Isotope effects and medium effects on sulfuryl transfer reactions

Article abstract of DOI:10.1021/ja0163974

Kinetic isotope effects and medium effects have been measured for sulfuryl-transfer reactions of the sulfate ester p-nitrophenyl sulfate (pNPS). The results are compared to those from previous studies of phosphoryl transfer, a reaction with mechanistic similarities. The N-15 and the bridge O-18 isotope effects for the reaction of the pNPS anion are very similar to those of the p-nitrophenyl phosphate (pNPP) dianion. This indicates that in the transition states for both reactions the leaving group bears nearly a full negative charge resulting from a large degree of bond cleavage to the leaving group. The nonbridge O-18 isotope effects support the notion that the sulfuryl group resembles SO₃ in the transition state. The reaction of the neutral pNPS species in acid solution is mechanistically similar to the reaction of the pNPP monoanion. In both cases proton transfer from a nonbridge oxygen atom to the leaving group is largely complete in the transition state. Despite their mechanistic similarities, the phosphoryl- and sulfuryl-transfer reactions differ markedly in their response to medium effects. Increasing proportions of the aprotic solvent DMSO to aqueous solutions of pNPP cause dramatic rate accelerations of up to 6 orders of magnitude, but only a 50-fold rate increase is observed for pNPS. Similarly, phosphoryl transfer from the pNPP dianion to tert-amyl alcohol is 9000-fold faster than the aqueous reaction, while the sulfuryl transfer from the pNPS anion is some 40-fold slower. The enthalpic and entropic contributions to these differing medium effects have been measured and compared.

Full text of DOI:10.1021/ja0163974

9338
J. Am. Chem. Soc. 2001, 123, 9338-9344
Isotope Effects and Medium Effects on Sulfuryl Transfer Reactions
Richard H. Hoff,† Paul Larsen, and Alvan C. Hengge*
Contribution from the Department of Chemistry and Biochemistry, Utah State UniVersity,
Logan, Utah 84322-0300
ReceiVed June 12, 2001
Abstract: Kinetic isotope effects and medium effects have been measured for sulfuryl-transfer reactions of
the sulfate ester p-nitrophenyl sulfate (pNPS). The results are compared to those from previous studies of
phosphoryl transfer, a reaction with mechanistic similarities. The N-15 and the bridge O-18 isotope effects for
the reaction of the pNPS anion are very similar to those of the p-nitrophenyl phosphate (pNPP) dianion. This
indicates that in the transition states for both reactions the leaving group bears nearly a full negative charge
resulting from a large degree of bond cleavage to the leaving group. The nonbridge O-18 isotope effects
support the notion that the sulfuryl group resembles SO3 in the transition state. The reaction of the neutral
pNPS species in acid solution is mechanistically similar to the reaction of the pNPP monoanion. In both cases
proton transfer from a nonbridge oxygen atom to the leaving group is largely complete in the transition state.
Despite their mechanistic similarities, the phosphoryl- and sulfuryl-transfer reactions differ markedly in their
response to medium effects. Increasing proportions of the aprotic solvent DMSO to aqueous solutions of pNPP
cause dramatic rate accelerations of up to 6 orders of magnitude, but only a 50-fold rate increase is observed
for pNPS. Similarly, phosphoryl transfer from the pNPP dianion to tert-amyl alcohol is 9000-fold faster than
the aqueous reaction, while the sulfuryl transfer from the pNPS anion is some 40-fold slower. The enthalpic
and entropic contributions to these differing medium effects have been measured and compared.
In recent years there has been a growing realization of the
biochemical importance of sulfuryl transfer. Sulfation has a
crucial biological role in detoxification. In addition, sulfate
monoesters are found among all the classes of natural products,
including nucleotides, peptides and proteins, polysaccharides,
steroids, and lipids. Despite its biological importance, sulfate
ester chemistry has been the subject of much less study than its
phosphate ester chemistry counterpart. We report here kinetic
isotope effect measurements and medium effects studies of
sulfuryl-transfer reactions of p-nitrophenyl sulfate (pNPS). These
studies parallel prior work on phosphoryl transfer and add to
the body of knowledge of sulfuryl-transfer reactions and how
they compare with phosphoryl transfer.
Figure 1. Mechanisms for the hydrolysis (sulfuryl transfer to water)
of a sulfate monoester as the anion in the pH-independent region (A)
and the neutral species (B). The transition state of the anion mechanism
may in principle lie anywhere between the tight and loose extremes.
Some aspects of the sulfuryl-transfer mechanism are known
from previous studies. Sulfate ester hydrolysis is accelerated
under strongly acidic and strongly basic conditions, with a broad
pH-independent range between pH 4 and 12.1,2 Sulfate mono-
esters may exist in either of two protonation states, the anion
or the neutral species (Figure 1). There is evidence that different
mechanisms are followed by these species; the chemistry of
the anionic compound (A in Figure 1) will be summarized first.
18O tracer studies have shown that aryl sulfates undergo
hydrolysis by S-O bond fission in the pH-independent region.2
The reactions of alkyl sulfate esters in the pH-independent region
have not been studied. In very basic solutions (pH g 13), where
the rate of hydrolysis increases with [HO-], both aryl and alkyl
Most of the information about the mechanism in the pH-
independent region comes from linear free energy relationships
with aryl sulfates. These studies reveal a large sensitivity to
the pKa of the leaving group but a very small sensitivity to the
nucleophile, suggesting a concerted pathway having a loose
transition state.1,2,5-7 These LFER results are similar to those
for reactions of the analogous phosphate monoester dianions.
Accordingly, it has been suggested that these reactions likely
proceed by similar mechanisms and transition states. The
3,4
(3) Garner, H. K.; Lucas, H. J. J. Am. Chem. Soc. 1950, 72, 5497-
5501.
sulfate esters undergo hydrolysis via C-O bond cleavage.
* Corresponding author. Alvan C. Hengge, Department of Chemistry
and Biochemistry, Utah State University, Logan UT 84322-0300. Tele-
phone: 435-797-3442. Fax: 435-797-3390. E-mail: hengge@cc.usu.edu.
† Present address: Associate Professor, Department of Chemistry, U.S.
Military Academy, West Point, NY.
(1) Fendler, E. J.; Fendler, J. H. J. Org. Chem. 1968, 33, 3852-3859.
(2) Benkovic, S. J.; Benkovic, P. A. J. Am. Chem. Soc. 1966, 88, 5504-
5511.
(4) Kaiser, E. T.; Panar, M.; Westheimer, F. H. J. Am. Chem. Soc. 1963,
85, 602-607.
(5) Hopkins, A.; Day, R. A.; Williams, A. J. Am. Chem. Soc. 1983, 105,
6062-6070.
(6) D’Rozario, P.; Smyth, R. L.; Williams, A. J. Am. Chem. Soc. 1984,
106, 5027-5028.
(7) Bourne, N.; Hopkins, A.; Williams, A. J. Am. Chem. Soc. 1985, 107,
4327-4331.
10.1021/ja0163974 CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/30/2001

KIE and EIE on Sulfuryl-Transfer Reactions
J. Am. Chem. Soc., Vol. 123, No. 38, 2001 9339
possibility of a fully dissociative mechanism proceeding by way
of a free SO3 intermediate has been ruled out by stereochemical
studies; incubation of phenyl[(R)-16O,17O,18O]sulfate with an
acceptor alcohol in carbon tetrachloride solution was found to
form product of exclusively inverted configuration.8 Similarly,
metaphosphate is not typically an intermediate in hydrolysis
reactions of phosphate esters,9 although it does form in reactions
where tert-butyl alcohol is the phosphoryl acceptor.10 The small
deuterium isotope effect of 1.262 indicates that no general acid/
base catalysis takes place.
Figure 2. A diagram of p-nitrophenyl sulfate showing the positions
at which isotope effects were measured and the nomenclature used.
The significant negative entropy of activation of -18.5 eu2
for the pH-independent hydrolysis of p-nitrophenyl sulfate anion
raises the question of whether this reaction might proceed with
significantly more nucleophilic participation than the corre-
sponding phosphoryl transfer, which exhibits an entropy of
activation of +3.5 eu.11 Negative values for ∆Sq of this
magnitude are typically indicative of associative mechanisms
with significant nucleophilic participation. The reason for the
discrepancy between the entropies of activation of these two
reactions remains unexplained.
Figure 3. Isotopic isomers used for measurement of the bridge (A
and C) and nonbridge (B and C) 18O isotope effects.
In the acidic hydrolysis of sulfate esters, the reactive species
is the neutral molecule (B in Figure 1) and this form is much
more reactive than the anion. A reduced value for âlg and a
on the rate constant k. Previous isotope effect studies on
reactions of p-nitrophenyl phosphate (pNPP) allow for direct
comparisons to be made between the two reactions. We have
also examined medium effects on the sulfuryl-transfer reaction
of pNPS in neat tert-amyl alcohol and in DMSO/water mixtures.
Since medium effects on phosphate monoester hydrolysis have
been well characterized under these conditions, the results allow
a comparison of medium effects on reactions of the two types
of esters.
1
solvent isotope effect of 2.4312 led to the proposal that the
hydrolysis of the neutral species proceeds via a two-step pathway
involving equilibrium proton transfer to the incipient leaving-
group oxygen atom followed by O-S bond cleavage. This is
mechanism B in Figure 1. The intermediacy of free SO3 in the
acid hydrolysis is often assumed but has never been proven.
Also consistent with previous data is mechanism C, in which a
proton is transferred to the leaving group in the same step as
sulfuryl transfer.
Experimental Section
18O-labeled sulfuric acid (95% H2SO4 in H2O) 95% atom percent
18O, was purchased from Isotec. Natural abundance p-nitrophenol was
a commercial product, recrystallized from toluene.
The potassium salt of pNPS was obtained from a commercial source
but was found to contain excessive free phenol. This commercial
potassium salt after recrystallization from aqueous KOH was used for
some work, but most kinetic work was done using locally synthesized
salts. Two different literature methods17,18 were used to synthesize
pNPS, in an effort to optimize the process for the later synthesis of
labeled compounds.
Unlabeled pNPS was converted to the tetra-N-butylammonium salt
by ion exchange using DOWEX-50A cation-exchange resin. A 1:1
stoichiometry was confirmed by integration of the proton NMR.
The isotopic isomers of p-nitrophenyl sulfate needed to measure the
isotope effects are shown in Figure 3. [14N]-p-nitrophenol, [15N]-p-
nitrophenol and [15N,18O]-p-nitrophenol were synthesized as described
previously.19 [14N]-p-nitrophenol, and [15N,18O]-p-nitrophenol were
mixed to reconstitute the natural abundance of 15N, and then the mixture
was sulfurylated to produce p-nitrophenyl sulfate using the method
It has been noted that the addition of organic cosolvents,
especially DMSO, to aqueous solutions of phosphate monoesters
increases the rate constant for hydrolysis by the dianion by up
to 106-fold.13 An analogous effect resulting from binding has
been postulated to be a possible contributor to enzymatic
catalysis.13 Whether sulfate ester anion hydrolysis is subject to
similar medium effects has not been determined, although the
acid-catalyzed hydrolysis of alkyl sulfate esters is accelerated
in moist dioxane compared to the rate in aqueous solution.14,15
To further understand the mechanism and the nature of the
transition state for sulfuryl transfer we have measured kinetic
isotope effects for the acid-catalyzed and the pH-independent
hydrolysis of pNPS. The substrate is shown in Figure 2, with
the positions indicated where isotope effects were measured.
The notation used to express isotope effects is that of Northrop16
where a leading superscript of the heavier isotope is used to
indicate the isotope effect on the following kinetic quantity;
for example 15k denotes k14/k15, the nitrogen-15 isotope effect
referred to above. This mixture of isotopic isomers (A and C, Figure
18
3) was used for determination of
k
.
bridge
(8) Chai, C. L. L.; Hepburn, T. W.; Lowe, G. J. Chem. Soc., Chem.
Commun. 1991, 1403-1405.
(9) Herschlag, D.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111, 7579-
7586.
Measurement of the isotope effect in the nonbridge oxygen atoms
requires the isomers B and C in Figure 3. 14N-labeled pNPS (C) was
synthesized from [14N]-p-nitrophenol using the sulfur trioxide-pyridine
complex method.18 [18O3]-chlorosulfonic acid was prepared from [18O4]-
sulfuric acid (obtained from Isotec) using the method of Williamson,
from sulfuric acid and phosphorus pentachloride.20 The [18O3]-chloro-
sulfonic acid thus obtained was reacted with 15N-labeled p-nitrophenol
to form isomer B (Figure 3), using the method of Burkhardt and
Lapworth.17 The percent of 18O incorporation in the product was
(10) Friedman, J. M.; Freeman, S.; Knowles, J. R. J. Am. Chem. Soc.
1988, 110, 1268-1275.
(11) Kirby, A. J.; Jencks, W. P. J. Am. Chem. Soc. 1965, 87, 3209-
3216.
(12) Kice, J. L.; Anderson, J. M. J. Am. Chem. Soc. 1966, 88, 5242-
5245.
(13) Abell, K. W. Y.; Kirby, A. J. Tetrahedron Lett. 1986, 27, 1085-
1088.
(14) Batts, B. D. J. Chem. Soc. (B) 1966, 547-551.
(15) Burstein, S.; Lieberman, S. J. Am. Chem. Soc. 1958, 80, 5235-
5239.
(16) Northrop, D. B. In Determining the Absolute Magnitude of Hydrogen
Isotope Effects; Cleland, W. W., O’Leary, M. H., Northrop, D. B., Eds.;
University Park Press: Baltimore, MD, 1977; pp 122-152.
(17) Burkhardt, G. N.; Lapworth, A. J. Chem. Soc. 1926, 684-687.
(18) Benkovic, S. J.; Dunikoski, L. K. J. Biochemistry 1970, 9, 1390-
1397.
(19) Hengge, A. C.; Edens, W. A.; Elsing, H. J. Am. Chem. Soc. 1994,
116, 5045-5049.
(20) Williamson, A. Proc. Royal Soc. London 1854/55, 7, 11-15.

9340 J. Am. Chem. Soc., Vol. 123, No. 38, 2001
Hoff et al.
measured by matrix assisted laser desorption/ionization time-of-flight
mass spectrometry, using a nitrogen laser at 337 nm and a 2.1 m linear
TOF mass spectrometer. The results are shown in Figure 1S of the
Supporting Information. Integration of the results showed that 64.6%
of the product was triply labeled in the nonbridge position, with the
remainder doubly labeled. This percentage was used in the mathematical
correction of the observed isotope effect.
Isotope Effect Determinations. The protocols for carrying out these
experiments were the same as those previously described for p-
nitrophenyl phosphate.19 In this method the reactions are allowed to
proceed to partial completion and stopped, followed by separation of
the product and residual substrate. These are then subjected to isotopic
analysis by isotope ratio mass spectrometry to determine the nitrogen
isotope ratios.
The 15N isotope effects were measured using natural abundance
pNPS. The 18O isotope effects were measured by the remote-label
method.22 In these experiments the nitrogen atom in the reactant is used
as a reporter for the bridging oxygen atom or the nonbridging oxygen
atoms. These experiments yield an observed isotope effect that is the
product of the effect due to 15N and that due to 18O substitutions. The
observed isotope effects from these experiments were then corrected
for the 15N effect and for incomplete levels of isotopic incorporation
in the starting material as previously described.23
Kinetics Experiments. The potassium salt of pNPS was used in all
aqueous kinetic experiments. Concentrations of pNPS used for kinetic
runs were 5-13 mM. Thermostatically controlled water baths or dry
blocks were used to maintain reaction temperatures over the course of
the experiments.
Isotope effects were measured for the hydrolysis of pNPS under
three conditions: in 100 mM CHES buffer at pH 9.0 at 85 °C, in 1.0
N HCl at 65 °C and 21 °C, and in 10 N HCl at 15 °C. Reactions were
begun with 100 µmol of pNPS. Experiments were performed in
triplicate with individual experiments stopped at fractions of reaction
ranging from 30 to 60%. Extents of reaction were measured by assaying
for p-nitrophenol in 0.1 N NaOH at 400 nm. The reactions at pH 9
were stopped by chilling the solutions on ice; the acidic reactions were
stopped by titration to pH 5. After the reaction solutions were titrated
to pH 5 they were extracted three times with an equivalent volume of
ethyl ether to quantitatively remove the p-nitrophenol product. The
aqueous layer was evaporated briefly under vacuum to remove dissolved
ether, sufficient 10 N HCl was added to bring the solution to at least
1.0 N in HCl, and the solutions were heated overnight to 80 °C to
completely hydrolyze the remaining pNPS. They were then titrated back
to pH 5 and extracted with ether, the p-nitrophenol in this ether fraction
representing the residual substrate from the initial reaction. The ether
fractions were dried over magnesium sulfate and filtered, and the solvent
was removed by rotary evaporation. The p-nitrophenol was sublimed
under vacuum at 90 °C, and 1.0 mg samples were prepared for isotopic
analysis using an ANCA-NT combustion system in tandem with a
Europa 20-20 isotope ratio mass spectrometer.
Reactions performed in 10 N HCl provided some challenges in terms
of kinetic control. At this concentration of acid, the hydrolysis rate
was very high; hence, time spent in mixing the reactant into the solution
was critical. The pNPS was dissolved in a small measured volume of
water. The volume of water was calculated to bring a measured volume
of concentrated HCl to 10 N when the two were combined. To prevent
a large temperature rise due to the production of heat upon mixing,
the solutions were cooled to a temperature below the desired reaction
temperature such that when the reactants were combined the heat
evolved would not raise the temperature of the mixture above the
desired temperature. To stop these reactions, the 10 N acid solution
was rapidly combined with 10 times its volume of ice-cold 1.0 N NaOH.
A small (<5%) excess of base was used to ensure the resulting solution
would be basic, where the reaction is orders of magnitude slower. This
cold solution was assayed for p-nitrophenol and then titrated to pH 5
and handled as described above to isolate product from unreacted pNPS.
Isotope effects were calculated from the isotopic ratios at partial
reaction in the p-nitrophenol product (Rp), in the residual substrate (Rs),
and in the starting material (Ro). Equations 1 and 2 were used to
calculate the observed isotope effect either from Rp and Ro or from Rs
and Ro respectively at fraction of reaction f.21 Thus each experiment
yields two independent determinations of the isotope effect.
The pH-independent aqueous reaction was examined in 100 mM
TRIS buffer at pH 9.0 at temperatures from 35 to 95 °C. Over time,
aliquots were removed and added to measured portions of 0.1 N NaOH,
and reaction progress was monitored spectrophotometrically by tracking
the evolution of the p-nitrophenol hydrolysis product at 400 nm (ꢀ )
18 320).24 The plot of absorbance versus time over the first 1-5% of
reaction was analyzed by assuming first-order kinetics using the initial
rates method. To determine the initial substrate concentration, an aliquot
of the reaction mixture was subjected to complete hydrolysis in a
measured amount of 1.0 N HCl, which was heated at 80 °C for 4 h
and then assayed for p-nitrophenol.
Reactions in tert-amyl alcohol were performed using the tetrabuty-
lammonium salt of pNPS. In cases where the reactions were very slow,
concentrations as high as 80 mM were used to obtain measurable
changes in the composition of the reaction mixture. Evolution of free
phenol was quantified by adding aliquots of the reaction solution to
measured portions of 50% aqueous ethanol, which was 0.1 N in NaOH.
The solutions were then analyzed spectrophotometrically for phenolate
ion at 405 nm (the λmax is slightly shifted in the ethanolic solution).
Rate constants were measured using the initial rates method.
Reactions in DMSO/water mixtures were carried out using the
potassium salt of pNPS. Reactions were 5 mM in pNPS, 0.1 M in CHES
buffer, pH 9, with ionic strength of 1.0 maintained using tetraethyl-
ammonium chloride. Fractions of DMSO ranged up to 95%, and rates
were measured at temperatures from 60 to 90 °C. The pKa of the buffer
and of the pNPS, and the effective pH of the solution, will vary
somewhat with the DMSO content of the solution. However the very
broad pH-independent range for hydrolysis and the very low pKa of
pNPS give confidence that the anion is the reactive species present.
Rate constants were measured using the initial rates method, assaying
for released p-nitrophenol by adding aliquots of the reaction mixtures
to 0.1 N NaOH.
Results
Values of the first-order rate constants for the reactions of
pNPS at pH 9.0 in aqueous solution, in 95% DMSO/water
mixtures, and in neat tert-amyl alcohol were determined by
following the liberation of p-nitrophenolate anion spectropho-
tometrically. The rate constants for these reactions gave linear
Eyring plots (Figures 2S and 3S) which were used to calculate
the enthalpies and entropies of activation. The enthalpy of
activation for each reaction was calculated from the slope. The
free energy of activation for each reaction was calculated at 35
°C using the equation ∆Gq ) -RT ln(kTh/kT), and the entropy
of activation was then calculated using the relation ∆Sq ) (∆Hq
- ∆Gq)/T. Table 1 shows the rate and activation parameter data
for these reactions. The pH-independent hydrolysis rate of pNPS
at 35 °C, and the activation parameters are in reasonable
agreement with previously reported data.2
isotope effect ) log(1 - f)/log(1 - f (Rp/Ro))
isotope effect ) log(1 - f)/log [(1 - f) (Rs/Ro)]
(1)
(2)
RO was determined by two methods: from unreacted pNPS by
isotope ratio mass spectrometry, and, as a control, from p-nitrophenol
isolated after complete hydrolysis of a sample of pNPS using the same
isolation and purification procedures used in the isotope effect
experiments. The agreement of these two numbers demonstrated that,
within experimental error, no isotopic fractionation occurs as a result
of the procedures used to isolate and purify the p-nitrophenol.
(22) O’Leary, M. H.; Marlier, J. F. J. Am. Chem. Soc. 1979, 101, 3300-
3306.
(23) Caldwell, S. R.; Raushel, F. M.; Weiss, P. M.; Cleland, W. W.
Biochemistry 1991, 30, 7444-7450.
(24) Kirby, A. J.; Varvoglis, A. G. J. Am. Chem. Soc. 1967, 89, 415-
423.
(21) Bigeleisen, J.; Wolfsberg, M. AdV. Chem. Phys. 1958, 1, 15-76.

KIE and EIE on Sulfuryl-Transfer Reactions
J. Am. Chem. Soc., Vol. 123, No. 38, 2001 9341
Table 1. Rates and Activation Parameters for Reactions of p-Nitrophenyl Sulfate Anion
aqueous
tert-amyl alcohol
5.07 × 10-11
95% DMSO
k at 35 °C, s-1
1.87 × 10-9
2.47 × 10-9
9.26 × 10-8
50
ksolv/kaqueous at 35 °C
∆Hq, kcal/mol
∆Sq, eu
1
0.027
24.4 ( 0.2
-19.5 ( 0.4
this work
24.6
-18.5
ref 2
25.0 ( 0.7
-25 ( 1
this work
23.2 ( 0.5
-16 ( 1
this work
source
Table 2. Kinetic Isotope Effects for the Aqueous Hydrolysis of pNPS and pNPPa
pNPS KIEs
pH 9.0, 85 °C
1 N HCl, 65 °C
1 N HCl, 21 °C
10 N HCl, 15 °C
18
k
0.9951 (3)
1.0210 (10)
1.0026 (1)
1.0067 (1)
1.0069 (2)
1.0002 (1)
1.0083 (2)
1.0097 (2)
1.0002 (1)
1.0098 (3)
1.0101 (2)
1.0004 (1)
nonbridge
18
k
bridge
15k
pNPP KIEs
dianion, 95 °C
monoanion, 95 °C
monoanion, 30 °C
18
k
k
0.9994 (5)
1.0189 (5)
1.0028 (2)
1.0184 (5)
1.0087 (3)
1.0004 (2)
1.0199 (3)
1.0094 (3)
1.0005 (1)
nonbridge
bridge
18
15k
18
a Results for the sulfate monoester are from this work, phosphate results were previously reported elsewhere.19,36 Values for
KIEs for all three nonbridge atoms labeled. Standard errors in the last decimal place(s) are shown in parentheses.
k
are the
nonbridge
shown. Six or more determinations of each isotope effect were
made. Isotope effects were measured for the pNPS monoanion
at pH 9.0, under conditions where the substrate was partially
protonated in 1.0 N HCl, and in 10 N HCl where a greater
fraction of the reactant is protonated. The values for the 18O
isotope effects have been corrected for the 15N effect and for
levels of isotopic incorporation.
It was not feasible to measure the KIEs for the hydrolysis of
the pNPS anion and of the conjugate acid at the same
temperature, due to the large difference in the reactivities of
the two species. The temperature difference was minimized to
make the temperature-induced differences in isotope effects as
small as possible. The KIEs for the hydrolysis of the anion were
measured at 85 °C, and those for the acid-catalyzed reaction in
1 N HCl were measured at 65 °C. To provide results in 1.0 N
HCl to compare both the very slow reaction of the anion and
the very fast reaction in 10 N HCl, the isotope effects in 1.0 N
HCl were measured at 65 and at 21 °C.
Figure 4. Dependence of the pseudo-first-order rate constant at 80
°C for hydrolysis of the anion of pNPS as a function of DMSO content
in DMSO/water mixtures.
The monoanion of pNPS was found to undergo sulfuryl
transfer to the solvent approximately 40 times slower in tert-
amyl alcohol than in water at 35 °C. This is in contrast to the
solvolysis of p-nitrophenyl phosphate, which is approximately
9000-fold faster in tert-amyl alcohol at 39 °C.25 The activation
parameters for the reactions of pNPS in the two solvents are
similar, in contrast to the pNPP phosphoryl-transfer reaction,
in which the entropy of activation changes from +3.5 in water
to +23 eu in tert-amyl alcohol.25
The effect of DMSO on the rate of hydrolysis of pNPS is
shown in Figure 4. The rate constant increases with the fraction
of DMSO, as has been observed with the hydrolysis of pNPP,13
but to a much smaller extent. At 80 °C only a 15-fold rate
increase is observed in 95% DMSO relative to the aqueous
reaction. Comparison of the extrapolated rate constants at 35
°C for the pNPS reaction in 95% DMSO with the aqueous rate
at the same temperature yields a 50-fold rate acceleration. By
comparison, a 106-fold increase is observed in the phosphoryl
transfer of pNPP at 39 °C.13
Discussion
Mechanistic Similarities between Phosphoryl and Sulfuryl
Transfer. LFER results suggest that the reactions of phosphate
monoester dianions and of sulfate ester anions are similar. In
both of these reactions the phosphoryl and sulfuryl groups are
fully deprotonated, and the major impetus for leaving-group
departure is internal electron donation from charge borne on
the nonbridge oxygen atoms. Another mechanistic similarity
exists between reactions of the phosphate monoanion and the
neutral sulfate ester. In both of these species, one of the
nonbridge oxygen atoms is protonated. In hydrolysis reactions
of both, it is proposed that the proton is transferred to the leaving
group during the reaction. The kinetic isotope effect data and
results from medium effects give further insights into the
similarities and differences of these sets of reactions.
Isotope Effects and the Sulfate Monoester Hydrolysis
Transition State. Reaction of the pNPS Anion. The kinetic
isotope effects at pH 9.0 represent the reaction in the pH-
independent range where pNPS is present as the anion. The
secondary 15k isotope effect of 1.0026 is comparable to the
equilibrium 15K effect for the deprotonation of p-nitrophenol
of 1.0023.26 This isotope effect is sensitive to the amount of
The isotope effects for the solution reactions of pNPS and
their standard errors are given in Table 2. The isotope effects
obtained from isotopic ratios of product, and those obtained from
isotopic ratios of residual substrate, agreed within experimental
error in all cases and were averaged together to give the results
(26) Hengge, A. C.; Cleland, W. W. J. Am. Chem. Soc. 1990, 112, 7421-
7422.
(25) Hoff, R. H.; Hengge, A. C. J. Org. Chem. 1998, 63, 6680-6688.

9342 J. Am. Chem. Soc., Vol. 123, No. 38, 2001
Hoff et al.
negative charge delocalized into the nitrophenolate group,19,26
and the observed value indicates that the leaving group bears
essentially a full negative charge in the transition state. In
phosphoryl-transfer reactions from the dianion of pNPP this
isotope effect can also be slightly larger than the equilibrium
effect for deprotonation.27 This is likely due to the fact that in
the transition state the leaving group, which resembles p-
nitrophenolate anion, remains in close proximity to the phos-
phoryl or sulfuryl group and the phenolic oxygen is not well-
solvated. This would result in greater charge delocalization (a
larger contribution from the quinonoid resonance form). Rela-
tively less charge will be delocalized in the phenolate anion in
aqueous solution as measured in the equilibrium isotope effect,
where charge on the phenolate oxygen atom is stabilized by
Isotope Effects and the Sulfate Monoester Hydrolysis
Transition State. The Reaction of Neutral pNPS. Isotope
effects were measured for the solution reaction of pNPS in acid
conditions for comparison with the reaction of the anion. At 35
°C the reaction is 105 fold faster in 1.0 N HCl than in the pH-
independent region at pH 9.0. The mechanism of reaction of
the neutral species has been postulated to proceed via the
zwitterionic species shown in part B of Figure 1.
Kinetic isotope effects are sensitive to the reaction temper-
ature, but due to the extreme difference in reactivity, these two
reactions could not be investigated at the same temperature. A
compromise was reached, with the pH 9 reaction studied at 85
°C and the 1.0 N HCl reaction at 65 °C.
In the reaction under acidic conditions the 15k isotope effect
is essentially unity, indicating that no significant negative charge
hydrogen bonding with solvent.
18
The
k
is a primary isotope effect as the S-O bond is
is developed on the nitrophenyl group in the transition state.
bridge
18
broken in the transition state. The value of 1.0210 for pNPS
hydrolysis is similar to the
The
k
isotope effect is reduced from its value in the
bridge
18
reaction of the anion by an amount close to the equilibrium 18
K
k
value of 1.0189 observed
bridge
for reaction of the pNPP dianion at the slightly higher
temperature of 95 °C. This indicates that the bond to the leaving
group is largely broken in the transition state, as is the case for
isotope effect for deprotonation of p-nitrophenol, which is
1.015330 (protonation will result in an inverse contribution to
18
k
of the same magnitude). Thus, the observed kinetic
bridge
18
phosphoryl transfer. Together the two isotope effects,
k
isotope effect is consistent with a transition state in which the
bridge
and 15k indicate that sulfuryl and phosphoryl transfer from the
p-nitrophenyl esters have very similar transition states both with
respect to the extent of leaving-group bond cleavage and the
amount of charge borne on the leaving group.
O-S bond is largely broken, and protonation of the leaving
18
group is essentially complete. Both 15k and
k
are very
bridge
similar to the values from the reaction of the pNPP monoanion
(Table 2).
An interpretation of the secondary isotope effect, 18knonbridge
,
The
k
isotope effect for this reaction will be
nonbridge
18
may be considered in part by analogy with a fairly considerable
influenced by the protonation state of the sulfuryl group. In the
mechanistically analogous reaction of the pNPP monoanion, the
isotope effects were measured at a pH (3.5) at which the
phosphoryl group (pKa 5.1) was essentially completely in the
monoprotonated state. In this case there is no isotopic fraction-
ation in the species that are present in the correct protonation
state for reaction. However the pKa of pNPS is too low to obtain
a large proportion of the neutral species. The pKa of unsubsti-
tuted sulfuric acid is on the order of -3,31,32 and the p-
nitrophenyl substituent will make this value even more negative.
Thus, even in 1 N HCl much less than 1% of pNPS will be in
the neutral form, and therefore the observed kinetic isotope effect
should be corrected by the full equilibrium isotope effect for
protonation. The 18K isotope effect for protonation of sulfate,
body of data for phosphoryl transfer. In phosphoryl-transfer
18
reactions,
k
is small and inverse for loose transition
nonbridge
states in which the phosphoryl group resembles metaphosphate
and is normal for reactions in which the transition state
resembles a pentacoordinate phosphorane.27 The change in
hybridization and in bonding to the nonbridge oxygen atoms is
directly analogous in sulfuryl and phosphoryl-transfer reactions.
This suggests that a similar trend might be expected for the
analogous sulfuryl-transfer reaction. Calculations at the
6-31++G** level of the expected equilibrium isotope effect
between pNPS and sulfur trioxide yield a value of 0.9910 (P.
Czyryca and A. Hengge, unpublished results). However we have
not been able to successfully model 18knonbridge for an associative
mechanism for comparison. Thus, firm conclusions cannot be
drawn from the value of 18knonbridge. However, the experimental
value of 0.9951 is consistent with expectations from the
calculation and from KIEs for analogous phosphoryl-transfer
reactions, for a transition state in which the sulfuryl group
resembles SO3. The value of 18knonbridge is more inverse than in
the reaction of the pNPP dianion. This suggests that nonbridge
S-O bond order increases more than nonbridge P-O bond order
in the corresponding phosphate ester reaction. The metaphos-
or of a sulfate ester, is unknown. However, reported values for
18
the
k
equilibrium isotope effect (EIE) for protonation
nonbridge
of phosphate, of phosphate esters, for carboxylic acids and for
formic acid, are all in the range of 1.015-1.02.33 Assuming
that the fractionation factor for the sulfate ester is similar to
that for a phosphate ester (1.016)34 yields a corrected isotope
effect of 1.023 for the acidic hydrolysis of pNPS in 1 N HCl at
21 °C. This is reasonably close to the value of 1.0199 measured
for the hydrolysis of the monoanion of pNPP at a similar
temperature (Table 2) and thus supports the hypothesis of a
similar mechanism for the two reactions, in which proton
transfer from the sulfuryl group is essentially complete in the
transition state.
-
phate-like PO3 unit has considerable contribution from a
resonance form with a positively charged phosphorus atom, two
negative charges on oxygen atoms, and diminished nonbridge
P-O bond order.28,29
The data for the reaction in 10 N HCl (Table 2) show that
Together, the isotope effects suggest a transition state like
the one shown diagrammatically in Figure 1 A. The bond to
the leaving group is largely broken, and the leaving group bears
essentially a full negative charge. The sulfuryl group resembles
sulfur trioxide.
18
while negligible changes result in 15k and in
k
, the
bridge
(30) Hengge, A. C.; Hess, R. A. J. Am. Chem. Soc. 1994, 116, 11256-
11263.
(31) Luder, W. F.; Zuffanti, S. The Electronic Theory of Acids and Bases,
2nd ed.; Dover Publications: New York, 1961.
(27) Hengge, A. C. In Insights from HeaVy-Atom Isotope Effects on
Phosphoryl and Thiophosphoryl Transfer Reactions; Frey, P. A., Northrop,
D. B., Eds.; IOS Press: Amsterdam, 1999; pp 72-84.
(28) Horn, H.; Ahlrichs, R. J. Am. Chem. Soc. 1990, 112, 2121-2124.
(29) Rajca, A.; Rice, J. E.; Streitweiser, A., Jr.; Schaefer, H. F. J. Am.
Chem. Soc. 1987, 109, 4189-4192.
(32) Shriver, D. F.; Atkins, P.; Langford, C. H. Inorganic Chemistry,
2nd ed.; W.H. Freeman and Company: New York, 1994.
(33) Rishavy, M. A.; Cleland, W. W. Can. J. Chem. 1999, 77, 967-
977.
(34) Knight, W. B.; Weiss, P. M.; Cleland, W. W. J. Am. Chem. Soc.
1986, 108, 2759-2761.

KIE and EIE on Sulfuryl-Transfer Reactions
J. Am. Chem. Soc., Vol. 123, No. 38, 2001 9343
sulfuryl group resembling neutral sulfur trioxide. If these two
processes have different solvent reorganization requirements,
then they may be the source of the variance in the entropies of
activation.
Sulfuryl Transfer from the pNPS Anion to tert-Amyl
Alcohol. In a study of the phosphoryl-transfer reaction of the
pNPP dianion in tert-amyl alcohol, it was found that the switch
from aqueous solution to tert-amyl alcohol was accompanied
by a 9000-fold increase in reaction rate at 39 °C.25 The rate
increase is due to entropic effects; while the activation enthalpies
in the two solvents are the same within experimental error, in
tert-amyl alcohol ∆Sq is +23 eu, compared to +3 in the aqueous
reaction.25 Previous stereochemical studies imply that in the
similar solvent and phosphoryl acceptor tert-butyl alcohol, the
phosphoryl reaction becomes completely dissociative, with a
metaphosphate intermediate.10 Such a mechanism is consistent
with a larger positive value for ∆Sq. By contrast, in the sulfuryl
transfer reaction of the pNPS anion the same solvent change
from water to tert-amyl alcohol is accompanied by a 40-fold
decrease in rate, and ∆Sq is somewhat more negative than the
value for the reaction in water. The slower rate is due to a
combination of enthalpic and entropic effects, but primarily the
latter (∆∆Hq ) 0.6 kcal/mol, T∆∆Sq ) 1.8 kcal/mol).
Because solvent can heavily influence entropies of activation,
it is difficult to assign a mechanistic cause to the more negative
value in tert-amyl alcohol, particularly since the difference is
modest. Clearly the data do not support a shift to a more
dissociative mechanism in the pNPS reaction, as in the phos-
phoryl transfer from pNPP in tert-amyl alcohol. The more
negative entropy of activation for the pNPS reaction in tert-
amyl alcohol than for that in water might be taken to imply a
transition state with more nucleophilic participation, but this is
counterintuitive with such a weak and sterically hindered
nucleophile.
Sulfuryl Transfer from the pNPS Anion in DMSO/Water
Mixtures. Figure 4 shows the effects of increasing fractions of
DMSO on the rate constant for hydrolysis of the pNPS anion
at 80 °C. While the general shape of the profile is similar to
that reported for the reaction of pNPP dianion, the rate
acceleration is very modest. The rate constants in water and in
95% DMSO/water were calculated at 35 °C from the Eyring
plot, for better comparison with previously reported data for
the pNPP reaction at 39 °C. This gives a very modest rate
acceleration of 50-fold at 35 °C, in contrast to the up to 106-
fold rate increase with 95% DMSO in the corresponding
phosphoryl-transfer reaction of pNPP. The internal energy of
bond fission ∆Hq for pNPP hydrolysis is significantly lowered
in 95% DMSO to 21 kcal/mol, from 31 kcal/mol in water. For
pNPS hydrolysis this parameter is negligibly affected (Table
1). These results suggest that disruption of solvation of the
monoanionic sulfuryl group of the sulfate ester gives rise to
much less reactant destabilization than the analogous effect upon
the dianionic phosphoryl group. This is a logical consequence
of the difference in charge between the two species. This
suggests that, while desolvation effects are important in phos-
phoryl transfer and may be a potential source for enzymatic
catalytic power for phosphatases,13 such effects are less
important for sulfuryl transfer and are not a significant potential
source for catalytic power.
Figure 5. Charge dispersion in the transition state of reaction of
phosphate monoester dianion, top, versus charge transfer in a sulfate
ester anion reaction, bottom. Data suggest that in each transition state
essentially a full negative charge resides on the leaving group, and
bond formation to the nucleophile is minimal.
18
k
isotope effect increases by considerably more than
nonbridge
the experimental uncertainty. This change is in the direction
expected if a greater (although still very small) fraction of the
pNPS present is in the correct protonation state for reaction,
and less of the inverse EIE for protonation is expressed in the
observed KIE. This indicates that the sulfuryl-protonated species
is the reactive form that then transfers the proton (probably via
the intermediacy of one or more water molecules) to the leaving
group.
One could hypothesize that given the very low pKa of the
sulfuryl group of pNPS, the pKa of the bridge oxygen atom
might not be too much lower, and thus protonation might not
occur on the sulfuryl group at all in the hydrolysis reaction. In
such a scenario the bridge oxygen atom is protonated directly
by hydronium ion to give the zwitterion. If this were the case,
18
then the
k
would be inverse, not normal.
nonbridge
If mechanism B in Figure 1 is followed, the EIE for
protonation and that for deprotonation would effectively cancel,
and 18knonbridge would be inverse, as for the anion reaction. Only
the reaction of the neutral form in which proton transfer occurs
in the transition state, with the imaginary frequency factor arising
from O-H bond cleavage giving a large normal contribution
to the isotope effect,35 will give the observed results. This is
shown in mechanism C.
Kinetic Data and Medium Effects. Aqueous Reaction of
the pNPS Anion. The rate of aqueous hydrolysis and the
activation parameters calculated here agree well with those
reported earlier.1,2 The significant negative entropy of activation
for the aqueous hydrolysis of the anion has always been an
anomaly when all of the evidence is combined into a coherent
model of the mechanism of this reaction. The individual
activation entropies reported in Table 1 are calculated at 35 °C;
however, the calculated values were very consistent across the
entire temperature range evaluated (data not shown). The KIE
data, in agreement with LFER studies, are fully consistent with
a transition-state geometry similar to that of the analogous
reaction of the pNPP dianion. If the different value for ∆Sq does
not have a mechanistic origin, then it likely results from differing
solvation effects. One possible origin for different solvation
effects is the difference in charge distributions that occurs during
the two reactions (Figure 5). In the loose transition state of the
pNPP dianion reaction, the -2 charge of the reactant becomes
dispersed between the phosphoryl group and the leaving group,
in which each bears a charge of about -1. In the transition
state of the sulfuryl-transfer reaction implied by the kinetic
isotope effects and prior linear free energy relationships, charge
is transferred rather than dispersed. The reactant’s charge of
-1 has been largely transferred to the leaving group, with the
Conclusions
(35) Melander, L.; Saunders, W. H. Rection Rates of Isotopic Molecules;
Wiley: New York, 1980.
(36) Czyryca, P. G.; Hengge, A. C. Biochim Biophys Acta 2001, 1547,
245-53.
The kinetic isotope effects support the notion that sulfuryl-
transfer reactions from aryl sulfate esters have similar mecha-

9344 J. Am. Chem. Soc., Vol. 123, No. 38, 2001
Hoff et al.
nisms and transition states very similar to those of their aryl
phosphate ester counterparts. The reaction of the anion of pNPS
and that of the dianion of pNPP both proceed via concerted
reactions characterized by loose transition states in which the
leaving group bears nearly a full negative charge. However,
the two reactions differ markedly in their response to medium
effects, which cause only very modest changes in the rate of
sulfuryl transfer. This further suggests that the substantial
difference between the activation entropies in the phosphoryl-
and sulfuryl-transfer reaction is due to solvation effects, and
not to a difference in mechanism. Neutral pNPS undergoes
hydrolysis by a mechanism analogous to that of the pNPP
monoanion, with a proton transferred from the sulfuryl group
to the leaving group simultaneous with cleavage of the S-O
bond.
Acknowledgment. The authors thank Prof. Bob Brown for
MALDI-TOF data. Financial support of this work came from
NIH Grant GM47297 and PRF Grant 35690-AC4 to A.C.H.,
and from the U.S. Army Advanced Civil Schooling Program
for support of R.H.H.
Supporting Information Available: MALDI/TOF mass
spectrum for 18O,15N-labeled pNPS; Eyring plots for the reaction
of the pNPS anion in aqueous solution and in tert-amyl alcohol;
Eyring plot for hydrolysis of the pNPS anion in 95% DMSO/
water solution; rate data for hydrolysis of the anion of pNPS in
DMSO/water mixtures at 80 °C, used to prepare Figure 4 (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
JA0163974