Proton-in-flight mechanism for the spontaneous hydrolysis of N-methyl O-phenyl sulfamate: Implications for the design of steroid sulfatase inhibitors

Article abstract of DOI:10.1021/jo300386u

The hydrolysis of N-methyl O-phenyl sulfamate (1) has been studied as a model for steroid sulfatase inhibitors such as Coumate, 667 Coumate, and EMATE. At neutral pH, simulating physiological conditions, hydrolysis of 1 involves an intramolecular proton transfer from nitrogen to the bridging oxygen atom of the leaving group. Remarkably, this proton transfer is estimated to accelerate the decomposition of 1 by a factor of 10¹¹. Examination of existing kinetic data reveals that the sulfatase PaAstA catalyzes the hydrolysis of sulfamate esters with catalytic rate accelerations of ～10⁴, whereas the catalytic rate acceleration generated by the enzyme for its cognate substrate is on the order of ～10¹⁵. Rate constants for hydrolysis of a wide range of sulfuryl esters, ArOSO₂X^-, are shown to be correlated by a two-parameter equation based on pK_a ^ArOH and pK_a^ArOSO2XH.

Full text of DOI:10.1021/jo300386u

Note
pubs.acs.org/joc
Proton-in-Flight Mechanism for the Spontaneous Hydrolysis of
N-Methyl O-Phenyl Sulfamate: Implications for the Design of Steroid
Sulfatase Inhibitors
David R. Edwards and Richard Wolfenden*
Department of Biochemistry and Biophysics, University of North Carolina, Chapel Hill, North Carolina 27599, United States
S
* Supporting Information
ABSTRACT: The hydrolysis of N-methyl O-phenyl sulfamate (1) has been
studied as a model for steroid sulfatase inhibitors such as Coumate, 667
Coumate, and EMATE. At neutral pH, simulating physiological conditions,
hydrolysis of 1 involves an intramolecular proton transfer from nitrogen to
the bridging oxygen atom of the leaving group. Remarkably, this proton
transfer is estimated to accelerate the decomposition of 1 by a factor of 1011.
Examination of existing kinetic data reveals that the sulfatase PaAstA
catalyzes the hydrolysis of sulfamate esters with catalytic rate accelerations of
∼104, whereas the catalytic rate acceleration generated by the enzyme for its
cognate substrate is on the order of ∼1015. Rate constants for hydrolysis of a wide range of sulfuryl esters, ArOSO2X−, are shown
to be correlated by a two-parameter equation based on pKaArOH and pKaArOSO2XH
.
eq 1 indicates rate constants of k1− = (1.09 0.07) × 10−5 and
k1 = (2.0 0.1) × 10−5 s−1 corresponding to reaction along the
high and low pH plateaus, respectively (Figure 1). The kinetic
ulfamate esters such as Coumate, 667 Coumate, and
SEMATE have been investigated as time-dependent and
irreversible inhibitors of human steroid sulfatase for the treat-
ment of hormone-dependent breast cancer and other ailments.1
Sulfatase inhibition appears to involve decomposition of
the parent sulfamate ester to generate a reactive electrophile
[HNS(O)2] that undergoes nucleophilic capture by
groups positioned in and around the enzyme’s active site.2
Here, we describe a mechanistic study on the hydrolysis of 1,
which serves as a simulant for the sulfamate esters being
developed as enzyme inhibitors. The N-methyl substituent of 1
precludes formation of the dianion PhOSO2N2−, simplifying
the interpretation of kinetic effects at high pH.3 Based on the
accumulated kinetic data, we propose a novel mechanism for
the spontaneous hydrolysis of sulfamate esters under physio-
logical conditions. We also present a physical organic des-
cription of sulfuryl transfer that should prove useful in
predicting the hydrolytic stability of the next generation of
sulfatase inhibitors.
Figure 1. Plot of log(kobs) versus pH for the hydrolysis of 1 at 60 °C,
I = 0.55 M (NaCl). The fitted parameters k1− = (1.09
0.07) ×
0.3 were
10−5 s−1, k1 = (2.0
0.1) × 10−5 s−1 and pKa = 9.1
obtained from a fit of the data to eq 1.
pKa of 9.1 0.3 is somewhat lower than that determined by
spectrophotometric titration at 25 °C where pKa = 9.7 0.1
(Figure S1, Supporitng Information). The activation parame-
ters for the hydrolysis of 1 at pH 5.9 were measured from
Pseudo-first-order rate constants for the hydrolysis of 1 were
determined by UV−vis spectrophotometry monitoring phenol
or phenoxide production at 60 °C from 4.8 < pH < 13.6 (I =
550 mM, NaCl). A nonlinear least-squares fit of the rate data to
Received: February 23, 2012
Published: April 9, 2012
© 2012 American Chemical Society
4450
dx.doi.org/10.1021/jo300386u | J. Org. Chem. 2012, 77, 4450−4453

The Journal of Organic Chemistry
Note
25 < T < 80 °C and determined to be ΔH⧧ = 18.7 0.5 kcal/
mol and ΔS‡ = −24 1 cal/mol·K (Table 1).
spontaneous hydrolysis of 2 at 25 °C (Table 1). Remarkably,
comparing this value to k1 indicates that substitution of one of
the N-methyl substituents in 2 with a hydrogen atom results in
a greater than 1011-fold rate acceleration at 25 °C. A rate
constant of 7 × 10−7 s−1 for the hydrolysis of PhOSO2NH2 in
50% aqueous acetonitrile at 25 °C can be calculated from the
reported activation parameters.6 This value is very similar to
that for 1, indicating that a single N-H group is sufficient to
achieve the full catalytic effect.
Table 1. Kinetic and Thermodynamic Data for the
Hydrolysis of Sulfamate Esters 1 and 2
ΔH⧧
ΔS‡ (cal/
mol·K)
(kcal/mol)
k25 (s−1
)
k60 (s−1
)
1
18.7 0.5
−24
1
2 × 10−7
(2.0 0.1) × 10−5
(1.1 0.1) × 10−5
7 × 10−16
1−
A solvent deuterium kinetic isotope effect of kH/kD = 1.8
0.1 was determined on k1 for hydrolysis at 60 °C (Figure S4,
Supporting Information). Solvent isotope effects of 2.6 and 3.9
have previously been reported for the hydrolysis of O-4-
nitrophenyl sulfamate [4-NO2C6H4SO2NL2] and EMATE for
reaction along the low pH plateau at 60 °C.7,8 These solvent
isotope effects were used to support the proposal of an
associative and bimolecular SN2 pathway. Notably, any
contribution of the N-L hydron to the observed kinetic
isotope effects was specifically ruled out as it was deemed not to
be involved in the slow step of sulfamate ester hydrolysis.8
However, the greater than 1011-fold difference in reactivity
between 1 and 2 seems to contradict this conclusion and
suggests that the sulfamate N-H group is actually engaged in
bonding interactions at the transition state that substantially
reduce the free energy for hydrolysis (ΔΔG⧧ = 15.6 kcal/mol).
We propose an alternative mechanism for the hydrolysis of 1
where the N-H proton is transferred, either directly or through
a network of intervening water molecules,9 to the phenoxy
leaving-group as in transition structures 3 and 4 (Figure 2).
a
2
38
3
−14
6
7 × 10−19
a
The rate constants k25 and k60 for hydrolysis of 2 at 25 and 60 °C
were calculated from the activation parameters using the Eyring
equation.
10−pH
⎡
log(kobs) = log k1
⎢
10−pH + 10−pKa
⎣
10−pKa
10−pH + 10−pKa
⎤
+ k1−
⎥
⎦
(1)
The appearance of the pH−rate profile (Figure 1) is
consistent with an equilibrium mixture of 1 and 1− undergoing
product formation by the two pathways k1 and k1− shown in
Scheme 1. Williams previously studied a series of N-methyl
Scheme 1. Hydrolysis of 1 and 1−
Figure 2. Possible transition structures for the hydrolysis of 1.
This mechanism accounts for the small βLG of −0.41 reported
for the spontaneous hydrolysis of O-aryl sulfamates [ArOSO2NH2].6
The magnitude of βLG for these reactions indicates a modest
degree of negative charge is accumulating on the aryloxy leaving-
group as the sulfamate progresses from ground state to
transition state.10 The transfer of the N-H proton to the
bridging oxygen atom of the leaving group partially neutralizes
the accumulating negative charge resulting from S−O bond
fission and provides for the appearance of a shallow Brønsted
coefficient or βLG. Similar conclusions have been drawn regard-
ing the hydrolysis of aryl phosphate monoester monoanions
(βLG = −0.27) and the acid-catalyzed hydrolysis of sulfate mono-
esters (βLG = −0.33).11,12
O-aryl sulfamates hydrolyzing at elevated pH via k1− and found
these to undergo spontaneous unimolecular decomposition
to −OAr and MeNSO2.4 Subsequent fast nucleophilic
capture of MeNSO2 by solvent (or amine buffer) completes
the transformation. The focus of the current study is on the k1
process, which for 1 predominates below pH ∼9 and provides
an appropriate model for hydrolysis under physiological con-
ditions of the O-aryl sulfamates being investigated as steroid
sulfatase inhibitors.
At 60 °C, the ratio of k1/k1‑ at ∼2 is small, and yet there is a
strong indication that the N-H group has a profound effect on
the reactivity of 1. Consider, for example, the hydrolysis of
N,N-dimethyl O-phenyl sulfamate (2), which does not possess
N−H groups. Spontaneous hydrolysis of 2 is too slow to
monitor at 60 °C (t1/2 = 3 × 107 years, vide infra), and kinetic
experiments were carried out at pH 5.9 from 240 < T < 279 °C.
Observed first-order rate constants for the hydrolysis of 2 were
determined by 1H NMR comparing the normalized signal
intensities for unreacted starting material and phenol.5 Fitting
the rate data for hydrolysis of 2 to the Eyring equation yielded
the activation parameters ΔH⧧ = 38 3 kcal/mol and ΔS⧧ =
−14 6 cal/mol·K (Table 1). The thermodynamic data allow
us to estimate a rate constant of k2 = 7 × 10−19 s−1 for the
Impressive second-order rate constants of ki/Ki ≈ 106
M−1 s−1 for irreversible inhibition of an aryl sulfatase from
P. aeruginosa (PaAstA) have been reported for the inhibitors
Coumate, 667 Coumate, and EMATE.2 The strong binding
affinities of these compounds (Ki < 1 μM) might be expected to
translate into high target selectivity; however, this does not
appear to be the case. Active site titrations reveal that up to six
equivalents of inhibitor are required for complete inactivation
of PaAstA.2 This suggests that off-target sulfamolyations are
occurring with solvent and/or nucleophilic groups on the enzyme’s
surface acting as acceptors for the electrophile HNS(O)2.
4451
dx.doi.org/10.1021/jo300386u | J. Org. Chem. 2012, 77, 4450−4453

The Journal of Organic Chemistry
Note
The high reactivity of the sulfamate esters likely contributes to their
poor target selectivity; for example, Coumate has a half-time in
dilute aqueous solution of only ∼2 h at pH 7 and 37 °C.
Shown in Figure 3 (■, dashed line) is a plot of log(ki) versus
pKaLG, where ki is the first-order rate constant for irreversible
Figure 4. (a) Plot of log(k, s−1) versus pKaROSO2XH for the hydrolysis
of 4-NO2-C6H4O-SO2X− at 25 °C. The data are fit to log(k) = (−0.70
0.05)pKaROSO2XH − (8.0 0.5) with r2 = 0.95 (13 data). Full details
are provided in the Supporting Information.
Figure 3. (a) Plot of log(ki) versus pKaLG ( , dashed line) for the
■
−
on pKaArOH and pKaArOSO2XH. The sulfuryl esters F5C6OSO3
irreversible inhibition of PaAstA by ArOSO2NH2 (pH 7, 37 °C). The
and 4-CH3C(O)C6H4OSO2C−HC6H5 are estimated to
decompose at 25 °C with rate constants of 5 × 10−9 and
4 × 106 s−1, respectively, which do not differ greatly from the
experimental values of 10−7 and 4 × 107 s−1.12b,16 The accuracy
of the predictions, estimated to be within 2 orders of magnitude,
is remarkable considering that the reactivities of the sulfuryl esters
span a range of almost 20 orders of magnitude. At present, there
is insufficient kinetic data on the hydrolysis of neutral sulfuryl
esters to construct an analogous correlation to that in Figure 4.
data were obtained from ref 2. (b) Plot of log(khydrolysis) versus pKaLG
●
( , solid line) for the uncatalyzed hydrolysis of ArOSO2NH2 at 37 °C.
The data for this series were obtained from ref 7. See Table S1 of the
Supporting Information.
inhibition of PaAstA at pH 7, 37 °C under saturating
concentrations of ArOSO2NH2.2 Also included in Figure 3 is
a plot of log(khydrolysis) versus pKaLG (●, solid line) for the
uncatalyzed hydrolysis of ArOSO2NH2 at 37 °C.13 The vertical
separation between the best-fit lines indicates that PaAstA
provides a ∼104−fold rate acceleration for the cleavage of the
S-OAr bond of simple sulfamate esters. This rate acceleration is
small when compared with the level of catalysis achieved with
sulfate monoesters, which are the enzyme’s natural substrates.
For example, a catalytic rate acceleration of kcat/kuncat = 16.7/
(2 × 10−14) = 8 × 1014 can be calculated for the hydrolysis of
4-methoxyphenyl sulfate.2,12b Steroid sulfatase inhibitors show-
ing higher ratios of ki/khydrolysis are likely to result in fewer
off-target sulfamoylations but it remains to be seen if this
strategy will also result in a reduction in potency.
ROSO2XH
ArOH
log(k) = (0.71 0.01)pKa
+ (4.7 2.5)
− (1.7 0.3)pKa
(2)
In summary, the spontaneous hydrolysis of neutral
sulfamates, containing one or more N-H groups, is shown to
proceed through a novel proton-in-flight mechanism (Figure 2).
The hydrolysis of 1 is accelerated by an impressive factor of 1011
relative to the hydrolysis of 2, and this effect is attributed to the
simultaneous neutralization of charge on the bridging oxygen and
nonbridging nitrogen atoms as a proton is transferred between
these two atoms at the transition state. This mechanism suggests
a rationale for the lack of irreversible inhibition observed with
N,N-dialkyl O-aryl sulfamates such as 2, which cannot react in a
like manner.17 Moreover, the observed rates of sulfuryl transfer, at
least for anionic esters with common leaving groups, can be
satisfactorily correlated with their pKa values. The data in Figure 4
suggests that a compound with a leaving group as activated as
Coumate’s (pKaArOH = 7.8)18 would require a pKa value of ∼5 in
order to attain a half-time in aqueous solution of 24 h at 25 °C.
It would be desirable to predict the hydrolytic stability of
sulfuryl esters before committing valuable resources toward
synthesis of steroid sulfatase inhibitors. A plot of log(k) versus
pKaROSO2XH for the hydrolysis of a series of anionic sulfuryl
esters with a common 4-nitrophenoxide leaving group at 25 °C
adheres to the equation log(k) = (−0.70 0.05)pKaROSO2XH
−
(8.0 0.5) (Figure 4).14 The 13 sulfuryl esters used to construct
the plot are 4-NO2C6H4OSO3−, 4-NO2C6H4OSO2N−CH2Ar,
4-NO2C6H4OSO2N−Me, 4-NO2C6H4OSO2C−(H)SO2CH3,
4-NO2C6H4OSO2C−(CH3)SO2CH3, and 4-NO2C6H4OSO2-
C−C6H5 (details are provided in the Supporting Informa-
tion).3,4,6,12b,15,16 The inset in Figure 4 shows a series of parallel
lines corresponding to the hydrolysis of sulfuryl esters ArOSO2X−
with different aryloxy leaving groups (where ArO = PhO,
4-Cl-PhO, 3-NO2-PhO, and 4-NO2-PhO). The best-fit lines
indicate an average gradient of 0.71 0.01 that is invariant with
respect to leaving group and y-intercepts that increase with
EXPERIMENTAL SECTION
■
N-Methyl O-phenyl sulfamate (1) was prepared as described and
determined to be 95% pure by comparison of the 1H NMR integrated
signal intensities of 1 to a known amount of pyrazine.4 The hydrolysis
of 1 (2.9 × 10−4 M) was carried out in 1.5 mL of polypropylene
centerfuge vials immersed in a temperature-controlled water bath.
Reaction pH was maintained with 0.15 M acetate, phosphate, Tris, or
CAPS buffer at I = 0.55 M (NaCl). Variations in buffer concentration
at constant pH did not show any observable effect on the rate
constants. Periodically, the reaction mixtures were transferred to 1 cm
path length quartz cuvettes, and the UV−vis spectra were obtained.
decreasing pKaArOH
.
Equation 2 can be derived from these data to estimate rate
constants for the decomposition of anionic sulfuryl esters based
4452
dx.doi.org/10.1021/jo300386u | J. Org. Chem. 2012, 77, 4450−4453

The Journal of Organic Chemistry
Note
Reaction progress was determined for reactions run below pH 9 by
monitoring phenol production at 280 nm (ε280 = 1418 Abs/M/cm).
Phenoxide production was monitored at 290 nm for higher pH
reactions, and an effective ε290 was determined under the exact
experimental conditions in these cases. Observed first-order rate
constants were calculated by a nonlinear least-squares fitting of the
absorbance versus time data to a standard first-order exponential
equation. Good first-order behavior was generally observed for greater
than three half-lives, and a comparison of the UV spectra before and
after complete hydrolysis demonstrated a 1:1 stoichiometry in all
cases. A rate constant for the hydrolysis of 1 at pH 5.9 and 25 °C was
determined by the method of initial rates. N,N-Dimethyl O-phenyl
sulfamate (2) was prepared as described and characterized by 1H
NMR.19 Hydrolysis of 2 (17 mM) in H2O was carried out in vacuum-
sealed quartz tubes containing 0.2 M potassium phosphate buffer at
pH 5.9. The sealed quartz tubes were inserted into stainless steel pipe
bombs and placed in thermally equilibrated ovens as described.20
Reaction progress was measured by diluting the reaction samples 5-
fold with D2O and then obtaining a 1H NMR spectrum on the
reaction mixture and integrating the signals corresponding to
PhOSO2NMe2 to PhOH. Control experiments reveal that hydrolysis
of 2 at pH 5.9 is independent of hydronium ion concentration and that
the spontaneous reaction extends up to at least pH 8.
(6) Spilane, W. J.; Thea, S.; Cevasco, G.; Hynes, M. J.; McCaw, C. J.
A.; Maguire, N. P. Org. Biomol. Chem. 2011, 9, 523.
(7) Spillane, W. J.; McCaw, C. J. A.; Maguire, N. P. Tetrahedron Lett.
2008, 49, 1049.
(8) Spillane, W. J.; Malaubier, J.-B. Tetrahedron Lett. 2010, 51, 2059.
(9) Shi, F.-Q.; Li, X.; Xia, Y.; Zhang, L.; Yu, Z.-X. J. Am. Chem. Soc.
2007, 129, 15503.
(10) An alternative and equally valid viewpoint would be that a small
amount of positive charge is lost from the aryloxy leaving group as the
sulfamate progresses from ground state to transition state.
(11) (a) Kirby, A. J.; Vargolis, A. G. J. Am. Chem. Soc. 1967, 89, 415.
(b) Grzyska, P. K.; Czyryca, P. G.; Purcell, J.; Hengge, A. C. J. Am.
Chem. Soc. 2003, 125, 13106.
(12) (a) Hoff, R.; Larsen, P.; Hengge, A. J. Am. Chem. Soc. 2001, 123,
9338. (b) Edwards, D. R.; Lohman, D. C.; Wolfenden, R. J. Am. Chem.
Soc. 2012, 125, 525.
(13) Rate constants for hydrolysis of ArOSO2NH2 at 37 °C were
calculated from the activation parameters reported in ref 7.
(14) A similar but less extensive plot appears in ref 16.
(15) (a) Guthrie, J. P. J. Am. Chem. Soc. 1980, 102, 5177. (b) Thea,
S.; Guanti, G.; Hopkins, A. R.; Williams, A. J. Org. Chem. 1985, 50,
5592.
(16) Davy, M. B.; Douglas, K. T.; Loran, J. S.; Steltner, A.; Williams,
A. J. Am. Chem. Soc. 1977, 99, 1196.
(17) (a) Woo, L. W. L.; Ganeshapillai, D.; Thomas, M. P.; Sutcliffe,
O. B.; Malini, B.; Mahon, M. F.; Purohit, A.; Potter, B. V. L
ChemMedChem 2011, 6, 2019. (b) Howarth, N. M.; Purohit, A.; Reed,
M. J.; Potter, B. V. L J. Med. Chem. 1994, 37, 219.
(18) Wolfbeis, O. S.; Fuerlinger, E.; Kroneis, H.; Marsoner, H.
Fresenius Z. Anal. Chem. 1983, 314, 119.
(19) Spillane, W. J.; Taheny, A. P.; Kearns, M. M. J. Chem. Soc., Perkin
Trans. 1 1982, 677.
(20) Wolfenden, R. Chem. Rev. 2006, 106, 3379.
ASSOCIATED CONTENT
■
S
* Supporting Information
Plot of absorbance versus pH for the UV−vis titration of 1.
Eyring plots for the hydrolysis of 1 and 2. Plot of observed rate
constant versus percent deuterium content used to calculate the
solvent kinetic isotope effect on 1 (Figure S4). Table of kinetic
constants for the hydrolysis of ArOSO2X− (Table S1). This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Email: water@med.unc.edu.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by National Institutes of Health
Grant No. GM-18325.
REFERENCES
■
(1) (a) Maltais, R.; Poirier, D. Steroids 2011, 76, 929. (b) Day, J. M.;
Purohit, A.; Foster, P. A.; Woo, L. W.; Potter, B. V.; Reed, M. J. Ann.
N.Y. Acad. Sci. 2009, 1155, 80.
́
(2) Bojarova, P.; Denehy, E.; Walker, I.; Loft, K.; DeSouza, D. P.;
Woo, L. W. L.; Potter, B. V. L; McConville, M. J.; Williams, S. J.
ChemBioChem 2008, 9, 613.
(3) For an example of a complex pH rate profile for the hydrolysis of
a sulfamate ester see: Thea, S.; Cevasco, G.; Guanti, G.; Williams, A. J.
Chem. Soc. Chem. Commun. 1986, 1582.
(4) Willams, A.; Douglas, K. T. J. Chem. Soc. Perkin Trans. 2 1974,
1727.
(5) We do not specify a mechanism for the spontaneous hydrolysis of
2 and consider that it could occur by S−O bond cleavage with direct
displacement of phenol. An alternative process involving hydrolysis of
2 via S−N bond fission to initially afford phenyl sulfate and
dimethylamine (or its ammonium ion) cannot be ruled out based
on the kinetic data because of the subsequent hydrolysis of phenyl
sulfate (ΔH⧧ = 32.9 kcal/mol ΔS⧧ −15 cal/mol/K, data from ref 12b)
to form phenol and inorganic sulfate is predicted to occur faster than
its formation.
4453
dx.doi.org/10.1021/jo300386u | J. Org. Chem. 2012, 77, 4450−4453