Transition State Differences in Hydrolysis Reactions of Alkyl versus Aryl Phosphate Monoester Monoanions

Article abstract of DOI:10.1021/ja036571j

Although aryl phosphates have been the subject of numerous experimental studies, far less data bearing on the mechanism and transition states for alkyl phosphate reactions have been presented. Except for esters with very good leaving groups such as 2,4-dinitrophenol, the monoanion of phosphate esters is more reactive than the dianion. Several mechanisms have been proposed for the hydrolysis of the monoanion species. ¹⁸O kinetic isotope effects in the nonbridging oxygen atoms and in the P-O(R) ester bond, and solvent deuterium isotope effects, have been measured for the hydrolysis of m-nitrobenzyl phosphate. The results rule out a proposed mechanism in which the phosphoryl group deprotonates water and then undergoes attack by hydroxide. The results are most consistent with a preequilibrium proton transfer from the phosphoryl group to the ester oxygen atom, followed by rate-limiting P-O bond fission, as originally proposed by Kirby and co-workers in 1967. The transition state for m-nitrobenzyl phosphate (leaving group pK_a 14.9) exhibits much less P-O bond fission than the reaction of the more labile p-nitrophenyl phosphate (leaving group pK_a = 7.14). This seemingly anti-Hammond behavior results from weakening of the P-O(R) ester bond resulting from protonation, an effect which calculations have shown is much more pronounced for aryl phosphates than for alkyl ones.

Full text of DOI:10.1021/ja036571j

Published on Web 10/01/2003
Transition State Differences in Hydrolysis Reactions of Alkyl
versus Aryl Phosphate Monoester Monoanions
Piotr K. Grzyska, Przemyslaw G. Czyryca, Jamie Purcell, and Alvan C. Hengge*
Contribution from the Utah State UniVersity, Department of Chemistry and Biochemistry,
Logan, Utah 84322-0300
Received June 9, 2003; E-mail: hengge@cc.usu.edu.
Abstract: Although aryl phosphates have been the subject of numerous experimental studies, far less
data bearing on the mechanism and transition states for alkyl phosphate reactions have been presented.
Except for esters with very good leaving groups such as 2,4-dinitrophenol, the monoanion of phosphate
esters is more reactive than the dianion. Several mechanisms have been proposed for the hydrolysis of
the monoanion species. 18O kinetic isotope effects in the nonbridging oxygen atoms and in the P-O(R)
ester bond, and solvent deuterium isotope effects, have been measured for the hydrolysis of m-nitrobenzyl
phosphate. The results rule out a proposed mechanism in which the phosphoryl group deprotonates water
and then undergoes attack by hydroxide. The results are most consistent with a preequilibrium proton
transfer from the phosphoryl group to the ester oxygen atom, followed by rate-limiting P-O bond fission,
as originally proposed by Kirby and co-workers in 1967. The transition state for m-nitrobenzyl phosphate
(
leaving group pK
a
14.9) exhibits much less P-O bond fission than the reaction of the more labile
) 7.14). This seemingly anti-Hammond behavior results from
p-nitrophenyl phosphate (leaving group pK
a
weakening of the P-O(R) ester bond resulting from protonation, an effect which calculations have shown
is much more pronounced for aryl phosphates than for alkyl ones.
Introduction
phosphates indicate that the leaving group is largely neutral in
the transition state, implying that P-O bond fission is ac-
The chemistry of phosphate monoesters forms the basis for
the regulation of a host of biochemical processes, accounting
in part for the intense interest in these compounds. Much of
what is known regarding the mechanisms of phosphoryl transfer
and their transition states has been inferred from experiments
using aryl phosphate esters. Conversely, most of the computa-
companied by protonation of the leaving group. Linear free
energy relationships with monoanionic phosphorylated pyridines
have been interpreted to indicate a loose transition state in which
4
metaphosphate is not an intermediate. The hydrolysis of the
monoanion of 2,4-dinitrophenyl phosphate is thought to be
5
concerted, but the possibility of a metaphosphate intermediate
tional work has been performed using methyl phosphate. We
report solvent isotope effects, and 18O isotope effects in the
has not been ruled out with esters having less activated leaving
groups. A stereochemical study of the hydrolysis of phenyl
phosphate monoanion indicates the reaction proceeds with
leaving group and in the phosphoryl group for the hydrolysis
of the alkyl phosphate m-nitrobenzyl phosphate (mNBP). The
results provide a basis for choosing among various mechanistic
proposals. In addition, comparison with data obtained previously
for p-nitrophenyl phosphate reveal differences in the transition
state of the hydrolysis of the aryl versus the alkyl ester.
Depending upon pH, phosphate monoesters exist either as
the monoanion or the dianion (except at very low pH values
where the neutral species exists). With the exception of esters
with highly activated leaving groups such as 2,4-dinitrophenol,
6
inversion. Such a result is consistent either with a concerted
mechanism, or with a discrete metaphosphate intermediate in a
preassociative mechanism. The results reported here do not
address this question, but rather, examine the timing and
synchronicity of proton transfer and P-O bond fission to the
leaving group.
1
Kirby postulated that protonation of the leaving group
accompanies deprotonation of the phosphoryl group (probably
via the intermediacy of one or more water molecules) and
considered the timing of the proton transfer (Figure 1). He
concluded that for less basic leaving groups, protonation occurs
simultaneously with leaving group departure and is partially rate-
limiting, whereas for more basic leaving groups, a bridge-
protonated intermediate forms followed by rate-limiting P-O
bond fission. Consistent with this proposal, a solvent isotope
1
the monoanion species is much more reactive than the dianion.
A large body of experimental evidence indicates that the
reactions of dianions are concerted with loose, metaphosphate-
2
like transition states. In reactions of monoanions, linear free
1
3
energy relationships and kinetic isotope effects with aryl
(
1) Kirby, A. J.; Varvoglis, A. G. J. Am. Chem. Soc. 1967, 89, 415.
2) Thatcher, G. R. J.; Kluger, R. AdV. Phys. Org. Chem. 1989, 25, 99; Hengge,
A. C. In ComprehensiVe Biological Catalysis: A Mechanistic Reference;
Sinnott, M., Ed.; Academic Press: San Diego, CA, 1998; Vol. 1, p 517.
3) Hengge, A. C.; Edens, W. A.; Elsing, H. J. Am. Chem. Soc. 1994, 116,
(
(4) Herschlag, D.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111, 7579.
(5) Admiraal, S. J.; Herschlag, D. J. Am. Chem. Soc. 2000, 122, 2145.
(6) Buchwald, S. L.; Friedman, J. M.; Knowles, J. R. J. Am. Chem. Soc. 1984,
106, 4911.
(
5
045.
13106
9
J. AM. CHEM. SOC. 2003, 125, 13106-13111
10.1021/ja036571j CCC: $25.00 © 2003 American Chemical Society

Phosphate Monoester Ester Hydrolysis
A R T I C L E S
Figure 3. Measured 18O and 15N KIEs for hydrolysis of the monoanions
1
5
of m-nitrobenzyl phosphate (top, this study) and p-nitrophenyl phosphate
bottom), measured at their respective pH-optima. For m-NBP, pH 4.0, 115
C. For pNPP, pH 3.5, 95 °C. Nonbridging oxygen atoms are designated
(
°
by “a”, bridging ones by “b”. The KIE of unity measured at the nitrogen
atom of m-nitrobenzyl phosphate is expected, since no charge delocalization
or other bonding changes involving this atom occur during the reaction.
Figure 1. Possible mechanisms for hydrolysis of monoanions of phosphate
monoesters. In mechanism A proton transfer from the phosphoryl group
(
probably via the intermediacy of a water molecule) yields an anionic
9
zwitterion intermediate. This may react in either concerted fashion (upper
pathway) or via a discrete metaphosphate intermediate in a preassociative
mechanism (bottom pathway). Mechanism B denotes proton-transfer
concerted with P-O bond fission. As with A, such a mechanism could
either occur with concerted phosphoryl transfer to the nucleophile (upper
pathway) or via a discrete metaphosphate intermediate in a preassociative
mechanism (bottom pathway).
In another computational study, the hydrolysis of methyl
phosphate monoanion was examined with the inclusion of
explicitly modeled solvating water molecules, and a dissociative
pathway was favored. Proton transfer to the leaving group and
P-O bond fission occur in the same step in the favored
mechanism, with proton transfer more advanced than P-O bond
1
0,11
fission. Calculations by others
favor the intermediacy of a
bridge-protonated intermediate in the hydrolysis of the methyl
phosphate monoanion, as originally proposed by Kirby and
Varvoglis. The bridge-protonated anionic zwitterions of methyl
and of phenyl phosphate were found to exist as viable
intermediates, whereas the corresponding anionic zwitterion of
10
2
,4-dinitrophenyl phosphate was not, implying that the transi-
tion from mechanisms A to B in Figure 1 occurs somewhere in
between. In a computational examination of the effect of the
protonation state on the hydrolysis mechanisms of phosphate
1
3
esters, it was concluded that a dissociative mechanism is
favored when the phosphoryl group is fully deprotonated,
whereas an associative one is favored for neutral esters.
We have measured the solvent deuterium and 18O kinetic
isotope effects (KIEs) on the hydrolysis of m-nitrobenzyl
Figure 2. In this proposed mechanism for monoanion hydrolysis the
protonated, neutral phosphate ester is attacked by hydroxide to form a
pentacoordinate phosphorane intermediate.
effect of 1.45 is found for hydrolysis of the monoanion of 2,4-
1
dinitrophenyl phosphate at 39 °C, whereas the figure for methyl
14
phosphate (leaving group pKa ) 14.9 ) for comparison with
previously reported data on the hydrolysis of p-nitrophenyl
phosphate is 0.87 at 100 °C.7
Computational results favoring several alternative mechanisms
for the hydrolysis of phosphate monoester monoanions have
been presented.8-11 One such proposal8 is an associative
mechanism for the hydrolysis of methyl phosphate monoanion,
in which the reactant is protonated and subsequently attacked
by hydroxide to form a pentacoordinate intermediate, the
breakdown of which is rate-determining (Figure 2). This
15
18
phosphate (pNPP) (leaving group pKa ) 7.14). The O KIEs
were measured by the remote label method, using the nitrogen
atom as a reporter for isotopic fractionation at the nonbridging,
15
or bridging, oxygen positions. Figure 3 shows the reactant
and the positions at which KIEs were measured.
The protonation or deprotonation of the phosphoryl group
18
will directly influence the knonbridge KIE. Deprotonation of
5
proposal has been controversial, with criticisms and counter-
18
phosphate monoesters is accompanied by a normal Knonbridge
arguments12 presented.
16
equilibrium isotope effect (EIE) of 1.015. In hydrolysis
18
reactions, knonbridge will also have a secondary contribution from
(
(
(
7) Bunton, C. A.; Llewellyn, D. R.; Oldham, K. G.; Vernon, C. A. J. Chem.
Soc. 1958, 3574.
8) Florian, J.; Warshel, A. J. Am. Chem. Soc. 1997, 119, 5473; Florian, J.;
Warshel, A. J. Phys. Chem. B 1998, 102, 719.
(12) Glennon, T. M.; Villa, J.; Warshel, A. Biochemistry 2000, 39, 9641.
(13) Wilkie, J.; Gani, D. J. Chem. Soc., Perkin Trans. 2 1996, 783.
(14) Sowa, G. A.; Hengge, A. C.; Cleland, W. W. J. Am. Chem. Soc. 1997,
119, 2319.
(15) Hengge, A. C. Acc. Chem. Res. 2002, 35, 105.
(16) Knight, W. B.; Weiss, P. M.; Cleland, W. W. J. Am. Chem. Soc. 1986,
108, 2759.
9) Hu, C.-H.; Brinck, T. J. Phys. Chem. A 1999, 103, 5379.
(
10) Bianciotto, M.; Barthelat, J.-C.; Vigroux, A. J. Phys. Chem. A 2002, 106,
6
521.
11) Bianciotto, M.; Barthelat, J.-C.; Vigroux, A. J. Am. Chem. Soc. 2002, 124,
573.
(
7
J. AM. CHEM. SOC.
9
VOL. 125, NO. 43, 2003 13107

A R T I C L E S
Grzyska et al.
1
50 mL of acetonitrile. To this mixture was introduced mercury (II)
chloride (10 g, 1.1 equiv) dissolved in acetonitrile (150 mL) via an
addition funnel. After addition was complete, the reaction was rapidly
brought to boiling and stirred vigorously for about 30 min. Water (100
mL) was injected to the reaction mixture, and the flask was cooled to
room temperature. The reaction mixture was filtered and the acetonitrile
removed by rotary evaporation. The remaining aqueous mixture was
titrated to pH 13 using sodium hydroxide and extracted with chloroform.
3
1
P NMR showed the presence of m-nitrobenzyl phosphate and
Figure 4. Isotopic isomers synthesized for measurement of the bridge (A
and C) and nonbridge (B and C) O isotope effects.
1
8
inorganic phosphate in a ratio of about 9: 1. The m-nitrobenzyl
phosphate was purified by anion exchange using DEAE Sephadex
eluting with an ammonium bicarbonate/water gradient, starting with
water and ending with 0.7 M ammonium bicarbonate. Fractions
containing m-nitrobenzyl phosphate were pooled and ammonium
bicarbonate removed by repeated rotary evaporation with isopropyl
alcohol. The product was titrated to pH 9 with cyclohexylamine, dried
and recrystallized from isopropyl alcohol/5% cyclohexylamine. The
product, dicyclohexylammonium m-nitrobenzyl phosphate was 10.9 g,
which is 78% yield from phosphorous acid. A sample was converted
to the sodium salt and analyzed by NMR: H NMR (400 MHz, D
δ 8.40 (s, 1H), 8.27 (d, 1H, J ) 8.2), 7.95 (d, 1H, J ) 7.7), 7.72 (t,
1H, J ) 8.0), 5.00 (d, 2H, J ) 5.9 H
phosphoric acid reference) δ 1.5
the hybridization change at phosphorus depending on whether
the transition state for phosphoryl transfer is loose or tight. Loose
transition states typical of monoester dianion reactions result
in small inverse secondary 18knonbridge KIEs, whereas the tight,
phosphorane-like transition states of triester reactions result in
15
1
8
17
normal knonbridge KIEs, up to 1.03.
1
8
The kbridge KIE is sensitive to the degree of fission of the
P-O bond in the transition state. The solvent isotope effects
give an indication of whether a proton is in flight in the transition
state of the rate-limiting step.
1
2
O)
COP), 31P NMR (162 MHz, 85%
2
Experimental Section
[14
N]-m-Nitrobenzyl Phosphate was prepared in a similar manner,
14
The materials, reagents, and solvents were commercial products and
were used as received unless otherwise noted. m-Nitrobenzyl alcohol
was distilled under reduced pressure. Acetonitrile was distilled from
phosphorus pentoxide. Carbon tetrachloride and phosphorus trichloride
were redistilled. Mercury (II) chloride was sublimed under reduced
pressure. Phosphorous acid was dried under reduced pressure. Triethyl-
amine was stored over potassium hydroxide. All solid reactants were
kept in desiccators over phosphorus pentoxide.
The commercially available bis(cyclohexylammonium) salt of p-
nitrophenyl phosphate was recrystallized before use. The disodium salt
of methyl phosphate was synthesized by a literature method.18 The
synthesis of bis(cyclohexylammonium) salt of m-nitrobenzyl phosphate
is described below.
using [ N]-m-nitrobenzyl alcohol.
14
Mixture of Isotopic Isomers A and C. [ N]-m-nitrobenzyl alcohol
15
18
and [ N, O]-m-nitrobenzyl alcohol were mixed to reconstitute the
natural abundance of N, and then the mixture was phosphorylated
using the method cited above to produce a mixture of isotopic isomers
A and C (Figure 4).
15
[15
N, Nonbridge- O
18
3
] m-Nitrobenzyl Phosphate. This was pre-
pared in the same manner as the natural abundance compound, on a
1
nitrobenzyl alcohol. The level of O incorporation was determined by
FAB-MS (see the Supporting Information for levels of isotopic
incorporation of all labeled compounds).
18
15
/100 scale, using [ O] labeled phosphorous acid and [ N]-m-
18
Isotope Effect Determinations. Isotope effects were measured by
The isotopic isomers of m-nitrobenzyl phosphate used for measure-
ment of kinetic isotope effects are shown in Figure 4. A mixture of
15
isotope ratio mass spectrometry. The N isotope effects were measured
using the natural abundance compound. The 18O isotope effects were
measured by the remote label method using the nitrogen atom as a
reporter for isotopic fractionation at the bridge or nonbridge oxygen
atoms. The experimental procedures used to measure these isotope
effects were similar to those previously reported to measure KIEs in
phosphoryl transfer reactions in which the leaving group is p-
nitrophenol.15 In summary, the reactions were carried out using ∼100
µmoles of mNBP at pH 4.0 in acetate buffer at 115 °C, stopped after
partial completion by chilling, and assayed by 31P NMR to determine
the precise fraction of reaction. The m-nitrobenzyl alcohol product from
this first portion of the reaction was separated from the unreacted ester.
The unreacted ester was completely hydrolyzed, and the m-nitrobenzyl
alcohol released in the second portion of the reaction was isolated.
Both samples of m-nitrobenzyl alcohol were purified and subjected-
to-isotope ratio analysis. The details of the experimental procedures
and the data analysis are given in Supporting Information.
isotopic isomers A and C was used for determination of 18
k
bridge. Isomers
15
B and C were mixed to reconstitute the natural abundance of N and
1
8
used for determination of
knonbridge. The isotopic abundance of the
mixtures was determined by isotope ratio mass spectrometry.
1
4
15
[
N] labeled -m-nitrobenzyl alcohol, [ N] labeled -m-nitrobenzyl
15
18
alcohol, and [ N, O] labeled -m-nitrobenzyl alcohol were synthesized
1
4
as described previously.
1
8
18
3 3
O-Labeled Phosphorous Acid (H P O ). A modification of a
19
published procedure was used. A 50 mL Claisen flask was fitted with
a condenser under dry nitrogen and placed in an ice bath. Carbon
tetrachloride (1.6 mL) and phosphorus trichloride (0.33 mL, 1 eq.) were
added and vigorously mixed. Into this mixture, [ O]-water (0.22 mL,
.8 eq.) was added over about 1 h in such a manner that overheating
1
8
2
was avoided. The ice bath was then removed and the reaction mixture
stirred for one more hour. Volatile materials were subsequently removed
under vacuum, leaving crystalline phosphorous acid (0.31 g, 95% yield).
Natural Abundance m-Nitrobenzyl Phosphate. This was synthe-
Solvent Deuterium Isotope Effects. Buffered solutions of the
reactants were prepared in parallel containing 20 mL of 25mM
phosphate ester in 250mM succinic acid buffer at pH 4.0. This solution
was divided into two separate flasks and dried in vacuo at ambient
temperature. The dry mixtures of buffer and phosphate ester were
2
0
sized by a modification of a published procedure. In a 500 mL three
neck flask fitted with reflux condenser and large stirrer, m-nitrobenzyl
alcohol (25 g, 6 equiv), 1,8-diazobicyclo-[5.4.0]-undec-7-ene (19.9 g,
4
.1equiv), and phosphorous acid (2.68 g, 1 equiv) were mixed with
reconstituted in either H
buffer was found to obey the expected formula pD ) meter reading +
.4. Samples of the reaction mixtures were removed at time intervals
and assayed. The hydrolyses of methyl phosphate and of m-nitrobenzyl
phosphate were monitored by 31P NMR spectroscopy. The hydrolysis
of p-nitrophenyl phosphate was followed by UV-vis spectrophotometry
2 2 2
O or D O. The pH for the D O-reconstituted
(
(
(
(
17) Anderson, M. A.; Shim, H.; Raushel, F. M.; Cleland, W. W. J. Am. Chem.
Soc. 2001, 123, 9246.
0
18) Grzyska, P. K.; Czyryca, P. G.; Golightly, J.; Small, K.; Larsen, P.; Hoff,
R. H.; Hengge, A. C. J. Org. Chem. 2002, 67, 1214.
19) Voigt, D.; Gallais, F. In Inorganic Syntheses: New York, 1953; Vol. 4, p
5
5.
20) Obata, T.; Mukaiyama, T. J. Org. Chem. 1967, 32, 1063.
1
3108 J. AM. CHEM. SOC. VOL. 125, NO. 43, 2003
9

Phosphate Monoester Ester Hydrolysis
A R T I C L E S
at 400 nm by periodically adding an aliquot of the reaction mixture to
cuvette containing 0.1N NaOH. Reactions were followed to completion,
and obeyed pseudo first-order kinetics. Rate constants were determined
from fits of these data to the first-order rate equation.
Choice of Buffers. Acetate and succinate were used as buffers in
experiments at elevated temperatures because both exhibit a negligible
temperature dependence on pK
a
(acetate: ∆pK
a
/°C ) 0.0002; succi-
2
1
nate: ∆pK /°C ) 0.0).
2
Computational Methods. The Gaussian 98 program2 was used for
the geometry optimizations and the calculations of the appropriate
Hessians. The B3LYP/6-31+G* method was employed, and all
calculations applied to the molecules in the gas phase. Geometry
optimization calculations were performed on the structures of p-
nitrophenyl phosphate, p-nitrophenyl acetate, and p-nitrophenol. The
optimizations were followed by calculations of Hessians, then the
calculated force constants were used to calculate vibrational frequencies
of the isotopically substituted species. Equilibrium isotope effects were
2
2
3
calculated according to the Bigeleisen equation. The Isoeff program,
developed by Prof. P. Paneth, was used in these calculations.
18O Tracer Experiments. Solutions were prepared containing 50
mM m-nitrobenzyl phosphate, or inorganic phosphate, and 250 mM
succinate or acetate buffer at pH 4.0 containing 30% of 18O-labeled
Figure 5. 31P NMR after partial hydrolysis of mNBP monoanion at 115
°
C (top) showing remaining mNBP (left) and inorganic phosphate product
(right). The isotope shift is that expected from incorporation of a single
atom of 18O. The bottom spectrum shows inorganic phosphate after
incubation under the same conditions. Extended incubation times at higher
temperature results in additional incorporation of 18O into inorganic
phosphate (data not shown).
2
water. These solutions also contained 5% D O to provide a lock to
facilitate acquisition of NMR spectra. Reaction mixtures were sealed
in high-pressure NMR tubes, and placed in an oil bath at 115 °C for
1
4 h. After cooling, the reaction samples were neutralized with 4N
3
1
KOH, and 1 mg of EDTA was added before P NMR analysis of the
products. Integrations of overlapping peaks were obtained by Gaussian
curve fitting. The m-nitrobenzyl alcohol product was extracted with
ether and analyzed by GC-MS.
that nucleophilic attack occurs at phosphorus with P-O bond
fission at pH 4, whereas at pH values <2, attack at carbon (with
2
5
C-O bond fission) increasingly predominates. At pH 4, the
concentration of the monoanionic species is at its highest; at
lower pH, the concentration of the fully neutral species increases,
resulting in the mechanistic change.
a
pK Determination. Solutions of m-nitrobenzyl phosphate were
prepared at pH values ranging from 4.0 to 9.0, containing 25mM mNBP
in 100 mM acetate, 100 mM Bis-Tris or 100 mM Tris buffers. The pH
a
values were measured using a glass electrode at 25 °C. The pK value
A second pKa of 6.2 was found for mNBP, a value that is
typical for alkyl phosphate monoesters. The hydrolysis of mNBP
3
1
was calculated from a fit of the plot of P NMR chemical shift vs pH.
Thermodynamic Data. Reaction solutions containing 30 mL of
18
at pH 4.0 at 115 °C in water containing 30% H2 O shows that
2
5mM m-nitrobenzyl phosphate ester in 250mM succinic acid buffer
31
at about 67% of reaction, the P NMR signal for the inorganic
at pH 4.0 in thick-walled glass vessels with a pressure cap were placed
in an oil bath maintained at appropriate temperature. The experiments
were carried out at 95, 110, 115, 120, 125, and 130 °C. The pH was
checked again at the end of the reaction and was found to be the same
phosphate product exhibits a second peak ∼0.02 ppm upfield,
a shift of the magnitude expected from the incorporation of a
1
8
26
single atom of
O
(Figure 5, top). Deconvolution and
31
integration of these peaks showed that the inorganic phosphate
within ( 0.05 pH units. P NMR was used to follow the reaction rates
for four or more half-lives. Rate constants were obtained from fits of
these data to the first-order rate equation. These rate constants were
used to construct the Eyring plot (see the Supporting Information) which
yielded a value for the enthalpy of activation of 40 ( 1 kcal/mol and
an entropy of activation of +28 ( 1 eu.
18
product contained 42% O. There was no detectable incorpora-
1
8
tion of O into unreacted mNBP (Figure 5, top). A control
experiment, in which inorganic phosphate was incubated under
the same conditions for the same time duration, showed that
18
1
3% of the phosphate had incorporated O from exchange with
solvent (Figure 5, bottom). These results show that under these
conditions exchange with inorganic phosphate occurs more
slowly than hydrolysis of mNBP, but at a rate that is sufficient
Results and Discussion
General Mechanistic Considerations. The hydrolysis of
benzyl phosphate has been studied by Westheimer and by Van
Etten.2 In that study, isotope tracer results at 75 °C showed
18
to account for the excess O observed in the phosphate product.
4,25
18
No incorporation of O into the m-nitrobenzyl alcohol product
could be detected by mass spectrometry. Thus, the reaction of
(
21) Stoll, V. S.; Blanchard, J. S. In Guide to Protein Purification; Deutscher,
M. P., Ed.; Academic Press: San Diego, CA, 1990; Vol. 182, p 27.
22) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M.
A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.; Stratmann,
R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A. D.; Kudin,
K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi,
R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski, J.;
Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick, D. K.;
Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.; Ortiz,
J. V.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.;
Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J. L.; Head-Gordon,
M.; Replogle, E. S.; Pople, J. A. Gaussian 98,; Gaussian, Inc.: Pittsburgh,
PA, 1998.
mNBP at pH 4.0 proceeds by the same mechanism as that
2
5
(
previously documented for benzyl phosphate, namely, attack
at phosphorus with P-O bond fission.
Isotope Effects. Solvent Isotope Effects. The solvent
deuterium KIEs (kH2O/kD2O) measured in this study are shown
in Table 1. The value of 0.81 for methyl phosphate at 120 °C
7
is close to the previously reported value of 0.87 at 101 °C. All
of the kH2O/kD2O values are slightly inverse, more suggestive of
(25) Parente, J. E.; Risley, J. M.; Van Etten, R. L. J. Am. Chem. Soc. 1984,
106, 8156.
(26) Cohn, M.; Hu, A. Proc. Natl. Acad. Sci. 1978, 75, 200.
(
23) Bigeleisen, J. J. Chem. Phys. 1949, 17, 675.
(
24) Kumamoto, J.; Westheimer, F. H. J. Am. Chem. Soc. 1955, 77, 2515.
J. AM. CHEM. SOC.
9
VOL. 125, NO. 43, 2003 13109

A R T I C L E S
Grzyska et al.
Table 1. Solvent Deuterium Kinetic Isotope Effects for Hydrolysis
negative entropies of activation; for methyl phosphate monoan-
Reactions of Phosphate Monoesters Measured in This Studya
‡
‡
ion hydrolysis ∆S ) -2.2 eu; for phenyl phosphate ∆S )
reactant
kH2O/kD2O
1
-
0.6 eu. We obtained a value of +28 ( 1 eu for the entropy
p-nitrophenyl phosphate
0.96 ( 0.01
0.95 ( 0.01
0.94 ( 0.01
0.81 ( 0.01
of activation for the hydrolysis of the mNBP monoanion. The
reason for the unexpectedly large positive value is uncertain,
but it argues strongly against a triester-like mechanism.
The small inverse 18kbridge of 0.9981 ( 0.0002 for the
(
pH 3.5, 95°C)
p-nitrophenyl phosphate
pH 4.0, 95°C)
m-nitrobenzyl phosphate
pH 4.0, 110°C)
methyl phosphate
pH 4.0, 120°C)
(
(
hydrolysis of the monoanion of mNBP contrasts with the normal
18
3
kbridge of 1.0087 ( 0.0003 for pNPP. This KIE is the net value
(
resulting from the normal contribution arising from P-O(R)
bond fission, and the inverse value from protonation. The net
replacement of the phosphoryl group of the reactant with
a
The pH-optimum for the alkyl phosphates is 4.0; that for p-NPP is
3
.5.
1
8
hydrogen will result in a normal Kbridge EIE due to the loss of
bending and torsional modes. (For similar reasons, the 18
fractionation factors between water and alcohols are 2% to 3%
fractionation factors rather than the normal values expected
when a proton is in flight in the transition state (for example,
kH2O/kD2O ) 1.45 for hydrolysis of 2,4-dinitrophenyl phosphate
O
3
1
normal (1.02 to 1.03), in the direction from alcohol to water).
1
monoanion ). The magnitudes of such KIEs can be reduced
18
To estimate the magnitude of this contribution to kbridge, we
calculated the gas phase 18O fractionation factor between
p-nitrophenol and pNPP, and that between p-nitrophenol and
p-nitrophenyl acetate. The calculated value between p-nitro-
phenol and p-nitrophenyl acetate is 1.014. An experimental value
of 1.012 for this fractionation factor can be obtained from the
reported EIE between p-nitrophenolate anion and p-nitrophenyl
when there is a large difference in pKa between the proton donor
and acceptor (resulting in a very early or late transition state
for proton transfer),27 or by the contribution of heavy-atom
28
motion to the reaction coordinate. However, to our knowledge
there is no precedent for such factors to completely eliminate
the normal deuterium KIE expected from a proton in flight in
the transition state.
3
2
acetate in water (0.9730) and the EIE for protonation of
18O Isotope Effects. The 18
knonbridge of 1.0151 (Figure 3) for
3
2
p-nitrophenol, (0.9849). The close agreement between the
calculated and the experimental values supports the validity of
the computational method, which gives a calculated EIE between
p-nitrophenol and pNPP of 1.011. The experimental KIE for
the hydrolysis of the pNPP monoanion (1.0087) approaches this
value, consistent with a late transition state in which proton
transfer to the leaving group and P-O(R) bond fission are both
far advanced. In contrast, the small inverse kbridge in the
reaction of m-NBP suggests a transition state with less P-O(R)
bond fission. In such a case, the inverse contribution from
protonation will be less offset by loss of P-O(R) bond order
and loss of bending and torsional modes associated with the
phosphoryl group.
the hydrolysis of the monoanion of mNBP is indistinguishable
from the 18O equilibrium isotope effect (EIE) of 1.015 for the
16
deprotonation of a phosphate ester. Although the elevated
temperature necessitated by the slow reaction rate of mNBP
will slightly reduce the observed value of 18knonbridge this effect
is relatively small and does not affect the present discussion.
(
An estimation of the magnitude of this reduction can be inferred
1
8
18
from previously reported values for knonbridge for the monoanion
of pNPP at 95 °C (1.0184 ( 0.0003) and at 30 °C (1.0199 (
2
9
0
.0005) ) Protonation of the phosphoryl group such as in the
mechanistic proposal of Figure 2 would result in an inverse
18
knonbridge of similar magnitude (0.985). Even if the added proton
were partially removed in a subsequent rate-limiting transition
state, the phosphoryl group would still be in a higher state of
The conclusion that the transition state of the more reactive
pNPP has significantly greater P-O(R) bond fission than the
less reactive mNBP is counter to expectations based on the
Hammond postulate. This apparent conundrum is resolved by
recent computational results indicating that protonation of the
P-O(R) ester oxygen atom weakens this bond to a degree that
1
8
protonation than the reactant. Thus, the normal knonbridge is
consistent with a mechanism in which the phosphate reactant
becomes deprotonated, but not with the mechanism of Figure
2
.
We considered the possibility that the normal 18knonbridge could
1
0
be the result of a mechanism like that shown in Figure 2, but
in which the inverse EIE for protonation is masked by a large
normal (∼3%) KIE in a phosphorane-like transition state, as in
depends on the basicity of the ester group. Comparison of the
structures of the bridge-protonated anionic zwitterion forms of
methyl and phenyl phosphates showed that in phenyl phosphate
the P-O(R) ester bond is 0.24 Å longer than in the methyl
17
reactions of phosphotriesters. Such a mechanism might ser-
1
8
10
33
endipitously result in an observed net value for knonbridge
resembling that for deprotonation. We measured the entropy of
activation to ascertain the likelihood of a triester-type mecha-
nism. Large negative entropies of activation characterize phos-
photriester hydrolysis reactions. For example, for trimethyl
ester. Using Pauling’s rule and 1.6 Å as the distance for a
P-O single bond, this constitutes a 60% weaker bond in the
anionic zwitterion form of phenyl phosphate compared to that
of methyl phosphate. These authors did not model bridge-
protonated p-nitrophenyl phosphate, but presumably, the effect
would be, if anything, even more pronounced. The experimental
results here are consistent with the idea that protonation of the
ester oxygen atom gives aryl phosphates a significantly greater
‡
phosphate hydrolysis, ∆S ) -23 eu, and for triethyl phosphate,
‡
30
∆S
) -34 eu. The hydrolyses of phosphate monoester
monoanions, on the other hand, are characterized by very small
“
head start” on the way to P-O(R) bond fission relative to their
(
27) Melander, L.; Saunders: W. H. In Reaction Rates of Isotopic Molecules;
Wiley: New York, 1980; p 131.
(
28) Melander, L.; Saunders: W. H. In Reaction Rates of Isotopic Molecules;
Wiley: New York, 1980; p 152.
(31) Rishavy, M. A.; Cleland, W. W. Can. J. Chem. 1999, 77, 967.
(32) Hengge, A. C.; Hess, R. A. J. Am. Chem. Soc. 1994, 116, 11256.
(33) Pauling, L. In The Nature of the Chemical Bond., 3rd ed.; Cornell University
Press: Ithaca, NY, 1960; p 255.
(29) Czyryca, P. G.; Hengge, A. C. Biochim. Biophys. Acta 2001, 1547, 245.
30) Kluger, R.; Taylor, S. D. J. Am. Chem. Soc. 1990, 112, 6669.
(
13110 J. AM. CHEM. SOC.
9
VOL. 125, NO. 43, 2003

Phosphate Monoester Ester Hydrolysis
A R T I C L E S
alkyl counterparts, accounting for the transition state difference
implied by the KIE data.
difference in the extent of P-O(R) bond fission in the transition
states for hydrolysis of the aryl ester pNPP compared to the
alkyl mNBP. The data argue against a mechanism involving
protonation of the reactant, such as that in Figure 2.
We previously interpreted 18O and 15N KIE data for pNPP
3
in terms of a concerted mechanism (Figure 1B), on the basis of
15
a small N KIE indicative of a small amount of negative charge
on the leaving group in the transition state, suggesting that
proton transfer and P-O bond fission might not be fully
synchronous. However, the kH2O/kD2O KIE measured here does
not support such a view, and the small 15N KIE may reflect an
EIE between p-nitrophenol (which the leaving group resembles
in the very late transition state) and the pNPP reactant. On the
basis of the collected evidence, the switch from preequilibrium
proton transfer to concerted proton transfer/P-O bond fission
probably occurs somewhere between the leaving groups p-
nitrophenol (pKa ) 7.14) and 2,4-dinitrophenol (pKa ) 4.07).
In summary, the data presented here are most consistent with
a pre-equilibrium proton transfer from the phosphoryl group to
the ester oxygen atom, followed by rate-limiting P-O bond
fission, for both pNPP and mNBP. The results also support
computational predictions of differential effects of protonation
on the P-O(R) ester bond, which are reflected in a substantial
Acknowledgment. The authors thank the NIH for financial
support (GM47297 to A.C.H.). The authors gratefully acknowl-
edge the use of equipment in the Shimadzu Analytical Sciences
Laboratory at Utah State University, created with equipment
donated by Shimadzu Scientific Instruments, Inc. (Columbia,
MD).
Supporting Information Available: Detailed procedures for
O isotope effect measurements, and levels of isotopic incor-
18
poration of mNBP isotopic isomers. Structures in z-matrix
format and Gaussian 98 output with vibrational frequencies
calculated for isotopomers of p-nitrophenol, p-nitrophenyl
acetate, and p-nitrophenyl phosphate. This material is available
free of charge via the Internet at http://pubs.acs.org.
JA036571J
J. AM. CHEM. SOC.
9
VOL. 125, NO. 43, 2003 13111