Adenylate Deaminase (5′-Adenylic Acid Deaminase, AMPDA)-Catalyzed Deamination of 5′-Deoxy-5′-Substituted and 5′-Protected Adenosines: A Comparison with the Catalytic Activity of Adenosine Deaminase (ADA)

Article abstract of DOI:10.1002/ejoc.200300435

The enzyme adenylate deaminase (AMPDA) is able to catalyze the hydrolytic deamination of 5′-substituted and 5′-protected 5′ -deoxyadenosines, whereas limited or no activity is shown by adenosine deaminase (ADA) towards the same substrates. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003.

Full text of DOI:10.1002/ejoc.200300435

FULL PAPER
Adenylate Deaminase (5Ј-Adenylic Acid Deaminase, AMPDA)-Catalyzed
Deamination of 5Ј-Deoxy-5Ј-Substituted and 5Ј-Protected Adenosines: A
Comparison with the Catalytic Activity of Adenosine Deaminase (ADA)
Pierangela Ciuffreda,[a] Angela Loseto,[a] Laura Alessandrini,[a] Giancarlo Terraneo,[a] and
Enzo Santaniello*[a]
Keywords: Nucleosides / Enzyme catalysis / Biotransformations / Deamination
The enzyme adenylate deaminase (AMPDA) is able to cata-
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
lyze the hydrolytic deamination of 5Ј-substituted and 5Ј-pro- Germany, 2003)
tected 5Ј-deoxyadenosines, whereas limited or no activity is
shown by adenosine deaminase (ADA) towards the same
substrates.
Introduction
Adenosine deaminase (ADA, EC 3.5.4.4) and adenylate
deaminase (5Ј-adenylic acid deaminase, AMPDA, EC
3.5.4.6) catalyze the deamination of adenosine (1a) and 5Ј-
phosphate adenosine (adenylic acid, AMP, 1b) to the corre-
sponding inosines 2a and 2b (Scheme 1).
Figure 1. Structure of 2Ј,3Ј-O-diacetyladenosine (3) and 2Ј,3Ј-O-
isopropylideneadenosines 4aϪc
Compared to ADA, applications of AMPDA are less fre-
quent and only recently this enzyme has been proposed for
the biotransformation of substrates other than AMP 1b.[6]
Scheme 1. ADA- and AMPDA-catalyzed deamination of adenos- We have previously observed that AMPDA is able to cata-
ine (1a) and 5Ј-adenylic acid (1b)
lyze the fast deamination of 2Ј,3Ј-isopropylideneadenosine
4a and of 5Ј-substituted 2Ј,3Ј-isopropylideneadenosines like
4b and 4c that are not substrates for ADA.[5] It seemed
ADA can already be considered a valuable biocatalyst
interesting to investigate the activity of AMPDA on a few
for the biotransformation of a wide range of structurally
5Ј-substituted and 5Ј-protected 5Ј-deoxyadenosines 5aϪe
modified purine nucleosides,[1] provided that a hydroxy
and compare the activity of this deamination to inosines
group is present at the 5Ј-position.[2] Additionally, it has
6aϪe with that of the other deaminating enzyme, namely
been shown that some flexibility is possible at the 2Ј,3Ј-
ADA (Figure 2).
positions.[2,3] In previous papers,[4,5] we have shown that
steric hindrance at positions 2Ј and 3Ј is well tolerated by
both enzymes, since 2Ј,3Ј-O-diacetyladenosine (3) and 2Ј,3Ј-
O-isopropylideneadenosine (4a) may be converted into the
corresponding inosine derivatives (Figure 1).
[a]
`
Dipartimento di Scienze Precliniche LITA Vialba-Universita
degli Studi di Milano,
Via G. B. Grassi, 74, 20157 Milano, Italy
Figure 2. Structure of adenosine 5aϪe and inosine derivatives 6aϪe
DOI: 10.1002/ejoc.200300435
4748
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2003, 4748Ϫ4751

Catalytic Activity of Adenosine Deaminase (ADA)
FULL PAPER
Scheme 2. Reagents and conditions: (a) SOCl2, Py, CH3CN, 0 °C; (b) NaN3, DMF, 80 °C; (c) aq. MeOH, room temp; (d) LiEt3BH,
DMSO, room temp; (e) CH3SNa, DMF, room temp; (f) Candida antarctica, vinyl acetate, THF, room temp.
out that Robins and co-workers had already reported[8] that
an α-amilase from Aspergillus oryzae (Sigma, USA) showed
Results and Discussion
deaminating properties and was able to convert the thiome-
thyl derivative 5d into the corresponding inosine 6d. How-
ever, this ‘‘adenosine deaminase’’ activity was different from
that of calf intestine ADA that did not transform the nu-
cleoside 5d. Since the AMPDA used by us is an enzymatic
preparation from Aspergillus species purchased from Sigma
(USA), there is the possibility that the enzyme used by
Robins was not an adenosine deaminase, but rather an early
preparation of AMPDA. In any event, we have confirmed
that the enzymatic preparation of AMPDA used by us
corresponds to a 5Ј-adenylic acid deaminase, since it is able
to convert AMP (1b) to IMP (2b) in 10 min.
Synthesis of 5Ј-Modified Adenosines 5a؊e
For the synthesis of 5Ј-substituted 5Ј-deoxy derivatives
5aϪd, we started from the diastereomeric mixture of 5Ј-
chloro-5Ј-deoxy-2Ј,3Ј-O-sulfinyladenosines (7) prepared
from adenosine (1a) as described previously.[7] Treatment
of the above intermediate 7 with sodium azide in dimethyl
formamide at 80 °C directly afforded the deprotected azide
5a (45 %). Compounds 5bϪd were prepared according to
Robins and co-workers.[7] Thus, hydrolysis with aqueous
methanol of the sulfinate moiety of 7 afforded the com-
pound 5b (98 %), that was in turn, reduced with lithium
triethylborohydride (Superhydride) to the 5Ј-deoxyadenos-
ine (5c, 44 %) or transformed with sodium methanethiolate
to the 5Ј-deoxy-5Ј-methylthioadenosine (5d, 50 %).
The acetate 5e was preferably prepared by the previously
described enzymatic acetylation of adenosine 1a by means
of the Candida antarctica lipase,[4] since any chemical ap-
proach would invariably yield mixtures of O- and N-acet-
ates. The overall synthetic picture is summarized in
Scheme 2.
Table 1. ADA- and AMPDA-catalyzed deamination of adenos-
ines 5aϪe
Substrate
AMPDA Reaction
time (min)[a]
Product
ADA
(yield,%)[b]
(yield,%)[c]
5a
5b
5c
5d
5e
150
30
10
3
6a (94)
6b (92)
6c (91)
6d (95)
6e (97)
Ϫ
26
25
Ϫ
AMPDA- and ADA-Catalyzed Deamination of 5Ј-Modified
Adenosines 5a؊e
10
22
[a] Time to complete reaction. [b] Yield of isolated products. [c] Reac-
tion stopped at 24 h.
The enzymatic deamination of compounds 5aϪe was car-
ried out in a 3 % DMSO aqueous solution, conditions that
do not influence ADA and AMPDA activity.[5] The com-
mercially available preparation of AMPDA was less active
than ADA (0.107 versus 1.5 units/mg solid) and we decided
to use similar activity for both enzymes. However, due to
problems of solubility in the medium for AMPDA, we fi-
Conclusions
Our results confirm that there might be a few differences
nally adopted an enzyme/substrate ratio (mg/mg) of 1:1 and in the binding pocket of the two deaminating enzymes
0.1:1 for AMPDA and ADA, respectively. Results show that ADA and AMPDA in spite of the similarity of the catalytic
all adenosines 5aϪe were substrates for AMPDA-catalyzed mechanism that has been proposed on the basis of kinetic
conversion into the corresponding inosines 6aϪe in times considerations and the fact that deoxycoformycin deriva-
varying between 3 and 150 minutes. For ADA-catalyzed de- tives are potent inhibitors of both enzymes.[9] This is re-
aminations, we followed the enzymatic transformation inforced by the observation that the two enzymes share
within a reaction time of 24 h. No transformation of com- some conserved sequence elements that may correspond to
pounds 5a and 5d was observed, whereas 5b, 5c and 5e were amino acid residues at the catalytic site.[10] Whereas the tri-
only partially deaminated (Table 1). It should be pointed dimensional structure is known for ADA,[11] the same infor-
Eur. J. Org. Chem. 2003, 4748Ϫ4751
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4749

P. Ciuffreda, A. Loseto, L. Alessandrini, G. Terraneo, E. Santaniello
FULL PAPER
6.1 Hz, 1 H, H-3Ј), 4.53 (ddd, J ϭ 4.4, 6.1, 7.0 Hz, 1 H, H-4Ј), 4.88
(dd, J ϭ 4.0, 4.0 Hz, 1 H, H-2Ј), 5.99 (d, J ϭ 4.0 Hz, 1 H, H-1Ј),
8.22 (s, 1 H, H-2), 8.27 (s, 1 H, H-8) ppm.
mation is not available for AMPDA and therefore a clear
picture of the catalytic events for this enzyme is not avail-
able at present. From the results reported in this paper,
AMPDA seems to be a biocatalyst that is more versatile
than ADA in the nucleoside field, since it is able to catalyze
fast conversion of all the compounds 5aϪe to the corre-
sponding inosine derivatives 6aϪe. This result is even more
intriguing if one considers that the residues present at the
5Ј position in the ribose moiety of the above adenosines are
virtually apolar, whereas at the pH conditions of the en-
zyme activity, the physiological substrate of AMPDA, i.e.
AMP, bears the anionic form of a phosphate anion at the
5Ј position. Further studies are needed in order to investi-
gate the structure-activity relationship of AMPDA in
more detail.
Enzymatic Deamination of Adenosine Derivatives 5aϪe to Inosine
Derivatives 6aϪe: Compounds 5aϪe (0.02 g) in phosphate buffer
(50 m, 6 mL; pH 7.4 for ADA and pH 6.5 for AMPDA) contain-
ing 3 % DMSO were treated with ADA (2 mg) or AMPDA (20 mg)
for the time indicated in Table 1. The progress of reactions was
monitored by HPLC using the phosphate buffer pH 6.0/acetonitrile
ratio as follows: 5a, 70:30; 5b and 5c, 90:10, 5d, 80:20; 5e 85:15.
When the reaction was complete, the solution was lyophilized to
afford inosines 6aϪe as white solids.
5Ј-Azido-5Ј-deoxyinosine (6a): Attempts to crystallize the product
failed [ref.[14] 161 °C (dec.)]. 1H NMR (CD3OD): δ ϭ 3.57 (dd, J ϭ
4.0, 12.4 Hz, 1 H, H-5Јb), 3.81 (dd, J ϭ 7.0, 12.4 Hz, 1 H, H-5Јa),
4.34 (dd, J ϭ 4.0, 4.0 Hz, 1 H, H-3Ј), 4.54 (ddd, J ϭ 4.0, 4.0,
7.0 Hz, 1 H, H-4Ј), 4.86 (dd, J ϭ 4.0, 4.0 Hz, 1 H, H-2Ј), 6.00 (d,
J ϭ 4.0 Hz, 1 H, H-1Ј), 8.08 (s, 1 H, H-2), 8.24 (s, 1 H, H-8) ppm.
Experimental Section
5Ј-Chloro-5Ј-deoxyinosine (6b): M.p. 182Ϫ184 °C (from MeOH),
1
[ref.[15] 191 °C (from MeOH)]. H NMR (CD3OD): δ ϭ 3.92 (dd,
General Remarks: Melting points were recorded with a Stuart
Scientific SMP3 instrument and are uncorrected. 1H NMR spectra
were recorded at 303 K with a Bruker AM-500 spectrometer
equipped with an Aspect 3000 computer, a process control and an
array processor. The 1H NMR chemical shifts are reported in parts
per million, using the signal for residual solvent protons (δ ϭ 3.30
ppm for CD3OD) as an internal reference and coupling constants
(J) are given in Hertz. NMR signals were assigned using 1H-homo-
decoupling and COSY experiments. The progress of all reactions
and column chromatography were monitored by TLC and HPLC.
HPLC analyses were carried out with a Jasco HPLC instrument
with an Uvidec 100 II UV detector using an Alltech Hypersil BDS
C18 (4.6 mm ϫ 250 mm). The eluent was a phosphate buffer pH
6.0/CH3CN (70:30) with a (flow rate 1 mL/min; detection at
260 nm). Thin-layer chromatography (TLC) was performed using
Merck silica gel 60 F254 precoated plates with a fluorescent indi-
cator. Flash chromatography[12] was performed using Merck silica
gel 60 (230Ϫ400 mesh) and appropriate mixtures of CH2Cl2 and
MeOH as eluents.
J ϭ 4.0, 11.4 Hz, 1 H, H-5Јb), 3.93 (dd, J ϭ 4.0, 11.4 Hz, 1 H, H-
5Јa), 4.26 (ddd, J ϭ 4.0, 4.0, 4.0 Hz, 1 H, H-4Ј), 4.37 (dd, J ϭ
4.0,4.0 Hz, 1 H, H-3Ј), 4.73 (dd, J ϭ 4.0, 4.0 Hz, 1 H, H-2Ј), 6.02
(d, J ϭ 4.0 Hz, 1 H, H-1Ј), 8.06 (s, 1 H, H-2), 8.24 (s, 1 H, H-
8) ppm.
5Ј-Deoxyinosine (6c): M.p. 188Ϫ190 °C (from MeOH), 1H NMR
(CD3OD): δ ϭ 1.41 (d, J ϭ 6.0 Hz, 3 H, CH3), 4.05 (dd, J ϭ 4.7,
5.4 Hz, 1 H, H-3Ј), 4.10 (dq, J ϭ 5.4, 6.0 Hz, 1 H, H-4Ј), 4.65 (dd,
J ϭ 4.7, 4.7 Hz, 1 H, H-2Ј), 5.95 (d, J ϭ 4.7 Hz, 1 H, H-1Ј), 8.05
(s, 1 H, H-2), 8.18 (s, 1 H, H-8) ppm. C10H12N4O4 (252.23): calcd.
C 47.62, H 4.80, N 22.21; found C 47.72, H 4.74, N 22.10.
5Ј-Deoxy-5Ј-methylthioinosine (6d): M.p. 218Ϫ220 °C (from
1
MeOH); {ref.[16] 223Ϫ224 °C (from water)}. H NMR (CD3OD):
δ ϭ 2.12 (s, 3 H, SCH3), 2.86 (dd, J ϭ 5.4, 14.0 Hz, 1 H, H-5Јb),
2.92 (dd, J ϭ 5.4, 14.0 Hz, 1 H, H-5Јa), 4.22 (ddd, J ϭ 5.4, 5.4,
5.4 Hz, 1 H, H-4Ј), 4.30 (dd, J ϭ 5.4, 5.4 Hz, 1 H, H-3Ј), 4.71 (dd,
J ϭ 5.4, 5.4 Hz, 1 H, H-2Ј), 6.01 (d, J ϭ 5.4 Hz, 1 H, H-1Ј), 8.05
(s, 1 H, H-2), 8.27 (s, 1 H, H-8) ppm.
All reagents were obtained from commercial sources and used
without further purification. Enzymes were obtained as follows:
adenosine deaminase from calf intestinal mucosa (Sigma, type II,
1.5 units/mg solid), 5Ј-adenylic acid deaminase from Aspergillus
species (Sigma, 0.107 units/mg solid).
5Ј-O-Acetylinosine (6e): M.p. 228Ϫ230 °C (from MeOH), 1H NMR
(CD3OD): δ ϭ 2.06 (s, 3 H, OCOCH3), 4.25 (ddd, J ϭ 3.2, 4.0,
4.0 Hz, 1 H, H-4Ј), 4.30Ϫ4.35 (m, 2 H, H-3Ј and H-5Јb), 4.37 (dd,
J ϭ 3.2, 12.0 Hz, 1 H, H-5Јa), 4.68 (dd, J ϭ 4.7, 4.7 Hz, 1 H, H-
2Ј), 6.01 (d, J ϭ 4.7 Hz, 1 H, H-1Ј), 8.04 (s, 1 H, H-2), 8.20 (s, 1
H, H-8) ppm. C12H14N4O6 (310.26): calcd. C 46.45, H 4.55, N
18.06; found C 46.52, H 4.48, N 18.12.
The starting nucleoside 1a was purchased from Aldrich. 5Ј-Chloro-
5Ј-deoxyadenosine (5b), 5Ј-deoxyadenosine (5c) and 5Ј-deoxy-5Ј-
methylthioadenosine (5d) were prepared according to literature
procedures.[7] The enzymatic preparation of 5Ј-O-acetyladenosine
(5e) has been described previously.[4]
Acknowledgments
5Ј-Azido-5Ј-deoxyadenosine (5a): Sodium azide (2.3 g, 35 mmol)
was added portionwise to a solution of 5Ј-chloro-5Ј-deoxy-2Ј,3Ј-O-
sulfinyladenosine (7) (2 g, 7 mmol) in DMF. The suspension was
heated at 80 °C for 5 h, and the mixture cooled to room tempera-
ture. Excess sodium azide was removed by filtration. The resulting
solution was diluted with ethyl acetate, washed with water, dried
with Na2SO4, and the solvents evaporated to give a solid residue.
The crude azide 5a was purified by flash chromatrography
(CH2Cl2/MeOH, 9:1) to give pure azide 5a (0.9 g, 3.15 mmol, 45
%) as an amorphous solid. Attempts to crystallize the product
failed.[13] 1H NMR (CD3OD): δ ϭ 3.58 (dd, J ϭ 4.4, 12.7 Hz, 1 H,
H-5Јb), 3.84 (dd, J ϭ 7.0, 12.7 Hz, 1 H, H-5Јa), 4.33 (dd, J ϭ 4.0,
`
This work has been financially supported by Universita degli Studi
di Milano (Fondi FIRST).
[1]
M. Ferrero, V. Gotor, Chem. Rev. 2000, 100, 4319Ϫ4347.
A. Bloch, M. J. Robins, J. R. Jr. McCarthy, J. Med. Chem.
[2]
1967, 10, 908Ϫ912.
[3]
´
´
G. Maury, T. Daiboun, A. Elalaoui, C. Genu-Dellac, C. Per-
igaud, C. Bergogne, G. Gosselin, J. L. Imbach, Nucleosides &
Nucleotides 1991, 10, 1677Ϫ1692.
P. Ciuffreda, S. Casati, E. Santaniello, Tetrahedron 2000, 56,
3239Ϫ3243.
[4]
4750
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2003, 4748Ϫ4751

Catalytic Activity of Adenosine Deaminase (ADA)
FULL PAPER
[5]
[11] [11a]
P. Ciuffreda, A. Loseto, E. Santaniello, Tetrahedron 2002, 58,
D. K Wilson, F. A. Quiocho, Biochemistry 1993, 32,
[11b]
5767Ϫ5771.
1689Ϫ1694.
T. J. Marrone, T. P. Straatsma, J. M. Briggs,
[6]
D. K. Wilson, F. A. Quiocho, J. A. McCammon, J. Med. Chem.
1996, 39, 277Ϫ284.
W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43,
2923Ϫ2925.
W. Jahn, Chem. Ber. 1965, 98, 1705Ϫ1708.
A. Hampton, M. Bayer, V. S. Gupta, S. Y. Chu, J. Med. Chem.
1968, 11, 1229Ϫ1232.
A. K. M. Anisuzzaman, R. L. Whistler, Carbohyd. Res. 1978,
61, 511Ϫ518.
A. L. Margolin, D. R. Borcherding, D. Wolf-Kugel, N. A. Mar-
golin, J. Org. Chem. 1996, 59, 7214Ϫ7218.
M. J. Robins, F. Hansske, S. F. Wnuk, T. Kanai, Can. J. Chem.
[12]
[7]
1991, 69, 1468Ϫ1474.
S. F. Wnuk, J. D. Stoeckler, M. J. Robins, Nucleosides & Nucleo-
[13]
[14]
[8]
tides 1994, 13, 389Ϫ403.
C. Frieden, L. C. Kurz, H. R. Gilbert, Biochemistry 1980, 19,
5303Ϫ5309.
[9]
[15]
[16]
[10]
Z. Chang, P. Nygaard, A. C. Chinault, R. E. Kellems, Biochem-
J. Baddiley, J. Chem. Soc. 1951, 1348Ϫ1351.
istry 1991, 30, 2373Ϫ2380.
Received July 14, 2003
Eur. J. Org. Chem. 2003, 4748Ϫ4751
www.eurjoc.org
 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4751