
.6 mm).
′,3′-O-Isop r op ylid en e-5′-S-p r op a r gyl-5′-t h ioa d en os-
2
3
(C-2 and C-8); 149.1 and 155.9 (C-4 and C-5). HRMS (CI/NH )
in e 3. To a dry oxygen-free methanol solution (15 mL) of
thioacetate 2 (2 g, 4.26 mmol) was added 1 equiv of sodium
methoxide (226 mg, 4.26 mmol) at 0 °C under argon, and
stirring was continued for 8 h at room temperature. To the
resulting solution was added propargyl bromide (0.32 mL, 4.26
mmol) in 3 mL of methanol at 0 °C. The reaction mixture was
allowed to warm to room temperature, 1 equiv (226 mg) of
sodium methoxide was then added, and stirring continued at
calcd C13
(C H O
13 15 3
H
15
O
3
S + H 322.0973, found 322.0979. Anal.
S) C, H, N.
5′-S-Allen yl-5′-t h ioa d en osin e 5. A total of 160 mg (0.5
mmol) of 4 was dissolved in 5 mL of anhydrous tert-butyl
alcohol at 50°, and 1 equiv (56 mg, 0.5 mmol) of potassium
tert-butoxide was then added. The reaction was followed by
analytical reversed-phase HPLC (elution methanol/water 5/5).
After complete transformation of 4 into 5, the reaction mixture
was cooled to room temperature, and 3 mL of water was added.
tert-Butyl alcohol was evaporated under reduced pressure, and
the pH of the resulting solution was carefully adjusted to pH
7 at 0 °C. Crude 5 was purified on a column of hydrophobic
resin HP20SS (elution: methanol/water 3/7). Evaporation of
methanol and lyophilization gave pure 5 (90 mg, 55%). mp:
4
0° for 1 h. Methanol was evaporated under reduced pressure,
and the crude product obtained was extracted with ethyl
acetate/water. The crude material was purified by flash column
chromatography on silica gel (ethyl acetate/methanol 9/1) to
give pure 3 (1.10 g, 72%) as a white foam. mp: 65 °C. MS (DCI/
+
20
D
NH
156 cm . H NMR, CDCl
s, 6H, (CH C); 2.17 (dd, 1H, CCH, J ) 2.67); 3.02 (d, 2H,
H-5′, J ) 6.87); 3.26 (d, 2H, CH CC, J ) 2.67); 4.48 (ddd, 1H,
H-4′, J ) 3.44, 6.87, 6.87); 5.12 (dd, 1H, H-3′, J ) 6.1, 3.44);
.51 (dd, 1H, H-2′, J ) 2.29, 6.1); 5.98 (s, 2H, NH ); 6.09 (d,
3
): 362 (MH) . [R] -21° (c 0.23, CHCl
3
). IR(KBr) 21.05;
, δ (ppm), J (Hz): 1.40 and 1.62 (2
-
1 1
3
3
2
D
0
3
)
2
90-92 °C. [R] -18° (c 0.18; CHCl
3
/MeOH: 4/1). IR (KBr)
-
1
1
2
1944 cm (weak). H NMR, CDCl
2.93 (dd, 1H, H-5′R, J ) 14.5, 4.96); 3.05 (dd, 1H, H-5′â, J )
14.5, 4.96); 4.21 (m, 1H, H-4′); 4.22 (dd, 1H, H-3′, J ) 6.1, 4.95);
3
, ꢀ CD OD δ (ppm), J (Hz):
3
5
2

2
750 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17
Guillerm et al.
4
.49 (dd, 1H, H-2′, J ) 4.95, 4,2); 4.94 (d, 2H, CH
2
dCdCHS,
Assa y of Ad oHcy Hyd r ola se Activity. AdoHcy hydrolase
activity was assayed in the synthetic direction by measuring
J ) 6.48); 5.70 (dd, 1H, CH
H-1′, J ) 4.2); 8.05 and 8.15 (2 s, 2H, H-2 and H-8). C NMR,
CDCl , ꢀ CD OD, δ (ppm): 34.2 (C-5′); 72.1 (C-4′); 74.2 (C-3′);
1.0 (CH dCdCHS); 83.0 (C-2′); 87.0 (CH dCdCHS); 89.3 (C-
′); 119.4 (C-6); 139.1 and 152.2 (C-2 and C-8); 148.7 and 155.3
dCdCHS). HRMS (CI/NH ) calcd
S + H 322.0973, found 322.0979. Anal. (C13 S)
2
dCdCHS, J ) 6.48); 5.86 (d, 1H,
1
3
14
14
the rate of formation of [8- C]-AdoHcy from [8- C]-Ado and
2
0
3
3
Hcy as described previously by Della Ragione et al.
1
4
8
1
2
2
Enzyme was mixed with 10 µM [8- C]-Ado (17 800 dmp), 5
mM D,L Hcy, 20 mM potassium phosphate buffer pH 7.5, 1
mg/mL BSA, and 1 mM EDTA in a final volume of 50 µL. The
assay mixture was incubated at 37 °C for 10 min and the
reaction stopped by addition of 150 µL of 15 mM HCl. A total
of 150 µL of the sample was then applied to a microcolumn
(0.5 mL) of Cellex P equilibrated with 10 mM HCl. The
(C-4 and C-5); 205.6 (CH
2
3
C
13
H
15
O
3
15 3
H O
C, H, N.
5
′-S-P r op yn yl-5′-th ioa d en osin e 6. A total of 225 mg (0.62
mmol) of 4 was dissolved in 10 mL of anhydrous tert-butyl
alcohol at 50 °C, and 3.5 equiv (224 mg, 2.18 mmol) of
potassium tert-butoxide was added. The reaction mixture was
stirred for 2.5 h at 50 °C. After the reaction mixture was cooled
to room temperature, 5 mL of water was added. tert-Butyl
alcohol was evaporated under reduced pressure, and the pH
of the resulting solution was carefully adjusted to pH 7 at 0
1
4
remaining [8- C]-Ado was eluted first with 5 mL of 10 mM
1
4
HCl and then the [8- C]-AdoHcy formed with 5 mL of 0.5 N
HCl. The eluates were directly poured into scintillation vials
with 7 mL of Ultima Flow AP (Packard) for counting. Under
these conditions, the velocity of AdoHcy hydrolase showed
m
normal Michaelis-Menten kinetics with the K value for Ado
2
1
°
C. Crude 6 was purified on a column of hydrophobic resin
being 1 µM, in accord with the literature value.
+
HP20SS (elution: methanol/water 3/7). Evaporation of metha-
nol and lyophilization gave pure 6 (140 mg, 60%). mp: 95-97
Deter m in ation of NAD an d NADH Con ten t of AdoHcy
Hyd r ola se. NAD and NADH present in enzyme before and
+
2
D
0
1
°
C. [R] +22.5° (c 0.020; CHCl
3
/MeOH: 4/1). H NMR,
OD, δ (ppm), J (Hz): 1.87 (s, 3H, CCCH ); 3.02
dd, 1H, H-5′R, J ) 13.72, 6.5); 3.13 (dd, 1H, H-5′â, J ) 13.72,
after inactivation with 4, 5, and 6 were mesured as previously
described6
a,22
by monitoring the intrinsic fluorescence emitted
CDCl
(
6
3
, ꢀ CD
3
3
by NADH at 460 nm upon excitation at 340 nm. NADH was
+
.1); 4.31 (m, 1H, H-4′); 4.32 (dd, 1H, H-3′, J ) 2.27, 5.2) 4.39
measured directly whereas NAD was first converted to NADH
(dd, 1H, H-2′, J ) 5.2, 3.6); 6.04 (d, 1H, H-1′, J ) 3.6); 8.19
by Baker’s yeast alcohol dehydrogenase in the presence of
ethanol.
1
3
and 8.29 (2 s, 2H, H-2 and H-8). C NMR, CDCl
δ (ppm): 4.4 (CCCH ); 39.2 (C-5′); 68 (CCCH ); 73.8 (C-4′); 74.8
C-3′); 84.8 (C-2′); 89.0 (CCCH ); 90 (C-1′); 119.4 (C-6); 141.2
and 153.8 (C-2 and C-8); 150 and 157.3 (C-4 and C-5). HRMS
3 3
, ꢀ CD OD,
3
3
(a ) Sa m p le P r ep a r a tion . AdoHcy hydrolase (20 µM) was
incubated in 200 µL of assay buffer with (600 µM) of 4, 5, or
(
3
+
6 at 37 °C. After total inactivation, both NAD and NADH
(
CI/NH
3
) calcd C13
S) C, H, N.
′-S-P r op ion yl-5′-th ioa d en osin e 7. To a dry oxygen-free
H
15
O
3
S + H 322.0973, found 322.0837. Anal.
were released from the enzyme by addition of 500 µL of ethanol
and centrifuged (10000g for 5 min) to sediment precipitated
material. The precipitate was washed twice with 200 µL of
ethanol, and the pooled supernatants were lyophilized in the
13 15 3
(C H O
5
methanol solution (15 mL) of thioacetate 2 (1 g, 2.13 mmol)
was added 2 equiv of sodium methoxide (226 mg, 4.26 mmol)
at 0° under argon. Stirring was continued for 8 h at room
temperature. To the resulting corresponding debenzoylated
thiolate, a solution of propionic anhydride (0.38 mL, 3 mmol)
in 3 mL of methanol was added at 0 °C. The reaction mixture
allowed to warm to room temperature, and stirring was
continued for 3 h. Methanol was evaporated under reduced
pressure, and the crude product obtained was taken in 3 mL
of formic acid/water solution (8/2). After heating at 40° for 1.5
h, formic acid was evaporated and the crude material neutral-
2
dark. The residue was dissolved in 0.4 mL of H O and 0.6 mL
of 100 mMphosphate buffer, pH 8.8. The resulting solution was
+
analyzed for NAD and NADH.
(b) Effect of Ad en osin e on th e In a ctiva tion of Ad oHcy
Hyd r ola se by 4, 5, a n d 6. AdoHcy hydrolase (0.1 µg) was
preincubated in 50 µL of assay buffer for 20 min at 37 °C with
the inactivators 4, 5, or 6 alone or in the presence of adenosine
(14-28 µM). Remaining activity was assayed by the transfer
of a 10 µL portion of the incubation mixture to 40 µL of
standard assay mixture.
ized (2 N NH
4
OH). Crude 7 was purified on a column of
An a lysis of Rea ction P r od u cts. AdoHcy hydrolase (28
µM) was incubated with 4, 5, or 6 (0.2 mM) in 200 µL of 20
mM ammonium acetate, 1 mM EDTA, 1 mg/mL BSA, pH 7.5,
without or in the presence of (2.5 mM) D,L-homocysteine. After
inactivation of the enzyme was completed, the reaction mix-
tures were filtered through an ultrafree-MC, 10000 Millipore
filter. The filtrates were analyzed by liquid chromatography/
electrospray ionization-mass spectrometry. The filtrates were
injected into an HPLC column (Spherisorb, C18, 5 µM, 250 ×
4.6 mm, Interchrom). The column was eluted at a flow rate of
hydrophobic resin HP20SS (elution: methanol/water 3/7).
Evaporation of methanol and lyophilization gave 7 (430 mg,
+
20
6
0
0%). mp: 60 °C. MS (DCI/NH
3
): 340 (MH) . [R] -6.1° (c
D
1
.392, MeOH). IR (KBr): 1650 cm-1. H NMR, DMSO-d
, δ
6
(
ppm), J (Hz): 1.10 (t, 3H, J ) 7.6); 2.6 (q, 2H, J ) 7.6); 3.15
(dd, 1H, H5′R, J ) 13.7, 7.3); 3.40 (dd, 1H, H5′â, J ) 13.7,
5
.2); 3.90 (m, 1H, H-4′); 4.10 (m, 1H, H-3′); 4.80 (ddd, 1H, H-2′,
J ) 5.6, 5,7, 5.6); 5.40, 5.60 (2d, D
OH); 5.90 (d, 1H, H-1′, J ) 5.7); 7.35 (s, 2H, NH
.40 (2 s, 2H, H-2 and H-8). 13C NMR, DMSO-d
, δ (ppm): 9.5
CH -CH C(O)S); 30.9 (CH -CH C(O)S); 36.9 (C-5′); 72.6 (C-3′
and C-4′); 74.2; 83.0 (C-2′); 87.6 (C-1′); 119.2-156.0 (C-
adenine); 199.3 C(O). Anal. (C13 S) C, H, N.
2
O exchangeable 3′- and 2′-
2
); 8.15 and
8
6
1 mL/min in gradients of 20-80% (MeOH/H
2
O) over 8 min
(
3
2
3
2
and 40-60% (MeOH/H O) over 30 min. The effluent was flow-
2
split with 1/10 of the flow directed toward the ESI mass
spectrometer and the residual 9/10 directed toward the UV
detector. ESI-MS data were recorded in the positive ion mode.
In these conditions, retention time for AdoHcy, Ado, 4, 5, and
17 5 4
H N O
En zym e P u r ifica tion . The recombinant human placental
AdoHcy hydrolase was purified from cell-free extracts of J M
6
were 5.3, 7.5, 26.4, 28.7, and 30.3 min, respectively, and the
1
2
09 Escherichia coli transformed with the plasmid pPROK cd
0 according to the procedure described by Yuan et al.6a for
compounds were characterized by their molecular peaks
+
(
MH ). In experiments with 5 and 6, in addition to Ado,
the first steps of purification, including ammonium sulfate
fractionation, DEAE Cellulose, and Sephacryl S200. The
purification was continued by FPLC. Aliquots of S200 purified
AdoHcy hydrolase (2 mL) were recycled twice on a Mono QHR
another peak of m/z 271 was detected at 11.4 min retention
time. It has been attributed to 5′-thioadenosine.
Ch em ica l Sta bility of 5 a n d 6. Compounds 5 and 6 proved
to be stable in assay buffer for several days at 37 °C (HPLC
control) but were subjected to facile acid hydrolysis. Compound
6 was quantitatively transformed into 5′-S-propionyl-5′-thio-
adenosine 7 when treated with aqueous HCOOH (80%) at 40
°C for 1 h. Under the same conditions, 5 led to the formation
of 7 and a substantial amount of 5′-thioadenosine (NMR
characterization).
1
0/10 column equilibrate 1 with 10 mM Tris/HCl, 1 mM
dithiothreitol, pH 7.8 and developed with a gradient from 0 to
.35 M NaCl at a flow rate of 4 mL/min. Active fractions were
pooled, concentrated by ultrafiltration, and dialyzed against
00 vol of 50 mM sodium phosphate, pH 7.4, 1 mM dithio-
0
1
threitol. The homogeneity of the AdoHcy hydrolase preparation
was checked by ESI-MS, the major molecular species being
detected at 47604.8 ( Da for each enzyme subunit (Figure 5,
panel A).
Molecu la r Mod elin g. The structure of AdoHcy has been
extracted from the Brookhaven Database (PDB entry 1A7A).

Synthesis of AdoHcy Hydrolase Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17 2751
Inhibitor 5 has been built with the molecular modeling
consisted of 0.1% TFA in H
2
O (solvent A) and 0.08% TFA in
software QUANTA 98.23 Its geometry has been optimized using
CH CN (solvent B); elution was performed at a flow rate of
3
2
4,25
the CHARMm force field.
The compound has been docked
250 µL/min using a biphasic gradient of 2% B (5 min), 2-60%
B (60 min), 60-80% B (5 min), and 80% B (5 min), successively.
The effluent was flow-split via a stainless steel Valco tee
(Supelco) with 1/15 of the flow directed toward the ESI mass
spectrometer and the residual 14/15 directed toward the UV-
detector. Fractions of the residual effluent were hand-collected.
manually from the information on the position and interactions
of the inhibitor cocrystallized in the 1A7A structure. When
unfavorable steric interactions occurred, the amino acid side
chains were manually reoriented. No subsequent geometry
optimization has been performed, as we were mainly interested
in a visual investigation of the possibility of a reaction between
the water molecule present in the enzymatic active site and
the C-6′ carbon of 5.
Ma ss Sp ectr om etr y. Abbr evia tion s: MS (mass spectrom-
etry), ESI (electrospray ionization), MALDI (matrix-assisted
laser desorption ionization), LC (liquid chromatography).
Samples of AdoHcy hydrolase were desalted by a five
dilution-concentration step dialysis (5 × 60 min) in 10 mM
ammonium acetate (pH 7.0) using Centricon PM3 concentra-
tors (Millipore). This desalting procedure allows the removal
of the inorganic salts that broaden the signal in mass
spectrometric analysis.
ESI-MS data were recorded in the positive ion mode using
a VG BioQ triple-quadrupole mass spectrometer. Scanning was
performed in the range 200-2000 m/z at a scan rate of 6 s/scan,
c
with a declustering voltage (V ) of 35 V. Before coupling the
chromatography system to the mass spectrometer, an external
calibration of the mass spectrometer was performed using the
multiply charged ions produced by horse heart myoglobin (2
µM in 49.5:49.5:1 (v/v) H
2
O-CH CN-HCOOH).
3
Ackn owledgm en t. The authors express their thanks
to Prof. Michael S.Hershfield and Dr. Francisco Arredon-
do (Duke University School of Medicine) for providing
a sample of E. coli transformed with a plasmid encoding
for human placental AdoHcy hydrolase and CNRS for
financial support.
Accu r a te Ma ss Mea su r em en t of Ad oHcy Hyd r ola se.
AdoHcy hydrolase (purified enzyme or enzyme inactivated by
5
) was diluted to 5 µM in a 1:1 H
2
3
O-CH CN mixture (v/v),
acidified with ca. 1% HCOOH. Mass data were collected in
the positive ion mode. Scanning was performed in the range
5
00-2000 m/z with a declustering voltage (V
c
) of 45 V.
Refer en ces
Calibration of the instrument is performed using the multiply
charged ions produced by horse heart myoglobin (2 µM in 49.5:
4
(
1) De la Haba, G.; Cantoni, G. L. The Enzymatic Synthesis of
S-Adenosyl-L-homocysteine from Adenosine and Homocysteine.
J . Biol. Chem. 1959, 234, 603-608.
9.5:1 (v/v) H
2
O-CH CN-HCOOH).
3
(
2) (a) Banerjee, A. K. 5′-Terminal Cap Structure in Eucaryotic
Messenger Ribonucleic Acids. Microbiol. Rev. 1980, 44, 175-
En zym a tic P r oteolytic Clea va ge of Ad oHcy Hyd r o-
la se. An aliquot of approximately 150 µg of AdoHcy hydrolase
purified enzyme or enzyme inactivated by 5) was incubated
2
05. (b) Ueland, P. M. Pharmacological and Biochemical Aspect
(
of S-Adenosylhomocysteine and S-Adenosylhomocysteine Hy-
drolase. Pharmacol. Rev. 1982, 34, 223-253. (c) Hasobe, M.;
McKee, J . G.; Borchardt, R. T. Relationship between Intracel-
lular Concentration of S-Adenosylhomocysteine and Inhibition
of Vaccinia Virus Replication and Inhibition of Murine L 929
Cell Growth. Antimicrob. Agents Chemother. 1989, 33, 828-834.
with either 1.2 µg of Endo-LysC (Boehringer-Mannheim) or 5
µg of Trypsin (Promega). Endo-LysC specifically cleaves pep-
tide bonds C-terminally at Lys; Trypsin specifically cleaves
peptide bonds C-terminally at Arg and Lys. For both enzymes,
the proteolysis was carried out overnight at 35 °C in 25 mM
NH HCO (pH 8.5); note that the final volume of the sample
4 3
never did exceed 60 µL. Proteolysis was stopped by freezing
the samples at -20 °C.
Ma tr ix-Assisted La ser Desor p tion Ion iza tion -Ma ss
Sp ectr om etr y (MALDI- MS). Cleavage products of AdoHcy
hydrolase were mass-analyzed without prior separation on a
BIFLEX MALDI-time of flight (MALDI-TOF) mass spectrom-
eter (Bruker Daltonics, Bremen, Germany), equipped with the
SCOUT high-resolution optics and a gridless reflector. Mass
data were collected in the reflector positive mode at a
maximum accelerating potential of 20 kV. Ionization was
accomplished with a pulsed nitrogen laser beam (λ ) 337 nm)
at a repetition rate of 3 Hz. External calibration of the mass
spectrometer was performed in the range 1000-2500 m/z by
using the singly charged ions produced by four peptides
(
d) Ransohoff, R. M.; Narayan. P.; Ayers, D. F., Rottman, F. M.,
Nilsen. T. W. Priming of Influenza mRNA Transcription is
Inhibited in CHO Cells Treated with the Methylation Inhibitor
Neplanocin A. Antiviral Res. 1987, 7, 317-327.
(
3) (a) Liu, S.; Wolfe, M. S.; Borchardt, R. T. Rational Approaches
to the Design of Antiviral Agents Based on S-Adenosyl-L-
homocysteine Hydrolase as a Molecular Target. Antiviral Res.
1
992, 19, 247-265. (b) Yuan, C.-S.; Liu, S.; Wnuk, S. F.; Robins,
M. J .; Borchardt R. T. Design and Synthesis of S-Adenosylho-
mocysteine Hydrolase, Inhibitors as Broad-Spectrum Antiviral
Agents. In Advances in Antiviral Drug Design: De Clercq, E.,
Ed.; J AI Press: Geenwich CT, 1996; Vol. 2, pp 41-88.
4) Palmer, J . L.; Abeles, R. H. The Mechanism of Action of
S-Adenosylhomocysteinase. J . Biol. Chem. 1979, 254, 1217-
1226.
(
(5) Yuan, C.-S.; Liu, S.; Wnuk, S. F.; Robins, M. J .; Borchardt, R.
T. Rational Approaches to the Design of Mechanism-based
Inhibitors of S-Adenosylhomocysteine Hydrolase. Nucleosides
Nucleotides 1995, 14, 439-447.
(Angiotensin, P-Substance, Bombesin and ACTH).
(
6) (a) Yuan, C.-S.; Yeh, J .; Liu, S.; Borchardt, R. T. Mechanism of
Inactivation of S-Adenosyl-L-homocysteine Hydrolase by (Z)-4′,5′-
Didehydro-5′-Deoxy-5′-Fluoro- adenosine. J . Biol. Chem. 1993,
A saturated solution of R-cyano-4-hydroxycinnamic acid in
acetone was used as a matrix. Samples were prepared as
follows: first a layer of fine crystals of matrix was obtained
by spreading and fast evaporating 0.5 µL of matrix solution.
A 0.5 µL droplet of aqueous HCOOH (5% in water) was then
deposited on this first matrix crystals layer. A droplet (0.5 µL)
of the cleavage products obtained by proteolysis of AdoHcy
hydrolase was subsequently deposited. Finally, a 0.2 µL droplet
of matrix saturated in a 1:1 (v/v) H
added. The preparation was dried under vacuum and washed
three times with 1 µL of aqueous HCOOH.
Liqu id Ch r om atogr aph y/Electr ospr ay Ion ization -Mass
Sp ectr om etr y (LC/ESI-MS). The chromatography was car-
ried out on an Alliance HPLC system (Waters 2690) equipped
with a diode array detector (Waters 996). In this study,
absorbances were monitored at 214 nm with a chromatography
manager Millennium 2.15.
After enzymatic proteolytic cleavage of AdoHcy hydrolase,
an aliquot of 35 µL of the peptides mixture was loaded on a
Nucleosil 300-5C18 column (Macherey Nagel, 2.1 mm × 125
mm). The solvent system used for elution of the peptides
2
68, 17030-17037. (b) Yuan, C.-S.; Wnuk, S. F.; Liu, S.; Robins,
M. J .; Borchardt, R. T. (E)-5′,6′-Didehydro-6′-Deoxy-6′-Fluoro-
homoadenosine: A Substrate that Measures the Hydrolytic
Activity of S-Adenosylhomocysteine Hydrolase. Biochemistry
1994, 33, 12305-12311. (c) Wnuk, S. F.; Yuan, C.-S.; Borchardt,
R. T.; Balzarini, J .; De Clercq, E.; Robins, M. J . Anticancer and
Antiviral Effects and Inactivation of S-Adenosyl-L-homocysteine
Hydrolase with 5′-Carboxaldehydes and Oximes Synthesized
from Adenosine and Sugar-Modified Analogues. J . Med. Chem.
2
3
O-CH CN mixture was
1
997, 40, 1608-1618. (d) Huang, H.; Yuan, C.-S.; Wnuk, S. F.;
Robins, M. J .; Borchardt, R. T. The Mechanism of Inactivation
of Human Placental S-Adenosylhomocysteine Hydrolase by (E)-
4
′,5′-Didehydro-5′-methoxyadenosine (DMAO) and Adenosine 5′-
carboxaldehyde Oxime (ACAO). Arch. Biochem. Biophys. 1997,
3
43, 109-117.
(
7) (a) Parry, R. J .; Muscate. A.; Askonas, L. J . 9-(5′,6′-Dedeoxy-â-
D-ribo-hex-5′-ynofuranosyl)adenine, a Novel Irreversible Inhibi-
tor of S-Adenosylhomocysteine Hydrolase. Biochemistry 1991,
3
0, 9988-9997. (b) Wnuk, S. F.; Mao, Y.; Yuan, C.-S.; Borchardt,
R. T.; Andrei, G.; Balzarini, J .; De Clercq, E.; Robins, M. J .
Discovery of Type II (Covalent) Inactivation of S-Adenosyl-L-
Homocysteine Hydrolase Involving its “Hydrolytic Activity”:
Synthesis and Evaluation of Dihalohomovinyl Nucleoside Ana-

2
752 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 17
logues Derived from Adenosine. J . Med. Chem. 1998, 41, 3078-
Guillerm et al.
(17) Turner, M. A.; Yuan, C.-S.; Borchardt, R. T.; Hersfield, M. S.;
Smith, G. D.; Howell, P. L. Structure Determination of Sele-
nomethionyl-S-Adenosylhomocysteine Hydrolase using Data at
a Single Wavelength. Nat. Struct. Biol. 1988, 5, 369-375.
(18) (a) De Clercq, E.; Descamps, J .; Verhelst; G.; Walker, R. T.;
J ones, A. S.; Torrence, P. F.; Shugar, D. Comparative Efficacity
of Different Antiherpes Drugs Against Different Strains of
Herpes Simplex Virus. J . Infect. Dis. 1980, 141, 563-574. (b)
De Clercq, E.; Hol y´ , A.; Rosenberg, I.; Sakuma, T.; Balzarini,
J .; Maudgal, P. C. A Novel Selective Broad-Spectrum Anti-DNA
Virus Agent. Nature 1986, 323, 464-467.
3
083. (c) Yuan, C.-S.; Wnuk, S. F.; Robins, M. J .; Borchardt, R.
T. A Novel Mechanism-based Inhibitor (6′-Bromo-5′,6′-didehy-
dro-6′-deoxy-6′-Fluoro-adenosine) That Covalently Modifies Hu-
man Placental S-Adenosylhomocysteine Hydrolase. J . Biol.
Chem. 1998, 273, 18191-18197. (d) Yang, X.; Wnuk, S. F.;
Robins, M. J .; Borchardt, R. T. Mechanism of Human S-
Adenosylhomocysteine Hydrolase by 5′,5′,6′,6′-Tetrahydro-6′-
deoxy-6′-halohomoadenosines Biochemistry 2000, 39, 15234-
1
5241.
8) Muzard, M.; Guillerm, D.; Vandenplas, C.; Guillerm, G. 5′-Deoxy-
′-Difluoromethy-thioadenosine, Potent Enzyme-Activated
(
5
a
(
19) Ault-Riche, D. B.; Yuan, C.-S.; Borchardt, R. T. A Single
Mutation at Lysine 426 of Human Placental S-Adenosylho-
mocysteine Hydrolase Inactivates the Enzyme. J . Biol. Chem.
Acylating Agent of S-Adenosyl-L-Homocysteine Hydrolase. Bioorg.
Med. Chem. Lett. 1997, 7, 1037-1040.
(
9) Muzard, M.; Vandenplas, C.; Guillerm, D.; Guillerm, G. The
Mechanism of Inactivation of S-Adenosyl-L-Homocysteine Hy-
drolase by Fluorinated Analogues of 5′-Methylthioadenosine. J .
Enzyme Inhib. 1998, 13, 443-456.
1
994, 269, 31472-31478.
(
20) (a) Della Ragione, F.; Pegg, A. E. Effect of Analogues of
5
1
′-Methylthioadenosine on Cellular Metabolism. Biochem. J .
983, 210, 429-435. (b) Gurannowski, A.; Montgomery, J . A.;
Cantoni, G. L.; Chiang, P. K. Adenosine Analogues as Substrates
and Inhibitors of S-Adenosylhomocysteine Hydrolase. Biochem-
istry 1981, 20, 110-115.
(
10) Vandenplas, C.; Guillerm, D.; Guillerm, G. A New Series of
Mechanism-based Inhibitors of S-Adenosyl-L-Homocysteine Hy-
drolase from Beef Liver. Nucleosides Nucleotides 1999, 18, 569-
5
70.
(
21) Hersfield, M. S.; Aiyar, V. N.; Premakumar, R.; Small, W. C.
S-Adenosylhomocysteine Hydrolase from Human Placenta. Af-
finity Purification and Characterisation. Biochem. J . 1985, 230,
(
11) Leopold, W. R.; Miller, J . A.; Miller, E. C. Comparison of Some
Carcinogenic, Mutagenic, and Biochemical Properties of S-Vinyl-
Homocysteine and Ethionine. Cancer Res. 1982, 42, 4364-4374.
12) (a) Baddiley, J .; J amieson, G. A. Synthesis of S-(5′-Deoxyad-
enosine-5′)-homocysteine, a Product from Enzymatic Methyla-
tions Involving “Active Methionine”. J . Chem. Soc. 1955, 1085-
4
3-52.
(
(
22) Hohman, R. J .; Guitton, M. C.; Veron, M. Inactivation of
S-Adenosyl-L-homocysteine Hydrolase by CAMP Results from
Dissociation of Enzyme-bound NAD . Proc. Natl. Acad. Sci.
+
1
089. (b) J ahn, W. Synthesis of 5′-Substituted Adenosine
Derivatives. Chem. Ber. 1965, 98, 1705-1708.
U.S.A. 1985, 82, 4578-4581.
(
(
(
(
13) Pourcelot, G.; Cadiot, P. Preparation of Propargyl, Allenyl and
(23) Quanta; Molecular Simulations, Incorporated, 9685 Scranton Rd,
San Diego, CA 92121-3752.
(24) Brooks, B. R.; Bruccoleri, R. E.; Olafson, B. D.; States, D. J .;
Swaminathan, S.; Karplus, M. CHARMM: A Program for
Macromolecular Energy, Minimisation, and Dynamics Calcula-
tions. J . Comput. Chem. 1983, 4, 187-217.
(25) Momany, F. A.; Rone, R. Validation of the General Purpose
QUANTA3.2/CHARMm Force Field. J . Comput. Chem. 1992, 13,
888-900.
Alkynyl Ethers and Thioethers. Bull. Soc. Chim. Fr. 1966, 9,
3
016-3033.
14) (a) Coulter-Karis, D. E.; Hershfield, M. S. Sequence of Full
Length CDNA for Human S-Adenosylhomocysteine Hydrolase.
Ann. Hum. Genet. 1989, 53, 169-175.
15) Kitz, K.; Wilson, I. B. Esters of Methanesulfonic Acid as
Irreversible Inhibitors of Acetylcholinesterase. J . Biol. Chem.
1
962, 237, 3245-3249.
16) Silverman, R. B. In Mechanism-based Enzyme Inactivators:
Chemistry and Enzymology: CRC Press: Boca Raton, FL, 1988;
p 11.
J M0108350