Kinetic Analysis of Diamine-Catalyzed RNA Hydrolysis

Article abstract of DOI:10.1021/jo961935u

The catalysis of various amines for the hydrolysis of RNA has been kinetically investigated, and the catalytic rate constants for each of the ionic states of these amines are determined. Ethylenediamine and 1,3-propanediamine are highly active under the physiological conditions, mainly because they preferentially take the catalytically active monocationic forms. The catalysis of these diamines is further promoted by the intramolecular acid-base cooperation of the neutral amine and the ammonium ion. In contrast, monoamines overwhelmingly exist at pH 7 as the inactive cations. Potential application of the catalysis by the diamines and the related oligoamines is discussed.

Full text of DOI:10.1021/jo961935u

J . Org. Chem. 1997, 62, 2155-2160
2155
Kin etic An a lysis of Dia m in e-Ca ta lyzed RNA Hyd r olysis
Makoto Komiyama* and Koichi Yoshinari†
Department of Chemistry and Biotechnology, Graduate School of Engineering, University of Tokyo,
Hongo, Tokyo 113, J apan
Received October 15, 1996X
The catalysis of various amines for the hydrolysis of RNA has been kinetically investigated, and
the catalytic rate constants for each of the ionic states of these amines are determined.
Ethylenediamine and 1,3-propanediamine are highly active under the physiological conditions,
mainly because they preferentially take the catalytically active monocationic forms. The catalysis
of these diamines is further promoted by the intramolecular acid-base cooperation of the neutral
amine and the ammonium ion. In contrast, monoamines overwhelmingly exist at pH 7 as the
inactive cations. Potential application of the catalysis by the diamines and the related oligoamines
is discussed.
In tr od u ction
DNA oligomers.6e,f The catalyses of the amines are
certainly relevant to biological applications.
Recently much interest has been focusing onto non-
1
This paper describes a full account on the RNA
hydrolysis by diamines and oligoamines. The catalyti-
cally active species are unambiguously characterized, and
the catalytic rate constant for each of the species is
determined. Dependencies of the catalytic activity on the
structure and the basicity of amine are quantitatively
clarified. Furthermore, catalytic turnover of the amine
is evidenced. The reaction mechanism is proposed in
terms of these kinetic results.
enzymatic hydrolysis of RNA, and varieties of organic
and inorganic catalysts were proposed.2
-5
One of the
goals of these studies is the preparation of sequence-
selective artificial ribonucleases, which cut a specific RNA
at the target site and thus are applicable to molecular
biology, therapy, and others.1
c,d
Conjugation of these
catalysts with sequence-recognizing moieties was already
reported.1
c,d,6
Undoubtedly, molecular design of eminent
catalysts for RNA hydrolysis is one of the most important
keys for further development of the field.
In a preliminary communication,2b we reported that
ethylenediamine, diethylenetriamine, and triethylene-
tetraamine are effective for RNA hydrolysis due to
intramolecular cooperation of two amino residues. Fur-
thermore, sequence-selective RNA scission was accom-
plished by the attachment of oligoamines to synthetic
Resu lts
Ca ta lysis of Dia m in es a n d Oligoa m in es for RNA
Hyd r olysis. Table 1 shows the catalytic activities of
various amines for the hydrolysis of adenylyl(3′-5′)-
adenosine (ApA) at pH 8 and 50 °C. Clearly, the
diamines are far more active than the monoamines: the
catalytic turnover of the amines is confirmed later. The
activity of ethylenediamine (N-2-N) is 13 times as great
as that of ethylamine, while N,N′-dimethylethylenedi-
amine is almost 60 fold more active than N-ethylmethyl-
amine. Two amino residues are required for the efficient
RNA hydrolysis (the differences in the activity between
the diamines and the monoamines are still more remark-
able at pH 7 as described below). In a series of unsub-
stituted alkylenediamines, the activity is in the following
order: N-2-N > N-3-N > N-1-N > N-4-N > N-5-N. 1,2-
Cyclohexanediamine is more active than the 1,3-coun-
terpart. In the absence of the amines, however, the
hydrolysis is virtually nil (the rate constant is evaluated
†
Current address: National Institute of Bioscience & Human-
Technology, MITI, Tsukuba Science City 305, J apan.
X
Abstract published in Advance ACS Abstracts, March 1, 1997.
(1) (a) Breslow, R. Acc. Chem. Res. 1995, 28, 146. (b) Kosonen, M.;
L o¨ nnberg, H. J . Chem. Soc., Perkin Trans. 2 1995, 1203. (c) Mes-
maeker, A. D.; H a¨ ner, R.; Martin, P.; Moser, H. Acc. Chem. Res. 1995,
2
8, 366. (d) Komiyama, M. J . Biochem. 1995, 118, 665 and references
therein. (e) Pyle, A. M.; Barton, J . K. Prog. Inorg. Chem. (Bioinorganic
Chemistry) 1990, 38, 414.
(
2) (a) Barbier, B.; Brack, A. J . Am. Chem. Soc. 1988, 110, 6880. (b)
Yoshinari, K.; Yamazaki, K.; Komiyama, M. J . Am. Chem. Soc. 1991,
13, 5899.
3) (a) Eichhorn, G. L. Inorganic Biochemistry; Eichhorn, G. L., Ed.;
Elsevier Scientific Publishing Co.: Amsterdam, 1973; Vol. 2, Chap.
1
(
3
1
4. (b) Matsumoto, Y.; Komiyama, M. J . Chem. Soc., Chem. Commun.
990, 1050. (c) Stern, M. K.; Bashkin, J . K.; Sall, E. D. J . Am. Chem.
Soc. 1990, 112, 5357. (d) Komiyama, M.; Matsumura, K.; Matsumoto,
Y. J . Chem. Soc., Chem. Commun. 1992, 640. (e) Morrow, J . R.;
Buttrey, L. A.; Shelton, V. M.; Berback, K. A. J . Am. Chem. Soc. 1992,
-
7
-1
to be 1 × 10 min from the pH-rate constant profile
1
14, 1903. (f) Irisawa, M.; Komiyama, M. J . Biochem. 1995, 117, 465.
which is a straight line of slope 1.0 from pH 9.5-13).
(
g) Irisawa, M.; Takeda, N.; Komiyama, M. J . Chem. Soc., Chem.
Varieties of oligoamines also effectively catalyzed ApA
hydrolysis. The rate constants at pH 8 and 50 °C
Commun. 1995, 1221. (h) Yashiro, M.; Ishikubo, A.; Komiyama, M. J .
Chem. Soc., Chem. Commun. 1995, 1793. (i) Sumaoka, J .; Uchida, H.;
Komiyama, M. Nucleic Acids, Symp. Ser. 1995, 34, 83.
-5
-1
([amine]
0
) 1.0 M) are 3.6 × 10 min for diethylene-
(4) For the hydrolysis of model phosphoesters, (a) Komiyama, M.;
-5 -1
Yoshinari, K. J . Chem. Soc., Chem. Commun. 1989, 1880. (b) Wall,
M.; Hynes, R. C.; Chin, J . Angew. Chem. Int. Ed. Engl. 1993, 32, 1633.
triamine, 4.6 × 10 min for triethylenetetraamine, and
(
3
c) Smith, J .; Ariga, K.; Anslyn, E. V. J . Am. Chem. Soc. 1993, 115,
62. (d) Tsuboi, A.; Bruice, T. C. J . Am. Chem. Soc. 1994, 116, 11614
and references therein.
5) For nonenzymatic hydrolysis of DNA: (a) Basile, L. A.; Raphael,
(6) (a) Zuckermann, R. N.; Schultz, P. G. J . Am. Chem. Soc. 1988,
110, 6592. (b) Matsumura, K.; Endo, M.; Komiyama, M. J . Chem. Soc.,
Chem. Commun. 1994, 2019. (c) Bashkin, J . K.; Frolova, E. I.; Sampath,
U. J . Am. Chem. Soc. 1994, 116, 5981. (d) Magda, D.; Miller, R. A.;
Sessler, J . L.; Iverson, B. L. J . Am. Chem. Soc. 1994, 116, 7439. (e)
Komiyama, M.; Inokawa, T. J . Biochemistry 1994, 116, 719. (f)
Komiyama, M.; Inokawa, T.; Yoshinari, K. J . Chem. Soc., Chem.
Commun. 1995, 77. (g) Reynolds, M. A.; Beck, T. A.; Say, P. B.;
Schwartz, D. A.; Dwyer, B. P.; Daily, W. J .; Vaghefi, M. M.; Metzler,
M. D.; Klem, R. E.; Arnold, L. J ., J r. Nucleic Acids Res. 1996, 24, 760.
(
A. L.; Barton, J . K. J . Am. Chem. Soc. 1987, 109, 7550. (b) Matsumoto,
Y.; Komiyama, M. Nucleic Acids, Symp. Ser. 1992, 27, 33. (c) Ko-
miyama, M.; Takeda, N.; Uchida, H.; Takahashi, Y.; Shiiba, T.;
Kodama, T.; Yashiro, M. J . Chem. Soc., Perkin Trans. 2 1995, 269. (d)
Takasaki, B. K.; Chin, J . J . Am. Chem. Soc. 1995, 116, 1121. (e)
Schnaith, L. M. T.; Hanson, R. S.; Que, L., J r. Proc. Natl. Acad. Sci.
U.S.A. 1994, 91, 569.
S0022-3263(96)01935-4 CCC: $14.00 © 1997 American Chemical Society

2
156 J . Org. Chem., Vol. 62, No. 7, 1997
Komiyama and Yoshinari
Ta ble 1. P seu d o-F ir st-Or d er Ra te Con sta n ts (in 10- m in -1) for th e Ap A Hyd r olysis in th e P r esen ce of Am in es a t p H 8
5
a n d 50 °Ca
catalyst
rate constant
catalyst
rate constant
Monoamines
ethylamine
diethylamine
N-ethylmethylamine
0.57
0.22
0.21
N,N,N′,N′-tetramethylethylenediamine
1,3-diaminopropane (N-3-N)
1.7
3.8
Diamines
diaminomethane (N-1-N)
ethylenediamine (N-2-N)
N-methylethylenediamine
N,N′-dimethylethylenediamine
2.6
7.4
7.9
1,4-diaminobutane (N-4-N)
1,5-diaminopentane (N-5-N)
1,2-cyclohexanediamine
1,3-cyclohexanediamine
none
2.0
1.6
6.8
3.9
12.9
b
0.01
a
Amine]0 ) 1.0 M. b Determined from the pH-rate constant profile which is a straight line between pH 9.5 and 13.
[
.0 × 10 min-1 for N,N′-bis(aminoethyl)-1,3-diamino-
-5
5
propane. Furthermore, poly[A] and poly[U] (as well as
other diribonucleotides) were hydrolyzed by these di-
amines and oligoamines.
In contrast, the corresponding DNA substrates, 2′-
deoxyadenylyl(3′-5′)2′-deoxyadenosine and poly[dA], were
not hydrolyzed by any of the amines to a measurable
extent. It is strongly indicated that the present RNA
hydrolysis involves an intramolecular attack by the 2′-
hydroxyl group of the ribose.
Totally hydrolytic character of the scission is confirmed
by the absence of byproducts such as adenine (formed
by the oxidative cleavage of the ribose) and the phos-
phoramide (due to the nucleophilic attack by the amine
toward the phosphorus atom) on HPLC and 31P-NMR:
see the Experimental Section. In the hydrolysis of
diribonucleotides, the ratio of the 2′-monophosphate of
ribonucleoside to the 3′-monophosphate in the products
was virtually 1:1, ruling out the possibility of contamina-
tion by ribonuclease (the enzyme if any should produce
the 3′-monophosphate alone).7 Divalent metal ions do
not participate in the catalysis, since EDTA showed no
effect on the hydrolysis rate (note that all the reactions
in this paper were achieved in the presence of 2 mM
EDTA).
F igu r e 1. The pH dependencies of kc,appar (in min- M-1) for
the ApA hydrolysis by N-2-N (a) and N-3-N (b) at 50 °C: the
solid lines are the theoretical ones calculated by use of the
1
a
catalytic rate constants and the pK values in Table 2. The
broken lines show the contribution of each of the species (left:
monocation; right, neutral species).
Active Sp ecies for th e RNA Hyd r olysis. The
pseudo-first-order rate constant of the ApA hydrolysis
(kobsd) by N-2-N and 1,3-propanediamine (N-3-N) at a
constant pH linearly increased with the amine concen-
tration. Thus, the apparent second-order catalytic rate
constant (kc,appar) was determined from the slope of the
straight line. Any intermolecular cooperation of two (or
more) amine molecules is excluded.
All the experimental points in Figure 1, parts a and b,
satisfactorily fit the theoretical lines (the solid ones). The
broken lines show the contributions of each of the
catalytic species. It is unambiguously confirmed that the
profiles cannot be fit in terms of only one catalytic
component.
The pH dependencies of kc,appar are depicted in Figure
1
. The profiles for both N-2-N (a) and N-3-N (b) are
The pH-kc,appar profiles for the diamine-catalyzed hy-
drolysis of adenosine 2′,3′-cyclic monophosphate (A>p),
poly[A], and poly[U] are also bimodal (see the Supporting
Information, Figures 1-5). The rate constants for the
hydrolysis of poly[A] and poly[U] were evaluated in terms
of the decrease of number of the phosphodiester linkages
in these polymers: details are described in the Experi-
mental Section.8
composed of two components, indicating that two ionic
states of these diamines (the neutral species and their
monocations as shown below) are responsible for the
catalysis. Thus, these pH-kc,appar profiles were analyzed
c
in terms of eq 1. Here, k ’s are the partial second-order
catalytic rate constants for each of the neutral species,
the monocations, and the dications.
kc,appar ) k
f
+ k
f +
c,monocation monocation
2
In the catalysis by ethylamine (NH Et), however, the
c,neutral neutral
plot of logarithm of kc,appar vs pH was a straight line of
slope 1.0 up to pH 9.5, with a gradual saturation at the
higher pH. Only the neutral species is catalytically active
k
f
(1)
c,dication dication
The molar fraction f of each of the species in the
reaction mixtures was calculated by using the pK values
(
a 2
the pK of NH Et is 10.4).
a
of the diamines, which were independently measured by
a potentiometric titration under the reaction conditions.
(
8) The rates of hydrolysis of poly[A] and poly[U] almost linearly
increased with [N-2-N] , indicating that the reactions do not involve
0
significant complexation between the RNA and N-2-N. A deviation due
to a polyion complex formation was observed only below pH 7 (see ref
2b).
(
7) Dugas, H.; Penney, C. Bioorganic Chemistry; Springer-Verlag:
New York, 1981.

Kinetic Analysis of Diamine-Catalyzed RNA Hydrolysis
J . Org. Chem., Vol. 62, No. 7, 1997 2157
Ta ble 2. Sta tistica lly Cor r ected Secon d -Or d er Ca ta lytic
-
4
-1
-1
Ra te Con sta n ts (kc/q in 10 m in
M
) of Am in es for
a ,b
RNA Hyd r olysis a t 50 °C
catalytic rate constant
substrate
ApA
catalyst
monocation
neutral
N-1-N
N-2-N
N-3-N
N-4-N
N-5-N
NH2Et
N-2-N
N-3-N
N-2-N
NH2Et
N-2-N
0.0 (97)
0.44 (61)
1.5 (7.4)
1.3 (1.6)
2.1 (0.8)
0.0 (∼100)
22 (61)
0.55 (3.1)
2.9 (0.2)
7.5 (0.01)
8.0 (0.001)
15 (0.0003)
14 (0.04)
45 (0.2)
110 (0.01)
2.6 (0.2)
14 (0.04)
2.1 (0.2)
A>p
5.4 (7.4)
1.4 (61)
poly[A]
0.0 (∼100)
poly[U]
1.7 (61)
a
The numbers in parentheses are the molar fractions f (in %)
of the corresponding species at pH 7. They were evaluated by
using the pKa values of the amines: N-1-N, <3 and 8.5; N-2-N,
6
.8 and 9.4; N-3-N, 8.1 and 9.8; N-4-N, 8.8 and 10.1; N-5-N, 9.1
b
and 10.4; NH2Et, 10.4. The catalytic rate constants of the neutral
diamines are corrected by a statistic factor of 1/2 (see the legend
for Figure 2).
Ta ble 3. Th e Ra te Con sta n ts a t p H 7 a n d 50 °C for th e
Ap A Hyd r olysis by th e Am in es a s w ell a s th e
a ,b
Con tr ibu tion s of Ea ch of th e Ion ic Sp ecies
rate constants (10- min-1
5
)
catalyst
by monocation
by neutral amine
total
N-1-N
N-2-N
N-3-N
N-4-N
N-5-N
NH2Et
0.0
2.7
1.1
0.21
0.17
0.0
0.34
0.12
0.02
0.00
0.00
0.056
0.34
2.8
1.1
0.21
0.17
0.056
F igu r e 2. Statistically corrected log-log plots of the catalytic
-1
-1
rate constant (in min
M ) vs the basicity of the amino
residue for the hydrolysis of (a) ApA and (b) A>p at 50 °C:
the open circles refer to the monocations, and the closed circles
are for the neutral diamines. The kind of amine is designated
by the number of the methylene carbon atoms (4 for N-4-N,
a
Amine]0 ) 1 M. b The rate constants assignable to each of
[
for example). The statistical corrections were made in k and
c
the species (in the two middle columns) were calculated by using
the equation: kobsd ) kc × [amine]0 × (molar fraction of the ionic
species).
pK by using p (the number of equivalent protons which can
be transferred from the protonated form of the amine) and q
a
(
the number of sites which can accept a proton in the neutral
9
form).
Ca ta lytic Ra te Con sta n ts of Neu tr a l Dia m in es
a n d Th eir Mon oca tion s. The catalytic rate constants
for each of the ionic species, determined by fitting the
pH-rate constant profiles to eq 1 (vide ante), are listed
in Table 2. The catalytic rate constants for the hydrolysis
of A>p, poly[A], and poly[U] are also presented. Here,
the values for the neutral diamines are statistically
Dep en d en ce of th e Ca ta lytic Activity on th e Ba si-
city of Am in o Resid u e. The closed circles in Figure
a show the plot of the catalytic rate constant for the
neutral species of the diamines for ApA hydrolysis (pre-
sented in the fourth column in Table 2) vs the pK
for
the first protonation of the diamine (in footnote a in the
table). Here, both k and pK have been statistically cor-
rected (the details are described in the legend for Figure
2
a
corrected by a factor 1/2, since they have two equivalent
c
a
amino residues.9 In all cases, the dicationic diamines are
9
2
). A fairly linear straight line of a slope 0.75 is ob-
totally inactive (kc,dication ) 0). At least one of the two
amino residues must be neutral for the catalysis, indicat-
ing an essential role of a base catalyst for the RNA
hydrolysis.
By using the kinetic parameters for the ionic species
of the amines and their molar fractions in the mixtures,
both of which are shown in Table 2, the rate constants
of ApA hydrolysis at pH 7 by these amines are evaluated
tained. The catalytic activities of these neutral diamines
are governed only by the basicities of amino residues.
The relationship between the catalytic rate constant
of the monocations (in the third column in Table 2) and
the pK for the second protonation of the amine is
a
depicted by the open circles. Apparently, the points for
the monocations of N-2-N and N-3-N (numbered as 2 and
3
) considerably deviate from the straight line (for the
(Table 3). The contribution of each of the species in the
neutral diamine) toward the positive side (5 fold for
N-2-N and 2 fold for N-3-N).10 Thus, the catalytic
activities of the monocations are greater than are ex-
pected simply from the basicities of the neutral amino
residues. Intramolecular cooperation of the protonated
amino residue with the neutral amino residue is inferred.
In contrast, the points for the monocations of N-4-N and
N-5-N deviate from the line toward the negative side.
catalysis is also presented there. Apparently, N-2-N and
N-3-N exceed the other amines in the activity. It is
noteworthy that the catalyses by N-2-N and N-3-N are
mostly performed by their monocationic species. Al-
though the neutral diamines are intrinsically more active
than the corresponding monocations, their concentrations
at pH 7 are too small to make a significant contribution
in the catalysis. The same conclusion is obtained for the
hydrolysis of A>p, poly[A], and poly[U].
(10) The deviations correspond to the differences in the activities
between these monocations and “hypothetical” diamines whose pK ’s
for the first protonation are the same with the second protonation of
N-2-N and N-3-N.
a
(
9) J encks, W. P. Catalysis in Chemistry and Enzymology; Dover
Publications, Inc: New York, 1969; pp 173-175.

2
158 J . Org. Chem., Vol. 62, No. 7, 1997
Komiyama and Yoshinari
F igu r e 4. Time-conversion curve for the hydrolysis of ApA
by N-2-N at pH 8 and 80 °C: [ApA]
0
) 0.27 M and [N-2-N]
0
)
0
.1 M. The open circles show the experimental results, and
the solid line is the theoretical one calculated by use of kobsd
)
-4
-1
1
.9 × 10 min under the assumption that the reaction shows
the pseudo-first-order kinetics until the completion.
Tu r n over of Oligoa m in e a s Ca ta lyst. When the
ApA hydrolysis by N-2-N was achieved under the condi-
tions that [ApA]
hydrolysis with respect to N-2-N exceeded 100 mol %
Figure 4). A satisfactory first-order kinetics was ob-
served, confirming that the rate constant was kept
constant throughout the reaction (the hydrolysis in the
absence of N-2-N is virtually nil). The efficient turnover
of N-2-N as catalyst has been concretely evidenced.
0
> [N-2-N]
0
, the conversion of the
(
F igu r e 3. Polyacrylamide gel electrophoresis patterns for the
3
2
-6
hydrolysis of a 30-mer RNA ( P-labeled at the 3′-end: 10
M) by N-2-N at pH 8 and 50 °C for 4 h. Lane 1, [N-2-N] ) 1
M; lane 2, 0.1 M; lane 3, 0.01 M; lane 4, 0.001 M; lane 5,
control. The T and H lanes are for the digestion by ribo-
nuclease T and the alkaline hydrolysis, respectively. The
0
1
1
sequence of the RNA is presented in the text. All the cleavage
sites are the primary ones.
Discu ssion
A similar Br o¨ nsted type plot for the hydrolysis of A>p
is presented in Figure 2b. In this case, only the point
for the monocation of N-2-N positively deviates from the
straight line (the slope 0.55), which correlates all the
neutral diamine catalysts.11 The monocationic N-3-N
shows a negative deviation from the line.12
D O Solven t Isotop e Effects. Significant D O sol-
2 2
vent isotope effects were observed for the ApA hydrolysis
by the monocation of N-2-N and by its neutral species:
k (in H O)/k (in D O) ) 2.0 and 1.4, respectively. The
c 2 c 2
catalyses by both of the species undoubtedly involve a
proton-transfer in the rate-limiting step.
Scission of RNA Oligom er by N-2-N. Figure 3
depicts the polyacrylamide gel electrophoresis pattern for
the N-2-N-catalyzed hydrolysis of a linear 30-mer RNA
F a ct or s Gover n in g t h e Ca t a lyt ic Act ivit ies of
Am in es. The remarkable catalytic activities of N-2-N
and N-3-N for RNA hydrolysis under the physiological
conditions (Tables 1-3) are primarily ascribed to the
abundance of catalytically active species in the reaction
mixtures. Considerable portions of these diamines (61%
for N-2-N and 7.4% for N-3-N: Table 2) exist at pH 7 as
the catalytic monocations, because of the electrostatic
suppression of the second protonation by the positively
a
charged ammonium ions (the corresponding pK ’s are 6.8
and 8.1, respectively). Thus, N-2-N is the most active of
all the amines. Furthermore, the intramolecular acid-
base cooperation between the neutral amine and the
ammonium cation promotes the catalysis of N-2-N and
N-3-N by 2-5 fold (see Figure 2). This effect compen-
sates the decrease in the intrinsic catalytic activity (a
smaller basicity) of the neutral amino residue, which is
caused by the electrostatic effect described above. As the
(5′-GGA GGU CCU GUG UUC GAU CCA CAG AAU
UCG-3′) at pH 8 and 50 °C. The pyrimidine-pyrimidine
sites are a little bit preferentially cleaved (lane 2: [N-2-
N]
since the scission takes place throughout the RNA chain
when [N-2-N] ) 1 M (lane 1: the secondary scissions
0
) 0.1 M).13 However, the preference is not so strict,
result, the overall catalytic activities () kc,monocation
×
f
monocation) of these monocations are significant.
0
In contrast, monoamines are only poorly active at pH
, since they mostly take the protonated forms which are
are probably involved here). The scission efficiency
decreases, as expected, with decrease in the amine
concentration and is virtually nil at 1 mM (lane 4).
7
catalytically inactive (the fraction of the active species
for NH Et is only 0.04%). The diamines N-4-N and
2
(
11) The intramolecular nucleophilic attack by the 2′-OH in ApA
N-5-N also dominantly exist as the inactive dications. No
activity of the monocation of N-1-N is ascribed to the
hydrolysis is replaced with the intermolecular one by water in A>p
hydrolysis. Thus, the activation entropy term is more critical in A>p
hydrolysis. Only N-2-N, which has a minimal conformational freedom,
shows the intramolecular cooperation there.
marginal basicity of its neutral amino residue (the pK
a
values of the amines are presented in the footnote a for
Table 2).
(12) The negative deviations of these monocations from the straight
lines are probably ascribed to conformational effects. When the
ammonium ions in these molecules electrostatically interact with the
negatively charged phosphates of RNA, their neutral amino residues,
which function as the general base catalysts, cannot be placed at the
most appropriate position for the catalysis.
The catalytic activity of the monocation of N-2-N
slightly increases when the nitrogen atoms are methyl-
ated: N-methylethylenediamine and N,N′-dimethyl-
ethylenediamine are 1.1 and 1.7 times as active as the
(
13) The preference might be ascribed to smaller steric hindrance.

Kinetic Analysis of Diamine-Catalyzed RNA Hydrolysis
J . Org. Chem., Vol. 62, No. 7, 1997 2159
mational freedom of the molecules.12 Direct nucleophilic
attack by the neutral amino residue toward the phos-
phorus atom is ruled out by (1) a considerable D O
2
solvent isotope effect, (2) no detection of the phosphor-
amide intermediate, and (3) no measurable catalysis of
the amines for DNA hydrolysis. The argument is further
supported by the fact that the 1:1 combination of (2-
2
aminoethyl)trimethylammonium ion and NH Et showed
no enhancement in the catalytic activity due to mutual
cooperation. Apparently, the acid-base cooperation must
be an intramolecular process.
The hydrolysis of ApA and A>p by the neutral di-
amines, which is dominant in highly alkaline solutions,
proceeds via the general base catalysis by one of the two
amino residues. The slopes of the Br o¨ nsted type plots
in Figure 2 (0.75 for ApA hydrolysis and 0.55 for A>p
F igu r e 5. Proposed mechanism for RNA hydrolysis by di-
amines.
1
6
hydrolysis) are consistent with the mechanism.
parent N-2-N (Table 1).14 Thus, versatile moieties can
Con clu sion
be attached to N-2-N without loss of its catalytic activity.
Ca ta lytic Mech a n ism of RNA Hyd r olysis by Di-
a m in es. The proposed mechanism for the catalysis by
the monocations of N-2-N and N-3-N in RNA hydrolysis
is depicted in Figure 5. The formation of 2′,3′-cyclic
monophosphate of the terminal ribonucleotide is rate-
limiting in the overall reaction.7 The neutral amino
residue pulls a proton from the 2′-hydroxyl group of the
ribose to promote its attack toward the phosphorus atom.
The acid catalysis by the protonated amino residue in
the same molecule assists the reaction. The cooperation
of the two types of catalyses can proceed either simul-
taneously or in a stepwise manner. The resultant 2′,3′-
cyclic monophosphate intermediates are hydrolyzed by
a similar cooperation.11 The mechanism is in accord with
the acid-base cooperation of two imidazoles in ribo-
Diamines and oligoamines, which involve two (or more)
amino residues bound by (CH or (CH chains, ef-
2
)
2
2 3
)
ficiently hydrolyze RNA under the physiological condi-
tions. These amines are repeatedly used as catalysts
without deterioration. They are simple in structures and
are chemically stable so that a wide range of practical
applications are promising.
It is noteworthy that the oligoamines have been
successfully used as the catalytic sites for sequence-
6e,f
specific artificial ribonucleases.
For example, the
conjugates of N-2-N and DNA oligomers hydrolyzed the
substrate RNA exactly at the target site under mild
conditions (pH 8 and 50 °C). There, the concentration
of the artificial enzyme (and thus of the N-2-N residue)
is only 100 µM. The N-2-N residue is efficiently con-
centrated at the target phosphodiester linkage due to a
proximity effect, when the conjugate and the substrate
RNA form a complex. Further increase of the catalytic
activity should be plausible if the molecular design is still
more precise. The simplicity and stability of the oligo-
amines are undoubtedly advantageous for the purpose.
7
nuclease A, except for the use of two aliphatic amino
residues in place of the imidazoles and is also consistent
with the successful turnover as the catalyst (Figure 4).
The possibility that the ammonium ions in the mono-
cations of N-2-N and N-3-N function as simple electro-
1
5
static catalysts is unlikely in terms of the following
results. First, the catalytic rate constant for the ApA
Exp er im en ta l Section
hydrolysis by (2-aminoethyl)trimethylammonium ion
+
Ma ter ia ls. Dinucleotides, poly[A], poly[U], and poly[dA]
were purchased from Sigma or Yamasa. The 30-mer RNA was
prepared on a DNA synthesizer using the phosphoramidite
[H
2
N(CH
2
)
2
N(CH
3
)
3
], which has a positive charge as does
the monocation of N-2-N but has no acidic proton, is 4.4
times as small as that of the monocationic N-2-N (after
the latter is statistically corrected). The difference is
exactly identical with the estimated contribution (5-fold
promotion) of the intramolecular acid-base cooperation
32
32
method and was P-labeled at the 3′-end by using [ P]pCp
from Amersham) and ligase (from Takara). The choice of the
sequence is rather arbitrary. Amines, D O, DCl (in D O), and
(
2
2
other reagents were commercially obtained. Water was ion-
exchanged and sterilized immediately before use. Throughout
the present study, great care was taken to avoid a contamina-
tion by natural ribonucleases.
a
in the monocationic N-2-N (the pK of (2-aminoethyl)-
trimethylammonium is 6.6, which is virtually identical
with that (6.8) for the second protonation of N-2-N).
Furthermore, the ammonium ions in the monocations of
N-4-N and N-5-N do not assist the catalysis by the
neutral amino residues (see Figure 2). Apparently, the
length of the alkylene chain between the amino residues
has much more a critical effect than is expected for the
simple electrostatic catalysis. The intramolecular co-
operation is inefficient in the diamines having longer
chains (N-4-N and N-5-N), due to the increased confor-
(2-Aminoethyl)trimethylammonium iodide was synthesized
from N-[(dimethylamino)ethyl]acetamide and methyl iodide in
methanol. The acetyl residue, used for the protection of the
primary amine in N,N-dimethylethylenediamine, was removed
by 2 N HCl.
Kin etic Mea su r em en ts. Hydrolyses of dinucleotides (0.1
mM) were performed at 50 °C in pH 8 Tris buffers (50 mM)
containing EDTA (2 mM) and were followed by a reversed-
phase HPLC: Merck LiChrospher 100 RP18(e) column; eluent,
pH 3.4 phosphate buffer containing 4% acetonitrile. The
diamine-catalyzed reactions were followed for 3-4 half-lives,
although the reactions by monoamines were monitored only
for 0.5-1 half-life. The reactions satisfactorily obeyed the
a
(14) The pK values are 6.7 and 9.7 for N-methylethylenediamine,
and 6.6 and 9.6 for N,N′-dimethylethylenediamine. The increase of the
catalytic activity is probably associated with the enhanced apolar
interaction with the substrate, since their pK
protonation are rather smaller than the value for N-2-N (6.8).
15) Dalby, K. N.; Kirby, A. J .; Hollfelder, F. J . Chem. Soc., Perkin
Trans. 2 1993, 1269.
a
’s for the second
(16) Bender, M. L.; Bergeron, R.; Komiyama, M. The Bioorganic
Chemistry of Enzymatic Catalyses; J ohn & Wiley Sons: New York,
1984; Chap. 5.
(

2
160 J . Org. Chem., Vol. 62, No. 7, 1997
Komiyama and Yoshinari
pseudo-first-order kinetics. Duplicate or triplicate experiments
were carried out, and the rate constants for these reactions
coincided with each other within (3%. For all the amines,
the pseudo-first-order rate constant kobsd of the diribonucleotide
hydrolysis at a constant pH was proportional to the concentra-
tion of amine. Thus the apparent catalytic rate constant
phates, and A>p. No signal assignable to the phosphoramide
and other products was detected.17
2
D O Solven t Isotop e Effects. The isotope effects for the
ApA hydrolysis by N-2-N were measured according to eq 1, in
which the fractions of the neutral species and the monocation
were evaluated by using the pK
in D O). In order to complete D/H exchange, the specimens
were prepared by evaporating the sample solutions in D O to
dryness, followed by the addition of a required amount of D O.
a
values of N-2-N (7.5 and 10.0
(k
c,appar) was determined from the slope of the straight line of
2
k
obsd vs [amine] . The pK values of the amines were measured
0
a
2
at 50 °C by potentiometric titration.
2
The amine-catalyzed hydrolysis of poly[A], poly[U], and
poly[dA] (the nucleotide concentration 5 mM) at pH 8 and 50
The pD values were determined by the equation: pD ) pH
meter reading +0.4.18
°
C was followed (2-4 half-lives) by ion-exchange HPLC using
Toso DEAE-NPR columns: a linear gradient from 0 to 0.4 M
NaClO at pH 9 in 20 mM Tris buffer. When the polymeric
Ack n ow led gm en t. The authors should like to thank
Prof. Kazunari Taira of the University of Tsukuba for
valuable comments. The assistance by Dr. Shinya
Sawata and Mr. Takuya Inokawa is also acknowledged.
This work was supported by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Science, and
Culture, J apan, and by the Mitsubishi Science Founda-
tion.
4
substrates were degraded to smaller fragments, the peaks in
the HPLC gradually shifted toward the shorter retention time
region (the retention time of the fragmental product monoto-
nously increases with the increase in its molecular weight).
The magnitude of the shift of the HPLC peak was converted
to the decrease in the number of the phosphodiester linkages
in the polymeric substrates, by using the calibration curves
(
the relationship between the retention time and the degree
Su p p or tin g In for m a tion Ava ila ble: pH-rate constant
profiles for the hydrolysis of A>p, poly[A], and poly[U] by N-2-
N, for the hydrolysis of A>p by N-3-N, and for the hydrolysis
of polymerization) which were obtained with the use of
authentic samples of different degrees of polymerization.
2
b
3
2
The hydrolysis products of the 30-mer synthetic RNA ( P-
2
of poly[A] by NH E (5 pages). This material is contained in
-
6
labeled at the 3′-end: 10
denaturing polyacrylamide gel electrophoresis.
M) were analyzed by a 20%
libraries on microfiche, immediately follows this article in the
microfilm version of the journal, and can be ordered from the
ACS; see any current masthead page for ordering information.
3
1
P -NMR Stu d y on th e Rea ction P r od u cts. At 50 °C,
2
O solutions of ApA (10 mM) were treated with N-2-N (1.0
D
3
1
M). The reaction mixtures were directly subjected to P-NMR
spectroscopy (a J NM GX-400 FT-NMR spectrometer) with
J O961935U
H
3
PO
4
(in D
2
O) as external standard. At either pD 8 or pD
(17) Hosseini, M. W.; Lehn, J .-M. J . Am. Chem. Soc. 1987, 109, 7047.
(18) Glasoe, P. K.; Long, F. A. J . Phys. Chem. 1960, 64, 188.
1
1, the products were adenosine, its 2′- and 3′-monophos-