Kinetic analysis of the temperature dependence of the template-directed formation of oligoguanylate from the 5′-phosphorimidazolide of guanosine on a poly(C) template with Zn2+

Article abstract of DOI:10.1246/bcsj.74.927

A kinetic study of the temperature dependence of the template-directed formation of oligoguanylate (oligo(G)) on polycytidylic acid (poly(C)) from the 5′-phosphorimidazolide of guanosine (ImpG) has been carried out in the presence of Zn²⁺ at 40

Full text of DOI:10.1246/bcsj.74.927

.
.
© 2001 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 74, 927—935 (2001)
927
Chemical
iety of Ja-
Chemical
iety of Ja-
Kinetic Analysis of the Temperature Dependence of the Template-Direct-
′
ed Formation of Oligoguanylate from the 5 -Phosphorimidazolide of
Guanosine on a Poly(C) Template with Zn2ꢀ
Prebiotic Formation of RNA at 40–80 ˚C
K. Kawamura et al.
01
Kunio Kawamura* and Masatosi Umehara
Department of Applied Chemistry, Osaka Prefecture University, Sakai, Osaka 599-8531
(Received October 3, 2000)
SJA8
9-2673
46
ober
A kinetic study of the temperature dependence of the template-directed formation of oligoguanylate (oligo(G)) on
polycytidylic acid (poly(C)) from the 5′-phosphorimidazolide of guanosine (ImpG) has been carried out in the presence
of Zn2ꢀ at 40–80 ˚C. It is surprising that a large amount of oligo(G) was formed at 40–60 ˚C and a small amount of oli-
go(G) was detected even at 80 ˚C. The rate constants of the template-directed formation of oligo(G) were determined at
40–60 ˚C, and the same trend, the rate constants increase with the chain length, was observed as that at lower tempera-
tures. Besides, second-order rate plots of the hydrolyses of ImpG and oligo(G) were consistent with pseudo-second-or-
der processes in the presence of 0.04 M Zn2ꢀ. The apparent activation energy determined from the Arrhenius plots de-
creases in the order hydrolysis of oligo(G) > formation of 4-mer ≈ formation of 3-mer ≈ hydrolysis of ImpG > formation
of 2-mer. The kinetic analysis and computer simulations demonstrate the importance of the rate of 2-mer formation to
determine the efficiency of the oligo(G) formation.
0
1
4
The discovery of the catalytic activity of RNA suggests that
RNA or RNA-like molecules had a central role on the primi-
tive earth.1–5 If the RNA world hypothesis is correct, then
RNA-like molecules formed spontaneously under primitive
earth conditions. A number of successful studies of relevance
with the RNA world hypothesis have been carried out.6–10
On the other hand, it is widely believed that hydrothermal
systems, such as hydrothermal vents in the deep ocean, played
an important role for the prebiotic synthesis of biologically es-
sential molecules.11–20 Based on the discovery of thermophilic
organisms in association with hydrothermal conditions, it has
been deduced that life on earth might have originated in hydro-
thermal systems.21–26 This hypothesis has been supported by
phylogenetic analyses of modern microorganisms, which sug-
gest that the last common ancestor (LCA) was a hyperthermo-
phile. Besides, there has been a challenging proposal on LCA
that was not a hyperthermophilic organism.27 Despite much
effort towards LCA, the nature of LCA is still unclear.28
In contrast to the importance of hydrothermal prebiotic syn-
thesis, the rapid disappearance of prebiotic molecules suggests
that the emergence of primitive organisms at high temperatures
was difficult.29–35 For example, although it is generally consid-
ered that RNA is unstable under hydrothermal conditions,
there has been fewer systematic studies on the stability of RNA
at high temperatures.34,35 Thus, we have been studying the sta-
bility of RNA molecules in aqueous solution at high tempera-
tures.36–41 The experiments have provided quantitative data
that nucleotides are notably unstable in aqueous solution at
high temperatures. Although this fact may support that the
emergence of the RNA world was not possible in hydrothermal
systems, the accumulation of RNA molecules in the RNA
world should be determined by both the rates of the formation
and the decomposition of RNA. Thus, parallel investigations
on the kinetics of the temperature dependence of the prebiotic
formation and decomposition of RNA are important to evalu-
ate the RNA world hypothesis from the viewpoint of hydro-
thermal reactions.
There have been a number of successful studies of the con-
densation reaction of activated nucleotides to form RNA oligo-
nucleotides in the presence42–45 and absence of RNA tem-
plates.46–50 However, the experiments have only been carried
out at temperatures up to 37 ˚C.51 Thus, it is important to in-
vestigate the scope of these successful studies at higher tem-
peratures. That is to say, a kinetic analysis of the temperature
dependence of the template-directed synthesis could explore
the knowledge on the prebiotic chemistry of RNA.52
Besides, a variety of kinetic investigations of prebiotic syn-
thesis of oligonucleotides have been carried out to provide in-
sight into the mechanism of oligonucleotide formation on
RNA templates53–59 and montmorillonite clay catalysts.60,61
Thus, a kinetic analysis of the prebiotic formation of RNA at
elevated temperatures would provide a basis to know whether
the chemical evolution of RNA has occurred under hydrother-
mal environments or not. However, there has been no kinetic
analysis on the condensation of activated nucleotide in the
presence and absence of templates at temperatures higher than
23 ˚C.53–59 In this study, the kinetics of the formation of oli-
go(G) from the poly(C)-template-directed reaction in the pres-
ence of Zn2ꢀ 62–64 was investigated at 40–80 ˚C. The rate con-
stants of the formation of oligo(G) and those of the hydrolyses
of oligo(G) and ImpG were determined. Based on a compari-
son between the rates of the formation and the decomposition
of oligo(G), it was deduced that both the rates of the formation
of 2-mer and the decomposition of oligo(G) are important fac-

928 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Prebiotic Formation of RNA at 40–80 ˚C
gen to stop further hydrolysis. The rates of the oligo(G) hydroly-
sis were followed in the absence and presence of a poly(C)
template.
tors to determine the yield of long oligonucleotides. Further,
kinetic simulations were performed to visualize the importance
of 2-mer formation. Based on the results, the reason for the
difficulty regarding template-directed formation of oligo(G) in
hydrothermal environments is going to be discussed.
′
Tm Studies of ImpG, 5 -GMP, and Oligo(G) with Poly(C).
The absorption spectra in a solution containing the same electro-
lytes as used in the kinetic investigations were determined using a
Shimadzu UV-2500PC. The oligo(G) solution was prepared by
the same procedure as used for the hydrolytic study of oligo(G).
Determination of Rate Constants and Kinetic Simulation.
The computer program SIMFIT was used to evaluate the rate con-
stants and the kinetic simulations of the template-directed forma-
tion of oligo(G).9
Experimental
Materials and Equipment. ImpG was synthesized using a
modification of the procedure of Joyce et al. (purity 97%).65,66
Polycytidylic acid (poly(C)), ribonuclease A (RNase A), ribonu-
clease T1 (RNase T1), ribonuclease T2 (RNase T2), and P1,P2-
bis(5′-guanosyl)diphosphate (G5′ppG) were purchased from SIG-
MA. All other reagents used were of analytical grade.
Results
Hydrolysis of ImpG in the Presence of Zn2ꢀ. The hy-
drolysis of ImpG was investigated in the presence of Zn2ꢀ to
determine if it followed pseudo-first- or second-order kinet-
ics. Because the hydrolysis of adenosine 5′-triphosphate obeys
pseudo-second-order kinetics in the presence of several kinds
of metal ions67,68 it is plausible that the ImpG hydrolysis obeys
pseudo-second-order kinetics. A hydrolysis study was carried
out in the presence of 0.004 M and 0.04 M Zn2ꢀ at 40–80 ˚C.
The reaction curves are shown in Fig. 1, in which a small
amount of G5′ppG is observed. The first-order rate plots of the
disappearance of ImpG were not best fitted at 0.04 M Zn2ꢀ
(Fig. 2a), while the hydrolysis of ImpG at 0.004 M Zn2ꢀ was
consistent with a pseudo-first-order process. Besides, the sec-
ond-order rate plots were well fitted for the disappearance of
ImpG at 0.04 M Zn2ꢀ (Fig. 2b). Although detail analyses
should be necessary to determine the reaction mechanism,
these facts indicate that the hydrolysis mainly involves a sec-
ond-order rate process at 0.04 M Zn2ꢀ. Thus, the rate con-
stants were determined on the basis of a pseudo-second-order
model by SIMFIT. The rate constants of the ImpG hydrolysis
and the formation of G5′ppG from ImpG are summarized in
Table 1. A best fit was obtained using a pseudo-second-order
model, in which the sum of the errors was a few-times smaller
than that obtained using a first-order model. The half-lives of
ImpG, calculated from the rate constants shown in Table 1,
High-performance liquid chromatography (HPLC) was per-
formed by a HPLC system LC10A (Shimadzu, Japan) with a
DNA-NPR anion-exchange column from TOSOH Co., Japan us-
ing a gradient of 0.3–1.5 M (1 M ꢁ 1 mol dmꢂ3) NaCl at pH 11
with 0.02 M 2-amino-2-hydroxymethyl-1,3-propanediol (Tris)
buffer and a ODS-2 column from GL Science Co., Japan using a
gradient of 0.005 M NaH2PO4 in water at pH 3.5 mixed with 0.01
M NaH2PO4 in 40% CH3OH at pH 4.0.
Kinetic Measurements. Oligomerization of ImpG on
Poly(C) Template at Elevated Temperatures. The template-
directed formation of oligo(G) on a poly(C) template from ImpG
was performed in a solution containing 0.015 M ImpG, 0.025 M
poly(C), 1 M NaCl, 0.2 M MgCl2, 0.1 M 2-[4-(2-hydroxyethyl)-1-
piperazinyl]ethanesulfonic acid (HEPES), 0.04 M ZnCl2, at pH
8.0. A 200 µL of the ImpG solution was added to a 500 µL plastic
vial with a screw cap, which bears up to 120 ˚C, and the vial was
settled to a cartridge heater controlled at 40–80 ˚C. The reaction
was monitored at regular intervals over 24 h and the sample solu-
tion was immediately quenched in liquid nitrogen at the end of the
reaction.
The sample was incubated with RNase A to cleave poly(C) in
the presence of a 0.1 M EDTA solution to mask Zn2ꢀ. RNase A
hydrolysis was performed in 0.1 mL of the reaction mixture for 18
h at 37 ˚C using 300 units. The sample was analyzed by the an-
ion-exchange and the reversed-phase HPLC. The hypochromicity
correction factor was regarded as approximately 1 under the
HPLC conditions.53 The extent of 3′,5′-linked oligo(G) was deter-
mined using RNase T2 or RNase T1. The hydrolysis with RNase
T2 was carried out in 0.1 mL of the sample solution for 18 h at 37
˚C using 30 units, and that with RNase T1 was performed in 0.1
mL for 18 h at 37 ˚C using 200 units of enzyme. The extent of oli-
go(G)s was determined based on anion-exchange HPLC and re-
versed-phase HPLC analyses.
Hydrolysis of ImpG. The rates of ImpG hydrolysis were de-
termined in a solution containing 0.015 M ImpG, 1M NaCl, 0.2 M
MgCl2, 0.1 M HEPES, 0.04 M, or 0.004 M ZnCl2 at pH 8.0. The
hydrolysis reaction was monitored for 24 h and the samples were
analyzed by reversed-phase HPLC.
Hydrolysis of Oligo(G). A standard solution containing oli-
go(G) was prepared from the incubation of poly(G) in 0.1 M
NaOH for 40 min at 37 ˚C and the treatment in 1 M HClO4 to
cleave 2′,3′-cyclic phosphate.42 The mixture was neutralized with
NaOH solution to avoid any further hydrolysis of oligo(G).
The hydrolysis of standard oligo(G) was monitored in a solu-
tion containing 1 M NaCl, 0.2 M MgCl2, 0.1 M HEPES, 0.04 M
ZnCl2, at pH 8.0 in the presence of 0.025 M poly(C) for 144 h at
60–100 ˚C. At regular time intervals, an aliquot of the sample so-
lution was withdrawn and immediately quenched in liquid nitro-
Fig. 1.
Reaction curves for the hydrolysis of ImpG in the
presence of Zn2ꢀ. [NaCl] ꢁ 1.0 M, [MgCl2] ꢁ 0.2 M,
[HEPES] ꢁ 0.1 M, [ZnCl2] ꢁ 0.04 M, [ImpG] ꢁ 0.015 M,
pH ꢁ 8.0, 40 ˚C. Lines, ꢀ, ImpG; ꢁ, pG; ꢂ , G5′ppG.

K. Kawamura et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 929
nius plots is 82 kJ molꢂ1 at 0.04 M Zn2ꢀ and 62 kJ molꢂ1 at
0.004 M Zn2ꢀ.
Kinetics of the Oligomerization of ImpG at 40–80 ˚C.
The reactions at 40 and 50 ˚C yielded oligo(G)s longer than
30-mer, and those at 60 and 80 ˚C yielded oligo(G)s longer
than 20-mer (Figs. 3a–d). The high efficiency of the oligo(G)
formation at elevated temperatures is unexpected, since 15–20-
mer was the longest at 37 ˚C in a previous study.51 The extent
of ImpG, 5′-GMP, and G5′ppG in the monomer fraction on an-
ion-exchange HPLC was determined on the basis of an analy-
sis by reversed-phase HPLC, in which the total extent of 4-mer
and longer oligomers was calculated at 40–80 ˚C (Fig. 4). The
reaction curve at 60 ˚C indicates that the oligo(G) formed from
ImpG was hydrolyzed after the template-directed formation of
oligo(G). Because the yield of oligo(G)s was notably low at
80 ˚C, kinetic analyses were carried out at 40–60 ˚C.
The regioselectivity of 3′,5-phosphodiester bond formation
was investigated by selective enzymatic hydrolysis with RNase
T1 and RNase T2 (Table 2). The regioselectivity at 40–60 ˚C is
consistent with that at lower temperature;63 this fact indicates
that the regioselectivity is not very sensitive at 0–60 ˚C.
The reaction curves of the disappearance of ImpG and the
formation of oligo(G) at 40–60 ˚C were fitted by SIMFIT us-
ing the reaction pathways defined in Eq. I-1–5 by varying the
pseudo-second-order rate constants. The reaction model is ba-
sically the same as that previously applied, except for using the
second-order hydrolysis model of ImpG,53,60,61 where subscript
“n” designates the length of oligo(G).
2 ImpG
ImpG ꢀ pG
ImpG ꢀ (pG)2 → (pG)3
ImpG ꢀ (pG)3 → (pG)4
...
→ ImpG ꢀ pG
→ (pG)2
(I-1)
(I-2)
(I-3)
(I-4)
Fig. 2.
(A) Pseudo-first-order rate plots and (B) pseudo-
second-order rate plots for the hydrolysis of ImpG in the
presence of Zn2ꢀ. [NaCl] ꢁ 1.0 M, [MgCl2] ꢁ 0.2 M,
[HEPES] ꢁ 0.1 M, [ZnCl2] ꢁ 0.04 M, [ImpG] ꢁ 0.015 M,
pH ꢁ 8.0. Lines, ꢀ, 40 ˚C; ꢁ, 50 ˚C; ꢂ, 60 ˚C; ꢃ, 80 ˚C.
ImpG ꢀ (pG)nꢂ1 → (pG)n
(I-5)
The rate constants were determined by continuous iteration to
a convergent result using SIMFIT, in which a constant magni-
tude of k4–kn was defined as that postulated in previous stud-
ies.58,59 The rate constants at 40–60 ˚C are summarized in Ta-
ble 3.
Kinetics of Oligo(G) Hydrolysis. The reaction curves of
the oligo(G) formation in the presence of a poly(C) template
indicate that the hydrolysis of oligo(G) occurs after 6 h at 60
˚C (Fig. 4). In order to evaluate the importance of the oligo(G)
hydrolysis, the kinetics of the oligo(G) hydrolysis in both the
presence and absence of poly(C) was studied at 60–100 ˚C.
Table 1.
The Rate Constants for the ImpG Hydrolysis in
the Presence of Zn2ꢀ a)
T/˚C
khy/sꢂ1 Mꢂ1 b)
kG5′ppG/sꢂ1 Mꢂ1 c)
40
50
60
80
(3.09 ꢃ 0.06) ꢄ 10ꢂ3
(8.14 ꢃ 0.14) ꢄ 10ꢂ3
(2.62 ꢃ 0.14) ꢄ 10ꢂ2
(1.04 ꢃ 0.01) ꢄ 10ꢂ1
(1.09 ꢃ 0.26) ꢄ 10ꢂ4
(2.33 ꢃ 0.45) ꢄ 10ꢂ4
(1.28 ꢃ 0.42) ꢄ 10ꢂ3
(4.06 ꢃ 0.42) ꢄ 10ꢂ3
a) Reactions were performed in a solution containing 0.015
M ImpG, 0.04 M ZnCl2, 1.0 M NaCl, 0.2 M MgCl2, and
0.1 M HEPES at pH 8.0. The rate constants were deter-
mined using SIMFIT. The reaction curves were fitted by
pseudo-second-order rate model. b) The rate constants for
the hydrolysis of ImpG. c) The rate constants for the for-
mation of G5′ppG.
Table 2.
The Extent (%) of 3′,5′-Linked Oligonucleo-
tides Formed from ImpG Condensation on Poly(C)
Template in the Presence of ZnCl2
a)
b)
T/˚C
RNase T1
RNase T1
RNase T2
40
50
60
95
95
95
96
96
98
83
90
90
were 22 min at 60 ˚C and 5.6 min at 80 ˚C in the presence 0.04
M Zn2ꢀ, while those were 3.1 h at 60 ˚C and 53 min at 80 ˚C in
the presence of 0.004 M Zn2ꢀ. This fact indicates that the hy-
drolysis of ImpG at 0.04 M Zn2ꢀ was about 9-times more ac-
celerated than that at 0.004 M Zn2ꢀ. The apparent activation
energy for the hydrolysis of ImpG determined from the Arrhe-
a) No treatment before enzymatic hydrolysis. b)
Samples with heating 1 min at 90 ˚C prior to enzy-
matic hydrolysis.

930 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Prebiotic Formation of RNA at 40–80 ˚C
Fig. 3.
HPLC profile of the template-directed formation of oligo(G) from ImpG on poly(C) template. [NaCl] ꢁ 1.0 M, [MgCl2]
ꢁ 0.2 M, [HEPES] ꢁ 0.1 M, [ZnCl2] ꢁ 0.04 M, [ImpG] ꢁ 0.015 M, [poly(C)] ꢁ 0.025 M, pH ꢁ 8.0. (a), 40 ˚C, 6h; (b), 50 ˚C,
6 h; (c), 60 ˚C, 6 h; (d), 80 ˚C, 10 min.
Table 3.
The Rate Constants (sꢂ1 Mꢂ1) for the ImpG Condensation at Elevated Temperaturesa)
Length of oligo(G)
40 ˚C
50 ˚C
60 ˚C
2
3
(1.19 ꢃ 0.06) ꢄ 10ꢂ3
(6.17 ꢃ 0.37) ꢄ 10ꢂ2
(9.11 ꢃ 0.42) ꢄ 10ꢂ2
(1.65 ꢃ 0.11) ꢄ 10ꢂ3
(1.96 ꢃ 0.18) ꢄ 10ꢂ1
(2.61 ꢃ 0.17) ꢄ 10ꢂ1
(2.77 ꢃ 0.16) ꢄ 10ꢂ3
(3.35 ꢃ 0.26) ꢄ 10ꢂ1
(6.45 ꢃ 0.55) ꢄ 10ꢂ1
average 4 <
a) Reaction conditions are the same as shown in Table 1. The rate constants were determined using SIMFIT.
The rate of the oligo(G) hydrolysis was followed by the de-
scribed procedure.37 Oligo(G)s up to approximately 20-mer in
length, which correspond to moderate oligo(G)s in length
formed by the template-directed reaction, were prepared by the
alkaline hydrolysis of poly(G) (described in the experimental
section). The number of phosphodiester bonds remaining in
oligo(G) was followed at regular time intervals during oli-
go(G) hydrolysis. The value (N), which is proportional to the
number of phosphodiester bonds and normalized as being 100
at time 0, was determined during oligo(G) hydrolysis and the
time courses of the N value were monitored (Fig. 5).37 The re-
action curves indicate that the hydrolysis of oligo(G) was nota-
bly inhibited in the presence of a poly(C) template, even at 100
˚C. Further, the hydrolysis did not obey the pseudo-first-order
rate, while the pseudo-second-order rate plots gave a good fit.
Thus, the rate constants were determined on the basis of pseu-
do-second order kinetics by SIMFIT (Table 4).
In addition, the half-life of the degradation of the oligo(G)
formed by the template-directed reaction at 60 ˚C is 15 h (Fig.
4), which is similar to that of oligo(G) in the absence of
poly(C) (t1/2 ꢁ 13 h), while t1/2 in the presence of poly(C) is
130 h.

K. Kawamura et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 931
dicated that the ImpG hydrolysis mainly involves a pseudo-
second-order process at 0.04 M Zn2ꢀ. This finding is in con-
trast to the fact that the hydrolysis of imidazole-activated nu-
cleotides catalyzed by Mg2ꢀ follows pseudo-first-order kinet-
ics.69 By hydrolytic studies of the nucleoside 5′-triphosphate,
it was confirmed that the hydrolysis obeyed pseudo-second-or-
der kinetics in the presence of metal ions.67,68 Thus, it is plau-
sible that ImpG hydrolysis is catalyzed by complexation in-
volving two ImpG molecules. The reaction mechanism should
be deduced by detail studies on the concentration dependence
of Zn2ꢀ. Besides, the formation of G5′ppG at 0.04 M Zn2ꢀ has
a similar activation energy (87 kJ molꢂ1) to that of the ImpG
hydrolysis. This fact may indicate that both the ImpG hydroly-
sis and the formation of G5′ppG proceed through a common in-
termediate.
On the other hand, the hydrolysis of oligo(G) followed a
pseudo-second order rate. It has also been known that the hy-
drolysis of ribonucleotides is catalyzed in the presence of
Zn2ꢀ.70–73 In the study, a kinetic analysis of the hydrolysis of
dinucleotides showed that the maximum rate was observed at
Zn2ꢀ/phosphate ꢁ 1 : 2.72 These facts support the pseudo-sec-
ond-order hydrolysis of oligo(G) in the presence of Zn2ꢀ. The
fact that both the hydrolyses of ImpG and oligo(G) are consis-
tent with pseudo-second-order processes at 0.04 M Zn2ꢀ indi-
cate that the acceleration of the hydrolyses of ImpG and oli-
go(G) by Zn2ꢀ involves a common mechanism.
Fig. 4.
The time course of the total amount of oligo(G) 4-
mer and longer. [NaCl] ꢁ 1.0 M, [MgCl2] ꢁ 0.2 M,
[HEPES] ꢁ 0.1 M, [ZnCl2] ꢁ 0.04 M, [ImpG] ꢁ 0.015 M,
[poly(C)] ꢁ 0.025 M, pH ꢁ 8.0. plots and solid lines, ꢀ,
40 ˚C; ꢁ, 50 ˚C; ꢂ, 60 ˚C; ꢃ, 80 ˚C. The dashed lines
were reproduced by the simulation using Eq. II-1–6.
The hydrolysis of oligo(G) was inhibited in the presence of
poly(C) at 60–100 ˚C. This fact is consistent with the fact that
the melting point (Tm) of poly(G)•poly(C) complex is nearly
100 ˚C,74–77 which is potentially affected in the presence of
metal ions. Actually, it was observed that the absorbance at
250–270 nm gradually increased at 10–70 ˚C. This is due to
the fact that the oligo(G) solution involved different lengths of
oligo(G)s, but the Tm value of oligo(G)•poly(C) complex
should be lower than that of poly(C)•poly(G) complex, which
is nearly 100 ˚C.74–77 These facts support that the inhibition of
the oligo(G) hydrolysis in the presence of poly(C) at 60–100
˚C is due to the partial complexation between oligo(G) and
poly(C). In addition, the coincidence between the half-life of
the oligo(G) formed by the template-directed reaction and that
of the standard oligo(G) in the absence of poly(C) at 60 ˚C in-
dicates that the hydrolysis of oligo(G) formed by the template-
directed reaction followed the hydrolysis in the absence of
poly(C).
Fig. 5.
The reaction curves of the disappearance of phos-
phodiester bond of oligo(G) in the presence and absence of
poly(C). [NaCl] ꢁ 1.0 M, [MgCl2] ꢁ 0.2 M, [HEPES] ꢁ
0.1 M, [ZnCl2] ꢁ 0.04 M, [oligo(G)] ꢁ 1.6–2.6 mM, pH
ꢁ 8.0. ꢁ, ꢃ, ꢄ, no poly(C); ꢀ, ꢂ, ꢅ, [poly(C)] ꢁ 0.025
M; ꢀ, ꢁ, 60 ˚C; ꢂ, ꢃ, 80 ˚C; ꢅ, ꢄ, 100 ˚C.
Oligo(G) Formation with Poly(C) Template at Elevated
Temperatures. The formation of higher oligo(G)s at 40–80
˚C than that at 37 ˚C in the previous study51 is surprising. Fur-
ther, the high yield of long oligo(G)s indicate that biologically
important interactions play effectively at these temperatures.
The reason of high efficiency of the formation of long oli-
go(G)s may be due to that the reaction conditions are some-
what different between the previous and present studies. ImpG
was used as an activated-monomer instead of 2-MeImpG and
Zn2ꢀ was added as a catalyst to form long oligo(G)s in the
present study.
Table 4.
The Rate Constants (sꢂ1
M
ꢂ1) for the Oligo(G)
Hydrolysis at Elevated Temperaturesa)
T/˚C
no poly(C)
0.025 M poly(C)
60
80
100
(1.09 ꢃ 0.07) ꢄ 10ꢂ2
(7.41 ꢃ 0.33) ꢄ 10ꢂ2
(8.37 ꢃ 0.44) ꢄ 10ꢂ1
(9.38 ꢃ 0.70) ꢄ 10ꢂ4
(1.46 ꢃ 0.03) ꢄ 10ꢂ2
(6.32 ꢃ 0.15) ꢄ 10ꢂ2
a) The reaction conditions are the same as shown in Table
1. The reaction curves were fitted using pseudo-second-
order model for the hydrolysis of oligo(G).
The trend in the rate constants of the oligo(G) formation, in
which the rate constants increase from 2-mer through 4-mer
and then level off, is similar to that of the template-directed re-
action of 2-MeImpG at lower temperatures53 and that of the
Discussion
Hydrolysis of ImpG and Oligo(G). Kinetic analyses in-

932 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Prebiotic Formation of RNA at 40–80 ˚C
spontaneous formation of oligonucleotides in the presence of
the montmorillonite clay catalyst.60,61 This fact indicates that
the association between ImpG and elongating oligo(G)s occurs
on the poly(C) template, even at 40–60 ˚C. In the present tem-
plate-directed reaction, a stacking interaction between ImpG
and elongating oligo(G) may be promoted in the presence of
Zn2ꢀ. Actually, it is well known that 3′,5′-linked oligo(G)
formed selectively in the presence of Zn2ꢀ, which has been
considered to be evidence that Zn2ꢀ stabilizes the double-helix
between poly(C) and ImpG and causes the high regioselectivi-
ty of 3′,5′-phosphodiester bond formation.62 Further, it is sur-
prising that the regioselective formation of 3′,5′-linked oli-
go(G) at 40–60 ˚C is similar to that observed at lower
temperatures.63 These facts support that the reaction mecha-
nism deduced at lower temperatures53–59 is able to be applied at
40–60 ˚C.
Besides, the Tm value of ImpG with poly(C) is considered to
be much lower than the temperatures used in this study, since
the absorbance at 254 nm was less changed at 10–60 ˚C. This
fact is consistent with the Tm value of a mixture of 2-MeImpG
with poly(C)78 and that of GMP with poly(C).79 The fact that
the template-directed synthesis is possible at higher tempera-
tures than the Tm of the mixture of activated-monomer and the
template. This is interesting because the same trend was also
observed in previous studies.51,78
Fig. 6.
Arrhenius plots for the template-directed reaction
and hydrolysis of ImpG and oligo(G). ꢀ, ImpG hydroly-
sis; ꢁ, 2-mer formation; ꢂ, 3-mer formation; ꢃ, 4-mer
and longer formation; ꢅ, hydrolysis of oligo(G) in the ab-
sence of poly(C); ꢄ, hydrolysis of oligo(G) in the presence
of poly(C).
Table 5. Apparent Activation Energy for the Reac-
tions Relevance to the Template-Direcetd Oligo(G)
Formation
There has been less attention on the temperature depen-
dence of the biologically important interactions, such as hy-
drogen bonding and stacking interaction at elevated to high
temperatures. However, the kinetics shown in the study sup-
port that interactions between ImpG, growing oligo(G) on the
poly(C) template definitely exist at elevated temperatures. The
interactions in the present system are probably stronger than
those appeared in the template-directed reaction using 2-Me-
ImpG at 37 ˚C in the previous study,51 since the higher oli-
go(G)s formed in the present study. Further, it has been dis-
covered that the enhancement of the helix formation between
template and oligonucleotides enables the ligation of 2′,5′-
linked oligoribonucleotides on a 2′,5′-linked template.80 Thus,
investigations on the association of ImpG with oligo(G) on
poly(C) template could be important for a detail analysis of the
template-directed reaction in the presence and absence of
Zn2ꢀ.
The residual amount of poly(C) during the template-direct-
ed reaction of oligo(G) was 69% (6 h), 44% (12 h), 37% (18
h), and 31% (24 h) at 60 ˚C and 13% (6 h), 10% (10 h), and 8%
(19 h) at 80 ˚C. At these temperatures, the oligomerization re-
action was almost completed at 6 h. Therefore, it is deduced
that the loss of poly(C) did not notably affect the template-di-
rected formation of oligo(G), but did affect the hydrolytic de-
composition of oligo(G).
Reactions
ImpG hydrolysis
GppG formation
Ea/kJ molꢂ1
82
87
Oligo(G) formation
2-mer
3-mer
36
74
85
4-mer and longer
Oligo(G) hydrolysis
with poly(C)
no poly(C)
112
109
mined in this study are appropriate to discuss the temperature
dependence of the formation and decomposition of oligo(G)
using the prebiotic reactions. The rate constants can be com-
pared directly, since all of the reactions follow pseudo-second-
order kinetics. The slope for the formation of 3-mer and 4-
mer, and that for the ImpG hydrolysis are virtually similar,
while the slope of the formation of 2-mer is about 2-times
smaller than that of others. The ratio of k4/k2 determined by
the experiments is 77 (40 ˚C), 158 (50 ˚C), and 233 (60 ˚C),
and the estimated k4/k2 value from the Arrhenius plots is 660 at
80 ˚C. In addition, the hydrolysis rate of oligo(G) has the larg-
est activation energy among the rate constants in the present
template-directed reaction system. The comparison suggests
that the 2-mer formation rate is an important factor to deter-
mine the yield of oligo(G)s at higher temperatures along with
the oligo(G) hydrolysis. The small slope for the 2-mer forma-
tion indicates that the associate formation of between two
ImpG molecules on poly(C) is more sensitively interfered at
higher temperatures than that between ImpG and long oli-
go(G)s.
Comparison of the Formation and Decomposition of Oli-
go(G). To evaluate the kinetic analysis in the present study,
the temperature dependence of the rate constants was evaluat-
ed. The Arrhenius plots of the rate constants for the formation
of oligo(G), the hydrolysis of ImpG, and the hydrolysis of oli-
go(G) in the presence and absence of poly(C) are illustrated as
a function of Tꢂ1 (Fig. 6) and the apparent activation energy is
determined (Table 5). The rate constants involve some errors,
but the Arrhenius plots indicates that the rate constants deter-

K. Kawamura et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 933
On the other hand, the apparent activation energy of the hy-
drolysis of oligo(G) is independent in the presence and ab-
sence of poly(C). This fact indicates that the contribution of
the entropy change is the main factor of the inhibition of the
oligo(G) hydrolysis by the formation of oligo(G)•poly(C)
complex.
The temperature dependence of the rate constants is ex-
pressed by Eq. II-1–6
ln khy,ImpG
ln khy,oligo(G),no poly(C)
ln khy,oligo(G),with poly(C)
ln k2
ln k3
ln k4av
ꢁ 25.6 ꢂ 9.82 ꢄ 103 Tꢂ1 (II-1)
ꢁ 35.7 ꢂ 1.34 ꢄ 104 Tꢂ1 (II-2)
ꢁ 32.7 ꢂ 1.31 ꢄ 104 Tꢂ1 (II-3)
ꢁ 7.28 ꢂ 4.40 ꢄ 103 Tꢂ1 (II-4)
ꢁ 25.6 ꢂ 8.85 ꢄ 103 Tꢂ1 (II-5)
ꢁ 30.2 ꢂ 1.02 ꢄ 104 Tꢂ1 (II-6)
where khy,ImpG is the rate constant of ImpG hydrolysis, khy,oli-
go(G),no poly(C) and khy,oligo(G),with poly(C) are the rate constants of oli-
go(G) hydrolysis without and with poly(C), k2, k3, and k4av are
the rate constants of the formation of 2-mer, 3-mer, and 4-mer
and longer. To evaluate whether Eq. II-1–6 provide good re-
productions of the experimental data or not, the total amounts
of the oligomers of 4-mer and longer determined by the exper-
iments were compared with those calculated using the Eq. II-
1–6 at 40–80 ˚C. For the hydrolysis of oligo(G) at 60 ˚C, the
rate constants in the presence of poly(C) in the time range 0–
2.8 h and those in the absence of poly(C) in the time range 2.8–
28 h, respectively. The calculated data (dashed lines in Fig. 4)
indicate good agreement with the experimental points. The
yield of 4-mer and longer at 80 ˚C was also calculated from the
equations, in which a fairly good estimation was performed,
even though the equations were determined from the tempera-
ture dependence at 40–60 ˚C. Conclusively, these analyses and
considerations support the reliability of the rate constants de-
termined by this study and the temperature dependence ex-
pressed by Eq. II-1–6.
Importance of the Rate of 2-Mer Formation. The tem-
perature dependence of the rate constant of the formation of 2-
mer (Fig. 6) has shown that the rate of 2-mer formation is an
important factor to determine the yield of oligo(G). Briefly,
the importance of the 2-mer formation rate on the template-di-
rected reaction was pointed out in a previous study.58 Here, the
time course of the extent of oligo(G) was simulated on the ba-
sis of the proposed rate constants (Table 3 and 4) and the influ-
ence of the k2 was demonstrated. The simulation was per-
formed using the rate constants at 40 ˚C and a new value of k2,
in which the original k2 value was multiplied by factors of 1/5,
1/25, 5, or 25. By the simulations, the total concentration of 4-
mer and higher (Fig. 7a) and the concentration of 10-mer (Fig.
7b) were visualized. Further, the relationship between the
yield of oligo(G) and the oligo(G) length was evaluated (Fig.
7c). The simulations show that the yield of oligo(G) increases
with increasing k2, and that it is highest when the k2 value is
multiplied by a factor of 5. On the other hand, the yield of 10-
mer is maximum using the original k2 value. This relationship
is more ambiguously shown in Fig. 7c, in which longer oligo-
mer forms at k2 ꢁ 1/5 and 1/25 than at the original k2 value.
The simulation indicates that the total yield of oligomers does
not simply increase with increasing the ratio of k4/k2, which is
Fig. 7.
Computer simulations for the template-directed
synthesis using SIMFIT. The rate constants used are of the
experimental data of 40 ˚C except 2-mer formation. (a)
The total concentration of 4-mer and longer; (b) the con-
centration of 10-mer; (c) the oligomer concentration vs.
length of oligo(G). Different k2 values are used and the ra-
tios of the k2 values to the original value are shown.
in contrast to the previous prediction.58
The simulations support the reliability of the reaction rate
constants determined in the study. Further, it has been demon-
strated that the k2 is an important factor to determine the yield

934 Bull. Chem. Soc. Jpn., 74, No. 5 (2001)
Prebiotic Formation of RNA at 40–80 ˚C
14 S. L. Miller and J. L. Bada, Nature, 334, 609 (1988).
15 “Special Issue-Marine Hydrothermal Systems and the Ori-
gin of Life,” ed by N. G. Holm, Origins Life Evol. Biosphere, 22,
5-242 (1992).
16 W. L. Marshall, Geochim. Cosmochim. Acta, 58, 2099
(1994).
of oligo(G)s and the difficulty of the template-directed forma-
tion at higher temperatures is due to decreasing the relative
magnitude of the 2-mer formation rate to other pathways.
Conclusions
The present study showed that a template-directed synthesis
is possible up to 60 ˚C and that a small amount of oligo(G)
forms even at 80 ˚C. In these reactions, the oligo(G) formed
by the template-directed reaction is gradually hydrolyzed at 60
˚C and rapidly at 80 ˚C. The rate constants for the template-di-
rected formation of oligo(G) at 40–60 ˚C were determined.
Besides, the kinetics of the hydrolytic degradation of ImpG
and oligo(G) suggested that both of the hydrolyses are consis-
tent with pseudo-second-order processes at 0.04 M Zn2ꢀ.
Based on a comparison of these reaction rates, it was described
that the low yield of oligo(G) at 80 ˚C is not only due to the hy-
drolysis of oligo(G), but also the relatively slow rate of the 2-
mer formation.
17 N. G. Holm and E. M. Anderson, Planet Space Sci., 43,
153 (1995).
18 S. W. Fox, Geochim. Cosmochim. Acta, 59, 1213 (1995).
19 E. Imai, H. Honda, K. Hatori, A. Brack, and K. Matsuno,
Science, 283, 831 (1999).
20 E. Imai, H. Honda, K. Hatori, and K. Matsuno, Origin Life
Evol. Biosphere, 29, 249 (1999).
21 N. R. Pace, Cell, 65, 531 (1991).
22 T. Oshima, Trans. Mat. Res. Soc. Jpn., 19B, 1069 (1994).
23 T. Oshima, “Seimei ha Nessui kara Hajimatta,” Tokyo Ka-
gaku Dojin, Tokyo, (1995).
24 P. Forterre, C. R. Sciences de la vie/Life sciences, Acad.
Sci. Paris, 318, 415 (1995).
Thus, these results suggest that a higher yield in the produc-
tion of oligo(G) would be possible if the degradation of oli-
go(G)•poly(C) complex can be inhibited and/or the template-
directed synthesis can be started using primers longer than 2-
mer. In other words, it is concluded that the template-directed
formation of RNA from activated monomers was difficult un-
less the process started with 2-mer or longer, and unless there
was some mechanism for protection from the hydrolysis of oli-
go(G) under hydrothermal environments. The present study
and future investigations would provide support to design a
RNA world model in vitro, which could have survived under
hydrothermal environments.
25 P. Forterre, Cell, 85, 789 (1996).
26 T. Oshima, Viva Origino, 25, 109 (1997).
27 N. Galtier, N. Tourasse, and M. Gouy, Science, 283, 220
(1999).
28 S. L. Miller and A. Lazcano, J. Mol. Evol., 41, 689 (1995).
29 L. L. Campbell and B. Pace, B. J. Appl. Bacteriol., 31, 24
(1968).
30 R. H. White, Nature, 310, 430 (1984).
31 Y. Qian, M. Engel, S. A. Macko, S. Carpenter, and J. W.
Deming, Geochim. Cosmochim. Acta, 57, 3281 (1993).
32 H. C. Helgeson and J. P. Amend, Thermochimi. Acta, 245,
89 (1994).
33 J. Bada, S. L. Miller, and M. Zhao, Origins Life Evol. Bio-
sphere, 25, 111 (1995).
34 R. Larralde, M. Robertson, and S. L. Miller, Proc. Natl. Ac-
ad. Sci. U.S.A., 92, 8158 (1995).
35 M. Levy and S. L. Miller, Proc. Natl. Acad. Sci. U.S.A., 95,
7933 (1998).
36 K. Kawamura, A. Yosida, and O. Matumoto, Viva Origino,
25, 177 (1997).
We thank Professor Taketoshi Nakahara for the use of
HPLC apparatus and Dr. Hiroyasu Ogino for the use of UV-vis
spectrophotometer in Osaka Prefecture University. Professor
G. von Kiedrowski in Ruhr-Universitaet Bochum for gener-
ously provided SIMFIT. This research was supported by the
Sumitomo Foundation 1998, Japan.
37 K. Kawamura, N. Kameyama, and O. Matumoto, Viva
Origino, 27, 107 (1997).
References
38 K. Kawamura, Nippon Kagaku Kaishi, 1998, 255.
39 K. Kawamura, Chem. Lett., 1999, 125.
40 K. Kawamura, Nucleic Acids Symp. Ser., 42, 289 (1999).
41 K. Kawamura, Bull. Chem. Soc. Jpn., 73, 1805 (2000).
42 R. Lohrmann and L. E. Orgel, J. Mol. Biol., 142, 555
(1980).
43 T. Inoue and L. E. Orgel, J. Mol. Biol., 162, 201 (1982).
44 T. Inoue and L. E. Orgel, Science, 219, 859 (1983).
45 L. E. Orgel, J. Theor. Biol., 123, 127 (1987).
46 H. Sawai, K. Kuroda, and H. Hojo, H. Bull. Chem. Soc.
Jpn., 62, 2018 (1989).
47 H. Sawai, K. Higa, and K. Kuroda, J. Chem. Soc., Perkin
Trans. 1, 1992, 505.
48 J. P. Ferris and G. Ertem, Science, 257, 1387 (1992).
49 G. Ertem and J. P. Ferris, Nature, 379, 238 (1996).
50 J. P. Ferris, A. R. Hill, J. R. Liu, and L. E. Orgel, Nature,
381, 59 (1996).
1
2
M.D. Been and T. R. Cech, Science, 239, 1412 (1988).
C. Guerrier-Takada, K. Gardiner, T. Marsh, N. Pace, and S.
Altman, Cell, 35, 849 (1983).
G. F. Joyce, A. W. Schwartz, S. L. Miller, and L. E. Orgel,
Proc. Natl. Acad. Sci. U.S.A., 84, 4398 (1987).
3
4
5
6
7
T. R. Cech, Proc. Natl. Acad. Sci. U.S.A., 83, 4360 (1986).
W. Gilbert, Nature, 319, 618 (1986).
A. A. Beaudry and G. F. Joyce, Science, 257, 635 (1992).
“The RNA World,” ed by R. F. Gesteland and J. F. Atkins,
Cold Spring Harbor Laboratory Press, New York (1993).
8
9
A. D. Ellington and J. W. Szostak, Nature, 346, 818 (1990).
A. Terfort and G. von Kiedrowski, Angew. Chem., Int. Ed.
Engl., 31, 654 (1992).
10 D. Sievers and G. von Kiedrowski, Nature, 369, 221
(1994).
11 H. Yanagawa and F. Egami, Proc. Jpn. Acad., Ser. B, 54,
331 (1978).
12 H. Yanagawa and K. Kojima, J. Biochem., 97, 1521 (1985).
13 J. A. Baross and S. E. Hoffman, Origins Life, 15, 327
(1985).
51 H. Fakhrai, T. Inoue, and L. E. Orgel, Tetrahedron, 40, 39
(1984).
52 R. Stribling and S. L. Miller, J. Mol. Evol., 32, 289 (1991).
53 A. Kanavarioti, C. F. Bernasconi, D. J. Alberas, and E. E.

K. Kawamura et al.
Bull. Chem. Soc. Jpn., 74, No. 5 (2001) 935
Baird, J. Am. Chem. Soc., 115, 8537 (1993).
54 A. Kanavarioti, Origins Life Evol. Biosphere, 24, 479
(1994).
55 A. Kanavarioti, T. B. Hurley, and E. E. Baird, J. Mol. Evol.,
41, 161 (1995).
56 A. Kanavarioti and E. E. Baird, J. Mol. Evol., 41, 169
(1995).
57 A. Kanavarioti, J. Org. Chem., 63, 6830 (1998).
58 A. Kanavarioti, C. F. Bernasconi, and E. E. Baird, J. Am.
Chem. Soc., 120, 8575 (1998).
59 A. Kanavarioti, L. F. Lee, and S. Gangopadhyay, Origins
Life. Evol. Biosphere, 29, 473 (1999).
60 K. Kawamura and J. P. Ferris, J. Am. Chem. Soc., 116,
7564 (1994).
61 K. Kawamura and J. P. Ferris, Origin Life Evol. Biosphere,
29, 563 (1980).
62 R. Lohrmann, P. K. Bridson, and L. E. Orgel, Science, 208,
1464 (1999).
63 P. K. Bridson and L. E. Orgel, J. Mol. Biol., 144, 567
(1980).
369 (1992).
67 H. Sigel, F. Hofstetter, R. B. Martin, R. M. Milburn, V.
Scheller-Krattiger, and K. H. Scheller, J. Am. Chem. Soc., 106,
7935 (1984).
68 H. Sigel, Inorg. Chim. Acta, 198–200, 1 (1992).
69 A. Kanavarioti, C. F. Bernasconi, D. L. Doodokyan, and D.
J. Alberas, D. J. J. Am. Chem. Soc., 111, 7247 (1989).
70 J. J. Butzow and G. L. Eichhorn, Biopolymers, 3, 95
(1965).
71 J. J. Butzow and G. L. Eichhorn, Biochemistry, 10, 2019
(1971).
72 H. Ikenaga and Y. Inoue, Biochemistry, 13, 577 (1874).
73 J. J. Butzow and G. L. Eichhorn, Nature, 254, 358 (1975).
74 M. N. Lipsett, J. Biol. Chem., 239, 1256 (1964).
75 M. J. Chamberlin, Fed. Proc., 24, 1446 (1965).
76 F. Pochon and A. M. Michelson, Proc. Natl. Acad. Sci.
U.S.A., 53, 1425 (1965).
77 D. M. Gray, J.-J. Liu, R. Ratliff, and F. S. Allen, Biopoly-
mers, 20, 1337 (1981).
78 H. T. Miles and J. Frazier, J. Mol. Biol., 162, 219 (1982).
79 R. J. H. Daves and N. Davidson, Bipolymers, 10, 1455
(1971).
64 J. H. G. van Roode and L. E. Orgel, J. Mol. Biol., 144, 579
(1980).
65 G. F. Joyce, T. Inoue, and L. E. Orgel, J. Mol. Evol., 176,
278 (1984).
80 H. Sawai, S. Totsuka, K. Yamamoto, and H. Ozaki, Nucleic
Acids Res., 26, 2995 (1998).
66 J. P. Ferris and G. Ertem, Origins Life Evol. Biosphere, 22,