Inactivation of human S-adenosylhomocysteine hydrolase by covalent labeling of cysteine 195 with thionucleoside derivatives

Article abstract of DOI:10.1016/j.bmcl.2004.09.051

Thionucleosides 1-4 were synthesized for selectively targeting ¹⁹⁵Cys of human placental AdoHcy hydrolase. A new series of 5′-thioadenosine derivatives 1-4 were synthesized for selectively targeting ¹⁹⁵Cys of human AdoHcy hydrolase. Their incubation with the enzyme resulted in time- and concentration-dependent inactivation, without major modifications of the NAD⁺/NADH ratio. The electrospray mass analysis of the inactivated enzyme with 1, 2, 3, and 4b showed that inhibition was accompanied by the formation of a specific and covalent labeling of each AdoHcy hydrolase subunit. Proteolytic cleavage (endo-Lys-C) and subsequent peptide characterization of the labeled enzyme revealed that ¹⁹⁵Cys was the residue modified during the inactivation process.

Full text of DOI:10.1016/j.bmcl.2004.09.051

Bioorganic & Medicinal Chemistry Letters 14 (2004) 5803–5807
Inactivation of human S-adenosylhomocysteine hydrolase by
covalent labeling of cysteine 195 with thionucleoside derivatives
a
b
Georges Guillerm,a Murielle Muzard,a, Cedric Glapski and Serge Pilard
*
´
aLaboratoire de Chimie bioorganique, UMR 6519, UFR Sciences, B.P. 1039, 51687 Reims Cedex 2, France
b
´
Plate-Forme Analytique, Universite de Picardie Jules Verne, 33 rue Saint-Leu, 80039 Amiens, France
Received 1 July 2004; revised 6 September 2004; accepted 17 September 2004
Available online 6 October 2004
Abstract—A new series of 50-thioadenosine derivatives 1–4 were synthesized for selectively targeting 195Cys of human AdoHcy
hydrolase. Their incubation with the enzyme resulted in time- and concentration-dependent inactivation, without major modifica-
tions of the NAD+/NADH ratio. The electrospray mass analysis of the inactivated enzyme with 1, 2, 3, and 4b showed that inhi-
bition was accompanied by the formation of a specific and covalent labeling of each AdoHcy hydrolase subunit. Proteolytic cleavage
(endo-Lys-C) and subsequent peptide characterization of the labeled enzyme revealed that 195Cys was the residue modified during
the inactivation process.
ꢀ 2004 Elsevier Ltd. All rights reserved.
1. Introduction
AdoHcy hydrolase.6 These two details led us to consider
that specific covalent modification of 195Cys should lead
to the inactivation of human AdoHcy hydrolase. This
possibility was examined with a series of thionucleosides
1–4.
S-Adenosylhomocysteine (AdoHcy) hydrolase catalyses
the interconversion of AdoHcy into adenosine (Ado)
and L-homocysteine (Hcy).1 Inhibition of this enzyme
results in intracellular accumulation of AdoHcy which
in turn provokes feed back inhibition of S-adenosyl-
methionine-dependent methylation reactions (i.e., viral
mRNA methylation) which are essential for viral repli-
cation.2 AdoHcy hydrolase also controls levels of Hcy,
which appear to be a risk factor for cardiovascular dis-
eases.3 Therefore, AdoHcy hydrolase has emerged as a
target for molecular design of antiviral agents as well
as therapeutic inhibitors that may restore normal plas-
ma levels of Hcy.
NH2
N
NH2
N
N
N
N
N
OH OH
O
N
N
HS
S S
O
O
N
N
N
OH OH
OH OH
NH2
N
N
1
2
NH2
N
N
NH2
N
N
N
S
5'
N
O
N
O
The recent success in the determination of the X-ray
structures of rat liver4 and human5 AdoHcy hydrolase
has led to the identification of the essential amino acids
involved in the different steps of the catalytic cycle.
Among them, the 194Cys residue in rat liver enzyme
MeS S
OH OH
4a (5'R)
OH OH
3
4b (5'S)
(
195Cys in human enzyme) was proposed to modulate
the oxidation state of the bound cofactor NAD+.4a This
proposal is consistent with the results obtained by Yuan
et al. in their pioneer study on chemical modification
and site-directed mutagenesis of cysteine residues in
These sulfur-containing nucleosides fulfill the basic
requirements of being active-site-directed inhibitors for
selectively targeting 195Cys. Disufide bond formation
resulting from their interaction with 195Cys at the active
site might provoke covalent inactivation of AdoHcy
hydrolase (Fig. 1).
*
Corresponding author. Tel.: +33 326 913 238; fax: +33 326 913
166; e-mail: murielle.muzard@univ-reims.fr
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.09.051

5804
G. Guillerm et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5803–5807
NH2
N
NH2
N
N
N
N
N
N
N
195Cys
S
S
HS
O
O
NH2
N
NH2
∆M = 282 Da
OH OH
OH OH
N
N
N
N
N
1
N
N
RS S
195Cys
S
S
O
or
195Cys
S
SMe
O
OH OH
∆M = 282 Da
∆M =46 Da
OH OH
2 or 3
NH2
NH2
N
N
N
195Cys
SH
N
N
195Cys
S
N
S
N
O
N
5'
O
HS
OH OH
∆M = 295 Da OH OH
4
Figure 1. Proposed mechanisms for inactivation of AdoHcy hydrolase by 1–4.
We describe here the synthesis of thionucleosides 1–4
and their interactions with human placental AdoHcy
hydrolase.
was fully acetylated and condensed with N6,N9-bis(tri-
methylsilyl)adenine.10 Next, compound 7 was deacetyl-
ated and the selective protection of the 20- and 30-
positions was achieved in two steps to generate the de-
sired allofuranosyladenine derivative 8 in a 22% overall
1
2. Chemistry
yield. The b-D configuration of 8 was confirmed by H
0
0
NMR (J1 ,2 = 3Hz and Dd = 0.2ppm between the chem-
ical shifts of the two CH3ꢀs of the 20,30-isopropylide-
nated nucleoside).11 The epoxynucleoside 9 prepared in
two steps from 8 was treated with thiourea leading to
the corresponding epithionucleoside with inversion of
50-Deoxy-50-thioadenosine 1 was obtained from Ado by
a described procedure.7 The corresponding disulfide 2
was prepared by iodine oxidation of 1.8 The disulfide
3, which was synthesized according to the general proce-
9
dure described by Decout and co-workers, was a gift
configuration at C5 (Scheme 2).12
0
´
from these authors. The epithionucleosides 4a and 4b
were obtained from an unique precursor: the N6-benzoyl-
20,30-O-isopropylidene-9-(b-D-allofuranosyl)adenine 8,
which was prepared from diacetone-D-glucose 5 by the
procedure outlined in Scheme 1.
Removal of the isopropylidene protecting group led to
thiirane 4a13 in 15% yield from 8. Its 50 epimer 4b14
was obtained by the same methodology via the silylated
allofuranosyladenine derivative 10 in 19% overall yield
from 8.
After standard inversion of stereochemistry at C3 and
acidic cleavage of the two 1,2- and 5,6-isopropylidene
groups, the D-allofuranoside derivative thus obtained
NHBz
NH2
N
N
N
N
N
N
N
O
S
O
N
AcO
AcO
O
O
O
OH
O
O
a, b
e
c, d
OAc
O
O
OH OH
a, b
NHBz
N
O
8
c
(5'R) 4a
OAc OAc
O
9
NHBz
NH2
N
NHBz
NHBz
5
6
N
N
N
N
N
N
N
N
N
N
N
TBDMSO
N
HO
HO
AcO
AcO
N
S
O
N
N
N
N
MsO
N
O
O
O
O
O
f
g,h
d,e
O
O
OH OH
O
O
O
O
OAc OAc
(5'S) 4b
7
8
10
Scheme 1. Reagents and conditions: (a) PDC, Ac2O, CH2Cl2, then
NaBH4, EtOH, 83%; (b) CH3CO2H, Ac2O, APTS, 79%; (c) N6,N9-
bis(trimethylsilyl)adenine, SnCl4, dichloroethane, 70%; (d) NaOH 2M,
EtOH, 70%; (e) dimethoxypropane, acetone, APTS then CH3CO2H/
H2O 7/3, 70%.
Scheme 2. Reagents and conditions: (a) TsCl, pyridine, 60%; (b) NaH,
THF, 89%; (c) thiourea, MeOH, 55%; (d) HCO2H/H2O 4/1 then
MeONa, MeOH, 50%; (e) TBDMSCl, imidazole, pyridine then MsCl,
pyridine, 50%; (f) TBAF, THF; (g) thiourea, MeOH, 77%; (h)
HCO2H/H2O 4/1 then MeONa, MeOH, 50%.

G. Guillerm et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5803–5807
5805
3. Results and discussion
which compound 4b proceeds to inactivate AdoHcy
hydrolase.
Recombinant human placental AdoHcy hydrolase puri-
fied to homogeneity was used in this study.15 AdoHcy
hydrolase (10nM) was assayed in the synthetic direction
in the presence of [8-14C]-Ado or [2,8-3H]-Ado (15lM,
300Bq) and Hcy (5mM) in 20mM potassium phosphate
buffer pH7.5, 1mM EDTA. Incubation of the enzyme
with 1, 2, 3, 4a, and 4b resulted in time- and concentra-
tion-dependent inactivation. Compounds 1–4 are stable
under assay conditions whereas thiol 1 became contam-
inated with 15% of disulfide 2 after 6h at 37ꢁC. In each
case, the inactivation was irreversible since enzyme
activity could not be restored after dialysis (24h) against
assay buffer. We confirmed that the action of each of the
thionucleosides 1–4 was active site directed by means of
protection experiments with Ado. Using the Kitz and
Wilson method16 a double reciprocal plot of the initial
pseudo-first-order inactivation rate constant 1/kapp ver-
sus 1/[I] gave the Ki and kinact values listed in Table 1.
The mechanism of inactivation was further investigated
using electrospray ionization mass spectrometry (ESI-
MS) analysis of the protein. An accurate molecular
weight of AdoHcy hydrolase subunit for the native or
inactivated protein with 1–4 was determined from the
wide distribution of highly charge states observed in
positive ion ESI-MS analysis under acidic conditions.18
An algorithm (MaxEnt) based on the maximum entropy
method was used to produce true molecular mass spec-
tra from multiply charged electrospray spectra. The na-
tive enzyme had a subunit molecular weight of
47,597 6Da (Fig. 3).
The mass modifications carried by each AdoHcy hydro-
lase subunit upon interaction of the enzyme with inhib-
itors 1–4 are listed in Table 2 and illustrated in Figure 4
for compound 3.
The effects of 1–4 on the NAD+/NADH content of Ado-
Hcy hydrolase were also determined after complete inacti-
vation of the enzyme. Compounds 1, 2, 3, and 4a led to
minor changes in the initial NAD+/NADH ratio
whereas 30% depletion of the enzymeꢀs NAD+ was ob-
served with 4b (Fig. 2).
Inhibition of the enzyme with thionucleosides 1, 2, 3,
and 4b was accompanied by a specific covalent linkage
and no further native enzyme was detected after total
inactivation.
These results support the mechanistic proposal de-
scribed in Figure 1 involving the formation of a covalent
disulfide bound between a cysteine of the active site and
each of the sulfur-containing inhibitors 1, 2, 3, and 4b.
To confirm the implication of the 195Cys in the inactiva-
tion process, AdoHcy hydrolase inhibited by the
This is indicative that the redox activity of the enzyme is
not involved in the inactivation process with 1, 2, 3, and
4a but it might partially participate in the pathway by
Table 1. Ki and kinact values for the inhibitory effect of 1–4 on AdoHcy
hydrolase
47593
kinact, minÀ1
100
Compd
Ki, lM
1
4
96
0.04
0.05
0.05
0.10
0.03
2
3
19
4a
4b
105
50
%
AdoHcy hydrolase (2lM) was incubated with inhibitor at various
concentrations and various times in 20mM potassium phosphate
buffer pH7.5, 1mM EDTA at 37ꢁC. Residual activity was determined
as described and a double reciprocal plot of the initial pseudo-first-
order inactivation rate constant versus 1/[I] gave the Ki and kinact
values.16
0
mass
44000 44500 45000 45500 46000 46500 47000 47500 48000 48500 49000 49500 50000 50500 51000 51500
Figure 3. Reconstructed ESI mass spectrum (MaxEnt) obtained for
purified AdoHcy hydrolase.
100
NADH
75
Table 2. Mass excess observed for each subunit of AdoHcy hydrolase
inactivated by 1–4
50
NAD+
25
Inhibitor
DM observed, Da 2
NAD+/NADH
1
283
281
46
0
E
1
2
3
4a 4b
2
3
Figure 2. Variation of NAD+/NADH content upon incubation with
inhibitors: AdoHcy hydrolase (20lM) was incubated with 600lM of
1–4 in 20mM potassium phosphate buffer pH7.5, 1mM EDTA at
37ꢁC until total inactivation. NAD+ and NADH in native AdoHcy
hydrolase (E) and after inactivation (1, 2, 3, 4a, and 4b) were measured
by a fluorescence method.17
4a
4b
None
293
Native AdoHcy hydrolase or enzyme inactivated by 1–4 was diluted to
1lM in a H2O/CH3CN mixture (v/v) acidified with ca. 1% HCOOH
and was analyzed by ESI-MS.19

5806
G. Guillerm et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5803–5807
References and notes
47639
100
1. Palmer, J. L.; Abeles, R. H. J. Biol. Chem. 1979, 254, 1217.
2. Narayan, P.; Rottman, F. M. In Advances in Enzymology;
Meister, A., Ed.; John Wiley & Sons: New York, 1994; pp
255–285.
3. (a) Chambers, J. C.; Seddon, M. D.; Shah, S.; Kooner, J.
S. J. R. Soc. Med. 2001, 94, 10; (b) Refsun, H.; Ueland, P.
M.; Nygard, O.; Vollset, S. E. Annu. Rev. Med. 1998, 49,
31; (c) Schnyder, G.; Roffi, M.; Pin, R.; Flammer, Y.;
Lange, H.; Eberli, F. R.; Meier, B.; Turi, Z. G.; Hess, O.
M. New Engl. J. Med. 2001, 345, 1593.
4. (a) Hu, Y.; Komoto, J.; Huang, Y.; Gomi, T.; Ogawa, H.;
Takata, Y.; Fujioka, M.; Takusagawa, F. Biochemistry
1999, 38, 8283; (b) Komoto, J.; Huang, Y.; Gomi, T.;
Ogawa, H.; Takata, Y.; Fujioka, M.; Takusagawa, F. J.
Biol. Chem. 2000, 275, 32147; (c) Huang, Y.; Komoto, J.;
Takata, Y.; Powell, D. R.; Gomi, T.; Ogawa, H.; Fujioka,
M.; Takusagawa, F. J. Biol. Chem. 2002, 277, 7477.
5. Turner, M. A.; Yuan, C. S.; Borchardt, R. T.; Hershfield,
M. S.; Smith, G. D.; Howell, P. L. Nature Struct. Biol.
1998, 5, 369.
%
0
44000 44500 45000 45500 46000 46500 47000 47500 48000 48500 49000 49500 50000 50500 51000 51500
mass
Figure 4. Reconstructed ESI mass spectrum (MaxEnt) obtained for
AdoHcy hydrolase inhibited by 3.
disulfide 2 was submitted to proteolytic cleavage by
endo-Lys-C.20 The cleavage products (i.e., a mixture of
peptides C-terminally cleaved at Lys residues) were mass
analyzed by LC/ESI-MS21 and compared to those ob-
tained from native AdoHcy hydrolase.
´
6. Yuan, C. S.; Ault-Riche, D. B.; Borchardt, R. T. J. Biol.
Chem. 1996, 271, 28009.
7. Pignot, M.; Pljevaljcic, G.; Weinhold, E. Eur. J. Org.
Chem. 2000, 549.
8. Kuhn, R.; Jahn, W. Chem. Ber. 1965, 98, 1699.
9. (a) Chambert, S.; Gautier-Luneau, I.; Fontecave, M.;
The 195Cys containing peptide was characterized in the
HPLC chromatogram of the digested native AdoHcy
hydrolase
(FDNLYG195CRESLIDGIK,
m/z =
´
Decout, J. L. J. Org. Chem. 2000, 65, 249; (b) Chambert,
S.; Decout, J. L. Org. Prep. Proced. Int. 2002, 34, 27.
922.01Da, doubly charged ion, molecular weight
1841.89Da). This peptide was absent in the peptide
map in the experiments carried out with inactivated en-
zyme, but a new peptide was detected in the correspond-
ing HPLC chromatogram. This peptide was analyzed
using on line ESI-MS detection. The peptidic fragment
(m/z = 1062.55Da, doubly charged ion, molecular
weight 2123.08Da) was attributed to the same peptide
that bears an adduct of 281Da: FDNLYG195CRES-
LIDGIK +281Da. This result strongly suggests that
195Cys was the residue modified in the inactivation proc-
ess as illustrated in Figure 1. In similar experiments car-
ried out with the disulfide 3, we confirmed that an excess
mass of 46Da was bound to the same peptidic fragment
(FDNLYG195CRESLIDGIK +46Da), a result consist-
ent with a chemical modification of 195Cys into a methyl
disulfide adduct (Fig. 1).
´
10. (a) Vorbruggen, H.; Ho¨fle, G. Chem. Ber. 1981, 114, 1256;
¨
(b) Vorbruggen, H.; Krolikiewicz, K.; Bennua, B. Chem.
¨
Ber. 1981, 114, 1279.
11. (a) Rayner, B.; Tapiero, C.; Imbach, J. L. Carbohydr. Res.
1976, 47, 195; (b) Lerner, L. M.; Mennitt, G. Carbohydr.
Res. 1994, 259, 191.
12. Bordwell, F. G.; Andersen, H. M. J. Am. Chem. Soc. 1953,
75, 4959.
13. Compound 4a: 1H NMR (DMSO-d6, 250MHz, d ppm,
0
JHz, 323K) 2.40(d, 1H, J6 a,5 5.0, H6 a); 2.60(d, 1H,
0
0
0
0
0
0
0
0
0
0
J6 b,5 5.0, H6 b); 3.37 (dt, 1H, J5 ,4 8.0, J5 ,6 5.0, H5 ); 3.55
0
0
0
0
0
0
(dd, 1H, J4 ,5 8.0, J4 ,3 5.0, H4 ); 4.35 (dd, 1H, J3 ,2 10.0,
0
0
J3 ,4 5.0, H3 ); 4.75 (dd, 1H, J2 ,3 10.0, J2 ,1 5.0, H2 ); 5.90
0
0
0
0
0
0
0
0
0
0
(d, 1H, J1 ,2 5.0, H1 ); 8.15 and 8.40(s, 2*1H, H 2 and H8).
13
0
C NMR (DMSO-d6, 62.5MHz, d ppm) 21.6 (C6 ); 36.2
0
(C5 ); 73.5 (C4 ); 79.2 (C3 ); 87.5 (C2 ); 88.1 (C1 ); 119.2
0
0
0
0
20
(C5); 139.7 (C6); 149.4 (C8); 152.7 (C2); 156.1 (C4). ½aꢀD
+22.5 (c 6 · 10À3, CHCl3/MeOH 4/1). MS m/z 296 (22,
MH+); 264 (60); 157 (55); 136 (100). Anal. Calcd for
C11H13N5O3SÆ0.5H2O: C, 43.41; H, 4.64; N, 23.01. Found:
C, 43.22; H, 4.27; N, 22.46.
4. Conclusion
1
14. Compound 4b: H NMR (DMSO-d6, 250MHz, d ppm, J
We have identified a new series of inhibitors of AdoHcy
hydrolase, which covalently modify the enzyme by inter-
actions with 195Cys. These results confirm the crucial
role of 195Cys at the catalytic center of AdoHcy hydro-
lase. Docking experiments with the epithionucleosides
4a and 4b into the active site of AdoHcy hydrolase are
under progress to understand their different behaviors.
0
Hz, 323K) 2.35 (d, 1H, J6 a,5 5.0, H6 a); 2.60(d, 1H, J6 b,5
0
0
0
0
0
5.0, H6 b); 3.47 (m, 1H, H5 ); 3.57 (dd, 1H, J4 ,5 9.0, J4 ,3
0
0
0
0
0
0
0
3.0, H4 ); 4.25 (dd, 1H, J3 ,2 5.0, J3 ,4 3.0, H3 ); 5.00 (dd,
0
0
0
0
0
0
0
0
0
0
0
0
1H, J2 ,3 = J2 ,1 = 5.0, H2 ); 5.55 (d, 1H, J1 ,2 5.0, H1 );
8.15 and 8.40(s, 2*1H, H 2 and H8). 13C NMR (DMSO-d6,
0
0
0
62.5MHz, d ppm) 24.1 (C6 ); 34.4 (C5 ); 72.2 (C4 ); 73.3
0
0
0
(C3 ); 87.2 (C2 ); 88.7 (C1 ); 119.2 (C5); 140.0 (C6); 149.5
20
(C8); 152.6 (C2); 156.1 (C4). ½aꢀD À50( c 3 · 10À3, CHCl3/
MeOH 4/1). MS m/z 296 (6, MH+); 264 (23); 157 (31); 136
(100). Anal. Calcd for C11H13N5O3SÆ0.5H2O: C, 43.41; H,
4.64; N, 23.01. Found: C, 43.39; H, 4.36; N, 23.02.
15. Yuan, C. S.; Yeh, J.; Squier, T. C.; Rawitch, A.;
Borchardt, R. T. Biochemistry 1993, 32, 10414.
Acknowledgements
´
´
The authors thank Pr. Jean Luc Decout, Universite Jo-
seph Fourier—Grenoble I, for providing us a sample of
50-deoxyadenosin-50-yl methyldisulfide and le Conseil
16. Kitz, K.; Wilson, I. B. J. Biol. Chem. 1962, 237, 3245.
17. Hohman, R. J.; Guitton, M. C.; Veron, M. Proc. Natl.
Acad. Sci. U.S.A. 1985, 82, 4578.
´ ´
General de la Marne for a doctoral grant for C.G.

G. Guillerm et al. / Bioorg. Med. Chem. Lett. 14 (2004) 5803–5807
5807
18. Guillerm, G.; Guillerm, D.; Vandenplas-Witkowki, C.;
Rogniaux, H.; Carte, N.; Leize, E.; Van Dorsselaer, A.;
De Clercq, E.; Lambert, C. J. Med. Chem. 2001, 44,
2743.
25mM NH4HCO3 buffer pH8.5 at 35ꢁC overnight.
Proteolysis was stopped by freezing the samples at À20ꢁC.
21. The chromatography was carried out on an Alliance
HPLC system (Waters 2695) equipped with a UV detector.
Absorbances were monitored at 210nm. The peptide
mixture was loaded on a Nucleosil 100-5C18 column,
150 · 4.6mm (Macherey Nagel). Elution was performed at
a flow rate of 1mL/min using a biphasic gradient of A
(0.1% TFA in H2O) and B (0.08% TFA in CH3CN): 2% B
(5min), 2–60% B (60min), 60–80% B (5min), and 80% B
(5min). The effluent was flow-split via a peek tee with 1/5
of the flow directed toward the ESI source of the ZQ mass
spectrometer and the residual 4/5 directed toward the UV
detector. LC/ESI-MS data were recorded in the positive
ion mode with a capillary voltage of 3kV and a cone
voltage of 45V. Scanning was performed in the range 200–
2200Da at a scan rate of 2s/scan.
19. ESI-MS data were obtained in the positive ion mode on a
quadrupole instrument (Waters-Micromass ZQ). The
source and desolvation temperatures were kept at 80and
150ꢁC, respectively. Nitrogen was used as a drying and
nebulizing gas at flow rates of 250and 50L/h, respectively.
The capillary voltage was 3kV and the cone voltage 40V.
The mass range was 400–2400Da and the spectra were
recorded at 4s/scan in the profile mode. Calibration of the
instrument was performed using the multiply charged ions
produced by horse heart myoglobin. Data acquisition and
processing were performed with MassLynx V4.0software.
20. 150lg of AdoHcy hydrolase, native or inactivated by 2,
was incubated with 1.2lg of endo-Lys-C (Roche) in