Potential utility of adenosine 5′-ester prodrugs to enhance its plasma half-life: Synthesis and molecular docking studies

Article abstract of DOI:10.1007/s00044-014-1299-z

Adenosine, the adenine nucleoside, has demonstrated various pharmacological properties related to the treatment of relevant clinical diseases. With respect to this, one of the most fascinating biological activities of adenosine is its capacity for reversing hepatic fibrosis, which has been established in several in vitro and in vivo studies. Although adenosine seems to be a privileged compound, it lacks of metabolic stability to be considered as a drug candidate. For this reason, in this preliminary study, six prodrugs were developed in order to enhance the plasma half-life of adenosine, in which the esterification at 5′ position of adenosine was considered. According to previous works, the increase in steric hindrance at this position could develop unfavorable interactions inside adenosine deaminase (ADA) catalytic domain, which is reported as the main enzyme implicated in adenosine metabolism. Besides, molecular docking was employed to verify if the hindrance at carbinol group in the prodrugs is enough to diminish its oxidation and also to predict if compounds would be metabolized by a promiscuous esterase. Finally, the in vitro assays corroborated the theoretical findings and also indicated that the compounds are less metabolized than adenosine by ADA; in the case of compounds containing proline and thioproline progroups (4d and 4e, respectively), the reduction was more than three-fold of decrease.

Full text of DOI:10.1007/s00044-014-1299-z

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-014-1299-z
ORIGINAL RESEARCH
Potential utility of adenosine 50-ester prodrugs to enhance its
plasma half-life: synthesis and molecular docking studies
•
Eduardo Hernandez-Vazquez Victoria Chagoya
´
´
Received: 31 March 2014 / Accepted: 10 November 2014
Ó Springer Science+Business Media New York 2014
Abstract Adenosine, the adenine nucleoside, has dem-
onstrated various pharmacological properties related to the
treatment of relevant clinical diseases. With respect to this,
one of the most fascinating biological activities of adeno-
sine is its capacity for reversing hepatic fibrosis, which has
been established in several in vitro and in vivo studies.
Although adenosine seems to be a privileged compound, it
lacks of metabolic stability to be considered as a drug
candidate. For this reason, in this preliminary study, six
prodrugs were developed in order to enhance the plasma
half-life of adenosine, in which the esterification at 50
position of adenosine was considered. According to pre-
vious works, the increase in steric hindrance at this position
could develop unfavorable interactions inside adenosine
deaminase (ADA) catalytic domain, which is reported as
the main enzyme implicated in adenosine metabolism.
Besides, molecular docking was employed to verify if the
hindrance at carbinol group in the prodrugs is enough to
diminish its oxidation and also to predict if compounds
would be metabolized by a promiscuous esterase. Finally,
the in vitro assays corroborated the theoretical findings and
also indicated that the compounds are less metabolized
than adenosine by ADA; in the case of compounds con-
taining proline and thioproline progroups (4d and 4e,
respectively), the reduction was more than three-fold of
decrease.
Keywords Adenosine ꢀ Prodrug ꢀ Molecular docking ꢀ
ADA
Introduction
Adenosine, an endogenous purine nucleoside, has an important
role in the normal metabolic machinery of the cell. In this
context, the participation of adenosine in the physiology of
central nervous and circulatory systems is well documented
whereas previous works have established that adenosine has
anti-inflammatory, cardioprotective, neuroprotective, and
´
antioxidant properties (Cronstein et al., 1983; Hasko et al.,
1998; Michel et al., 1999; Yoo et al., 2005; Mu¨ller and Jac-
obson, 2011). Moreover, recent studies suggest that adenosine
may be implicated in reversing fibrosis by involving several
mechanisms such as the enhancement of collagenolytic
activity, the decrease of tissue inhibitor of metalloproteinases 1
and 2 (TIMPs), the modulation of cell redox state, the main-
tenance of liver energy availability, and the preservation of an
´
adequate mitochondrial function of hepatic cells (Hernandez-
´
Mun˜oz et al., 1990;. Hernandez-Munoz et al., 1997; Hernan-
dez-Mun˜oz et al., 2001; Mendieta-Condado et al., 2007).
Despite of all the pharmacological activities attributed
to adenosine, it has a serious pharmaceutical problem:
adenosine shows a short plasma half-life (0.6–1.4 s), which
is attributed to an extensive metabolism carried out by the
enzymes adenosine deaminase (ADA) and adenosine
Electronic supplementary material The online version of this
article (doi:10.1007/s00044-014-1299-z) contains supplementary
material, which is available to authorized users.
´
´
E. Hernandez-Vazquez (&)
Departamento de Farmacia, Facultad de Quımica, UNAM,
´
´
04510 Mexico, DF, Mexico
e-mail: eduardo.hervaz@gmail.com
¨
kinase (AK) (Mosher et al., 1989). ADA metabolizes
adenosine to inosine, which has lower affinity to adenosine
receptors than adenosine itself, while AK converts adeno-
sine to AMP (Fig. 1) (Jacobson and Gao, 2006).
V. Chagoya
Departamento de Bioenergetica, Instituto de Fisiologıa Celular,
´
UNAM, 04510 Mexico, DF, Mexico
´
´
123

Med Chem Res
In order to improve the plasma half-life of adenosine,
the synthesis of some adenosine prodrugs (4a–4f) by the
esterification of different acids at 50-hydroxyl group of the
ribose and the performance of molecular docking studies in
ADA and AK are presented (Fig. 2). Molecules were
designed in accordance to previous studies, in which it has
been demonstrated that bulky alkyl groups other than OH
at 50-position of adenosine have a loss in affinity for ADA
(Ciuffreda et al., 2003). In addition to the theoretical
reduction of metabolism by steric hindrance, the use of
progroups with antioxidant or hepatoprotective activities,
such as glycine (Senthilkumar et al., 2003; Ligumsky et al.,
1995), alanine (Ishizaki et al., Ishizaki-Koizumi et al.
2002), thioproline (Gulizar, 2004; Suo et al., 2006), and its
non-sulfured analog proline, could offer further benefits in
the protection of hepatic injury during fibrosis and cirrho-
sis, without significant loss in aqueous solubility.
(120 mL), 2,2-dimethoxypropane (5.6 mL, 45.6 mmol) was
added. After 3 days of vigorous stirring, the solution was
neutralized with a saturated solution of Na2CO3 and then
extracted with chloroform (3 9 50 mL). The organic layers
were mixed and dried using Na2SO4. The solvent was
removed under reduced pressure and the remainder solid was
suspended with ether and washed with cold water obtaining a
white solid (3.11 g, 10.13 mmol, 90 %); mp 202–203 °C; IR
(KBr)mmax: 3,271, 3,109, 2,987, 2,938, 1,686, 1,605, 1,476,
1
1,378, 1,300, 1,208, 1,116; H NMR (400 MHz, CDCl3)
d = 1.38 (s, 3H, CH3), 1.65 (s, 3H, CH3), 3.81 (dd, 1H,
J = 12.6, 2 Hz, H-50), 3.98 (dd, 1H, J = 12.6, 1.6 Hz, H-50),
4.55 (br, 1H, H-40), 5,11 (t, 1H, J = 6.4 Hz, H-30), 5.21 (t,
1H, J = 5.6 Hz, H-20), 5.90 (d, 1H, J = 4.4 Hz, H-10), 6.23
(s, 2H, NH2), 7.91 (s, 1H, H-8), 8.31 (s, 1H, H-2); FABMS m/
z 308 [M]? (88).
General procedure 2: preparation of 50 20,30-O-
isopropylidenadenosine esters (3a–3e)
Experimental
20,30-O-isopropylidenadenosine 2 (3.9 mmol) and the
appropriate N-Boc-protected amino acid (3.9 mmol) were
dissolved in 40 mL of anhydrous chloroform. Then, dicy-
clohexylcarbodiimide (4.6 mmol) and 4-dimethylamino-
pyridine (1.15 mmol) were added and stirred under
anhydrous atmosphere for 3 h. The suspension was filtered
and the solution was washed with 20 mL of saturated
solution of Na2CO3 and then with 20 mL of an aqueous
solution of citric acid at 20 % w/v. The organic layer was
dried with Na2SO4 and the solvent was removed under
reduced pressure. The products were purified by column
chromatography (CHCl3–MeOH, 95:5).
Solvents and reagents were purchased from Sigma-Aldrich
and were used without purification. Melting points were
determined on a Bu¨chi melting point apparatus, model
B-540 and are uncorrected. The progress of reactions and
the purity of final products were monitored by thin layer
chromatography (TLC), whereas UV and iodine oxidation
were used as revealing agents. H and 13C NMR spectra
1
were obtained in DMSO-d6 and CDCl3 solutions at
1
400 MHz (for H) and 100 MHz (for 13C) on Varian MR-
400 (9.2 T) instrument and on a Varian VNMRS-400
(9.2 T) instrument. Chemical shifts (d) are reported in ppm
relative to the solvent signal. Signal splitting patterns are
described as singlet (s), doublet (d), triplet (t), quartet (q),
broad signal (bs), doublet of doublet (dd), or multiplet (m).
Mass spectrometry spectra were obtained using a Thermo-
electron model DFS instrument.
[(3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methyl
aminoacetate (3a)
White solid in 87 % yield; mp 110–112 °C; 1H NMR
(400 MHz, CDCl3) d = 1.41 (s, 3H, CH3C), 1.43 (s, 9H,
(CH3)3), 1.62 (s, 3H, CH3C), 3.36 (d, 2H, J = 6 Hz, Bo-
cNH–CH2), 4.28 (dd, 1H, J1 = 11.8 Hz, J2 = 6.4 Hz,
H-50a), 4.37 (dd, 1H, J = 11.6 Hz, J2 = 4.8 Hz, H-50b),
4.46 (m, 1H, H-40), 5.10 (dd, 1H, J = 6 Hz, J2 = 3.6 Hz,
H-30), 5.16 (s, 1H, BocNH), 5.50 (dd, 1H, J = 6.4, 1.6 Hz,
Synthesis of [(3aR,4R,6R,6aR)-6-(6-amino-9H-purin-
9-yl)-2,2-dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-
yl]methanol (2)
To a suspension of adenosine (3 g, 11.23 mmol) and p-tol-
uensulfonic acid (2.33 g, 13.55 mmol) in dry acetone
Fig. 1 Major metabolic
pathways of adenosine
catabolism involving adenosine
deaminase (ADA) and
adenosine kinase (AK)
123

Med Chem Res
Fig. 2 A bulky group at 50-hydroxyl of adenosine could minimize ADA recognition and metabolism. Esterification with amino acids could
achieve this and are metabolically removable
H-20), 6.02 (s, 2H, NH2), 6.10 (d, 1H, J = 1.2 Hz, H-10),
7.89 (s, 1H, H-8), 8.34 (s, 1H, H-2); FABMS m/z 465 [M]?
(44).
5.71 (s, 2H, NH2), 6.10 (s, 1H, H-10), 7.87 (s, 1H, H-8),
8.35 (s, 1H, H-2); FAB-MS m/z 505 [M]? (100).
[(3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methyl
(4R)-1,3-thiazolidine-4-carboxylate (3e)
[(3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methyl
3-aminopropanoate (3b)
White solid in 77 % yield; mp 108–109 °C; 1H NMR
(400 MHz, CDCl3) d: 1.33–1.42 (m, 9H, (CH3)3), 1.36 (s,
3H, CH3C), 1.57 (s, 3H, CH3C), 3.18 (br, 2H, S–CH2CH),
4.36 (m, 3H, Boc–CHa and H-50), 4.42 (m, 1H, H-40), 4.55
(d, 1H, J = 8.8 Hz, BocN–CHb), 4.73 (br, 1H, S–CH2CH),
5.17 (br, 1H, H-30), 5.55 (br, 1H, H-20), 6.23 (d, 1H,
J = 2 Hz, H-10), 6.71 (br, 2H, NH2), 8.22 (s, 1H, H-8),
8.19 (s, 1H, H-2). FABMS m/z 523 [M]? (32).
White oil in 87 % yield; 1H NMR (400 MHz, CDCl3)
d = 1.41 (s, 3H, CH3C), 1.43 (s, 9 H, (CH3)3), 1.62 (s, 3H,
CH3C), 2.48 (m, 2H, BocNHCH2CH2), 3.36 (m, 2H, Bo-
cNHCH2CH2), 4.28 (dd, 1H, J = 11.8, 6.4 Hz, H-50a),
4.37 (dd, 1H, J = 11.6, 4.8 Hz, H-50b), 4.47 (m, 1H, H-40),
5.10 (dd, 1H, J = 6, 3.6 Hz, H-30), 5.16 (s, 1H, BocNH),
5.50 (dd, 1H, J = 6.4, 1.6 Hz, H-30), 6.02 (s, 2H, NH2), 6.1
(d, 1H, J = 2 Hz, H-10), 7.89 (s, 1H, H-8), 8.34 (s, 1H,
H-2); FABMS m/z 479 [M]? (16).
Standard procedure 3: preparation of 50adenosine esters
(4a–4f)
[(3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methyl
(2S)-2-aminopropanoate (3c)
The corresponding N-Boc-protected ester (1.91 mmol) and
trifluoroacetic acid (2 mL) were stirred vigorously at room
temperature for 2 days. The excess of acid was removed
under reduced pressure and the residue was dissolved in
20 mL of chloroform and washed with 2 9 15 mL of
NaHCO3 at 10 % (w/v). The organic layer was dried over
Na2SO4 and the solvent was removed under reduced
pressure. The purification was performed by column
chromatography (CHCl3–MeOH, 90:10).
White solid in 75 % yield; mp 87–89 °C; 1H NMR
(400 MHz, CDCl3) d = 1.33 (d, 3H, J = 7.2 Hz,
CH3CHNH), 1.40 (s, 3H, CH3C), 1.42 (s, 9H, (CH3)3), 1.62
(s, 3H, CH3C), 3.48 (m, 1H, CH3CHNH), 4.36 (dd, 1H,
J = 11.6, 6 Hz, H-50a), 4.4 (dd, 1H, J = 11.6, 4.4 Hz,
H-50b), 4.47 (m, 1H, H-40), 5.07 (dd, 1H, J = 6.4, 3.2 Hz,
H-30), 5.22 (br, 1H, BocNH), 5.52 (dd, 1H, J = 6.4,
4.0 Hz, H-20), 5.81 (s, 2H, NH2), 6.07 (d, 1H, J = 2 Hz,
H-10), 7.89 (s, 1H, H-8), 8.36 (s, 1H, H-2); FABMS m/
z 479 [M]? (62).
[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methyl aminoacetate (4a)
Yellow oil in 50 % yield; IR (neat)m:max3,324, 3,155, 3,120,
2,932, 1,679, 1,668, 1,604, 1,206, 1,134; 1H NMR
(400 MHz, DMSO-d6) d: 3.63 (s, 2H, NH2–CH2), 4.16 (dd,
1H, J = 11.6, 6.4 Hz, H-50a), 4.25 (dd, 1H, J = 11.8,
4.8 Hz, H-50b), 4.32 (m, 1H, H-40), 5.02 (dd, J = 6.0,
3.2 Hz, H-30), 5.42 (dd, J = 6.2, 2.8 Hz, H-20), 6.16 (d,
J = 2.4 Hz, H-10), 7.16 (br, 2H, NH2–CH2), 7.31 (br, 2H,
Adenosine-NH2), 8.14 (s, 1H, H-8), 8.28 (s, 1H, H-2); 13C
NMR (100 MHZ, DMSO-d6) d = 42.2 (NH2CH2), 64.5
(C-50), 81.1 (C-30), 83.5 (C-20), 83.9 (C-40), 89.4 (C-10),
119.5 (C-5), 140.2 (C-8), 149.2 (C-4), 153.2 (C-2), 156.6
[(3aR,4R,6R,6aR)-6-(6-Amino-9H-purin-9-yl)-2,2-
dimethyltetrahydrofuro[3,4-d][1,3]dioxol-4-yl]methyl
(2S)-pyrrolidine-2-carboxylate (3d)
White solid in 92 % yield; mp 113–115 °C; 1H NMR
(400 MHz, CDCl3) d = 1.35, 1.39 (s, s, 9H, (CH3)3), 1.44
(s, 3H, CH3C), 1.63 (s, 3H, CH3C), 1.86 (m, 4H, Bo-
cNCH2CH2CH2 and BocNCH2CH2CH2), 3.42 (m, 2H,
BocN–CH2), 4.29 (m, 1H, BocN–CH), 4.45 (m, 2H, H-50),
4.50 (m, 1H, H-40), 5.06 (m, 1H, H-30), 5.47 (m, 1H, H-20),
123

Med Chem Res
(C-6), 170.6 (C = O); FABMS m/z 325 [M]? (64);
HRFABMS m/z (pos): 325.1254 C12H17O5N6 (calcd
325.1255).
13C NMR (100 MHZ, DMSO-d6) d = 23.4 (NH–
CH2CH2CH2), 28.1 (NH–CH2CH2CH2), 46.0 (NH–
CH2CH2CH2), 59.1 (NH–CH), 66.2 (C-50), 81.6 (C-30),
86.3 (C-20), 88.3 (C-40), 91.2 (C-10), 119.6 (C-5), 140.5 (C-
8), 149.7 (C-4), 152.7 (C-2), 156.2 (C-6), 169.2 (C=O).
FABMS m/z 365 [M]? (26); HRFABMS m/z (pos):
365.1551 C15H21O5N6 (calcd 365.1568).
[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methyl 3-aminopropanoate
(4b)
Yellow oil in 51 % yield; IR (neat)mmax: 3,339, 3,199,
1,677, 1,422, 1,203, 1,132; H NMR (400 MHz, DMSO-
[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methyl (4R)-1,3-
thiazolidine-4-carboxylate (4e)
1
d6) d: 2.48 (t, 2H, J = 2.8 Hz, NH2CH2CH2), 3.30 (m, 2H,
NH2CH2CH2), 4.24 (dd, 1H, J = 11.8, 6.4 Hz, H-50a), 4.32
(dd, 1H, J = 11.6, 4.8 Hz, H-50b), 4.42 (m, 1H, H-40), 5.17
(dd, J = 6.2, 3.6 Hz, H-30), 5.56 (dd, 1H, J = 6.4, 2.4 Hz,
H-20), 6.22 (d, 1H, J = 2 Hz, H-10), 6.71 (br, 2H, Adeno-
sine-NH2), 8.19 (s, 1H, H-8), 8.22 (s, 1H, H-2); 13C NMR
(100 MHZ, DMSO-d6) d = 34.3 (NH2CH2CH2), 36.2
(NH2CH2CH2), 63.8 (C-50), 81.7 (C-30), 83.9 (C-20), 84.4
(C-40), 90.2 (C-10), 113.8 (C-5), 140.0 (C-8), 149.3 (C-4),
152.9 (C-2), 156.4 (C-6), 171.0 (C=O); FABMS m/z 339
[M]?(40); HRFABMS m/z (pos): 339.1439 C13H19O5N6
(calcd 339.1411).
Yellow oil in 47 % yield; IR (neat)mmax: 3,199, 3,071, 2,897,
1,678, 1,672, 1,473, 1,441, 1,194, 1,129; 1H NMR
(400 MHz, DMSO-d6) d = 2.89 (dd, 1H, J = 10.4, 6.8 Hz,
S–CH2aCH), 3.16 (dd, 1H, J = 10.4, 7.2 Hz, S–CH2bCH),
4.01 (t, 1H, J = 7.2 Hz, S–CH2CH), 4.10 (d, 1H,
J = 9.2 Hz, NH–CHa–S), 4.30 (d, 1H, J = 9.2 Hz, NH–
CHb–S), 4.31 (m, 1H, H-40), 4.42 (dd, 1H, J = 12, 4.8 Hz,
H-50a), 4.49 (dd, 1H, J = 12, 4 Hz, H-50b), 4.57 (t, 1H,
J = 5.2 Hz, H-30), 4.93 (t, 1H, J = 5.2 Hz, H-20), 5.62 (s,
1H, NH-thiazolidine), 5.86 (s, 2H, Adenosine-NH2), 6.06
(d, 1H, J = 4.8 Hz, H-10), 8.21 (s, 1H, H-8), 8.25 (s, 1H,
H-2); 13C NMR (100 MHZ, DMSO-d6) d = 41.3 (S–
CH2CH), 59.2 (NH–CH2–S), 69.5 (C-50), 70.4 (S–CH2–
CH), 76.0 (C-30), 78.8 (C-20), 87.1 (C-40), 94.0 (C-10), 118.2
(C-5), 145.1 (C-8), 154.9 (C-4), 157.8 (C–C-2), 161.3 (C-6),
176.2 (C=O); FABMS m/z 383 [M]? (10); HRFABMS m/
z (pos): 383.1135 C14H19O5N6S (calcd 383.1132).
[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methyl (2S)-2-
aminopropanoate (4c)
Yellow oil in 51 % yield; IR (neat)mmax: 3,314, 3,096, 2,985,
1,677, 1,201, 1,135; 1H NMR (400 MHz, DMSO-d6) d: 1.26
(d, 3H, J = 13.6 Hz, CH3CH), 3.17 (m, 1H, CH3CH), 4.14
(m, 1H, H-40), 4.28 (t, 1H, J = 5.2 Hz, H-30), 4.38 (dd, 1H,
J = 12, 6.4 Hz, H-50a), 4.47 (dd, 1H, J = 11.8, 3.6 Hz,
H-50b), 4.68 (t, 1H, J = 5.2 Hz, H-20), 5.93 (d, 1H,
J = 4.8 Hz, H-10), 7.88 (br, 2H, Adenosine-NH2), 8.25 (s,
1H, H-8), 8.43 (s, 1H, H-2); 13C NMR (100 MHZ, DMSO-
d6) d = 17.1 (CH3CH), 31.1 (CH3CH), 65.9 (C-50), 70.6 (C-
30), 73.2 (C-20), 81.9 (C-40), 88.3 (C-10), 119.5 (C-5), 141.5
(C-8), 149.5 (C-4), 151.2 (C-2), 155.1 (C-6), 168.0
(C = O); FABMS m/z 339 [M]? (50); HRFABMS m/
z (pos): 339.1406 C13H19O5N6 (calcd 339.1411).
4-{[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methoxy}-4-oxobutanoic
acid (4f)
Clear oil in 53 % yield; IR (neat)mmax: 3,353, 3,109, 2,993,
1,703, 1,678, 1,204, 1,135, 1,080. 1H NMR (400 MHz,
DMSO-d6) d = 2.52–2.60 (m, 4H, HOOC–CH2CH2), 4.31
(m, 1H, H-40), 4.40 (dd, 1H, J = 12, 4.8 Hz, H-50a), 4.51 8
(dd, 1H, J = 12, 4.4 Hz, H-50b), 5.12 (dd, 1H,
J = 6.2 = 2.8 Hz, H-30), 5.87 (dd, J = 5.8, 2.4 Hz, 1H,
H-20), 6.32 (d, 1H, J = 4.8 Hz, H-10), 7.23 (bs, 2H, Adeno-
sine-NH2), 8.20 (s, 1H, H-8), 8.28 (s, 1H, H-2); 13C NMR
(100 MHZ, DMSO-d6) d = 27.3(HOOC–CH2CH2–COOR),
32.0(HOOC–CH2CH2–COOR), 65.6(C-50), 79.4(C-30), 82.0
(C-20), 85.3 (C-40), 84.5 (C-10), 118.7 (C-5), 142.5 (C-8),
149.7 (C-4), 152.2 (C-2), 162.4 (C-6),173.4 (C=O), 177.2
(C=O); FABMS m/z 368 [M]? (24); HRFABMS m/z (pos):
368.1210 C14H18O7N5 (calcd 368.1201).
[(2R,3S,4R,5R)-5-(6-Amino-9H-purin-9-yl)-3,4-
dihydroxytetrahydrofuran-2-yl]methyl (2S)-pyrrolidine-2-
carboxylate (4d)
Yellow oil in 47 % yield; IR (neat)mmax: 3,338, 3,214,
2,946, 1,681, 1,651, 1,204, 1,133; 1H NMR (400 MHz,
DMSO-d6) d = 1.88 (m, 3H, NH–CH2CH2CH2a), 2.22 (m,
1H, NH–CH2CH2CH2b), 3.19 (m, 2H, NH–CH2), 4.12 (m,
1H, H-40), 4.27 (t, 1H, J = 5.2 Hz, NH–CH), 4.41 (m, 2H,
H-50), 4.57 (t, 1H, J = 6 Hz, H-30), 4.68 (t, 1H,
J = 5.2 Hz, H-20), 5.89 (d, 1H, J = 5.2 Hz, H-10), 7.45 (br,
2H, Adenosine-NH2), 8.14 (s, 1H, H-8), 8.34 (s, 1H, H-2);
Biological assays
The activity of ADA was determined by measuring the
ammonialiberation. A concentration of 4 mmol/L adenosine
123

Med Chem Res
was used as substrate (Kocic et al., 2012; Pedersen and
Berry, 1977). The AK assay is based on the adsorption of
phosphorylated compounds onto DEAE-cellulose paper as
previously reported (Willem de Jong and Kalkman, 1973).
dockings runs were carried out with both enzymes. Finally,
˚
resulting docking orientations lying within 2.0 A in the
root-mean square deviation (rmsd) tolerance of each other
were clustered together and represented by the result with
the most favorable free energy of binding. The final ana-
lysis was performed using the best energy conformation
from the most populated conformation cluster using
AutoDockTools and Pymol 0.99.
Docking study methodology
Protein setup
The X-ray structures of the human ADA and human
carboxylestearase 1 were taken from the Protein Data Bank
under entries 3IAR and 1MX1, respectively, and Sybyl 8.0
was used in all the study. Ligands, solvent, and ions (except
the zinc cation present in ADA) were removed from the
original PDB files. All hydrogens were added to the
enzyme structures and AMBER FF02 atom charges were
assigned using the Biopolymer module of Sybyl. The
structures were relaxed by energy minimization, employing
the AMBER FF02 force field with a dielectric constant of
Results and discussion
Chemistry
The amino acids glycine, b-alanine (S)-alanine, (S)-proline,
and (R)-thioproline were protected in the amine moiety
through the tert-butoxy carbamate formation as a pre-
ventive approach for avoiding auto-condensation in future
synthetic steps. This was achieved by the treatment of the
corresponding amino acid with a basic aqueous solution of
di-tert-butyl pyrocarbonate. For the same reason, the cis-
diol of adenosine was protected by forming the 20,30-O-
isopropylideneadenosine ketal (2). Intermediate 2 was
obtained by stirring adenosine with acetone in an acidic
media and 2,2-dimethoxypropane as desiccant (Scheme 1).
The subsequent coupling of 2 with the corresponding N-
Boc-protected amino acid by employing dicyclohexylcar-
bodiimide as coupling reagent or succinic anhydride,
afforded the esters 3a–3f in good yields (75–92 %). The
regioselective acylation in the 50 hydroxyl group versus the
exo-amino group of adenosine was verified with the ana-
˚
20 D and a gradient cutoff of 0.05 kcal/molA using the
molecular mechanics module. After minimization, the
resulting structures were saved in PDB format; non-polar
hydrogens were merged from the structures, and Gasteiger
atom charges on the atoms were added using AutoDock-
Tools 1.5.2. The grid maps representing the protein were
calculated with AutoGrid 4.0. Thus, grid maps were cal-
culated covering all the previously reported active site. The
dimensions of grids were: (a) for the ADA a box of
˚
40 9 40 9 40 A centered in de zinc cation of one catalytic
site and (b) for the human carboxylesterase 1 a box of
˚
50 9 50 9 50 A centered in the catalytic site of monomer
˚
6; in both cases, the grid was spacing of 0.375 A between
1
lysis of H NMR spectra of the compounds 3a–3f. The
the grid points.
signal between 5.71 and 6.71 ppm (a broad peak that
integrates for two hydrogen atoms) was found in all the
spectra and corresponds to the amino group of the adeno-
sine. Figure S1 in supplementary material shows an extract
Ligand setup
1
Ligands were constructed and optimized to a semi-empiric
level using AM1 model in Spartan 02 for Linux. After
minimization, the resulting structures were saved in Sybyl
mol2 format; non-polar hydrogens were merged from the
structures; torsion tree root and rotatable bonds were
identified. Finally, Gasteiger atom charges on the atoms
were added using AutoDockTools 1.5.2.
of the H NMR spectrum of compound 3b to illustrate the
regioselectivity of the reaction. This evidence suggests that
the amino group of adenosine did not participate in the
nucleophilic attack toward the carbonyl group of the amino
acids or succinic anhydride, and therefore, the coupling
was achieved exclusively at the 50 position of the ribose.
Moreover, the change on the chemical shift of the methy-
lene of the ribose to downfield in the spectra confirms the
regioselectivity of the coupling reaction, due to a decrease
in the electronic density of the ester bond with respect to
the hydroxyl of adenosine (deshielding effect).
Docking
Docking simulations were performed with Autodock 4.0
software. These simulations were carried out using the
Lamarckian Genetic Algorithm, with an initial population
of 150 randomly placed individuals, maximum number of
5,000,000 energy evaluations, mutation rate of 0.02,
crossover rate of 0.80. The remaining parameters were pre-
established in AutoDockTools. A total of 20 independent
The simultaneous removal of the dioxolane ketal and
Boc group was achieved by treating the esters 3a–3e (or
ketal removal in case of 3f) with trifluoroacetic acid, which
afforded the desired products 4a–4f (Scheme 1). Although
the yields were not good (ranging from 50 to 74 %), these
conditions were found to be the most appropriate to
123

Med Chem Res
Scheme 1 Synthesis of compounds 4a–4f. Reagents and conditions: a acetone, p-TsOH, 2,2-dimethoxypropane, rt, 89 %; b R0COOH, DCC,
4-DMAP, CHCl3, rt, 75–92 %; c TFA, rt, 48–74 %. d Succinic anhydride, AcOHcat, r.t. 3 h
achieve the simultaneous deprotection, in contrast to other
Table 1 Predicted logP and Docking DGbinding results from ADA
and hCE1
methodologies such as the treatment with hydrogen chlo-
ride in dioxane or water, or by microwave-assisted organic
synthesis (MAOS); these methodologies involved the ester
hydrolysis as major product. Also a recently reported Boc
deprotection method using fluorinated alcohols was
employed without significant reaction progress (Jason
et al., 2008). All the intermediates and final products
structures were characterized by its spectroscopic (1H
NMR, 13C NMR and IR) and spectrometric spectra
(FABMS and HRFABMS). In order to predict if the
esterification would not modify the pharmacokinetic
properties in a significant manner, the logP was calculated
using the molinspiration server (www.molinspiration.com).
The results are reported in Table 1; according to these data,
a similar solubility is expected because no significant
change in logP values with respect to adenosine was
observed.
Compound
Predicted logPa
DGbind (kcal/mol)
ADA
hCE1
Adenosine
-0.854
-1.555
-1.593
-1.154
-0.185
-0.213
-0.881
-7.58
-8.13
-8.04
-8.23
-9.16
-7.20
-5.95
ND
4a
4b
4c
4d
4e
4f
-7.30
-7.58
-7.65
-6.37
-2.70
ND
ND not docked
a
Molinspiration (www.molinspiration.com)
and was retrieved from the Protein Data Bank (PDB, entry
3IAR). Sybyl 8.0 was used to prepare the protein for the
docking procedure, while Autodock 4.0 software with a
Lamarckian genetic algorithm was used to perform the
study (Morris et al., 1998; Huey et al., 2007). Compounds
4a–4f, desoxyadenosine [(2R, 3S, 5R)-5-(6-amino-9H-pu-
rin-9-yl)-tetrahydro-2-(hydroxymethyl)furan-3-ol] and aden-
osine were docked in the catalytic domain of ADA.
Molecular docking studies
The main goal of this preliminary work is the enhancement
of the plasma half-life of adenosine caused by the esteri-
fication of different acids at the carbinolic alcohol. We
attempted to evaluate this situation by predicting the
binding affinity of compounds and measuring the DG value
predicted after carrying out a molecular docking study.
Human ADA was used as target for the docking analysis
Desoxyadenosine is the co-crystallized ligand found in
ADA with PDB entry 3IAR, and thus, we performed a
molecular docking to validate our methodology. The pose
of docked and crystallized desoxyadenosine is similar in
123

Med Chem Res
orientation and conformation, with a RMSD value of 0.197
(Fig. 3a). Additionally, we docked a set of previously
reported ADA inhibitors (Curtis et al., 1998), which
experimental Ki values were compared with their calcu-
lated DGbinding. A reasonable correlation was found
(R2 = 0.85) and confirms the reliability of the model (see
Supplementary material).
affinity to this enzyme (DG = -7.20 and -5.95 kcal/mol,
respectively) as compared with adenosine (DG = -7.58),
see Table 1. However, the remaining compounds showed a
more stabilized complex (lower DG value, see Table 1) in
the catalytic domain of ADA than adenosine; these data
result in a disagreement with previously published works in
which a bulky group near the carbinol diminishes the
affinity for ADA. Nevertheless, an enzyme inhibition
originated by the compounds interaction with ADA is not
discarded. This idea is reasonable if we consider that ara-
lkyladenines, which were used in the docking validation,
showed the same binding pose than adenosine in ADA
catalytic domain (see supplementary material). Neverthe-
less, further studies must be done to corroborate this
hypothesis since the molecular docking studies were
performed in a rigid structure of the protein, thus resulting
in a limited depiction of the enzyme-ligand binding
phenomena.
On the other hand, ADA is an enzyme which requires a
zinc cation for the catalytic activity and a water molecule
(between Asp 295 and Asp 296) to carry out the nucleo-
philic attack at the C6 of adenosine. In accordance to this,
the adenosine orientation and conformation after the
docking study were similar as those proposed by Kinoshita
(Kinoshita et al., 2005), where the adenine moiety is in the
S0 subsite (a hydrophilic domain) and the cis-diol points to
the F0 subsite (a hydrophobic domain) (Fig. 3b). The dis-
tance between the nitrogen of the amino group and the zinc
˚
cation is 2.61 A and the orientation of the adenosine seems
to be the appropriate for the enzymatic reaction. The
ligand-receptor complex is stabilized by the existence of
six hydrogen bonds: the exocyclic amino group with
Asp295, His238 and Glu217, the N1 of purine ring with
Glu217 and finally, N3 of the purine with Gly184 and N7
with Asp296. Also, the energy of the complex is negative
(Table 1), which indicates that this interaction could be
thermodynamically favored. In summary, the molecular
docking of desoxyadenosine and adenosine with ADA
validate the molecular docking procedure used and thus
providing enough reliability for further studies.
Since thermodynamics of the molecular docking did not
predict an increase of half-life by loss of affinity, the
analysis of the conformations and orientations of the
compounds 4a–4d in the catalytic domain were performed.
The binding mode of 4c and 4d is shown in Fig. 4.
According to the docking studies, compounds bind to the
ADA catalytic domain as adenosine does, where the ade-
nine ring points to the S0 subsite; however, there are sig-
nificant differences between the conformations and
interactions between them. First, there exist a rotation in
compounds 4c and 4d with respect to adenosine, and the
complex is stabilized by the formation of different hydro-
gen bonds between Glu217 and Gly184 with the amino
group of purine ring, along with a new hydrogen bond
formed with the 20 position hydroxyl and Asp19; this
interaction could contribute to the increase in the binding
affinity for compounds 4c and 4d. Moreover, the amino
acid moiety also presents interactions with the F0 subsite
Having validated our molecular docking procedure, we
focused on predicting the affinity of compounds 4a–4f to
ADA employing the same methodology used to desoxy-
adenosine and adenosine. The first criterion to measure the
enhanced plasma half-life of the prodrugs was the
DG value. In this sense, a decrease in the metabolism by
ADA for 4e and 4f could be predicted due to a loss in
Fig. 3 a Pose of ADA (3IAR) co-crystallized ligand. Crystal
structure of deoxyadenosine is shown in red while the binding mode
obtained with the docking methodology is shown in blue
(RMSD = 0.197). b Binding mode of adenosine in the ADA binding
cavity. The adenine ring is oriented to the S0 subsite of the catalytic
domain. The orientation and distance are adequate for the possible
oxidation. H-bonds are shown as dashed black lines and the blue
sphere corresponds to the zinc cation (Color figure online)
123

Med Chem Res
(colored as purple amino acids in Fig. 4) that may con-
tribute to the higher stabilization of the ligand-receptor
complex. This conformational change has an important
impact on the disposition of compounds to suffer the oxi-
dative deamination as a result of the increase in the dis-
tance between the amino group of purine and the zinc
On the other hand, the design of molecules considered
the cleavage by ubiquitous esterases in order to release
adenosine, which is the biological active molecule. There
are several nucleotide prodrugs containing ester moieties
that are activated by esterases, therefore the possibility of
enzymatic cleavage is reasonable (Li et al., 2008). As a
way to predict the metabolic hydrolysis, molecular docking
was conducted with a liver carboxylesterase, and among all
the plasmatic esterases, human caboxylesterase 1 (hCE1)
seems to be the best candidate for the study. hCE1 is an
enzyme involved in the cleavage of a large variety of
compounds containing ester, amide, carbamate, and thio-
ester functionalities and is also involved in the activation of
various antiviral, anticancer, and antibiotic prodrugs con-
taining ester or lactone functionalities (Yang et al., 2011).
The catalytic domain of this enzyme consists in the
Glu354-His468-Ser221 residue triad and involves a clas-
sical ‘‘ping-pong’’ hydrolysis mechanism via a tetrahedral
intermediate (Redinbo and Potter, 2005). This esterase is
more abundant in liver, bowels, and plasma so it was
considered to predict the possibility of hydrolysis. The
mol2 file was retrieved from the Protein Data Bank with
the PDB entry 1MX1 (Bencharit, et al., 2003). Compounds
4a–4e were satisfactory docked in the hCE1 catalytic
domain by showing negative DG values (Table 2). With
these preliminary results, we can anticipate that compounds
can be hydrolyzed by this promiscuous esterase and thus
releasing adenosine. However, the analysis of the orienta-
tion was carried out in order to analyze the feasibility of the
hydrolysis by this enzyme, similarly to description per-
formed to ADA. The binding mode of compounds 4b and
4c is shown in Fig. 5, where the carbonyl group of the ester
is clearly susceptible to the nucleophilic attack by the
hydroxyl group of Ser221. As expected, the adenine moiety
˚
˚
cation (2.61 A for adenosine and 4.7 and 5.34 A for 4c and
4d, respectively). The change in conformation reduces the
probability of being substrate for a nucleophilic attack by
the water molecule responsible for adenosine oxidation
(localized between Asp295 and Asp296) despite of the
higher affinity to ADA. Moreover, the 6 position of purine
ring is not in a correct orientation for the nucleophilic
reaction, so that oxidation could not be possible. Com-
pounds 4a, 4b, and 4f showed similar results and charac-
teristics in binding mode described above (Figure S1 in
Supplementary material).
In addition to the change in conformation, compounds
4a–4c have an alternative binding mode in ADA catalytic
site (with a similar energetic value to that previously
described), where the amino acid moiety is coordinated
with the zinc cation in the S0 subsite, while the adenine
ring is in the F0 subsite. In this orientation, the molecule
is not suitable for the oxidative reaction and, in conse-
quence, a decrease in metabolism could be observed
(Figure S2 in Supplementary material). Using molecular
docking and the analysis of the binding modes of the
compounds in the catalytic site of ADA, a possible
explanation for the hypothetical decrease in the metabo-
lism of compounds for ADA was proposed, even though
the binding energy is lower than adenosine. Therefore,
according with this theoretical study, compounds 4a–4f
could be employed as conjugates designed to enhance
adenosine plasma half-life.
Fig. 4 Binding modes of compounds 4c (a) and 4d (b) in ADA catalytic domain. Purple amino acids belong to the F0 subsite. H-bonds are
shown as dashed black lines, the blue sphere corresponds to the zinc cation and the distances are expressed in Angstroms (Color figure online)
123

Med Chem Res
human plasma and the publication of these results is
forthcoming.
Table 2 ADA and AK metabolism for compounds 4a–4f
Compound
Metabolism (nmol*mg prot/min)
Adenosine deaminase
Adenosine kinase
In vitro preliminary metabolism assay
Adenosine
10.4
7.25
7.25
6.25
2.19
3.23
8.20
0.100
0.135
0.043
0.090
0.113
0.159
0.023
Even though docking studies indicated that compounds 4a–
4f could be less metabolized than adenosine by ADA, an
experimental corroboration is required. For this reason, a
preliminary enzyme-kinetic assay was performed using
ADA. Table 2 shows that adenosine biotransformation in
this assay was 10.4 nmol mg/min; nevertheless, com-
pounds 4a–4f clearly showed a significant reduction in
the bioconversion by ADA ranging from 8.20 to
2.19 nmol mg/min. These evidences undoubtedly corrob-
orate that the esterification of different acids in the 50
hydroxyl of ribose diminished the oxidative deamination of
the nucleoside, which was previously predicted by the
docking study. Apparently, the reduction of the metabolism
is due to an unfavorable orientation within the catalytic
pocket of ADA in case of compounds 4a–4d and affinity
loss in ADA catalytic domain for compounds 4e and 4f.
Indeed, these results are in agreement with those proposed
by Ciuffreda and coworkers (Ciuffreda et al., 2004), in
which modifications in the carbinolic moiety of ribose,
result in activity loss by ADA. It is noteworthy that the
bulkier compounds 4d and 4e (with a proline and thiopr-
oline progroups, respectively) have the best profile in the
study, with a reduction of 4.7 and 3.2 fold in adenosine
metabolism, respectively.
4a
4b
4c
4d
4e
4f
was oriented to the large and flexible pocket existing in the
catalytic domain of hCE1 (red amino acids); this domain
confers to the enzyme promiscuity properties, regarding to
substrate specificity (Imai, 2006). Compound 4a, 4d–4e
presents similar results, while compound 4f was not suc-
cessfully docked in a similar binding mode. Nevertheless,
the possibility of cleavage by other plasmatic or liver
esterases, such as human carboxylesterase 2, butyrylcho-
linesterase, or acetylcholinesterase is not discarded. Indeed,
the early treatment of compounds with acidic media for
some hours resulted in ester cleavage, so the probability of
occurring the same phenomenon in a biological system is
reasonable (data not shown). Finally, we propose a plau-
sible mode of hydrolysis of compounds 4a–4e supported by
the affinity to the catalytic site of hCE1 and the analysis of
the binding pose of the prodrugs within it. Currently, we
are examining the in vitro hydrolysis of compounds in
Similar findings were detected in the metabolism by
AK: the bioconversion of adenosine by this enzyme was
Fig. 5 Binding modes of compounds 4b (a) and 4c (b) in hCE1
catalytic site. The carbonyl-serine221 (hydroxyl group) distance is
shown as dashed black lines. Purple amino acids correspond to the
catalytic triad and red amino acids correspond to the large pocket of
the enzyme (Color figure online)
123

Med Chem Res
Cronstein BN, Kramer SB, Weissmann G, Hirschorn R (1983)
Adenosine: a physiological modulator of superoxide anion
generation by human neutrophils. J Exp Med 158:1160–1177
Curtis MA, Varkhedkar V, Pragnacharyulu PVP, Abushanab E (1998)
Adenosine deaminase inhibitors. Synthesis and biological eval-
uation of aralkyladenines (ARADS). Bioorg Med Chem Lett
8:1639–1642
0.14 nmol mg/min, whereas compounds 4b, 4c, and 4f
showed slower transformation (0.043, 0.090 and
a
0.023 nmol mg/min, respectively). In summary, these
findings suggest that the synthesized esters could increase
the feasibility of using adenosine in the treatment and
prophylaxis of several diseases, including cirrhosis; in
addition, the progroup’s specific activity could offer an
added benefit to this pathology.
Gulizar A (2004) Antioxidant effects of sulfur-containing amino
acids. Yonsei Med J 45:776–788
´
´
´
Hasko G, Nemeth ZH, Vizi ES, Salzman AL, Szabo C (1998) An
agonist of adenosine A3 receptors decreases interleukin-12 and
interferon-gamma production and prevents lethality in endo-
toxemic mice. Eur J Pharmacol 358:261–268
Conclusions
Hernandez-Munoz R, Diaz-Munoz M, Lopez V, Lopez-Barrera F,
´
Yanez L, Vidrio S, Aranda-Frausto A, Chagoya de Sanchez V
(1997) Balance between oxidative damage and proliferative
potential in an experimental rat model of CCl4-induced cirrho-
sis: protective role of adenosine administration. Hepatology 26:
1100–1110
Although adenosine has an important role in prophylaxis
and treatment of liver cirrhosis, its lack of metabolic sta-
bility limits its usage as an alternative therapy. For this
reason, six novel adenosine derivatives were synthesized in
order to minimize the adenosine metabolism by ADA. The
molecular docking studies suggested that the decrease in
metabolism was caused by two possible mechanisms:
(a) an inadequate orientation in the catalytic site of ADA,
which makes impossible the nucleophilic attack necessary
for the deamination reaction and (b) affinity lost to ADA
which was detected in the case of compounds 4e and 4f.
Also, according to the performed molecular docking
studies, it is expected that the ester type bond in the
compounds 4a–4f could be hydrolyzed by human carb-
oxylesterase, so they could be considered as bipartite
prodrugs. In accordance to theoretical studies, all of com-
pounds showed a lower metabolism by ADA than adeno-
sine. This result was also observed in the bioconversion by
AK. Therefore, these compounds demonstrated to be
interesting prototypes for further optimization in order to
increase the adenosine hepatoprotective potential, which
could be explored in further studies.
´
´
´
´
Hernandez-Mun˜oz R, Dıaz-Mun˜oz M, Suarez J, Chagoya de Sanchez
V (1990) Adenosine partially prevents cirrhosis induced by
carbon tetrachloride in rats. Hepatology 12:242–248
´
´
´
´
Hernandez-Mun˜oz R, Dıaz-Mun˜oz M, Suarez-Cuenca JA, Trejo-Solıs
´
´
´
C, Lopez V, Sanchez-Sevilla L, Yan˜ez L, de Sanchez VC (2001)
Adenosine reverses a preestablished CCl4-induced micronodular
cirrhosis through collagenolytic activity and stimulating hepato-
cyte cell proliferation in rats. Hepatology 34:667–687
Huey R, Morris GM, Olson AJ, Goodsell DA (2007) A semiempirical
free energy force field with charge-based desolvation. J Comput
Chem 28:1145–1152
Imai T (2006) Human carboxylesterase isozymes: catalytic proper-
ties and rational drug design. Drug Metab Pharmacokinet 21:
173–185
Ishizaki-Koizumi S, Sonaka I, Fujitani S, Nishiguchi S (2002)
Mechanism of the protective effect of L-alanine to D-galactos-
amine-induced hepatocellular injury: comparative studies of
L-alanine and pyruvate. Biochem Biophys Res Commun 291:
738–743
Jacobson KA, Gao ZG (2006) Adenosine receptors as therapeutics
targets. Nat Rev Drug Discov 5:247–264
Jason C, Jaime-Figueroa S, Jiang L, Wagner P (2008) Novel practical
deprotection of N-Boc compounds using fluorinated alcohols.
Synth Commun 38:3840–3853
Kinoshita T, Nakanishi I, Terasaka T, Kuno M, Seki N, Warizaya M,
Matsumura H, Inoue T, Takano K, Adachi H, Mori Y, Fujii T
(2005) Structural basis of compound recognition by adenosine
deaminase. Biochem 44:10562–10569
Acknowledgments This work was supported by Grants from CO-
NACyT. The authors thank Dr. Rafael Castillo for lending the soft-
ware and hardware.
Kocic G, Nikolic J, Jevtovic-Stoimenov T, Sokolovic D, Kocic H,
Cvetkovic T, Pavlovic D, Cencic A, Stojanovic D (2012) L-
Arginine intake effect on adenine nucleotide metabolism in rat
parenchymal and reproductive tissues. Sci World J 2012:
208–239
References
Bencharit S, Morton CL, Hyatt JL, Kuhn P, Danks MK, Potter PM,
Redinbo MR (2003) Crystal structure of human carboxylesterase
1 complexed with the Alzheimer’s drug tacrine: from binding
promiscuity to selective inhibition. Chem Biol 10:341–349
Ciuffreda P, Loseto A, Allessandrini L, Terraneo G, Santaniello E
(2003) Adenylate deaminase (50-adenylic acid deaminase,
AMPDA)-catalyzed deamination of 50-deoxy-50-substituted and
50-protected adenosines: a comparison with the catalytic activity
of adenosine deaminase (ADA). Eur J Org Chem 24:4748–4751
Ciuffreda P, Buzzi B, Alessandrini L, Santaniello E (2004) Activity of
adenosine deaminase (ADA) and adenylate deaminase (AMP-
DA) towards 6-chloropurine nucleosides modified in the ribose
moiety. Eur J Org Chem 21:4405–4409
Li F, Maag H, Alfredson T (2008) Prodrugs of nucleoside analogues
for improved oral absorption and tissue targeting. J Pharm Sci
97:1109–1134
Ligumsky M, Sestieri M, Okon E, Ginsburg I (1995) Antioxidants
inhibit ethanol-induced gastric injury in the rat: role of
manganese, glycine and carotene. Scand J Gastroenterol 30:
854–860
Mendieta-Condado E, Chagoya de Sanchez V, Hernandez-Mun˜oz R
(2007) Adenosine can accelerate the cell cycle during rat liver
regeneration induced by partial hepatectomy.
46:S142l-
J Hepatol
Michel PP, Marien M, Ruberg M, Colpaert F, Agid Y (1999)
Adenosine prevents the death of mesencephalic dopaminergic
123

Med Chem Res
neurons by a mechanism that involves astrocytes. J Neurochem
72:2074–2082
Morris GM, Goodsell D, Halliday RS, Huey R, Hart WE, Belew RK,
Olson AJ (1998) Automated docking using a Lamarckian genetic
algorithm and a empirical binding free energy function. J Comput
Chem 19:1639–1662
and antioxidant enzymes in the erythrocite of rats with alcohol-
induced liver injury. Cell Biochem Funct 22:123–128
Suo M, Mukaisho K, Shimomura A, Sugihara H, Hattori T (2006)
Thioproline prevents carcinogenesis in the remnant stomach
induced by duodenal reflux. Cancer Lett 237:256–262
Willem de Jong J, Kalkman C (1973) Myocardial adenosine kinase:
activity and localization determined with rapid radiometric
assay. Biochim Biophys Acta 320:388–396
¨
Mosher GH, Schrader J, Deussen A (1989) Turnover of adenosine in
plasma of human and dog blood. Am J Physiol 256:C799–C806
Mu¨ller CE, Jacobson KA (2011) Recent developments in adenosine
receptor ligands and their potential as novel drugs. Biochim
Biophys Acta 1808:1290–1308
Yang Y, Aloysius H, Inoyama D, Chen Y, Hu L (2011) Enzyme-
mediated hydrolytic activation of prodrugs. Acta Pharm Sin B.
1:143–159
Pedersen RC, Berry AJ (1977) Sensitive, optimized assay for serum
AMP deaminase. Clin Chem 23:1726–1733
Redinbo MR, Potter PM (2005) Mammalian carboxylesterases: from
drug targets to protein therapeutics. Drug Discov Today
10:313–325
Yoo BK, Choi JW, Yoon SY, Ko KH (2005) Protective effect of
adenosine and purine nucleos(t)ides against the death by
hydrogen peroxide and glucose deprivation in rat primary
astrocytes. Neurosci Res 51:39–44
Senthilkumar R, Sengottuvelan N, Nalini N (2003) Protective effect
of glycine supplementation on the levels of lipid peroxidation
123