Buffer catalyzed cleavage of uridylyl-3′,5′-uridine in aqueous DMSO: Comparison to its activated analog, 2-hydroxypropyl 4-nitrophenyl phosphate

Article abstract of DOI:10.1039/c4ob02682a

Buffer catalysis of the cleavage and isomerization of uridylyl-3′,5′-uridine (UpU) has been studied over a wide pH range in 80% aq. DMSO. The diminished hydroxide ion concentration in this solvent system made catalysis by amine buffers (morpholine, 4-hydroxypiperidine and piperidine) visible even at relatively low buffer concentrations (10-200 mmol L^-1). The observed catalysis was, however, much weaker than what has been previously reported for the activated RNA model 2-hydroxypropyl 4-nitrophenyl phosphate (HPNP) in the same solvent system. In the case of morpholine, contribution of both the acidic and the basic buffer constituent was significant, whereas with 4-hydroxypiperidine and piperidine participation of the acidic constituent could not be established unambiguously. The results underline the importance of using realistic model compounds, along with activated ones, in the study of the general acid/base catalysis of RNA cleavage.

Full text of DOI:10.1039/c4ob02682a

View Article Online
View Journal
Organic &
Biomol ec ular
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: L. Lain, H.
Lonnberg and T. A. Lönnberg, Org. Biomol. Chem., 2015, DOI: 10.1039/C4OB02682A.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/obc

View Article Online
DOI: 10.1039/C4OB02682A
Page 1 of 8
Organic & Biomolecular Chemistry
Journal Name
RSCPublishing
ARTICLE
Buffer catalyzed cleavage of uridylyl-3´,5´-uridine in
aqueous DMSO: comparison to its activated analog,
Cite this: DOI: 10.1039/x0xx00000x
2
-hydroxypropyl 4-nitrophenyl phosphate
a
a
a
L. Lain, H. Lönnberg and T. A. Lönnberg
Received 00th January 2012,
Accepted 00th January 2012
Buffer catalysis of the cleavage and isomerization of uridylyl-3´,5´-uridine (UpU) has been
studied over a wide pH range in 80% aq. DMSO. The diminished hydroxide ion concentration
in this solvent system made catalysis by amine buffers (morpholine, 4-hydroxypiperidine and
DOI: 10.1039/x0xx00000x
www.rsc.org/
-
1
piperidine) visible even at relatively low buffer concentrations (10 – 200 mmol L ). The
observed catalysis was, however, much weaker than what has been previously reported for the
activated RNA model 2-hydroxypropyl 4-nitrophenyl phosphate (HPNP) in the same solvent
system. In the case of morpholine, contribution of both the acidic and the basic buffer
constituent was significant, whereas with 4-hydroxypiperidine and piperidine participation of
the acidic constituent could not be established unambiguously. The results underline the
importance of using realistic model compounds, along with activated ones, in the study of the
general acid/base catalysis of RNA cleavage.
Introduction
cosolute and electrolyte effects is difficult, which complicates
distinguishing between various mechanistic alternatives. The
cleavage is subject to both general acid and general base
catalysis. In basic buffers, a simple general base catalyzed
deprotonation of the attacking 2´-OH, accompanied by a more
or less concerted breakdown of the dianionic phosphorane
Cleavage of RNA phosphodiester bonds by various small
molecular entities has been the object of considerable interest
during the past two decades. Mechanistic studies by Brönsted
1
buffers and metal ion complexes have been motivated by the
desire to understand the action of ribonucleases and
intermediate, undoubtedly takes place (Route A in Scheme
2
ribozymes, on one hand, and by the development of artificial
7,8
1
). The detailed nature of the general acid catalysis occurring
3
ribonucleases, on the other hand. 2-Hydroxypropyl 4-
in less acidic buffers, such as imidazole and morpholine
buffers, seems to be open to different interpretations. Either the
buffer acid catalyzes the breakdown of the monoanionic
phosphorane intermediate obtained without assistance of any
nitrophenyl phosphate (HPNP) has repeatedly been used as a
model compound in these studies, since the cleavage is much
faster than with dinucleoside-3´,5´-monophosphates and easy to
follow by UV spectrophotometry, allowing rapid acquisition of
accurate and extensive experimental data. This compound has
recently been used to study the cleavage of RNA-like
phosphodiester bonds by basic buffers in 80% aq. DMSO.
Since the autoprotolysis of water is markedly suppressed in this
solvent mixture, while the effect on the protonation of amines
remains modest, the hydroxide ion catalyzed cleavage that
8
buffer constituent (Route B in Scheme 1) or a specific acid /
general base catalyzed formation of the monoanionic
phosphorane intermediate is followed by its specific base /
4
7
general acid catalyzed breakdown (Route C in Scheme 1) .
Participation of the buffer acid and base must be sequential,
since no indication of second-order dependence of rate on the
buffer concentration has been observed.
5
6
severely disturbs kinetic studies on buffer catalysis at high pH
is largely eliminated. The results have been implied to shed
light on some mechanistic ambiguities related to general
acid/base catalyzed cleavage of uridylyl-3´,5´-uridine (UpU).
The buffer catalysis of the cleavage of the latter compound is
rather inefficient and high buffer concentrations have been used
The insightful studies of the group of Yatsimirsky with
HPNP also lend support for occurrence of two parallel
pathways; now both a first- and second-order dependence of
rate on the buffer concentration has been reported. The first-
order term refers to general base catalysis (Route A in Scheme
4
1) and the second-order term has been interpreted to result from
7,8
in the kinetic measurements.
Complete elimination of
This journal is © The Royal Society of Chemistry 2013
J. Name., 2013, 00, 1-3 | 1

View Article Online
DOI: 10.1039/C4OB02682A
Organic & Biomolecular Chemistry
Page 2 of 8
ARTICLE
Journal Name
stabilization of the immediate product of the general base buffers and indicators used and the values obtained are
catalyzed reaction by the acidic buffer constituent, in particular summarized in Table 1. For the indicators, the detection
by guanidinium and amidinium ions. However, even piperidine wavelengths are also included. The resulting pKw of 15.81,
and methyl amine buffers exhibit the second order term, when compared to the respective value of 18.38 at room
5
although its contribution to the total cleavage rate is less temperature, shows that the dissociation equilibrium of water
important than with guanidine or amidine buffers. To what is more sensitive to temperature in 80% aq. DMSO than in pure
extent do these observations apply to the cleavage of RNA water. In water, the pKw is decreased by only 1.58 units on
1
2
phosphodiester bonds remains to be answered. 4- going from 25 °C to 90 °C.
Nitrophenoxide ion is as a leaving group several orders of
Table 1 pKa values for various buffer acids and indicators as determined by
ladder titrations in 80% aq. DMSO; T = 90 °C; I(NaClO4) = 0.2 mol L .
magnitude less basic than the 5´-oxyanion of a ribonucleoside
and this difference may well be large enough to result in
mechanistic changes. The hydroxide ion catalyzed hydrolyses
of uridine 3´-alkylphosphates and 3´-arylphosphates, for
example, have different rate limiting steps: formation of a
phosphorane intermediate in the case of arylphosphates and
-1
pK
a
obs / nm
Buffers used for the ladder
-Chloroacetic acid
2
4.90 ± 0.07
6.47 ± 0.09
8.13 ± 0.03
10.99 ± 0.06
12.39 ± 0.08
15.81 ± 0.05
Morpholine
Piperidine
9
1,8-Diazabicyclo[5.4.0]undec-7-ene
breakdown of this species in the case of alkylphosphates.
2
,4,6-Trimethylphenol
Similarly, metal ion catalyzed cleavage of nucleoside 3´-
alkylphosphates and 3´-arylphosphates has been suggested to
Water
Other buffers
1
0
11
4-Cyanophenol
-Hydroxypiperidine
Indicators
2,4-Dinitrophenol
8.14 ± 0.02
7.76 ± 0.02
proceed by departure of alcohol
and aryloxy ion,
4
respectively. To answer this question, we have repeated some
of the previous measurements with HPNP using UpU as a
substrate.
3.72 ± 0.07
5.17 ± 0.03
7.900 ± 0.004
9.45 ± 0.01
10.69 ± 0.06
419
626
630
588
500
Bromocresol green
Bromothymol blue
Cresol red
2
,4-Dinitrodiphenylamine
The data given in Table 1 was obtained at buffer
-
1
concentrations ranging from 40 to 80 mmol L . In addition,
some measurements were carried out in the most concentrated
-
1
buffers used in the kinetic studies, i.e. at 200 mmol L
concentration. Deviations up to 0.2 units from the values
obtained at lower buffer concentrations were observed. These
deviations were likely caused by cosolute effects on protonation
of the buffer acid, indicator or both, but the overall effect could
not be broken down into separate contributions. Accordingly,
no corrections with respect to buffer concentration were done,
but the values obtained at low buffer concentrations were used
throughout the work. This should not have any influence on the
mechanistic conclusions drawn. In most concentrated buffers,
where the change of the pKa value is largest, the buffer-
dependent rate is already so high compared to the rate of the
hydroxide ion catalyzed reaction that slight changes in pH do
not appreciably bias the kinetic data of the buffer catalyzed
reaction.
Scheme 1
Results
Determination of the dissociation constants of water and the
buffer acids
The dissociation constants of water and the buffer acids
employed, were determined in 80% aq. DMSO at 90 °C
-
1
pH-rate profile for the cleavage and isomerization of 3´,5´-UpU
in 80% aq. DMSO
(
I(NaClO ) = 0.2 mol L ) by the ladder titration technique
4
5
previously reported by Baughman et al. First, the dissociation
constant of the most acidic indicator (2,4-dinitrophenol) was
determined in a dilute solution of a strong acid. A solution of
this indicator in an acidic buffer allowed determination of the
dissociation constant of the most acidic buffer acid (2-
chloroacetic acid). The same buffer was then used to determine
the dissociation constant of a somewhat less acidic indicator,
and the process was repeated using gradually more basic
buffers and indicators until the most basic indicator was
measured against a dilute sodium hydroxide solution. The
It is well known that 3´,5´-UpU undergoes over a wide pH-
range two transesterification reactions: isomerization to 2´,5´-
UpU and cleavage to uridine and 2´,3´-cyclic UMP that is
subsequently hydrolyzed to a mixture of 2´- and 3´-UMP
1
3
(
Scheme 2).
Fig. 1 shows the pH-rate profile for these
-
1
reactions in 80% aq. DMSO at 90 °C (I(NaClO ) = 0.2 mol L )
4
at buffer concentration zero (for the observed rate constants, see
the supplementary material). The rate profiles reported
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

View Article Online
DOI: 10.1039/C4OB02682A
Page 3 of 8
Journal Name
Organic & Biomolecular Chemistry
ARTICLE
1
3
previously for the same reactions in water are also included second-order dependence of rate on hydronium ion
for reference purposes. To facilitate the comparison of the rates concentration, the second-order reaction taking over on passing
of the hydroxide ion catalyzed reactions in water and in 80% pH 2.4. The observed first-order rate constant for the cleavage
aq. DMSO, the rate constants obtained for the cleavage reaction and isomerization may, hence, be expressed by eqns. (1) and
at pH > 6 are additionally plotted as a function of pOH (inset in (2), respectively, as long as the predominant ionic form is the
Fig. 1). Consistent with the situation in water, the reverse phosphodiester monoanion and the 2´-OH is not deprotonated.
reactions of the isomerization of the 2´,5´- and 3´,5´-UpU are The same equations were used to fit the data obtained in 80%
equally fast (k = k ) and both isomers are cleaved to 2´,3´- aq. DMSO.
1
-1
clv
cUMP as rapidly (k = k = k ). The isomerization rate was
2
3
iso
ꢅꢆꢇ
ꢃꢄ
푐푙푣
+
푐푙푣
+
푐푙푣
ꢂ
퐾ꢈ
푐푙푣
presented as k + k = k to allow direct comparison with the
푘표푏푠 = 푘
2
[ꢀ3푂 ] ꢁ 푘퐻 [H3O ] ꢁ 푘푊 ꢁ
(1)
(2)
1
-1
퐻
ꢉ ꢋꢌ]
[
ꢊ
previously reported plot.
푖푠표
푖푠표
+
푖푠표
+
푖푠표
푊
푘
= 푘퐻
2
[ꢀ 푂 ] ꢁ 푘 [H O ] ꢁ 푘
표푏푠
3 3
퐻
The values obtained for the partial rate constants are
recorded in Table 2. Although there are rather few data points
and the points referring to cleavage in the pH-range 6-9 are
rather scattered due to experimental difficulties in following the
very slow reactions (the longest half-lives were approximately
100 days), some conclusions appear possible. The first order
acid-catalyzed cleavage and isomerization are more prominent
clv
iso
in 80% aq. DMSO, kH being 30 times and kH 8 times as
1
3
large as the corresponding rate constant in water. In contrast,
no exhaustive evidence for the existence of the second-order
term could be obtained. In the case of cleavage, the value
obtained for kH2 is similar to the respective value previously
reported in water but with very large error limits. The pH-
independent region of the cleavage reaction is wider in 80% aq.
DMSO than in water and shifted to higher pH, but the cleavage
rate remains unchanged. The pH-independent isomerization, in
turn, is decelerated by a factor of 3 on going from water to aq.
DMSO.
Scheme 2
Table 2 Rate constants extrapolated to zero buffer concentration for the
partial reactions of cleavage and isomerization of UpU in 80% aq. DMSO; T
-
1
=
4
90 °C; I(NaClO ) = 0.2 mol L . The values in parentheses refer to the
-1
corresponding partial reactions in water at I(NaCl) = 0.1 mol L .
clv
iso
Cleavage (k
)
Isomerization (k )
-
2
2
-2 -1
a
k
k
H2/ 10 L mol s
-1 -1
8 ± 14 (7.7)
N/A (12.5)
-
3
H
/ 10 L mol s
-1
11 ± 5 (0.33)
3 ± 1 (2.9)
2.3 ± 0.7 (0.14)
14.6 ± 0.8 (1.8)
41.2 ± 0.9 (130)
-
8
k
W
/ 10
s
-1
-1
b
k
OH / L mol
s
N/A
a
Could not be determined reliably – a negative value was obtained by fitting
clv
to eqn. (2).
2 3
Figure 1 pH-rate profiles for the buffer-independent cleavage (, k = k = k )
b
iso
Isomerization was not detected due to the overwhelmingly faster cleavage.
and mutual isomerization (
DMSO; T = 90 °C; I(NaClO
, k + k-1 = k ) of 2´,5´- and 3´,5´-UpU in 80% aq.
) = 0.2 mol L . The dashed lines refer to the respective
reactions in water. The inset shows the rate constants for the cleavage reaction
as a function of pOH.
1
-1
4
1
3
At a given pH, the hydroxide ion catalyzed cleavage is
slower in aq. DMSO than in water, since the hydroxide ion
concentration at a fixed pH in water is 2500 times as high as in
The cleavage and isomerization are both acid catalyzed and
approximately equally fast at pH < 5. At higher pH, the
isomerization turns pH-independent, and remains so at least
until pH 9. The rate of the cleavage reaction passes through a
broad minimum at pH 6-7 and then turns hydroxide ion
catalyzed. Evidently the cleavage additionally proceeds by a
pH-independent mechanism, but only over a narrow pH-range
around pH 6.5. Accordingly, the overall shape of the profiles is
quite similar in water and in 80% aq. DMSO. In water, the
cleavage and isomerization of UpU show both first- and
8
0% aq. DMSO, owing to the 3.4 unit difference in the pKw
values. When the comparison is made at a fixed hydroxide ion
concentration, the reaction is 17 times as fast in 80% aq.
DMSO as in water. In contrast to the situation in water, no sign
of levelling-off of the rate to a zero-order dependence at high
hydroxide ion concentrations is observed. The cleavage was
-
-1
already too fast to be followed at [OH ] = 0.1 mol L ,
indicating that hydroxide ion catalysis still prevails beyond the
-
-
last reliable experimental point obtained at [OH ] = 0.01 mol L
1
.
This means that the pK value of the 2´-hydroxy function
a
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 3

View Article Online
DOI: 10.1039/C4OB02682A
Organic & Biomolecular Chemistry
Page 4 of 8
ARTICLE
Journal Name
must be higher than 14, while the corresponding value in water The reaction rates in both piperidine and 4-hydroxypiperidine
1
3
is 11.5.
buffers are directly proportional to total buffer concentration
and increase with the increasing content of the basic buffer
Cleavage and isomerization in amine buffers
constituent. When the rate constant for the buffer catalysis
clv
A
MIDINE AND GUANIDINE BUFFERS. As discussed above, (kbuf ) was broken down to contributions of the acidic and
amidine and guanidine buffers exhibited the most efficient basic components by fitting to eqn. (3)
4
buffer catalysis on using HPNP as a substrate. To allow direct
푟
푘푐푎푡 = ( ) 푘
푟+1
1
퐵퐻ꢌ
ꢁ (
) 푘퐵
(3)
comparison with these results, the first experiments on the
cleavage and isomerization of 3´,5´-UpU were carried out in
acetamidine and benzamidine buffers. Unfortunately, these two
amidines turned out to decompose too fast at 90 °C to allow
kinetic measurements. Even attempts to determine the rate
constants by the method of initial rates (based on withdrawal of
aliquots at intervals ranging from a few minutes to a few hours,
the half-life being several days) proved inconclusive.
Accordingly, those buffers catalyze the cleavage reaction much
less efficiently than what could be expected on the basis of the
several orders of magnitude acceleration observed with HPNP.
In such a case, detectable amounts of cleavage products should
have been formed in one minute.
푟+1
+
clv
-7
-1
where r = [BH ] / [B], the values of k
= (12 ± 1) × 10 L s
B
-
1
clv
-7
-1
-1
mol and kBH+ = (2 ± 1) × 10 L s mol were obtained for
clv
-7
-1
-1
piperidine catalysis, and k
= (36 ± 6) × 10 L s mol and
B
clv
-7
-1
-1
kBH+
= (5 ± 6) × 10 L s mol for 4-hydroxypiperidine
catalysis. In both cases, the value for acid-catalyzed
contribution is comparable to the uncertainty in the fitting.
Accordingly, no conclusive evidence for existence of general
acid catalysis could be obtained. Even if such a pathway exists,
its contribution compared to the general base catalyzed reaction
clv
is minor. Consistent with this conclusion, the values (k
–
obs
clv
kbkg
)
referring to all different buffer ratios fall on a single
straight line when plotted against the concentration of the basic
constituent (Fig. 3).
P
IPERIDINE AND 4-HYDROXYPIPERIDINE BUFFERS. Fig. 2
shows the observed first-order rate constants for the cleavage of
UpU as a function of buffer concentration in piperidine (A) and
4-hydroxypiperidine (B) at three different ratios of the basic
and acidic buffer constituent. While the cleavage clearly is
subject to buffer catalysis, the isomerization is not. The rate of
isomerization is increased by only 10-25% on going from
-
1
-1
buffer concentration 0.010 mol L to 0.20 mol L , except in
the most acidic 4-hydroxypiperidine buffer, in which case the
increase is 40%. The rate constants obtained for the buffer-
catalyzed and buffer-independent (background) cleavage and
isomerization by linear least-squares fitting are given in Table
3.
Table 3 Rate constants for the buffer catalyzed and buffer-independent
cleavage and isomerization of UpU in amine buffers; T = 90 °C; I(NaClO
4
) =
-
1
0
.2 mol L .
clv
clv
iso
buf
-7
iso
Buffer
[B]:
BH ]
pH
k
10
mol
buf
/
k
10
bkg
/
k
/
k
10
bkg
/
+
-7
-7 -1
-7 -1
[
L
s
10 L
s
-
1
-1
-1 -1
s
mol s
Piperidine
3:1
8.61
8.13
7.65
8.25
7.77
7.29
6.95
6.47
5.99
9 ± 1
8 ± 1
0.8 ±
.2
0.2 ±
.1
0.25 ±
.05
1.8 ±
0.3
1.8 ±
2.2 ± 0.5
2.5 ± 0.7
1.5 ± 0.6
1.9 ± 0.7
1.0 ± 0.7
4.5 ± 0.6
9 ± 3
2.0 ±
0.1
1.9 ±
0.1
2.1 ±
0.1
1.8 ±
0.1
2.1 ±
0.1
2.2 ±
0.1
2.3 ±
0.5
2.1 ±
0.3
0
1
1
:1
:3
0
4.6 ± 0.4
30 ± 3
17 ± 2
15 ± 2
9.4 ± 0.5
18 ± 1
10 ± 3
0
4
-Hydroxy-
3:1
piperidine
1
1
:1
:3
0
.2
0.6 ±
.3
0.21 ±
.07
0.3 ±
.1
0.5 ±
.4
0
Morpholine
3:1
0
1
1
:1
:3
10 ± 3
0
clv
4.7 ± 0.5
2.2 ± Figure 2 Observed first-order rate constants for the cleavage of UpU (kobs ) in
0.1 (A) piperidine and (B) 4-hydroxypiperidine buffers as a function of total buffer
0
+
concentration at [BH ]:[B] = 3:1 (), 1:1 (

) and 1:3 (

); T = 90 °C; I(NaClO
4
) =
-
1
0.2 mol L .
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

View Article Online
DOI: 10.1039/C4OB02682A
Page 5 of 8
Journal Name
Organic & Biomolecular Chemistry
ARTICLE
MORPHOLINE BUFFERS
.
The kinetic data obtained in the
somewhat more acidic morpholine buffers differ markedly from
those presented above for piperidine and 4-hydroxypiperidine
buffers. The common feature is that in all three buffers, the
cleavage rate is first-order in buffer concentration (Figs. 2 and
4
). With morpholine, however, the cleavage is faster in the
+
[
B]:[BH ] = 1:1 buffer than in 1:3 or 3:1 buffers. In addition,
the isomerization, faster than cleavage, is also subject to modest
buffer catalysis. As with piperidine, the effect of ionic strength
remains negligible. The rate constants for buffer catalysis are
summarized in Table 3.
Discussion
Buffer-independent reactions
RNA phosphodiester bonds are known to be cleaved over a
clv
clv
Figure 3 Rate constants for the buffer catalyzed cleavage of UpU (kobs – kbkg
in piperidine and hydroxypiperidine ( buffers plotted against the
concentration of the basic buffer constituent; T = 90 °C; I(NaClO
)
wide pH range via formation of
a
pentacoordinated
(

)
)
-1
phosphorane intermediate obtained by the attack of the 2´-
4
) = 0.2 mol L .
1
,2
hydroxy group on the phosphorus atom. Depending on pH,
the reaction is second- or first-order in hydronium ion
concentration, pH-independent or first-order in hydroxide ion
concentration, passing through various ionic forms of the
phosphorane intermediate (Scheme 3). Among the
oxyphosphoranes obtained, the monocationic, neutral and
monoanionic species are decomposed by two different
pathways: cleavage of the P-O5´ bond leads to cleavage of the
phosphodiester bond, while departure of O3´ results in
isomerization to a 2´,5´-bond. The dianionic phosphorane, in
turn, gives only cleavage products.
Both cleavage and isomerization were insensitive to ionic
strength. When the piperidine catalyzed reactions were studied
at ionic strength ranging from 0.1 to 0.2 mol L , no marked
change in the rate was observed.
-
1
Scheme 3
As indicated above (Fig. 1), the acid-catalyzed cleavage and
isomerization of UpU is faster in 80% aq. DMSO than in water,
but shows only first-order dependence of rate on hydronium ion
concentration. In other words, the reactions proceed only via
neutral phosphorane, not via a cationic species. This difference
can tentatively be attributed to 80% aq. DMSO being a poorer
solvent for the ionic species than water. The first-order acid
catalyzed reactions proceeding via a neutral (monoprotic)
phosphodiester are accelerated, whereas the second protonation,
clv
iso
Figure 4 Rate constants for the cleavage (A, kobs ) and isomerization (B, kobs
)
of UpU in morpholine buffers as a function of total buffer concentration at
BH ]:[B] = 3:1 (), 1:1 () and 1:3 (); T = 90 °C; I(NaClO ) = 0.2 mol L .
4
+
-1
[
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 5

View Article Online
DOI: 10.1039/C4OB02682A
Organic & Biomolecular Chemistry
Page 6 of 8
ARTICLE
Journal Name
giving a phosphodiester monocation, seems to be too difficult observation is that the buffer base competes with the phosphate
to contribute appreciably. group for the O2´ proton. In other words, the reaction proceeds
The rate of the pH-independent cleavage that proceeds via a by mechanism A in Scheme 1. The dianionic phosphorane
monoanionic phosphorane is approximately as fast as in water, intermediate obtained is too unstable to pseudorotate and,
while the isomerization is retarded by a factor of 3 compared to hence, no isomerization can take place. The marginally stable
the respective reaction in water. These uncatalyzed reactions do dianionic phosphorane collapses to 2´,3´-cyclic phosphate and
not involve net build-up of charges and their rate is therefore 5´-oxyanion by unimolecular rupture of the P-O5´ bond.
only modestly affected. The somewhat lower rate of the pH- Departure of the 5´-oxyanion may, in principle, be facilitated
independent isomerization might tentatively be attributed to by concomitant protonation, but if so, the protonation takes
lower stability of the phosphorane intermediate. In other words, place after the rate limiting formation of the marginally stable
pseudorotation that is a prerequisite for isomerization competes dianionic phosphorane and is, hence, kinetically invisible. One
less efficiently with the in-line departure of the 5´-linked might also speculate that deprotonation of the monoanionic
nucleoside.
phosphorane is subject to general base catalysis. This
The hydroxide ion catalyzed reaction is approximately one alternative, however, appears less likely, since isomerization is
order of magnitude faster in 80% aq. DMSO than in water, in not accelerated. Either the 2´ or the 3´-oxygen within the
spite of the fact that deprotonation of the 2´-hydroxy group phosphorane intermediate is always apical and deprotonation of
seems to be retarded in 80% aq. DMSO. No levelling off to a the nonbridging oxygen should accelerate collapse of the
constant value is observed, in contrast to the situation in water. intermediate, not only to cleavage products, but also to
Evidently, the 2´-oxyanion, although formed in a lesser extent, isomerization products.
is less solvated in 80% aq. DMSO than in water and, hence,
Catalysis by the somewhat less basic morpholine buffer
serves as a better nucleophile, resulting in overcompensation of differs in two respects from piperidine catalysis. Firstly,
+
the more difficult deprotonation. In aqueous solution, a rather catalysis of the cleavage reaction is in a 1:1 buffer ([BH ]:[B] =
+
high βnuc value of 0.75 has been reported for the hydroxide ion 1:1) twice as efficient as in a more acidic 3:1 ([BH ]:[B] = 3:1)
+
catalyzed cleavage of RNA phosphodiester linkages, lending or less acidic 1:3 ([BH ]:[B] = 1:3) buffer (Fig. 4A). Secondly,
1
4
support to this interpretation.
morpholine buffer catalyzes additionally isomerization, though
less clearly than cleavage (Fig. 4B). The catalysis of
isomerization is more marked in the more acidic 3:1 buffer than
Buffer catalyzed reactions
UpU undergoes in piperidine and 4-hydroxypiperidine buffers in the more basic 1:3 buffer, suggesting a general acid
cleavage and isomerization at comparable rates, the buffer- catalyzed mechanism. In one respect the experimental data,
independent isomerization being faster than cleavage (Table 3). however, appears controversial. The fact that the cleavage
Both buffer-independent reactions are also pH-independent reaction is most efficiently catalyzed in a 1:1 buffer suggests
suggesting that they proceed via a monoanionic phosphorane that both the buffer acid and buffer base participate in a single
obtained by proton transfer from 2´-OH to the phosphoryl transition state and, hence, the cleavage should be second-order
oxygen concerted with the attack of the developing 2´-oxyanion in buffer concentration. However, no indication of second-order
1
,2e,13,15,16
on the phosphorus atom (Scheme 4).
Different dependence could be obtained. Possibly, cosolute effects
opinions exist on whether the proton transfer between oxygen together with simultaneous appearance two different
1
7
atoms takes place directly or is mediated through a water mechanisms (discussed below) obscure the appearance of the
1
5b,18
molecule.
second-order term. Anyway, it appears clear that on going from
piperidine buffers to less basic morpholine buffers, general acid
catalysis starts to compete with the general base catalysis
(
mechanism C in Scheme 1), but no firm conclusions on the
detailed mechanism of this catalysis can be drawn. Possibly, the
monoanionic intermediate is under these slightly more acidic
conditions obtained by pre-equilibrium protonation of the
phosphodiester linkage followed by general base catalyzed
attack of the 2´-hydroxy function. Accordingly, the formation
of the phosphorane intermediate would be general acid
catalyzed (Scheme 5). More or less concerted with this, the
breakdown of the intermediate could be facilitated by general
base catalyzed deprotonation of the monoanionic phosphorane.
Since the uncatalyzed cleavage of the endocyclic P-O bonds is
Scheme 4
1
9
In spite of existence of a common intermediate, only much faster than uncatalyzed exocycylic cleavage , the
cleavage is subject to buffer catalysis. The reaction is first-order isomerization is less dependent on buffer catalysis than
in the concentration of the buffer base (Fig. 3), the contribution cleavage. In parallel with this general acid general base
of the buffer acid to the catalysis being a minor one or catalyzed reaction (that resembles Route C in Scheme 1),
nonexistent. The most straightforward explanation for this general base catalyzed cleavage via a dianionic phosphorane
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012

View Article Online
DOI: 10.1039/C4OB02682A
Page 7 of 8
Journal Name
Organic & Biomolecular Chemistry
ARTICLE
(
Route A) undoubtedly takes place. Competition between solution. NaOH solutions in 80% aq. DMSO were found to be
Routes A and B in Scheme 1 cannot be strictly excluded either, unstable, so NaOH was added as an aqueous solution and neat
but the higher catalytic efficiency of the 1:1 buffer compared to DMSO was then added to compensate. The error introduced
more and less acidic buffers cannot be explained on this basis.
should be minimal, as we determined that the ∆V for mixing
water and DMSO is very small. In all other cases, the stock
solutions were prepared in 80% aq. DMSO.
Determination of pK and pK
a
W
The UV spectra were measured with a Perkin Elmer Lambda-
5 instrument equipped with temperature control. The blank
3
solution was 0.2 M NaClO in 80% aq. DMSO. The total buffer
4
concentration used was between 40 and 80 mM. The cuvettes
were left in the instrument for 10 min to thermostate before
measuring the spectra; repeated measurements at different time
intervals showed that the spectra did not change significantly
after that. Two of the indicators used, namely 2,4-
dinitrodiphenylamine and 2,4-dichloro-4-nitroaniline, were
found to be unstable at 90 °C; in those cases, the solution was
instead prepared without the indicator and thermostated to 90
Scheme 5
Comparison to buffer catalyzed reactions of HPNP
The buffer catalyzed cleavage is much more modest with UpU
much than with HPNP. Accordingly, decreasing the hydroxide
ion concentration by going from water to 80% aq. DMSO did
not allow such detailed study of the buffer catalyzed reactions
as reported for HPNP. While a 1:1 piperidine buffer at a
concentration of 0.1 mol L accelerated the cleavage of HPNP
°
C, after which the indicator was quickly added, the cuvette
shaken, and the measurement carried out as quickly as possible.
As the temperature control in those cases was far from ideal, a
moderate systematic error is possible, but if so we are unable to
quantify it. Each buffer-indicator pairing was measured at least
three times.
For most buffers, the absorbance was measured at the
wavelength corresponding to the absorption maximum of the
deprotonated form, where absorption of the protonated form
4
-
1
3
by a factor of 10 compared to the background reaction, with
UpU the acceleration was only 4-fold. Another major
difference was that the cleavage of HPNP exhibited both first-
and second-order dependence of the rate on buffer
-
1
+
concentration. In the 0.1 mol L piperidine buffer ([BH ]:[B] =
:1), the second order term was responsible for 40% of the total
1
was
negligible.
The
only
exception
was
2,4-
cleavage. With UpU, no second-order dependence of cleavage
rate on buffer concentration could be observed in piperidine
buffer. Unfortunately the guanidine and amidine buffers that
exhibited the most prominent second-order dependence with
HPNP, could not be used in the present study due to their
inherent instability at elevated temperature.
dinitrodiphenylamine, in which case the most reproducible
point was not the absorption maximum, and the absorbance of
the acidic form was small but significant enough that it had to
be subtracted.
Kinetic measurements
UpU undergoes in piperidine, 4-hydroxypiperidine and
morpholine buffers cleavage and isomerization at comparable
rates. In piperidine buffers only cleavage was subject to buffer
catalysis. In morpholine buffers that are 1.7 logarithmic units
more acidic than piperidine buffers, isomerization, as well as
cleavage, appeared to be buffer catalyzed and some evidence
for simultaneous participation of buffer acid and buffer base
was obtained. With HPNP, isomerization has not been reported
to compete with cleavage. In fact, it has been previously shown
The reactions were carried out in sealed tubes immersed in a
thermostated water bath (90.0 ± 0.1 °C). For each buffer,
measurements were carried at 3:1, 1:1 and 1:3 [HB]:[B] ratio
and several total buffer concentrations ranging from 0.01 to 0.2
-1
mol L . The ionic strength of the solutions was kept constant at
-1
0
.20 mol L with sodium perchlorate. The initial UpU
-1
concentration was approximately 0.5 mmol L . Potassium p-
nitrobenzensulfonate was added at sub-millimolar
concentration to serve as an internal standard. The reactions
were typically followed over 1-2 half-lives. In the case of the
slowest reactions (t½ up to 100 days) shorter reaction times
were used, down to 20% of a half-life.
that
uridine
3´-(2-chlorophenyl)phosphate
undergoes
2
0
isomerization only at pH < 2.
Experimental section
The composition of the samples withdrawn at appropriate
time intervals was analyzed by HPLC on a Hypersil-Keystone
Aquasil C18 column (4 × 150 mm, 5 μm) using a mixture of
General Methods
-
1
All the reagents, including UpU, were commercially available acetonitrile and a 60 mmol L acetate buffer (pH 4.3)
-
1
and used as received. The NaOH solutions were prepared from containing 0.1 mol L ammonium chloride as an eluent. To
Dilute-it ampoules and replaced after 60 days to prevent bring the concentration of DMSO down to a level where it does
spoiling. All the protonated buffer solutions and the HClO4 not interfere with the elution of the other components, the
solution were titrated with these standard NaOH solutions. The samples were diluted 1:10 with water before injection. For the
deprotonated buffers, in turn, were titrated with the HClO4 first 8 min, isocratic elution with 100% buffer was used,
This journal is © The Royal Society of Chemistry 2012
J. Name., 2012, 00, 1-3 | 7

View Article Online
DOI: 10.1039/C4OB02682A
Organic & Biomolecular Chemistry
Page 8 of 8
ARTICLE
Journal Name
followed by a linear acetonitrile gradient from 0 to 30% during
Curr. Opin. Chem. Biol., 2005,
9
, 665-673; (f) T. Lönnberg, Chem.
-
1
22 min. A constant flow rate of 1 mL min and a detection
Eur. J., 2011, 17, 7140-7153; (g) K.-Y. Wong, H. Gu, S. Zhang, J.
A. Piccirilli, M. E. Harris and D. M. York, Angew. Chem. Int. Ed.,
2012, 51, 647-651; (h) H. Korhonen, N. Williams and S. Mikkola, J.
Phys. Org. Chem., 2013, 26, 182-186; (i) H. Gu, S. Zhang, K.-Y.
Wong, B. K. Radak, T. Dissanayake, D. L. Kellerman, Q. Dai, M.
Miyagi, V. E. Anderson, D. M. York, J. A. Piccirilli and M. E.
Harris, Proc. Natl. Acad. Sci. U.S.A., 2013, 110, 13002-13007.
wavelength of 260 nm were employed. The observed retention
times (tR / min) for UpU, its hydrolytic products and the
internal standard were as follows: 5.5 (uridine), 9.7 (p-
toluenesulfonate), 15.3 (2´,5´-UpU) and 16.5 (3´,5´-UpU). The
products were characterized by spiking with authentic samples.
The peak areas were converted to relative concentrations by
dividing them with the peak area of the internal standard (p-
toluenesulfonate). The molar absorptivities of the 2´- and 3´-
isomers of UpU were assumed to be identical and twice as high
as the molar absorptivity of uridine. Pseudo-first-order rate
constants for the cleavage and isomerization reactions were
calculated by numerical fitting of the experimental data to the
appropriate differential rate equations. Our previous experience
has shown data obtained by this method to be in good
agreement with those obtained by analytical fitting to the
3
(a) J. R. Morrow and O. Ironzo, Curr. Opin. Chem. Biol., 2004, 8,
192-200; (b) J. Weston, Chem. Rev., 2005, 105, 2151-21744; (c) N.
Mitic, S. J. Smith, A. Neves, L. W. Guddat, L. R. Gahan and H.
Schenk, Chem. Rev., 2006, 106, 3338-3363; (d) F. Mancin and P.
Tecilla, New J. Chem., 2007, 31, 800-817; (e) D. Desbouisa, I. P.
Troitskyb, M. J. Belousoffa, L. Spiccia and B. Graham, Coord.
Chem. Rev., 2012, 256, 897-937.
4
5
D. O. Corona-Martínez, O. Taran and A. K. Yatsimirsky, Org.
Biomol. Chem., 2010,
E. H. Baughman and M. M. Kreevoy, J. Phys. Chem., 1974, 78, 421-
23.
Z. Pawlak and R. G. Bates, J. Solution Chem., 1975,
(a) E. Anslyn and R. Breslow, J. Am. Chem. Soc., 1989, 111, 4473-
8, 873-880.
2
1
integrated rate equations.
4
Conclusions
6
7
4, 817-829.
In summary, it appears clear that care should be exercised on
drawing mechanistic conclusions concerning the hydrolysis of
RNA on the basis of the behavior of structurally simplified
models, such as HPNP. Replacement of the 5´-linked
nucleoside with an aryl group evidently increses the
electrophilicity of the phosphorus atom, but above all
destabilizes the oxyphosphorane intermediate. Replacement of
the cyclic ribofuranosyl moiety of the 3´-linked nucleoside with
an acyclic 1,2-propanediol, in turn, makes the intramolecular
nucleophilic attck on phosphorus entropically less favorable,
retarding the formation of phosphorane intermediate. The
overall effect on not only the kinetics, but also the timing of
various elementary processes and, hence, the mechanism, is
noteworthy.
4
482; (b) R. Breslow and D.-L. Huang, J. Am. Chem. Soc., 1990,
12, 9621-9623; (c) R. Breslow, S. D. Dong, Y. Webb and R. Xu, J.
1
Am. Chem. Soc., 1996, 118, 6588-6600.
8
(a) C. Beckmann, A. J. Kirby, S. Kuusela and D. C. Tickle, J. Chem.
Soc. Perkin Trans. 2, 1998, 573-582; (b) A. J. Kirby and R. E.
Marriott, J. Chem. Soc. Perkin Trans. 2, 2002, 422-427.
H. Lönnberg, R. Strömberg and A. Williams, Org. Biomol. Chem.,
9
1
2
004, 2, 2165-2167.
0 S. Mikkola, E. Stenman, K. Nurmi, E. Yousefi-Salakdeh, R.
Strömberg and H. Lönnberg, J. Chem. Soc. Perkin Trans. 2, 1999,
1
619-1625.
1
1 M. Komiyama, Y. Matsumoto, H. Takahashi, T. Shiiba, H. Tsuzuki,
H. Yajima, M. Yashiro and J. Sumaoka, J. Chem. Soc. Perkin Trans.
2
, 1998, 691-696.
2 A. V. Bandura and S. N. Lvov, J. Phys. Chem. Ref. Data, 2006, 35
5-30.
3 P. Järvinen, M. Oivanen and H. Lönnberg, J. Org. Chem., 1991, 56
396-5401.
4 J.-D. Ye, N.-S. Li, Q. Dai and J. Piccirilli, Angew. Chem. Int. Ed.,
007, 46, 3714-3717.
1
1
1
1
,
,
Acknowledgements
1
Financial support from the FP7 Marie Curie Actions is
gratefully acknowledged.
5
2
Notes and references
5 (a) M. Kosonen, E. Yousefi-Salakdeh, R. Strömberg and H.
Lönnberg, J. Chem. Soc. Perkin Trans. 2, 1998, 1589-1595; (b) B.
Gerratana, G. A. Sowa and W. W. Cleland, J. Am. Chem. Soc.,
a
Department of Chemistry, University of Turku, Vatselankatu 2, FIN-
2
0014 Turku, Finland. Email: tuanlo@utu.fi
†
Electronic Supplementary Information (ESI) available: observed rate
2
000, 122, 12615-12621.
constants for the cleavage and isomerization of UpU in morpholine, 4-
hydroxypiperidine, piperidine and 1,3,5-trimethylphenol buffers of
various concentrations and ratios of the acidic and basic buffer
constituents. See DOI: 10.1039/b000000x/
1
6 D. L. Kellerman, D. M. York, J. A. Piccirilli and M. E. Harris, Curr.
Opin. Chem. Biol., 2014, 21, 96-102.
1
1
7 Y. Yang and Q. Cui, J. Phys. Chem. B, 2009, 113, 4930-4939.
8 M. Boero, K. Terakura and M. Tateno, J. Am. Chem. Soc., 2002, 124
949-8957.
9 C. S. Lopez, O. N. Faza, B. A. Gregersen, X. Lopez, A. R. de Lera
and D. M. York, ChemPhysChem, 2004, , 1045-1049.
0 M. Oivanen and H. Lönnberg, Acta Chem. Scand., 1991, 45, 968-
71.
1 L. Lain, H. Lönnberg and T. Lönnberg, Chem. Eur. J., 2013, 19
2424-12434.
,
8
1
2
H. Lönnberg, Org. Biomol. Chem., 2011,
(a) D. M. Perrault and E. V. Anslyn, Angew. Chem. Int. Ed., 1997,
, 432-450; (b) M. Oivanen, S. Kuusela and H. Lönnberg, Chem.
Rev., 1998, 98, 961-990; (c) D.-M. Zhou and K. Taira, Chem. Rev.,
998, 98, 991-1026; (d) Y. Takagi, Y. Ikeda and K. Taira, Top.
9, 1687-1703.
1
2
2
5
3
6
9
1
,
Curr. Chem., 2004, 232, 213-251; (e) T. Lönnberg and H. Lönnberg,
1
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 2012