The thermodynamics of phosphate versus phosphorothioate ester hydrolysis

Article abstract of DOI:10.1021/jo0511997

Phosphorothioate esters are phosphate esters in which one of the nonbridging oxygen atoms has been replaced by sulfur. In the comparative hydrolysis reactions of phosphorothioate and phosphate esters, the sulfur substitution accelerates the rates of the monoesters while slowing the rates of diesters and of triesters. Previously measured enthalpies and entropies of activation for the hydrolysis reactions of the monoesters, p-nitrophenyl phosphate and p-nitrophenyl phosphorothioate, were compared to the activation parameters measured herein for the diesters, ethyl p-nitrophenyl phosphate and ethyl p-nitrophenyl phosphorothioate, and the triesters, diethyl p-nitrophenyl phosphate and diethyl p-nitrophenyl phosphorothioate. A consistent trend of a greater ΔH? for the phosphorothioate analogue was found in all three classes of ester. In the monoester case, a more positive ΔS? arising from a mechanistic difference (D_N + A_N for the phosphorothioate versus A_ND_N for the phosphate) compensates, resulting in a lower ΔG? for the phosphorothioate monoester. Spectroscopic investigations indicate there is no significant difference in bond order to the leaving group in phosphates, as compared to their phosphorothioate analogues, ruling this out as a contribution to the consistently higher enthalpies of activation.

Full text of DOI:10.1021/jo0511997

The Thermodynamics of Phosphate versus Phosphorothioate
Ester Hydrolysis
Jamie Purcell and Alvan C. Hengge*
Department of Chemistry and Biochemistry, Utah State University, Logan, Utah 84322-0300
hengge@cc.usu.edu
Received June 13, 2005
Phosphorothioate esters are phosphate esters in which one of the nonbridging oxygen atoms has
been replaced by sulfur. In the comparative hydrolysis reactions of phosphorothioate and phosphate
esters, the sulfur substitution accelerates the rates of the monoesters while slowing the rates of
diesters and of triesters. Previously measured enthalpies and entropies of activation for the
hydrolysis reactions of the monoesters, p-nitrophenyl phosphate and p-nitrophenyl phosphorothioate,
were compared to the activation parameters measured herein for the diesters, ethyl p-nitrophenyl
phosphate and ethyl p-nitrophenyl phosphorothioate, and the triesters, diethyl p-nitrophenyl
q
phosphate and diethyl p-nitrophenyl phosphorothioate. A consistent trend of a greater ∆H for the
phosphorothioate analogue was found in all three classes of ester. In the monoester case, a more
q
positive ∆S arising from a mechanistic difference (D
N
+ A for the phosphorothioate versus A D
N
N
N
q
for the phosphate) compensates, resulting in a lower ∆G for the phosphorothioate monoester.
Spectroscopic investigations indicate there is no significant difference in bond order to the leaving
group in phosphates, as compared to their phosphorothioate analogues, ruling this out as a
contribution to the consistently higher enthalpies of activation.
Introduction
TABLE 1. Correlation of Thio Effects for Alkaline
Hydrolysis Reactions of Phosphate and
Phosphorothioates are analogues of phosphate esters
in which a sulfur atom is present in place of one of the
nonbridging oxygen atoms. A change in the rate of
hydrolysis has been observed to accompany this sulfur
substitution, leading to a phenomenon known as the “thio
1,2
Phosphorothioate Esters
phosphate ester
range of thio effects (kO/kS)
triester
10-160
4-11
0.1-0.3
diester
monoester
O S
effect”. The thio effect is reported as k /k , the ratio of
the rate of phosphoryl transfer to that for thiophosphoryl
transfer. This sulfur substitution significantly increases
the rates of hydrolysis of phosphomonoester dianions,
Linear free energy relationships and kinetic isotope
effects have identified the mechanisms and yielded
information regarding the transition states of the reac-
tions of phosphate and phosphorothioate monoesters,
diesters, and triesters. Monoesters undergo hydrolysis via
loose transition states in which bond fission to the leaving
group is extensive, and nucleophilic participation in the
rate-limiting step is either small (in the case of phos-
O S
leading to an inverse thio effect (k /k < 1). However,
similar substitution has the opposite effect on the hy-
drolysis rates of diesters and triesters (Table 1). As a
result of this trend, the magnitudes and directions of
thio effects have been used to examine mechanistic
details of enzymatic phosphoryl transfer. An understand-
ing of the origin of the opposite thio effect for monoesters
versus diesters and triesters is needed to properly
interpret such data.
phates, where the mechanism is A
case of phosphorothioates, where the mechanism is D
+ A ). Acyclic diesters and triesters with good (i.e., aryl)
N N
D ) or absent (in the
N
N
1
0.1021/jo0511997 CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/13/2005
J. Org. Chem. 2005, 70, 8437-8442
8437

Purcell and Hengge
TABLE 2. Activation Parameters for Hydrolysis
Reactions of Phosphate and Phosphorothioate
Monoesters, Diesters, and Triesters at 312 K
q
q
q
∆
G (kcal/mol) ∆H (kcal/mol)
∆S (eu)
Monoester Dianionsa
pNPP10
29.5
30.6
+3.5
+29 ( 3
1
1
pNPPT
27.9 ( 1.0
37.0 (1.0
Diestersb
EtO-pNPP
EtO-pNPPT
27.4 ( 0.5
27.9 ( 0.4
14.91 ( 0.06
18.5 ( 0.1
-36.6 ( 0.1
-26.9 ( 0.3
Triestersb
paraoxon
parathion
20.0 ( 0.3
21.5 ( 0.2
12.69 ( 0.03
16.43 ( 0.08
-24.81 ( 0.05
-17.3 ( 0.1
a
Previously reported data. b This work.
In this work we report the activation parameters for
the hydrolysis reactions of the PdO and PdS analogues
of a monoester, diester, and triester, to ascertain the
thermodynamic contributions to the differences in activa-
tion energies. This information should provide further
understanding of the fundamental causes of the observed
thio effects. We have also carried out spectroscopic
investigations via 31P NMR and FTIR in order to deter-
mine whether sulfur substitution in a nonbridging posi-
tion has any effect on the scissile P-O bond.
FIGURE 1. More-O’Ferrall-Jencks diagram depicting the
general location of transition states of uncatalyzed phosphoryl
transfer reactions of phosphomonoesters, diesters, and tri-
esters. In general, the transition states become tighter as the
alkylation state of the ester increases.
leaving groups undergo hydrolysis by concerted reactions
via transition states with greater nucleophilic participa-
tion. In reactions of diesters, nucleophilic attack and
leaving group departure are more synchronous, in a
Results
N N
concerted, A D mechanism. Triester reactions exhibit
The activation parameters were calculated from Eyring
plots for the alkaline hydrolysis of the diester and triester
phosphate and phosphorothioate esters, giving the values
tight, phosphorane-like transition states with nucleo-
philic attack ahead of leaving group departure. This is
shown pictorially on the More-O’Ferrall-Jencks dia-
gram in Figure 1.
10
in Table 2. Data for the monoester hydrolysis of pNPP
11
and pNPPT are from previous work.
Thio effects have been used in attempts to discern
whether enzymatic phosphoryl transfer is associative or
dissociative, although the risks associated with such
The O-induced isotope effect on the 31P chemical
18
shifts of the monoesters and triesters were obtained from
the NMR of labeled versions of the four compounds,
3
interpretations of thio effects have been pointed out. A
18
containing 41% O incorporation in the bridging position.
number of proposals have been made to explain reactivity
differences between phosphates and their thiophosphate
analogues, such as the comparative electron-donating
ability of oxygen versus sulfur,4 the greater electrophi-
licity of phosphorus in phosphoryl versus thiophosphoryl
31
As an example of these spectra, Figure 2 shows the
P
NMR spectrum of the monoester pNPPT. Table 3 shows
the isotope shift data obtained for the phosphate and
phosphorothioate analogues of the monoester and tri-
ester.
The frequencies of the infrared stretching bands of the
P-O(nitrophenyl) bond in the monoesters and triesters
were measured as described in Experimental Section and
are shown in Table 4.
,5
6,7
compounds, differences in polarization of PdO and Pd
8
S bonds, and structural differences between the two
ester types, including interatomic distances and atomic
9
radii. However, none of these provides an explanation
for the thio effects observed across all three classes of
ester.
Discussion
(
1) Domanico, P.; Mizrahi, V.; Benkovic, S. J. In Mechanisms of
The activation parameters reveal that for all three
classes of phosphoesters (monoester, diester, and triester)
hydrolysis of the thiophosphate is accompanied by a
Enzymatic Reactions: Stereochemistry; Frey, P. A., Ed.; Elsevier: New
York, 1986; pp 127-137.
(
2) Herschlag, D. J. Am. Chem. Soc. 1994, 116, 11631-11635.
q
(
3) Holtz, K. M.; Catrina, I. E.; Hengge, A. C.; Kantrowitz, E. R.
higher enthalpy of activation (∆H ), but a more favorable
Biochemistry 2000, 39, 9451-9458.
q
entropy of activation (∆S ), than the corresponding
(
4) Williams, A.; Douglas, K. T. J. Chem. Soc., Perkin Trans. 2 1975,
010-1016.
5) Zhang, Y.-L.; Hollfelder, F.; Gordon, S. J.; Chen, L.; Keng, Y.-
F.; Wu, L.; Herschlag, D.; Zhang, Z.-Y. Biochemistry 1999, 38, 12111-
2123.
6) Ketelaar, J. A. A.; Gresmann, H. R.; Koopmans, K. Rec. Trav.
Chim. Pays-Bas 1952, 71, 1253-1258.
7) Neimysheva, A. A.; Savchik, V. I.; Ermolaeva, M. V.; Knunyants,
I. L. Bull. Acad. Sci. USSR. Div. Chem. Sci. Engl. Transl. 1968, 2104-
phosphate ester. In the diester and triester reactions, the
enthalpic contribution to the overall barrier is larger than
1
(
q
the T∆S term, resulting in slower reaction rates for the
1
phosphorothioate analogues. However, in the monoester
(
q
case, a significantly more positive (favorable) ∆S for the
(
q
phosphorothioate ester overcomes the larger ∆H , result-
2
7
6
108.
(
8) Chlebowski, J. F.; Coleman, J. E. J. Biol. Chem. 1974, 249, 7192-
(10) Kirby, A. J.; Jencks, W. P. J. Am. Chem. Soc. 1965, 87, 3209-
202.
3216.
(
9) Teichmann, H.; Hilgetag, G. Angew. Chem., Int. Ed. Engl. 1967,
(11) Catrina, I. E.; Hengge, A. C. J. Am. Chem. Soc. 1999, 121,
2156-2163.
, 1013-1023.
8438 J. Org. Chem., Vol. 70, No. 21, 2005

Phosphate/Phosphorothioate Hydrolysis Thermodynamics
17,19
TABLE 5. Isotope Shifts for ADP and ATP Isomers,
2
1
Inorganic Phosphate, and Paraoxon
∆
δP ) δ16Ã-δ18
Ã
formal
bond order
compound
(ppb)
8
[
[
â,γ-bridging 1 Ã] ATP
16.5
16.6
20.5
21.5
22.0
28.5
40.0
1.0
1.0
1.25
1.33
1.33
1.5
1
8
R,â-bridging Ã] ADP
inorganic phosphate trianion
1
8
[
[
[
[
â-nonbridging Ã] ADP
γ-nonbridging Ã] ATP
1
8
1
8
â-nonbridging Ã] ATP
1
8
nonbridging Ã] paraoxon
2.0
izability of sulfur, relative to oxygen, complicates any
rationalization of the differential effects of sulfur sub-
stitution on the rates of these compounds to a single
atomic property.
In the hydrolysis reactions of all three types of esters,
regardless of mechanistic differences, bond fission to the
leaving group always occurs in the rate-determining step.
This raises the question of whether differences in the
leaving group P-O bonds might be responsible for the
consistently higher enthalpy of activation in the phos-
phorothioate reactions. Computational and experimental
work has shown that sulfur substitution affects the P-O
bond order and the charge distribution between non-
3
1
18
FIGURE 2. P NMR spectrum of 41% [ O]-labeled pNPPT
1
8
31
showing the O-induced perturbation to the P chemical shift.
TABLE 3. Isotope-Induced 31P NMR Shifts (∆δP) for
Monoesters p-NPP and p-NPPT and Triesters Paraoxon
and Parathion
compound
∆δP ) (δ16O - δ18O) (ppb)
1
3,14
bridging atoms within the (thio)phosphoryl group.
If
pNPP
pNPPT
20.2
27.8
15.8
20.4
the bridging P-OR bonds were also affected, such that
the bond to the leaving group is stronger in the thio
analogues, then a higher enthalpy of activation might
result. To test this hypothesis, this bond was investigated
using NMR and FTIR. A comparison of the magnitudes
of the 18O-induced perturbation to the P NMR chemical
paraoxon (diethyl p-NPP)
parathion (diethyl p-NPPT)
TABLE 4. Stretching Frequency of the P-(O)R Bond in
31
Monoesters and Triesters
P
shift (∆δ ) in a labeled monoester and triester was used
compound
P-O(R) stretch (cm-1)
to investigate this bond, since these ester types give the
largest thio effects and were the most amenable to
isotopic synthesis.
pNPP
pNPPT
730
725
785, 764, 737
788, 770, 749
paraoxon (diethyl p-NPP)
parathion (diethyl p-NPPT)
Such isotope-induced NMR shifts result from the
combination of greater nuclear shielding provided by the
heavier nucleus, and anharmonicity, manifested in a
shortened bond length upon substitution of the heavier
isotope. Isotope shift data have been compiled for a large
number of compounds, and the cumulative data has led
to the observation that, within the same or similar
functional groups, the magnitude of the isotope shift is
ing in a faster hydrolysis and a thio effect of less than 1.
The significantly more positive ∆S in the phosphorothio-
q
ate monoester reaction results from a mechanistic dif-
ference. Phosphorothioate monoesters undergo hydrolysis
via a thiometaphosphate intermediate,1 making the
rate-limiting step a unimolecular dissociation, reflected
in the significantly more positive entropy of activation.
In contrast, the oxygen analogue reacts by a loose
transition state with little, but measurable, nucleophilic
participation.
Several reasons have been proposed to explain why
sulfur substitution should facilitate this mechanistic
change from a loose transition state in a concerted
reaction to a completely dissociative mechanism. One
proposal is that the sulfur atom is more able to donate
negative charge, thus facilitating leaving group expul-
,12
1
5,16
18
proportional to bond order.
Published O-induced
isotope shifts on 31P NMR spectra of phosphate esters
also show that the magnitude of this isotope-induced shift
(
∆δ
P
) is proportional to the P-O bond order. The mag-
nitude of ∆δ
P
increases from ∼0.016 ppm for P-O single
bonds to ∼0.041 ppm for double bonds (Table 5 and
Figure 3).1
7-19
The validity of the correlation is further
demonstrated by the linear relationship between the
(
13) Frey, P. A.; Sammons, D. Science 1985, 228, 541-545.
(
14) Liang, C.; Allen, L. C. J. Am. Chem. Soc. 1987, 109, 6449-
5
sion. However, replacing oxygen by sulfur in the alkaline
6
453.
hydrolysis of phenyl esters of phenylphosphonamidates
changes the mechanism in the opposite direction (from
E1cB to addition-elimination), a change that has been
rationalized on the poorer electron-donating ability of
(15) Risley, J. M.; Van Etten, R. L. In Isotope Effects in NMR
Spectroscopy; Diehl, P. F. E., Guenther, H., Kosfeld, R., Seelig, J., Eds.;
Springer-Verlag: Berlin, 1990; Vol. NMR 22, pp 81-168.
(
16) Jameson, C. J. In Encyclopedia of Nuclear Magnetic REsonance;
Grant, D. M., Harris, R. K., Eds.; John Wiley: London, 1995; Vol. 6,
4
pp 2638-2655.
sulfur. The lower electronegativity, but greater polar-
(
(
17) Cohn, M.; Hu, A. J. Am. Chem. Soc. 1980, 102, 913-916.
18) Cohn, M.; Hu, A. Proc. Natl. Acad. Sci. 1978, 75, 200-203.
(
12) Burgess, J.; Blundell, N.; Cullis, P. M.; Hubbard, C. D.; Misra,
R. J. Am. Chem. Soc. 1988, 110, 7900-7901.
(19) Lowe, G.; Sproat, B. S. J. Chem. Soc., Chem. Commun. 1978,
565-566.
J. Org. Chem, Vol. 70, No. 21, 2005 8439

Purcell and Hengge
TABLE 6. Literature Values for the 18O-Induced 31
P
NMR Shift, Used to Construct the Phosphorothioate
Calibration Graph in Figure 4
formal
bond order
compound
∆δP (ppb)
1
8
13,22
[
R,â-bridging O]ADPRS
22.0
1.0
1.33
1.5
1.5
2.0
2.0
1
8
13
S-carbamoylethyl-[ O]-phosphate
27.0
1
8
13,23
S-methyl-[ O] AMPS
35.0
1
8
13,23
[
[
R- O] AMPS
33.0
1
8
13,22
R-nonbridging O] ADPRS
37.0
1
8
24
endo-uridine-2′3′-cyclic-[ O]-
41.0
phosphorothioate
1
8
24
uridine 2′[ O] phosphorothioate
33.0
1.5
1.5
1.0
2.0
2.0
1.0
1
8
24
uridine 3′[ O] phosphorothioate
34.0
1
8
22
[
[
R,â-bridging O] ATPâS
21.0
1
8
22
â-nonbridging O] ATPâS
36.0
a
25
compound 5
compound 4
41.0
b
25
19.0
a
1
,3,2-Dioxaphosphorinane, 2-hydroxy-5,5-dimethyl-2-sulfide
b
(
Registry no. 15762-04-6). 2-[(5,5-Dimethyl-2-oxido-1,3,2-dioxa-
phosphorinan-2-yl)oxy]-5,5-dimethyl-2-sulfide (Registry no. 55474-
6-3).
1
FIGURE 3. Graph of the isotope shifts in Table 5. The line
represents a least-squares fit to the data and obeys the relation
∆
δ
P
) 23.8(bond order) - 8.4.
magnitude of ∆δ
in methyl, dimethyl, and trimethyl phosphates.
P
and the P-O stretching frequencies
20
The monoesters and triesters were synthesized as
isotopic isomers in which 18O was partially incorporated
into the bridging phenolic position, which is the scissile
18
P-O bond. Incorporation of O in phosphate esters
results in a small upfield shift in the 31P NMR signal.
The isotope shift data for both the monoester and triester
(
P
Table 3) show a larger ∆δ for the thio esters. This might
be taken as an indication of a stronger P-O bond to the
leaving group of the latter compounds, perhaps account-
ing for the higher enthalpies of activation. However, the
phosphate and phosphorothioate functional groups differ,
P
and we considered the possibility that the larger ∆δ in
the phosphorothioates might arise from differences in the
relative contributions of nuclear shielding and anhar-
monicity in the two sets of compounds. To examine this
possibility, we compiled published isotope shift values for
phosphorothioates and constructed a correlation for these
compounds, analogous to the correlation for phosphate
FIGURE 4. Graph of the isotope shifts in Table 6. The line
represents a least-squares fit to the data and obeys the relation
P
∆δ ) 17.8(bond order) + 4.4.
P
esters shown in Figure 3. This compilation of ∆δ data
To provide additional evidence of whether the differ-
ences in ∆δ reflect true differences in bond order, the
for a series of phosphorothioates with varying bond
orders to the labeled oxygen atom is shown in Table 6
and displayed graphically in Figure 4.
P
frequencies of the infrared stretching bands of the
P-O(nitrophenyl) bond in the monoesters and the tri-
esters were measured. The location of this band in the
monoester pNPP has previously been assigned to ∼730
A small but statistically significant difference is noted
between the two correlations, indicating that a bond order
of approximately unity exhibits a larger isotope shift in
phosphorothioates than in phosphates. This raises the
possibility that the difference in isotope shifts in Table 3
might arise from a difference in the contributions from
anharmonicity between the two functional groups, af-
firming the caution raised by others15 that correlations
of isotope shifts with bond order must be made within
functional groups. An alternate interpretation is that the
P-O bond orders in phosphorothioates are indeed higher
than the formal bond orders obtained from Lewis struc-
tures, and the different correlations result from the
formal bond orders not reflecting the true bond orders.
-
1
26
cm , with a slight dependence on solvent. The IR
results are shown in Table 4.
The close similarity of the phosphate-ester stretching
vibrations in the phosphate versus phosphorothioate
(
21) Sorensen-Stowell, K.; Hengge, A. C. J. Org. Chem. 2005, 70,
4
1
1
805-4809.
(22) Webb, M. R.; Trentham, D. R. J. Biol. Chem. 1980, 255, 1775-
778.
(
23) Iyengar, R.; Eckstein, F.; Frey, P. A. J. Am. Chem. Soc. 1984,
06, 8309-8310.
(24) Gerlt, J. A.; Wan, W. H. Biochemistry 1979, 18, 4630-4638.
(
25) Frey, P. A.; Reimschussel, W.; Paneth, P. J. Am. Chem. Soc.
1
986, 108, 1720-1722.
(
26) Cheng, H.; Nikolic-Hughes, I.; Wang, J. H.; Deng, H.; O’Brien,
(
20) Lowe, G.; Potter, B. V. L.; Sproat, B. S. J. Chem. Soc. Chem.
P. J.; Wu, L.; Zhang, Z. Y.; Herschlag, D.; Callender, R. J. Am. Chem.
Soc. 2002, 124, 11295-11306.
Commun. 1979, 733-735.
8440 J. Org. Chem., Vol. 70, No. 21, 2005

Phosphate/Phosphorothioate Hydrolysis Thermodynamics
monoesters and triesters supports the conclusion that
there is no significant difference in the P-OR ester bond
as a result of sulfur substitution. Using the formulation
of activation. With a preassociated nucleophile in the
milieu of the enzymatic active site, the mechanistic
difference between the reactions of the phosphate and
phosphorothioate monoesters might disappear, minimiz-
ing the difference in the entropies of activation. Assuming
the relative enthalpies of activation remain similar, then
a slower reaction of the phosphorothioate monoester
would result without any mechanistic change to a triester
type mechanism.
2
6
of Cheng et al. to relate differences in the stretching
frequency of this bond to changes in bond length, the ∼5
-
1
cm difference between the phosphate and phospho-
rothioate monoesters and the triesters will correspond
to differences in bond length of ∼0.004 Å. This suggests
that the differences in isotope shift arise from a source
other than altered bond orders to the leaving group. We
conclude that the effect of sulfur substitution in the
nonbridging position on the scissile ester bond is insig-
nificant and is too small to account for the notable
Conclusions
The hydrolysis of the phosphorothioate monoester,
diester and triester examined in this study each exhibit
a greater enthalpy of activation than their phosphate
ester counterpart. The inverse thio effect in the mo-
noester reaction results from a mechanistic difference
that results from the sulfur substitution, in which uni-
molecular dissociation becomes rate-limiting. The ac-
companying large positive entropy of activation over-
comes the increased enthalpy of activation, resulting in
a faster hydrolysis rate for the phosphorothioate mo-
noester. In diesters and triesters, the differences in the
entropies of activation are more modest and not large
enough to offset the increased enthalpic barrier, resulting
in slower hydrolysis rates for phosphorothioate diesters
and triesters. The consistently greater enthalpic barrier
in the reactions of the phosphorothioates does not result
from a difference in strength of the scissile ester bond.
The source for the trend in enthalpies across the three
classes of esters examined herein is not revealed by this
study.
q
differences in ∆H .
What, then, could account for the consistently higher
q
values of ∆H in the reactions of phosphorothioates?
Differences in the extent of leaving group bond fission
in the transition states of phosphates compared to
phosphorothioates are not a likely source. Linear free
energy relationships and kinetic isotope effects have
shown that this parameter is very similar for the respec-
tive reactions of the monoesters, and for diesters and
triesters, leaving group bond fission is slightly more
advanced for reactions of phosphates than phospho-
2
7
rothioates. Moving to a consideration of the (thio)-
phosphoryl group, the changes in bonding between
phosphorus and the nonbridging atoms during these
reactions are diametrically opposite in monoesters than
in the other two classes of esters. In monoesters, the
reaction coordinate is dominated by leaving group expul-
sion, and bond order between the nonbridging atoms and
phosphorus should increase to maintain constant bond
order to phosphorus in the metaphosphate-like transition
state. In diesters, and even more so in triesters, nucleo-
philic attack dominates the reaction coordinate; the
transition state resembles a phosphorane, and bond order
to the nonbridging atom(s) decreases. This makes it
unlikely that bonding changes involving the nonbridging
Experimental Procedures
Materials. All reagents and solvents were purchased com-
mercially and unless otherwise noted were used as received.
Pyridine was distilled from calcium hydride. Tetrahydrofuran
(
THF) was distilled from sodium. p-Nitrophenol is a com-
q
atoms and phosphorus account for the higher ∆H across
mercial product that was recrystallized from toluene. p-
16
18
all three ester classes. A reviewer raised the question of
whether the higher barrier might be the result of steric
or electrostatic properties of the sulfur atom that slow
nucleophilic attack. For this to be the case, an alternative
explanation would need to be found for the observations
of stereochemical racemization and the significant posi-
tive entropy of activation in the hydrolysis of phospho-
rothioate monoesters, which have been interpreted to
imply that unimolecular dissociation to form thiometa-
phosphate is rate-limiting. At this point, a satisfactory
explanation for the trend in the enthalpies of activation
remains elusive.
Nitrophenol containing a mixture of O and O in the phenolic
oxygen atom was synthesized by exchange with labeled water
under alkaline conditions as previously described. The
percentage of O incorporation in the p-nitrophenol was
measured using GC-MS and found to be 41%.
3
1
1
8
Synthesis. Diethyl p-Nitrophenyl Phosphate (Paraox-
on). Natural abundance paraoxon was synthesized by adding
a solution containing 8.2 mmol of p-nitrophenol, 0.8 mmol of
dimethyl aminopyridine (DMAP), and 8.2 mmol of triethyl-
amine in 5 mL of THF dropwise to a stirring solution of 8.2
mmol diethyl chlorophosphate in 10 mL of THF. The reaction
mixture was allowed to stir under nitrogen in an ice bath for
4
h. The reaction mixture was then quenched with 15 mL of
water, extracted with diethyl ether (3 × 25 mL), and dried
with magnesium sulfate. The crude product was purified using
a flash silica gel column with a 3:1 hexanes/ethyl acetate
elution mixture and dried down by rotary evaporation. The
product, a light yellow oil, was characterized by NMR and
found to be pure.
Relevance to Enzymatic Thio Effects. A reversal
of the thio effect in monoesters from <1 in uncatalyzed
reactions to >1 in enzymatic reactions has been used as
an indicator that such reactions (for example, alkaline
phosphatase) might proceed by a triester-like transition
28-30
Diethyl p-Nitrophenyl Phosphorothioate (Parathion).
Natural abundance parathion was synthesized using the same
method as above, but with the appropriate diethyl chlorothio-
phosphate substitution. Bridging isotopically labeled paraoxon
and parathion analogues were also synthesized via the same
state.
The results here suggest a possible alternative
explanation. The opposite thio effect in monoester hy-
drolysis results from a large difference in the entropies
1
6
18
(
27) Catrina, I. E.; Hengge, A. C. J. Am. Chem. Soc. 2003, 125,
method using [ O, O] p-nitrophenol in place of natural
abundance p-nitrophenol.
7
546-7552.
(28) Breslow, R.; Katz, I. J. Am. Chem. Soc. 1968, 90, 7376-7377.
(29) Han, R.; Coleman, J. E. Biochemistry 1995, 34, 4238-4245.
(30) Mushak, P.; Coleman, J. E. Biochemistry 1972, 11, 201-205.
(31) Hengge, A. C. J. Am. Chem. Soc. 1992, 114, 2747-2748.
J. Org. Chem, Vol. 70, No. 21, 2005 8441

Purcell and Hengge
Ethyl p-Nitrophenyl Phosphate (EtO-pNPP) and Ethyl
p-Nitrophenyl Phosphorothioate (EtO-pNPPT). The di-
esters EtO-pNPP and EtO-pNPPT were synthesized as previ-
or pNPPT, as the cyclohexylammonium salt. The spectrometer
-
1
was set at a resolution of 2 cm for 32 scans between 400
and 4000 wavenumbers. Bridge-labeled analogues (31%-41%
2
7
18
ously described.
O incorporation) of each compound were also evaluated as
p-Nitrophenyl Phosphate (pNPP) and p-Nitrophenyl
Phosphorothioate (pNPPT). The bis(cyclohexylammonium)
the cyclohexylammonium salts in KBr pellets, to confirm the
assignment of the P-(O)R stretching vibration. Incorporation
3
2
11
18
salts of monoesters p-NPP and p-NPPT were synthesized
of the heavier O atom into the bridging bond causes a slight
as previously described. The same method was employed with
O, O] p-nitrophenol in order to synthesize the bridge-labeled
decrease in the frequency of the stretching vibration. This
decrease causes the infrared stretch to appear at a slightly
lower wavenumber. The shift enables identification of the
specific ester stretch of interest. P-(O)R stretching vibrations
for phosphorothioate and phosphate monoesters are, respec-
16
18
[
monoesters that were used for spectroscopic investigations.
Kinetics. The activation parameters for the aqueous hy-
drolysis of the monoanion and dianion of the monoesters
p-NPP and p-NPPT were previously reported.11 The alkaline
hydrolysis of the diesters and triesters followed first-order
kinetics with respect to hydroxide concentration (data not
-1
18
tively: [p-NPPT (νP-(O)R stretch ) 725 cm , νP-( O)R stretch ) 712
cm )], [p-NPP (ν P-(O)R stretch ) 730 cm , ν P-(18O)R stretch ) 721
-
1
-1
-
1
cm )].
Solutions (20 mM) were made of the cyclohexylammonium
salt of both monoesters and of their bridge labeled analogues
3
3,34
shown), consistent with previous studies.
Diesters. Reactions were initiated by injection of 500 µL
of a 20 mM aqueous stock solution of EtO-pNPP into 2 mL of
3
1
in deuterium oxide and analyzed via P NMR to determine
1
8
31
1
.2 N NaOH, giving a final hydroxide concentration of 1 N.
the O isotope effect on the P chemical shift. Spectra were
recorded at 161.97 MHz, with an acquisition time of 0.6 s, a
sweep width of 8200 Hz, a pulse width of 8 µs, and a line
broadening of 0.50 Hz.
Reaction mixtures were placed in sealed Pyrex tubes and were
maintained at constant temperatures in a heating block (50,
57, 64, 69, 75, 80, 86, and 92 °C). The progress of the reaction
was followed by periodically adding a 100 µL aliquot of the
reaction mixture to 3 mL of 0.1 N NaOH and recording the
absorbance due to p-nitrophenolate production, at 400 nm.
Activation parameters were calculated from Eyring plots,
using second-order rate constants obtained using the initial
rates method.
For the thiophosphate diester, reactions were initiated by
injection of 200 µL of a 12 mM aqueous stock solution of EtO-
pNPPT into 1 mL of 1.2 N NaOH. These reactions were heated
similarly at seven temperatures (57, 63, 74, 79, 85, 90, and 92
Triesters. Solutions of the triesters parathion and paraoxon
were made in chloroform. IR spectra of the triesters were
obtained using KBr cells. Chloroform solutions of the bridge
labeled triesters, with the same isotopic incorporations as the
monoesters above, were analyzed similarly. However, due to
the complicated nature of the triester spectra between 600 and
-
1
800 cm for both phosphate and phosphorothioate species, it
was more difficult to identify one specific P-(O)R stretch in
this area. Therefore, we have included all three stretches that
exhibit isotopic shifts in the area in which P-(O)R stretching
vibrations are expected. Possible P-(O)R stretching vibrations
for phosphorothioate and phosphate triesters are, respec-
°
4
0
C). Reaction course was monitored from the absorbance at
00 nm of 50 µL of the reaction mixture added to 1.5 mL of
-
1
.1 N NaOH. Rate constants were also obtained from the initial
tively: [parathion (νP-(O)R stretch ) 788, 770, and 749 cm
;
νP-( O)R stretch
18
) 779, 758, and 737 cm-1)], [paraoxon (νP-(O)R stretch
rates method.
-
1
18
Triesters. Stock solutions (2 mM) of parathion and paraox-
on in 1,4-dioxane were prepared. Aliquots of these stock
solutions (50 µL) were then added to 3 mL of sodium hydroxide
to initiate hydrolysis. Five hydroxide concentrations (0.10,
) 785, 764, and 737 cm ; νP-( O)R stretch ) 778, 751, and 728
1
-
cm )]. FT-IR spectrometer parameters were identical to those
used in the monoester experiments.
3
1
P NMR experiments were performed on 20 mM triester
0
.25, 0.50, 0.75, 1.0 M) were used in order to determine second-
solutions with chloroform as solvent. Acquisition parameters
were identical to those employed in the monoester experi-
ments.
order rate constants. These reactions were followed to comple-
tion (>10 half-lives) with a Cary-1 Bio spectrophotometer at
400 nm to monitor p-nitrophenolate liberation. Rate constants
for aqueous hydrolysis were calculated for each compound at
several temperatures (10, 15, 20, 25, and 30 °C).
Acknowledgment. The authors thank the NIH for
financial support (GM47297). Acknowledgment is given
to Shimadzu Instruments, Inc. (Columbia, MD) for use
of the Shimadzu Analytical Sciences Laboratory at
USU.
Spectroscopic Experiments. All FT-IR spectra were
31
obtained using a Shimadzu FTIR-4800 spectrophotometer.
P
NMR experiments were conducted using a Bruker-400 MHz
spectrometer.
Monoesters. Pellets (KBr) were made from oven-dried
potassium bromide and approximately 1 mg of either pNPP
Supporting Information Available: Eyring plots for the
alkaline hydrolysis of the phosphate and phosphorothioate
diesters and triesters. This material is available free of charge
via the Internet at http://pubs.acs.org.
(32) Bourne, N.; Williams, A. J. Org. Chem. 1984, 49, 1200-1204.
(33) Khan, S. A.; Kirby, A. J. J. Chem. Soc. B 1970, 1172-1182.
(34) Kirby, A. J.; Younas, M. J. Chem. Soc. B 1970, 510-513.
JO0511997
8442 J. Org. Chem., Vol. 70, No. 21, 2005