ABSTRACT
We report thermodynamic values for binding of the guanosine nucleophile to the ribozyme derived from the
Anabaena
group I intron, and find that they are similar to those measured previously for
the structurally distinct
Tetrahymena
ribozyme. The free energy of binding guanosine 5
'
-monophosphate (pG) at 30
o
C is similar for the two ribozymes. The
[Delta]
H
o'
and
[Delta]
S
o'
for pG binding to the
Anabaena
ribozyme-RNA substrate complex (E
-
S) are 3.4
"
4 kcal/mol and 27
"
10 e.u., respectively. The negligible enthalpic contribution and positive
entropy change were found previously for the
Tetrahymena
ribozyme, and are considered remarkable for a hydrogen-bonding interaction between a nucleotide and a nucleic acid. These
thermodynamic values may reflect conformational changes or water release upon
pG binding that are comparable for the two ribozymes. In addition, the apparent
chemical steps of the two ribozyme reactions share similar activation energies
and a positive
[Delta]
S
}
. It now appears that such thermochemical values for guanosine binding and
activation may be intrinsic properties of the group I intron catalytic center.
An RNA enzyme or `ribozyme' derived from the group I intron of
Tetrahymena
displays unexpected thermodynamic parameters for interaction with its guanosine
substrate. Although guanosine binding clearly involves hydrogen bond formation
(
1
), it exhibits a negligible enthalpic change and a positive entropic change (
2
). Entropically-driven binding is common for nucleic acid-protein interactions, where much surface area is buried with
release of counterions and water molecules (
3
-
8
). However, such thermodynamic behavior is uncommon when protein enzymes bind
single nucleotides (
9
-
12
) and was essentially unprecedented for interactions between nucleic acids,
which usually occur with a negative [Delta]
S
(
13
-
15
). The positive [Delta]
S
for guanosine binding could be explained by either a ribozyme conformational
change or the release of water from the guanosine binding site. It was
therefore of interest to determine whether these properties were peculiar to
the
Tetrahymena
ribozyme, or common to group I introns.
Group I introns are found across phylogenetical boundaries (
16
,
17
). These introns are recognizable as a group because of common nucleotides and
secondary structures in their catalytic cores (
18
,
19
). Specifically, sequence conservation is found in regions that bind the
guanosine nucleophile (
1
) as well as those that position the P1 helix for nucleophilic attack by
guanosine (
20
,
21
).
These group I RNAs undergo self-splicing in a two step reaction in the absence of proteins. In the first
step, the intron binds exogenous guanosine (G) or guanosine 5'-monophosphate (pG) which is subsequently used as a nucleophile to
attack the 5' splice site, producing an intron-3' exon molecule and a free 5' exon that ends with a 3'-hydroxyl. In the second step, the 5' and 3' exons are ligated
together in a reaction that is chemically equivalent to the reverse of the
first step (
22
).
Derivatives of several group I introns have been made that catalyze a reaction
analogous to the first splicing step in an intermolecular fashion, cleaving a
short RNA substrate with multiple turnovers (Fig.
1
;
22
-
25
). These ribozymes allow a fundamental investigation of the kinetic and
thermodynamic properties of RNA intron chemistry (
26
-
31
). The current work makes use of a ribozyme derived from the pre-tRNA
Leu
intron of
Anabaena
, which belongs to the cyanobacteria family, a proposed progenitor of
chloroplasts (
32
-
34
). The
Anabaena
ribozyme (
24
) differs from the
Tetrahymena
ribozyme in size (248 and 389 bases, respectively) and presence of peripheral structure elements. Even within the catalytic core the two ribozymes differ in 22 of 63 nt.
Furthermore, because the
Anabaena
RNA functions naturally in a prokaryotic cellular environment, it is not at all
obvious that its activity
in vitro
should be the same as that of the
Tetrahymena
RNA, which functions naturally in a eukaryotic nucleolar environment.
The following terminology is used throughout the remainder of this manuscript:
k
obs
, an observed rate constant;
k
max
, single-turnover rate constant at given [E] and saturating [pG];
k
cat
(mt), multiple-turnover rate constant with saturating S and pG;
k
rel
, ratio of rate constants;
k
c
, the rate constant of the chemical step at saturating [pG]; IGS, internal guide
sequence; [Delta]
G
o',[Delta]
H
o' and [Delta]
S
o' represent values measured under the reaction conditions by the
methods herein, and the symbols [Delta]
G
o, [Delta]
H
o and [Delta]
S
o represent the true state functions; [Delta]
G
}
and [Delta]
H
}
and [Delta]
S
}
represent values for reaching the transition state of the reaction.
The L-8 HH ribozyme was prepared as described by Zaug
et al
. (
24
). Oligoribonucleotide substrates were synthesized and
32
P-radiolabeled according to previous procedures (
24
).
All kinetics were carried out in 25 mM HEPES (pH 7.5) and 15 mM MgCl
2
unless otherwise indicated. A subsaturating ribozyme concentration of 0.5 [mu]M was used for all single-turnover experiments [
K
m
(S) = 15 " 2 [mu]M). The ribozyme was first preincubated in 25 mM HEPES (pH 7.5), 15 mM MgCl
2
with G or pG for 15 min at 50oC followed by a 2 min incubation at the desired temperature. Reactions were initiated by addition of a trace amount of oligonucleotide substrate that had been preincubated at the same
temperature of the kinetics run. For reactions at lower pH, preincubation of
the ribozyme was done in 10 [mu]l of the same pH 7.5 buffer with guanosine, followed by the addition of the
oligonucleotide substrate in 40 [mu]l of a 25 mM Mes, 15 mM MgCl
2
buffer at the desired lower pH (6.5 or 5.5). Studies involving the ribozyme-substrate complex (E-S) utilized multiple-turnover reactions with an initial substrate concentration of
100 [mu]M [
K
m
(S) = 12 [mu]M]. Reaction tubes were submerged for high temperature experiments.
Typically six to eight portions (3-4 [mu]l each) were removed at specified times and the reaction was stopped
by adding an equal volume of stop buffer consisting of 30 mM EDTA, 10 M urea,
0.01% bromophenol blue, 0.025% xylene cyanol and 0.1* TBE electrophoresis buffer (1* TBE is 0.1 M Tris base, 0.083 M boric acid and 1 mM EDTA). The
reaction products were separated by electrophoresis on a 20% polyacrylamide
[29:1 acrylamide:bis-acrylamide]-8 M urea gel, and the ratio of substrate to product was quantitated
with a Molecular Dynamics PhosphorImager.
For single-turnover experiments, it had been shown that only 2-3% of the starting material was unreactive and correction of the data for this end point did not appreciably change the rate of
reaction; therefore the data shown in this report are all uncorrected. Values of
k
obs
were determined from the slopes of graphs of ln[S/(S + P)] versus time.
k
obs
was then plotted as a function of [pG], and the equation
k
obs
=
k
max
[pG]/{
K
m
(pG) + [pG]}, in which
k
max
is the rate constant for cleavage of the oligonucleotide substrate with
saturating pG, was used to determine the
K
m
(pG).
The focus of these studies was to examine the temperature dependence of the
K
d
(pG) and
k
c
values for the ribozyme reaction (equation
1
). Single-turnover conditions were used because they involve fewer steps than
multiple turnover reactions, thereby minimizing the possibility of kinetic
complexities that might make
K
d
(pG) differ from
K
m
(pG). At subsaturating concentrations of the oligoribonucleotide substrate (S),
k
max
was 0.40/min at 0.5 [mu]M E and 2 mM pG. This compares favorably with the value of 0.21/min reported
by Zaug
et al
. (
24
) at the same ribozyme and guanosine concentration; the difference may be
attributable to the guanosine nucleophile used by Zaug. In addition, we
measured the
K
m
for guanosine [
K
m
(G)] to be ~1.4 times larger (weaker binding to E) than that for guanosine 5'-phosphate [
K
m
(pG)] at 22 and 32oC. This is in accord with observations for the
Tetrahymena
ribozyme which showed slightly weaker binding of G both to free E and to E-S (
37
).
The reaction rate for the
Anabaena
ribozyme is limited largely by the chemical cleavage step at pH <= 7.5 (
24
), so the maximal
k
obs
under saturating [pG] reflects primarily the rate of chemistry. Therefore an
Arrhenius plot of this rate constant is taken to provide thermodynamic
parameters of activation for reaching the transition state of the chemical
step. For studies using subsaturating ribozyme (E + S), the ribozyme
concentrations were varied at the lowest and highest [pG] for the lower and
upper bounds of the temperature range used in this study (0 and 35oC). The observed rate was linear over a 12-fold concentration range of the ribozyme (0.5-6.0 [mu]M E), which indicated a bimolecular reaction with respect
to ribozyme and oligonucleotide substrate. Furthermore, all single turnover
data followed first order kinetics with no evidence of an initial lag or burst
which suggests a rapid pG binding equilibrium with no buildup of any additional intermediates.
To study the reaction with saturating substrate (E-S + pG), multiple turnover reactions had to be utilized. Substrate
concentration was also varied over a 10-fold range (50-500 [mu]M S) at the lowest and highest [pG] tested. Cleavage rate was
independent of [S] at the highest and lowest temperatures (10 and 55oC), which implies a unimolecular process with respect to E and S consistent
with a saturated ribozyme-substrate complex.
It has already been established that
K
m
(G) equals
K
d
(G) for guanosine binding to the
Tetrahymena
ribozyme (
37
). We tested whether such a relationship also held for the
Anabaena
ribozyme. Specifically, if
K
d
is equal to
K
m
, then by definition
K
m
is
k
-1
/
k
1
. This implies that the rate constant for the chemical step (
k
c
) makes a negligible contribution to the
K
m
(G) term.
K
m
= (
k
-1
+
k
c
)/
k
1
2
The equality of
K
m
and
K
d
can be tested by observing the change in
K
m
(G) or
K
m
(pG) upon changing the rate constant for chemistry (
k
c
) of the ribozyme reaction. Two methods were used to alter the rate of
chemistry, varying pH and incorporating a single deoxynucleotide at the cleavage site of the substrate (Table
1
). In the case of the deoxy substrate CU(dU)A
5,
we saw a dramatic drop of 10
-3.6
in the relative rate of cleavage with saturating pG (
k
rel
) with little change in the
K
m
(pG) (Fig.
2
). Furthermore, in spite of a 63-fold difference in the
k
rel
at pH 5.6 compared with 7.5, the respective
K
m
(G) values of 2.5 and 1.2 mM were similar (Table
1
). Both of these results suggest that
k
c
makes a minimal contribution to the
K
m
term, which supports the approximation that
K
m
(G) =
k
-1
/
k
1
=
K
d
(G).
With the approximation that
K
m
(pG) is equal to
K
d
(pG), the temperature dependence of
K
m
can be used to determine the [Delta]
H
o' and [Delta]
S
o' for pG binding using the van't Hoff equation:
ln
K
d
(pG) = [Delta]
H
o'/RT - [Delta]
S
o'/R
3
Thus a plot of 1/T versus ln
K
m
should give a slope that is proportional to [Delta]
H
o' and a y-intercept equal to -[Delta]
S
o'/R. One complication is that the
Anabaena
ribozyme begins to lose activity above 40oC (Fig.
3
A) under subsaturating RNA substrate conditions. Therefore thermodynamic
interpretation of the data at high temperatures becomes problematic, which
limits the useful temperature range to 2-35oC. However, even in this temperature range it is clear that the
slope of the van't Hoff plot is close to zero and the intercept gives rise to a
positive [Delta]
S
(Fig.
3
B); these results imply that the
Anabaena
ribozyme has thermodynamic values similar to those of the
Tetrahymena
ribozyme.
An Arrhenius plot for the reaction rate under multiple-turnover conditions with saturating [pG] gives a description of the
thermodynamics for reaching the transition state in the reaction of E-S-pG -> E-P. The Arrhenius plot was linear over a 45oC temperature range. The E
a
, [Delta]
H
}
and [Delta]
G
}
(at 30oC) for the
Anabaena
ribozyme (24, 23 and 18 kcal/mol, respectively) are quite close to those found
for the
Tetrahymena
ribozyme (29, 28 and 19 kcal/mol, respectively;
2
). The lower [Delta]
S
}
for
Anabaena
(17 " 5 e.u.) versus
Tetrahymena
(32 " 10 e.u.) may be due to the fact that the
Anabaena
kinetics was done at 5 mM higher MgCl
2
concentration. Past experiments with
Tetrahymena
have shown that high MgCl
2
concentrations decrease the [Delta]
S
}
(
2
). The noteworthy feature here is that the positive [Delta]
S
}
contributes to the stabilization of the transition state of the apparent
chemical step for both ribozymes, with a T[Delta]
S
}
at 30oC worth 5.2 kcal/mol for the
Anabaena
ribozyme versus 10 kcal/mol for
Tetrahymena
.
Group I introns share structural features in their catalytic cores, but differ
in their peripheral structures and show some variability in nucleotide sequence
even within their catalytic centers. Furthermore, they operate in nature in
different intracellular environments-prokaryotic, eukaryotic nucleolar and mitochondrial-so they may be `tuned' to have different intrinsic activities.
Thus, it seems reasonable that some mechanistic features established for the
much-studied
Tetrahymena
group I intron will be general for all group I RNAs, while others will be
inconstant. The only way to identify the common features is to test the
reactivity of different introns.
The current work shows that the unusual thermochemical parameters of pG binding
first found for the
Tetrahymena
ribozyme (
2
) are also characteristic of the
Anabaena
ribozyme. These include a near-zero enthalpic contribution, unexpected given that H-bonds are being formed, and a positive entropy change, which means
that the system becomes more disordered upon complex formation. In addition,
both ribozymes show a positive entropy of activation for the apparent chemical
step and thermodynamic coupling between binding of guanosine and of the
oligonucleotide substrate (cf. data herein for
Anabaena
, data of ref.
37
and
38
for
Tetrahymena
). Thus, these features may be general to group I introns, although more introns
will need to be characterized to test this hypothesis. These features join a
growing list of conserved properties which also includes similar affinity of
guanosine binding, stereoselectivity for the R
p
diastereomer of a phosphorothioate at the cleavage site in the RNA, and a log-linear pH-rate profile in the acid range, indicative of loss of a proton in
the transition state for the chemical step (
24
,
35
-
37
).
A positive entropy change, observed for G binding to both the
Anabaena
and
Tetrahymena
ribozymes, is common for a globular protein-substrate interaction but not for a nucleic acid-nucleic acid interaction. As previously discussed for the
Tetrahymena
ribozyme, the increased disorder upon G binding seen with the
Anabaena
ribozyme may be attributed to a conformational change in the ribozyme or, most
simply, to release of bound water from the G-binding site. An RNA aptamer that specifically binds adenosine has been
discovered by
in vitro
selection-amplification (
39
), and the structure of the aptamer-AMP complex has been determined by NMR spectroscopy (
40
,
41
). Study of the thermodynamics of this interaction would further illuminate the
generality of entropically driven binding of nucleosides by RNA.
L.K. is grateful to Lewis & Clark College for the junior sabbatical leave and wishes to thank Art Zaug,
Michael Tanner and Timothy McConnell for technical help and useful advice. We
thank Alice Sirimarco for preparation of the manuscript. This work was
supported by National Institutes of Health Grant GM28039 to T.R.C. T.R.C. is an
investigator of Howard Hughes Medical Institute and an American Cancer Society
Professor. We thank the W. M. Keck Foundation for support of RNA science on the
Boulder campus.
+
Permanent address: Department of Chemistry, Lewis & Clark College, Portland, OR 97219, USA
REFERENCES
Return