Hydrolysis of phosphomonoesters in nucleotides by cerium(IV) ions. Highly selective hydrolysis of monoester over diester in concentrated buffers

Article abstract of DOI:10.1039/a701481c

Phosphomonoesters in nucleotides are efficiently hydrolysed by Ce^IV ions under physiological conditions. The half-lives of the residues at pH 7.2 and 50 °C ([Ce^IV] = 10 mM) are around 10 min. Phosphomonoester hydrolysis by Ce^IV ions is faster than the hydrolysis of phosphodiesters. Significantly, the selectivity for monoester hydrolysis over diester hydrolysis is remarkably increased by using concentrated buffer solutions (TRIS and HEPES). In 500 mM TRIS buffer, pdA and dAp are hydrolysed 500- and 580-fold faster than is d(ApA), whereas the corresponding ratios in 50 mM TRIS buffer are 85 and 90 respectively. Selective removal of the terminal monophosphate from d(pApA) is achieved by Ce^IV in these concentrated buffers.

Full text of DOI:10.1039/a701481c

View Article Online / Journal Homepage / Table of Contents for this issue
Hydrolysis of phosphomonoesters in nucleotides by cerium(IV) ions.
Highly selective hydrolysis of monoester over diester in concentrated
buffers
Sachiko Miyama, Hiroyuki Asanuma and Makoto Komiyama *
Department of Chemistry and Biotechnology, Graduate School of Engineering,
University of Tokyo, Hongo, Tokyo 113, Japan
Phosphomonoesters in nucleotides are efficiently hydrolysed by CeIV ions under physiological conditions.
The half-lives of the residues at pH 7.2 and 50 ЊC ([CeIV] = 10 mM ) are around 10 min. Phosphomonoester
hydrolysis by CeIV ions is faster than the hydrolysis of phosphodiesters. Significantly, the selectivity for
monoester hydrolysis over diester hydrolysis is remarkably increased by using concentrated buffer
solutions (TR IS and H EPES). In 500 mM TR IS buffer, pdA and dAp are hydrolysed 500- and 580-fold
faster than is d(ApA), whereas the corresponding ratios in 50 mM TR IS buffer are 85 and 90 respectively.
Selective removal of the terminal monophosphate from d(pApA) is achieved by CeIV in these
concentrated buffers.
and adenosine triphosphate by using T4 kinase, and purified by
Introduction
reversed-phase HPLC. HEPES and TRIS were purchased from
Aldrich. Lanthanide salts from Soekawa Chemicals were used
without further purification. Water was purified by a Millipore
Milli-XQ, and sterilized immediately before use. The specific
resistance of the water was greater than 18.3 MΩ cmϪ1, con-
firming that the amounts of metal ions, if any, were sufficiently
minimized. Great care was also taken to avoid contamination
by natural enzymes.
Hydrolysis of monophosphates in nucleotides is one of the
most fundamental reactions in molecular biology and bio-
technology.1 Currently
a natural enzyme, phosphomono-
esterase, is used for the purpose. However, non-enzymatic
catalysts for the removal of terminal monophosphates, if avail-
able, should be useful for various applications.
Recently, remarkable catalyses by lanthanide ions and their
complexes of the hydrolysis of biologically important phospho-
diesters and phosphotriesters were reported.2–7 The cerium()
ion is the most active for the hydrolysis of both DNA 3 and
the 3Ј,5Ј-cyclic monophosphate of adenosine,4 whereas
TmIII, YbIII, and LuIII ions are superb for RNA hydrolysis.5
Lanthanide complexes for phosphotriester hydrolysis were also
elegantly designed.6 However, detailed and systematic studies
on the hydrolysis of phosphomonoesters (especially of 2Ј-
deoxyribonucleosides) by lanthanide ions have been rather less
well documented, although they should be formed as inter-
mediates in the course of these di- and tri-ester hydrolyses.8,9 In
order to accomplish a phosphomonoesterase-like function by
non-enzymatic systems, both the selectivity for the hydrolysis of
phosphomonoesters (with respect to phosphodiester hydrol-
ysis) and the catalytic activities must be sufficiently high. How-
ever, little is known on these subjects.
Hydrolysis of nucleotides
The reactions were carried out at 50 ЊC under air unless
otherwise noted, and were followed by reversed-phase HPLC
[a Merck LiChrosphere RP-18(e) ODS column; water–
acetonitrile = 92:8 or 96:4 (v/v)]. The initial concentration of
the substrate was 0.1 m. At an appropriate interval, 30 µl of
the reaction mixture was removed, and 10 wt% aqueous phos-
phoric acid (3 µl) was added to it. The specimen was treated
with a disposable pretreatment filter (Tosoh; W-3-2), and then
injected onto an HPLC column. The HPLC peaks were
assigned by coinjection with authentic samples.
All the reactions satisfactorily showed pseudo first-order
kinetics. The rate constants presented here are the averages of at
least duplicate runs which coincided with each other within 5%.
The reaction mixtures were homogeneous when the concen-
tration of TRIS buffer was greater than 250 m. Other mix-
tures contained some precipitates probably due to the form-
ation of metal hydroxide.
This paper reports that the CeIV ion is quite active for the
hydrolysis of monophosphates in nucleotides in neutral solu-
tions.10 Furthermore, the selectivity of CeIV for monophosphate
hydrolysis, with respect to the corresponding phosphodiester
hydrolysis, is greatly promoted by using concentrated TRIS and
HEPES buffers as the solvents [TRIS = tris(hydroxymethyl)-
methylamine and HEPES = NЈ-(2-hydroxyethyl)piperazine-N-
ethanesulfonic acid]. The highly selective catalysis is applied
to the removal of the terminal monophosphate from a
dinucleotide.
Kinetic analysis of d(pApA) hydrolysis
These reactions were analysed according to Scheme 1. The
k1
d(pApA)
d(ApA)
k2
fast
Experimental
d(pAp) + pdA
+ dA
2dA(+ 2p)
Materials
2Ј-Deoxyadenylyl (3Ј→5Ј)-2Ј-deoxyadenoside [d(ApA)], as well
as monophosphates of various ribonucleosides and 2Ј-deoxy-
ribonucleosides, was obtained from Sigma. The 5Ј-mono-
phosphate of d(ApA) [d(pApA)] was prepared from d(ApA)
Scheme 1
hydrolysis of the terminal monophosphate in d(pApA) (the rate
constant k1) gives d(ApA), whereas either [d(pAp) ϩ pdA ϩ
J. Chem. Soc., Perkin Trans. 2, 1997
1685

View Article Online
Table 1 Pseudo first-order rate constants for the hydrolysis of dAp,
pdA and d(ApA) at pH 7.2 (50 m HEPES buffer) and 50 ЊC a
Table 3 Rate constants for the hydrolysis of the terminal phospho-
monoester and the internal phosphodiester in d(pApA) by CeCl3 (10
m) at pH 8.0 and 50 ЊC.
k/10Ϫ2 minϪ1
Buffer
ka /10Ϫ2 minϪ1
Terminal k1
Catalyst
pdA
dAp
d(ApA)
Concentration/
m
b
CeCl3 ϩ O2
Ce(NH4)2(NO3)6
6.7
6.9
6.2
6.5
0.20
0.27
Kind
Internal k2
k1/k2
TRIS
50
500
50c
500
15.9
(4.1)
2.9
(2.0)
5.6
(6.7)
4.0
(3.2)
3.0
(0.048)
0.20
(0.004)
0.70
(0.20)
0.52
5.3
(85)
15
a
[CeIV salt] = 10 m.
Table 2 Rate constants for the hydrolysis of nucleotides at pH 7.2 (50
m HEPES buffer) and 50 ЊC by CeCl3 (10 m)
(500)
8.0
(34)
7.7
(97)
HEPES
Substrate
k/10Ϫ2 minϪ1
Substrate
k/10Ϫ2 minϪ1
(0.033)
dAp
dGp
dCp
dTp
6.2
2.4
5.2
4.5
pdA
pdG
pdC
pdT
6.7
2.8
7.3
7.4
a
The numbers in parentheses are the rate constants for the hydrolysis of
pdA and d(ApA) [kpdA in the column of k1 and kd(ApA) in the column of
k2]. b The kpdA/kd(ApA) ratios are presented in parentheses. c pH 7.2.
dA] or two dpAs should be produced when the internal
phosphodiester linkage is hydrolysed (the rate constant k2).11
The amounts of d(pApA), d(ApA) and dA during the reactions
were determined precisely by HPLC. Thus, the first-order rate
constant kd for the disappearance of d(pApA), which is equal to
the sum of k1 and k2, was determined from the corresponding
time-course. The rate constant k1 was evaluated from the time-
course of d(ApA). Finally, k2 was obtained by subtracting k1
from kd. The d(pAp) and pdA, formed in the k2 step, were
rapidly converted to the final product dA, and not much
accumulated in the mixtures.
In the present analyses, the process [d(pApA) → d(ApA)
→ 2dA] was not taken into consideration, since the hydrolysis
of d(ApA) to two dAs is much slower than the other reactions
(see Table 1) and thus the process did not contribute much to
the distribution of the products.
Titration of CeIV in the reaction mixtures
The amounts of CeIV, formed in situ from CeCl3 in the reaction
mixtures, were determined by a back-titration.3f Known
amounts of Ce(NH4)4(SO4)4 were added to the mixtures, and
then the whole solutions were titrated with FeSO4 using 1,10-
phenanthroline as an indicator.
Fig. 1 Time-courses for the hydrolysis of d(pApA) by CeCl3 (10 m)
under air at 50 ЊC and pH 8.0: (a) in 50 m TRIS buffer, (b) in 500 m
TRIS buffer, and (c) in 500 m HEPES buffer. ᭺, d(pApA); ᭹,
d(ApA); ᭾ dA. The solid lines are the theoretical ones calculated by use
of Scheme 1 (see Experimental). The attempts to calculate the theor-
etical lines for dA were not successful, since dA is generated via too
many pathways.
Results and discussion
Hydrolysis of nucleoside monophosphates by CeIV ions
The 3Ј- and 5Ј-monophosphates of 2Ј-deoxyadenosine (dAp
and pdA) were rapidly and quantitatively converted to 2Ј-
deoxyadenosine (dA), when they were treated with CeCl3 under
air at pH 7.2 and 50 ЊC (Table 1). The reactions were complete
within 1 h (the half-life is around 10 min). The rate constants
for the hydrolysis of dAp and pdA (kdAp and kpdA) are virtually
identical. The scission is hydrolytic, as confirmed by the absence
of by-products. Adenine should be released if the ribose was
cleaved by an oxidative pathway. Monophosphates of other
deoxyribonucleosides (as well as monophosphates of ribo-
nucleosides) were also hydrolysed (Table 2). The hydrolysis rate
is not significantly dependent on the kind of nucleotide, indicat-
ing that the catalytic species does not interact with the nucleic
acid base in the catalysis.
Removal of terminal monophosphate from d(pApA)
The terminal phosphate of d(pApA) was rapidly removed by
CeCl3 at pH 8.0 and 50 ЊC in 50 m TRIS buffer [Fig. 1(a)].
About 73 mol% of d(pApA) was converted to d(ApA) in only
10 min, during which dA was formed in 17 mol%. The selectiv-
ity for the formation of d(ApA) (monophosphate hydrolysis/
total hydrolysis) is 80% at the completion of the reaction. Note
that two moles of dA are formed when the internal phospho-
diester linkage in d(pApA) is hydrolysed. As described above,
most of dA in the present reactions is formed by the hydrolysis
of the internal phosphodiester linkage. By fitting these time-
courses to Scheme 1 (see Experimental section), the rate con-
stants k1 [for the hydrolysis of the terminal monophosphate in
d(pApA)] and k2 (for the hydrolysis of the internal phospho-
diester linkage) are determined to be 15.9 × 10Ϫ2 and 3.0 × 10Ϫ2
minϪ1, respectively.
In addition to CeCl3, CeIV(NH4)2(NO3)6 is also active in the
hydrolysis of monophosphates (Table 1). In the absence of
these cerium salts, however, no measurable hydrolysis of the
monophosphates occurred (the intrinsic half-lives of the
monophosphates are estimated to be 160 000 years).12 The
acceleration achieved here is nearly 1010-fold.
The hydrolysis of d(ApA) by either CeCl3 or Ce(NH4)2-
(NO3)6 is 30–130-fold slower than is the hydrolysis of dAp and
pdA. Thus, phosphomonoesters are hydrolysed by these metal
salts far more rapidly than are phosphodiesters.
Selective removal of the terminal monophosphate was also
achieved in 50 m HEPES buffer. The selectivity for the form-
ation of d(ApA) is 82% at a conversion of 98 mol% (see Table
3).
Active species for the phosphomonoester hydrolysis by CeCl3
The monophosphate hydrolysis by CeCl3 is ascribed to the
1686
J. Chem. Soc., Perkin Trans. 2, 1997

View Article Online
Table 4 Effects of buffer concentration on the rate constants for the
hydrolysis of pdA, dAp and d(ApA) by CeCl3 (10 m) at pH 8.0 and
50 ЊC
O
P
–O
O
A
O
O–
Buffer
ka /10Ϫ2 minϪ1
HO
Ce
O
P
O
Concentration/
O
O–
O
Kind
m
pdA
dAp
d(ApA)
Ce
O
A
OH
O
TRIS
50
250
500
50
4.1 (85)
3.5 (120)
2.0 (500)
7.6 (30)
6.7 (34)
3.6 (52)
3.2 (97)
4.3 (90)
4.2 (140)
2.3 (580)
8.0 (32)
6.2 (31)
3.6 (52)
3.5 (110)
0.048
0.029
0.004
0.25
0.20
0.069
0.033
OH
HEPES
50b
250
500
Fig. 2 Proposed mechanism for the hydrolysis of d(pApA). The mono-
phosphate is hydrolysed in preference to the internal phosphodiester.17
Improvement of selectivity for removal of terminal mono-
phosphate from d(pApA) using concentrated TRIS buffer
solutions
a
The numbers in parentheses are the ratios of the rate constant for
monophosphate hydrolysis to that for d(ApA) hydrolysis [kpdA/kd(ApA)
and kdAp/kd(ApA)]. b pH 7.2.
The time-course for the hydrolysis of d(pApA) in 500 m TRIS
buffer is presented in Fig. 1 (b). The selectivity for the formation
of d(ApA) is notably improved, compared with that in 50 m
TRIS buffer [Fig. 1(a)]. Almost 90% of d(pApA) is converted to
d(ApA) at 120 min, where only 7% of the starting material is
hydrolysed to dA via the hydrolysis of the internal phospho-
diester. The selectivity for the formation of d(ApA) is 93%.
With the use of 500 m HEPES buffer, the selectivity was 81%
at the conversion 98 mol% [Fig. 1 (c)], and was not much
dependent on the buffer concentration.
These time-courses are analysed according to Scheme 1. The
k1/k2 ratio in 500 m TRIS buffer is 15, which is more than
3-fold greater than the value (5.3 = 15.9 × 10Ϫ2/3.0 × 10Ϫ2) in 50
m TRIS buffer. The k1/k2 in 500 m HEPES buffer is 7.7
(Table 3).
catalysis by the CeIV ion, which is formed in situ from CeCl3
and molecular oxygen,3d–f based on the following results. (1)
The monophosphate hydrolysis did not take place at all when
molecular oxygen was removed from the mixture by repeated
freeze–thaw cycles (no CeIV was formed in the reaction mixtures
here). (2) About 50% of the CeIII ion was oxidized to CeIV in
the reaction mixtures, according to a titration. (3) The
hydrolysis by CeIV(NH4)2(NO3)6 took place efficiently even in
the absence of molecular oxygen. Any reaction mechanism in
which the catalytic species are derived from molecular oxygen is
rather unlikely.
All these kinetic features are essentially identical with those
reported previously for CeIV-induced DNA hydrolysis.3f This
strongly indicates that the present phosphomonoester
hydrolysis proceeds via the nucleophilic attack by CeIV-bound
hydroxide ion towards the phosphorus atom, as does the DNA
hydrolysis.13,14
Simultaneous coordination of terminal phosphomonoester and
internal phosphodiester to CeIV for d(pApA) hydrolysis
As shown in Table 3, the k2 values [for the hydrolysis of the
internal phosphate in d(pApA)] in 50 and 500 m TRIS buffers
are 63- and 50-fold greater than the corresponding rate con-
stants kd(ApA) for the hydrolysis of d(ApA), which has an anal-
ogous internal phosphodiester linkage. Similarly, the k1 values
(for the hydrolysis of the external phosphate) also exceed the
rate constants (kpdA) for the hydrolysis of monophosphate
in pdA. However, the k1/kpdA ratios (3.9 and 1.5, respectively,
in 50 and 500 m TRIS buffers) are not so great as the
k2/kd(ApA) ratios presented above. Thus, the k1/k2 ratios for the
hydrolysis of d(pApA) in TRIS buffers (5.3 and 15 in 50 and
500 m buffers) are considerably smaller than the kpdA/kd(ApA)
ratios (85 and 500: the numbers in parentheses), where large k2
values are mostly responsible for these facts. In HEPES buffer,
the k1/k2 ratios are also smaller than kpdA/kd(ApA), as shown in
Table 3.
This indicates that coordination of phosphodiester linkage in
d(pApA) toward the CeIV ion is promoted by the terminal
monophosphate. One of the most plausible mechanisms is
shown schematically in Fig. 2 (the CeIV ions should at least
partially form hydroxide clusters under the reaction con-
ditions). Both the terminal monophosphate and the internal
phosphodiester linkage are simultaneously coordinated to the
CeIV ions.17 With the aid of the monophosphate as a strong
ligand, the weakly coordinating diester linkage can efficiently
form a complex with CeIV and be hydrolysed. Otherwise, the
catalysis by CeIV on the diester hydrolysis should be predomin-
antly suppressed (or diminished) due to the competitive inhib-
ition by the buffer agents. The increase in the monoester/diester
selectivity with increasing concentration of the buffers (Table 4),
which mostly originates from the decrease in k2, is considerably
cancelled out by this effect. In contrast, the coordination of
terminal monophosphate (and thus its hydrolysis rate) is
affected to a lesser extent by the simultaneous coordination
with the adjacent phosphodiester.
Promotion of monoester/diester selectivity using concentrated
buffer solutions
Interestingly, the selectivity for the phosphomonoester
hydrolysis, with respect to the phosphodiester hydrolysis, not-
ably increases with an increase in the concentration of buffer
agent (Table 4). In 500 m TRIS buffer, for example, pdA
and dAp are hydrolysed 500- and 580-fold faster than is d(ApA)
(the numbers in parentheses). The corresponding ratios in 50
m TRIS buffer are only 85 and 90. In HEPES buffers, the
monoester/diester selectivity also monotonically increases with
increasing buffer concentration. The kpdA/kd(ApA) and kdAp
kd(ApA) ratios in 500 m HEPES buffer are 97 and 110,
respectively.
/
The increase in the monoester/diester selectivity with increas-
ing concentration of buffers is mostly ascribed to buffer-
induced suppression of phosphodiester hydrolysis. As shown in
Table 4, the rate constants for d(ApA) hydrolysis are decreased
by 12- and 8-fold respectively, when the concentrations of TRIS
and HEPES are increased from 50 to 500 m. However, the
values for the hydrolysis of pdA and dAp are decreased by only
1.9–2.3 fold. Presumably, the buffer agents function as compet-
itive inhibitors of the coordination to the CeIV ions. As a
result, the coordination of phosphodiester linkages to CeIV is
drastically suppressed, since they are monoanionic15 and
rather poor as ligands. On the other hand, the coordination
of monophosphates to CeIV is affected to a smaller degree
(they are dianionic in the mixtures15 and are better ligands).16
Thus, the suppression effect by the buffers is much more sig-
nificant in the diester hydrolysis. The proposed interactions
between CeIV ions and TRIS are supported by the fact that
the metal ions, which easily form metal hydroxide gels above
pH 7, are satisfactorily solubilized in concentrated (> 250 m)
TRIS buffers.
J. Chem. Soc., Perkin Trans. 2, 1997
1687

View Article Online
J. R. Morrow, L. A. Buttrey, V. M. Shelton and K. A. Berback,
J. Am. Chem. Soc., 1992, 114, 1903.
The greater monoester/diester selectivity for TRIS buffers
than for HEPES buffers is probably associated with stronger
binding of TRIS to CeIV. Consistently, the k1/k2 ratio is not
much changed when the HEPES concentration is increased
from 50 to 500 m (Table 3). Here, the ‘k2 suppression effect’
and ‘simultaneously coordinating effect’ compensate each
other.
A similar coordination was previously proposed on metal
ion-mediated RNA hydrolysis.18 In the proposal, a metal ion is
simultaneously bound to both the terminal phosphomonoester
and the internal phosphodiester, resulting in the increase of its
local concentration at the phosphodiester linkage. Thus, the
monophosphate residues accelerate the hydrolysis of the
adjacent phosphodiester linkage, exactly as observed here in
the CeIV-induced DNA hydrolysis.
6 For triester hydrolysis: (a) R. W. Hay and N. Govan, J. Chem. Soc.,
Chem. Commun., 1990, 714; (b) S. J. Oh, C. W. Yoon and J. W. Park,
J. Chem. Soc., Perkin Trans. 2, 1996, 329.
7 For the hydrolysis of other phosphate esters: (a) M. Matsumura
and M. Komiyama, J. Inorg. Biochem., 1994, 55, 153; (b) B. K.
Takasaki and J. Chin, J. Am. Chem. Soc., 1995, 117, 8582; (c) Ref. 2.
8 Hydrolysis of ribonucleoside 2Ј- and 3Ј-monophosphates by
lanthanide ions has been examined extensively: S. Kuusela and
H. Lönnberg, J. Phys. Org. Chem., 1993, 6, 347.
9 Monophosphate hydrolysis by metal ions other than lanthanide ions
has been studied: (a) J. M. Harrowfield, D. R. Jones, L. F. Lindoy
and A. M. Sargeson, J. Am. Chem. Soc., 1980, 102, 7733; (b)
D. H. Vance and A. W. Czarnik, J. Am. Chem. Soc., 1993, 115,
12 165; (c) J. S. Seo, N.-D. Sung, R. C. Hynes and J. Chin, Inorg.
Chem., 1996, 35, 7472 and references cited therein.
10 Pioneering works on the catalysis of lanthanide ions for the
hydrolysis of phosphomonoesters were carried out about 40 years
ago: (a) E. Bamann, H. Trapmann and F. Fischler, Biochem. Z.,
1953, 32, 89; (b) M. Shimomura and F. Egami, Bull. Chem. Soc.
Jpn., 1953, 26, 263. However, these studies were made only in
alkaline solutions, and thus the hydrolyses reported were much less
efficient than those presented in the present paper. The catalysis by
lanthanide ions is notably suppressed at a higher pH than 8,
probably due to formation of the precipitates of metal hydroxide.
11 The ratio of the products in the k2 process depends on which of the
two P᎐O bonds in the internal phosphodiester linkage is cleaved. In
the hydrolysis of d(ApA), two P᎐O linkages are cleaved by CeIV
at almost the same rates, providing dAp and pdA in nearly 1:1 ratio
(ref. 3f ). These monophosphates are promptly hydrolysed to dA.
12 The rate of the hydrolysis of methyl phosphate at pH 7 and 25 ЊC is
5 × 10Ϫ10 hϪ1: J. P. Guthrie, J. Am. Chem. Soc., 1977, 99, 3991.
13 The water molecule(s) bound to CeIV should assist the reaction as
a general acid catalyst, and the negatively charged transition state
for monophosphate hydrolysis is stabilized by the positive charge on
Activities of other lanthanide metal ions and non-lanthanide ions
Other lanthanide() ions (La, Pr, Nd, Tb, Dy) also hydrolyse
monophosphates of nucleotides. However, their activities are
rather small. Even NdIII, which is the most active among them,
is about 100-fold less active than CeIV. All the other lanthanide
ions as well as non-lanthanide ions investigated are inactive.
The activity of CeIV is overwhelmingly great.
In summary, CeIV ions efficiently hydrolyse phospho-
monoesters in nucleotides. The terminal monophosphates are
preferentially hydrolysed over the internal phosphodiester link-
ages, especially when concentrated TRIS buffers are used as the
media. The present results indicate the potential of CeIV to be
used as the catalytic center of artificial phosphomonoesterases.
The selectivity for monophosphate hydrolysis could be pro-
moted still more by designing a complex in which the phos-
phomonoester cannot significantly assist the binding of the
metal ion to the adjacent phosphodiester linkage.
the CeIV
.
14 When hydrogen peroxide was added to the reaction mixtures, the
substrate was gradually decomposed (the total peak area in HPLC
decreased) and varieties of unassignable products were formed.
Apparently, some radical species were formed there. The possibility
of the participation of some peroxide-like species in the catalysis,
proposed in ref. 3d for the DNA hydrolysis by CeCl3–O2 combin-
ation, is unlikely (see also refs. 3e and 3f ).
15 The pKa values of adenosine 5Ј-phosphate are 3.80 and 6.19, whereas
the values of the 3Ј-phosphate are 3.65 and 5.83: Kagaku Binran
Kiso-Hen, ed. by Chemical Society of Japan, Maruzen, 1984.
16 The binding constants for the complex formation of lanthanide()
ions with adenosine monophosphate are around 104-fold greater
than the corresponding values with dinucleotide: the former values
are from Stability Constants Supplement No. 1, Special Publication
25, The Chemical Society, London, 1971, whereas the latter values
are from M. Komiyama, N. Takeda, Y. Takahashi, Y. Matsumoto
and M. Yashiro, Nucleosides, Nucleotides, 1994, 13, 1297.
17 In Fig. 2, each of the monophosphate and the phosphodiester res-
idues is coordinated to one of the CeIV ions in the metal hydroxide
cluster. Alternatively, both of them can be coordinated to one
CeIV ion.
18 (a) J. J. Butzow and G. L. Eichhorn, Biochemistry, 1971, 10, 2019; (b)
H. Ikenaga and Y. Inoue, Biochemistry, 1974, 13, 577; (c) A. George,
P. Draganac and W. R. Farkas, Inorg. Chem., 1985, 24, 3627;
(d ) V. M. Shelton and J. R. Morrow, Inorg. Chem., 1991, 30, 4295;
(e) J. R. Morrow and V. M. Shelton, New J. Chem., 1994, 18, 371;
( f ) S. Kuusela, A. Azhayev, A. Guzaev and H. Lönnberg, J. Chem.
Soc., Perkin Trans. 2, 1995, 1197; (g) S. Kuusela, A. Guzaev and
H. Lönnberg, J. Chem. Soc., Perkin Trans. 2, 1996, 1895.
Acknowledgements
This work was partially supported by a Grant-in-Aid for Scien-
tific Research from The Ministry of Education, Science, and
Culture, Japan.
References
1 R. L. P. Adams, J. T. Knowler and D. P. Leader, The Biochemistry of
Nucleic Acids, 10th edn., Chapman and Hall, London, 1986.
2 Reviews: (a) M. Komiyama, J. Biochem., 1995, 118, 665; (b)
R. Breslow, Acc. Chem. Res., 1995, 28, 146.
3 (a) Y. Matsumoto and M. Komiyama, Nucleic Acids, Symp. Ser.,
1992, 27, 33; (b) M. Komiyama, K. Matsumura, K. Yonezawa
and Y. Matsumoto, Chem. Express, 1993, 8, 85; (c) T. Shiiba,
K. Yonezawa, N. Takeda, Y. Matsumoto, M. Yashiro and
M. Komiyama, J. Mol. Catal., 1993, 84, L21; (d ) B. K. Takasaki and
J. Chin, J. Am. Chem. Soc., 1994, 116, 1121; (e) M. Komiyama,
T. Shiiba, T. Kodama, N. Takeda, J. Sumaoka and M. Yashiro,
Chem. Lett., 1994, 1025; ( f ) M. Komiyama, N. Takeda,
Y. Takahashi, H. Uchida, T. Shiiba, T. Kodama and M. Yashiro,
J. Chem. Soc., Perkin Trans. 2, 1995, 269; (g) S. Hashimoto and
Y. Nakamura, J. Chem. Soc., Chem. Commun., 1995, 1413;
(h) J. Rammo, R. Hettich, A Roigk and H.-J. Schneider, J. Chem.
Soc., Chem. Commun., 1996, 105 and references cited therein.
4 J. Sumaoka, M. Yashiro and M. Komiyama, J. Chem Soc., Chem.
Commun., 1994, 1755.
Paper 7/01481C
Received 3rd March 1997
Accepted 27th May 1997
5 (a) J. Ciesiolka, T. Marciniec and W. Krzyzosiak, Eur. J. Biochem.,
1989, 182, 445; (b) M. Komiyama, K. Matsumura and
Y. Matsumoto, J. Chem. Soc., Chem. Commun., 1992, 640; (c)
1688
J. Chem. Soc., Perkin Trans. 2, 1997