A comparison of phosphonothioic acids with phosphonic acids as phosphatase inhibitors

Article abstract of DOI:10.1021/jm030106f

Phosphorothioates, analogues of phosphate esters in which a sulfur replaces an oxygen atom in the phosphoryl group, are competent surrogate substrates for a number of phosphatases. In some cases the thio analogues show similar binding (as estimated by K

Full text of DOI:10.1021/jm030106f

J . Med. Chem. 2003, 46, 3703-3708
3703
A Com p a r ison of P h osp h on oth ioic Acid s w ith P h osp h on ic Acid s a s
P h osp h a ta se In h ibitor s
Krzysztof Swierczek,† Arti S. Pandey, J ohn W. Peters,‡ and Alvan C. Hengge*
Utah State University, Department of Chemistry and Biochemistry, Logan, Utah 84322-0300
Received March 5, 2003
Phosphorothioates, analogues of phosphate esters in which a sulfur replaces an oxygen atom
in the phosphoryl group, are competent surrogate substrates for a number of phosphatases. In
some cases the thio analogues show similar binding (as estimated by Km) while other
phosphatases show quite different Km values for phosphate compared to phosphorothioate esters.
On this basis it was hypothesized that there might be different inhibitory tendencies by the
nonhydrolyzable analogues, phosphonothioic acids compared with phosphonic acids. A series
of phosphonothioic acids and corresponding phosphonic acids were synthesized and their
inhibitory properties were compared toward human placental and E. coli alkaline phosphatases,
the protein-tyrosine phosphatase from Yersinia, and the serine/threonine protein phosphatases
PP2C and lambda. Sulfur substitution for oxygen gives the phosphonothioic acids pKa values
that are close to those of phosphate esters, in contrast to the higher pKa values typical of
phosphonic acids. Despite different steric requirements and differences in charge distribution
in the anions of phosphonothioic acids compared with phosphonic acids, it was found that,
with some exceptions, differences in inhibitory properties were modest.
In tr od u ction
ological conditions. We have synthesized several such
compounds and evaluated their inhibitory properties
toward a number of phosphatases.
Protein phosphorylation plays a crucial role in the
regulation of many biochemical processes, including the
regulation of metabolism and signal transduction.1-5
The phosphorylation levels are controlled by the balanc-
ing actions of kinases and phosphatases. The crucial
roles played by phosphatases have led to considerable
interest in designing inhibitors. Phosphonic acids have
formed the basis for a number of nonhydrolyzable
phosphate ester substrate mimics. Such compounds
have a carbon-phosphorus bond in place of the scissile
P-O ester bond of the natural substrates of phos-
phatases. As a result of this replacement, phosphonic
acids have higher first and second pKa values than
phosphate esters. This has relevance due to the fact that
protein-tyrosine phosphatases and serine/threonine phos-
phatases that have been studied utilize the dianion
forms of their substrates in catalysis. Thus phosphonic
acids bearing one or two fluorine atoms at the R-meth-
ylene position, which lowers the pKa values6 (Table 1),
have been widely explored as the basis for phosphate
mimetics in inhibitor design. However, their superior
inhibitory properties compared to simple phosphonic
acids have been attributed, at least in some instances,
to direct interactions between the fluorine atoms and
active site resides rather than their reduced pKa.7
The substitution of a sulfur atom for one of the
nonbridging oxygen atoms in a phosphonic acid, yielding
a phosphonothioic acid, will also reduce the pKa values.
The resulting phosphonothioic acids are also mimics of
phosphate esters and are nonhydrolyzable under physi-
The corresponding sulfur analogues of phosphate
esters, phosphorothioates (RO-PSO2H2), have found
use as surrogate substrates for phosphate esters in the
study of a number of enzymatic phosphoryl transfer
reactions. Phosphorothioates may be made chiral by the
use of a sulfur atom and two isotopes of oxygen. Such
chiral phosphorothioates have been used to study a
number of enzyme-catalyzed phosphoryl transfer reac-
tions, the results of which have been reviewed.8 Aside
from stereochemical studies, phosphorothioates have
also been used in a number of other mechanistic inves-
tigations of enzymatic phosphoryl transfer reactions of
monoesters.9-11 In some kinetics studies, phospho-
rothioates exhibit Km values that are similar to those
of corresponding phosphate esters, while in other cases,
Km values are more disparate. For example, in a study
of several protein-tyrosine phosphatases (PTPases) us-
ing phosphate and phosphorothioate ester substrates it
was found that Km values for the two types of com-
pounds differed by 3-fold or less; for the Yersinia PTPase
the values were the same within experimental uncer-
tainty.9 In contrast, alkaline phosphatase from E. coli
exhibits a 10-fold higher Km for p-nitrophenyl phospho-
rothioate than for p-nitrophenyl phosphate.10
These results imply that the active sites of phos-
phatases vary in their abilities to accommodate phos-
phorothioates, which differ in the larger radius of sulfur
and longer P-S bond length. In addition to the different
steric demands of the thiophosphoryl group compared
to the phosphoryl group, there is also a difference in
charge distribution in their respective dianions. Experi-
mental and computational evidence indicates that one
negative charge in phosphorothioate dianions is borne
* To whom correspondence should be addressed. E-mail: hengge@
cc.usu.edu.
† On leave from the Silesian University of Technology, Gliwice,
Poland.
‡ Present address: Department of Chemistry and Biochemistry,
Montana State University, Bozeman, MT 59717.
10.1021/jm030106f CCC: $25.00 © 2003 American Chemical Society
Published on Web 07/18/2003

3704 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Swierczek et al.
Ta ble 1. Comparison of the Second pKa Values for Phenyl Phosphate and Related Phosphonic Acids6
F igu r e 1. In a phosphorothioate dianion, one negative charge
resides on the sulfur atom and the other is delocalized between
the two nonbridging oxygen atoms. In a phosphate ester the
charge is equally distributed among the three nonbridging
atoms.
by the sulfur atom while the second is delocalized
between the two oxygen atoms12,13 (Figure 1). Thus
F igu r e 2. The thiono and thiolo tautomers of phosphonothioic
acids (a) and phosphorothioic acids (b). The thiono-thiolo
equilibrium lies far to the left phosphonothioic acids (a). In
phosphorothioic acids (b) more of the thiolo form is typically
present, depending upon the solvent.
there is an unequal distribution of charge among the
three nonbridging atoms of the thiophosphoryl group
in contrast to the phosphoryl dianion. Experimental
evidence indicates that the dianion is the substrate for
alkaline phosphatases as well as protein phosphatases.
The differences in geometry and in charge distribution
between the anions of phosphonothioic and phosphonic
acids could potentially be exploited, in that phospho-
nothioic acids might display different relative specificity
in inhibition of phosphatases than corresponding phos-
phonic acids. To evaluate the potential inhibitory prop-
erties of phosphonothioic acids and to compare these
properties with those of the corresponding phosphonic
acids, we have synthesized a number of simple ana-
logues of both compounds (Table 2) and tested them as
inhibitors. Inhibition by these compounds was measured
with human placental alkaline phosphatase (PLAP) and
E. coli alkaline phosphatase, protein-tyrosine phos-
phatase from Yersinia (YOP), and the serine/threonine
protein phosphatases PP2C and lambda. These phos-
phatases represent three different classes of phos-
phatase active sites and catalytic mechanisms. The
alkaline phosphatases have a dinuclear zinc active site
and catalyze phosphoryl transfer by a two-step mech-
anism with a phosphoserine intermediate.14,15 The Yers-
inia enzyme has the active site typical of PTPases,
which lack a metal cofactor and also catalyze phosphoryl
transfer by a two-step mechanism, with a phospho-
cysteine intermediate.16 The serine/threonine protein
phosphatases utilize a dinuclear metal center, but
catalyze direct phosphoryl transfer to a metal-bound
water molecule.3
Ta ble 2. Phosphonothioic and Phosphonic Acids Compared for
Inhibitory Properties in This Study
The general phosphonothioic acid structure shown at
the top of Table 2 is shown in the thiono form. Neutral
phosphonothioic acids have two tautomeric structures
(Figure 2a). Studies indicate that the thiono-thiolo
equilibrium lies far on the side of the thiono form in
phosphonothioic acids (in which a carbon-phosphorus
bond is present). In compounds where only phosphorus-
oxygen bonds are present (Figure 2b) more of the thiolo
form is typically present, depending upon the solvent.17-19
The preparation of some of the compounds shown
required the development of some new synthetic meth-
odology. The majority of the thiophosphonic acids (1S-
6S) were synthesized as previously described.20 However
this method was not amenable for the synthesis of
compounds 7 and 8, with heteroatoms in the alkyl
group. Therefore, an alternative method was devised
that is based on modifications of a series of known

Phosphatase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3705
Sch em e 1
Ta ble 3. Millimolar Inhibition Constants (Ki) for the Compounds in Table 2 toward Human Placental Alkaline Phosphatase (PLAP)
and E. coli Alkaline Phosphatase (AP), PTPase from Yersinia (YOP), and the Serine/Threonine Protein Phosphatases PP2C
and Lambdaa
Ki values (mM)
alkaline phosphatases
PLAP E. Coli AP
19 ( 3
0.60 ( 0.06
0.20 ( 0.01
1.5 ( 0.3
0.33 ( 0.03
1.5 ( 0.2
PTPase
YOP
Ser/Thr phosphatases
PP2C lambda
inhibitor
1S
2S
3S
4S
5S
6S
7
no inhibition
no inhibition
no inhibition
no inhibition
no inhibition
no inhibition
no inhibition
no inhibition
no inhibition
0.61 ( 0.06
7.0 ( 0.8
2.7 ( 0.3
17 ( 10
20 ( 6
no inhibition
no inhibition
29 ( 4
0.33 ( 0.02
0.187 ( 0.008
0.210 ( 0.006
0.22 ( 0.02
0.015 ( 0.001
0.70 ( 0.07
0.20 ( 0.02
0.67 ( 0.08
precipitation
precipitation
0.41 ( 0.04
0.048 ( 0.003
0.57 ( 0.06
1.5 ( 0.2
no inhibition
0.20 ( 0.01
no inhibition
15.8 ( 0.7
30 ( 1
1.1 ( 0.1
0.17 ( 0.01
1.2 ( 0.2
0.40 ( 0.06
0.46 ( 0.04
11 ( 1
8
0.47 ( 0.03
1.01 ( 0.07
precipitation
precipitation
0.43 ( 0.01
0.13 ( 0.02
1O
2O
3O
4O
6O
1.55 ( 0.06
0.58 ( 0.02
7.0 ( 0.7
no inhibition
no inhibition
42 ( 4
no inhibition
no inhibition
0.30 ( 0.02
0.39 ( 0.01
no inhibition
a
The entry “no inhibition” designates cases where no reduction in rate was observed at 5 mM concentration of the inhibitor. Both
PP2C and lambda phosphatases are routinely assayed in the presence of Mn2+; in some cases phosphonates precipitate under these
conditions, precluding kinetic measurements.
reactions. The reaction of easily accessible diethyl esters
of phosphonic acid with an equimolar amount of sodium
ethanethiolate in refluxing ethanol gives high yields of
monodealkylation products in the form of their sodium
salts.21 Such salts react with oxalyl chloride in the
presence of a catalytic amount of dimethylformamide
to give monochlorides of phosphonic acid monoesters.22,23
These monochlorides subsequently undergo nucleophilic
substitution of chlorine by the ethanethiolate anion.24
The O,S-diethyl esters thus obtained can be easily
dealkylated with iodotrimethylsilane to the free acids,
which may be precipitated as their anilinium salts. This
approach, used to prepare compounds 7 and 8, is
outlined in Scheme 1.
This approach affords a route to pure phosphonothio-
ates without the need for purification of intermediates.
Moreover, we successfully applied this new approach for
synthesis of the previously reported propyl- and ben-
zylphosponothioic acids (2S and 3S),20 suggesting this
method is suitable for the general synthesis of phospho-
nothioic acids.
as well, but was not measured in this study. This effect
of sulfur substitution for oxygen is similar to that
reported in esters,27 where the second pKa for p-
nitrophenyl phosphorothioate (3.68 ( 0.08) is reduced
compared to that of p-nitrophenyl phosphate (5.0).
The Ser/Thr phosphatases PP2C and lambda have
been most often studied kinetically in the presence of
Mn2+. Unfortunately, phosphonates are known to form
insoluble polymeric complexes with a number of diva-
lent cations. Complexes of the propyl and benzyl phos-
phonates 2O and 3O precipitated from the Mn2+-con-
taining buffer solutions used with these two phospha-
tases, precluding Ki measurements. Phosphonates bear-
ing a carboxylate moiety (4O and 6O) were soluble un-
der the same conditions. No precipitation occurred with
any of the phosphonothioic acids. Their superior solubil-
ity in the presence of Mn2+, if found to extend to other
divalent metal ions, would make phosphonothioic acids
more amenable than phosphonic acids to kinetics stud-
ies requiring divalent metal ions to be supplied in
solution.
In h ibition Resu lts. The inhibitory properties re-
ported in Table 3 were evaluated using p-nitrophenyl
phosphate as the substrate. All of the inhibitors were
initially screened using a 5 mM concentration of the
inhibitor. Instances where no reduction of rate was
observed in this preliminary screen are designated as
“no inhibition” in Table 3. In all other cases, a value for
Ki is given with the standard error. The inhibition was
found to be competitive in all cases.
Resu lts a n d Discu ssion
The second pKa of propyl phosphonothioic acid 2S was
determined to be 6.64, while the reported pKa of the
oxygen analogue 2O is 8.18;25 we obtained an experi-
mentally indistinguishable value of 8.17. The second pKa
of the benzyl compound 3S was determined to be 6.41,
compared to 7.5526 for the oxygen analogue 3O. Thus,
the net effect of sulfur substitution is to depress the
second pKa value of phosphonates to values similar to
phosphate esters. The first pKa is expected to be lowered
Alk a lin e P h osp h a ta ses. No significant inhibition of
the alkaline phosphatase from E. coli by any of the

3706 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Swierczek et al.
phosphonothioic acids was observed. The phosphonic
acids 2O and 6O showed moderate inhibition, with no
detectable inhibition from the others when screened at
5 mM concentration. These results are consistent with
previous reports of only weak inhibition of E. coli AP
by phosphonates.28
dues.35 The metal ions in PP2C have fewer amino acid
ligands than in the PPP family, and the resulting looser
coordination is evidenced by kinetic analyses that reveal
Kmetal values in the millimolar range.36 Mn2+ and Fe2+
result in the highest activities; saturation kinetics
reveals concentrations of 5 mM or greater of Mn2+ are
required for maximum activity to be observed.36 In
contrast lambda phosphatase has a greater affinity for
Mn2+, with dissociation constants in the micromolar
range.37
Inhibition of PP2C by methyl phosphonic acid 1O was
∼4-fold better than by the thio analogue, but this
preference is reversed for compounds 4 and 6. The small
size of the methyl group gives compounds 1O and 1S
much more freedom to adopt a geometry to make
hydrogen bonds to the (thio)phosphonyl group, although
the methyl group does not provide any interactions to
directly assist or otherwise affect binding.
Experiments with lambda show the same preferential
inhibition by the oxygen analogues in the methyl
derivatives, which disappears in compounds with larger
substituents. Of the longer alkyl chain derivatives, 4S
is a slightly better inhibitor than 4O. The carboxynaph-
thyl 6S and 6O have very similar Ki values.
In general, PP2C was inhibited somewhat better than
lambda by the phosphonothioic acids. Compound 7,
incorporating a potential second metal ligand in com-
parison with the approximately isosteric 2S, shows
approximately a 17-fold better inhibition of lambda. In
contrast, the difference is about 2-fold with PP2C. Only
minor differences were observed in the inhibition by 8
and 3S of PP2C and of lambda.
Much better inhibition was observed by both phospho-
nothioic and phosphonic acids of the human placental
alkaline phosphatase (PLAP). PLAP shares the active
site residues and dinuclear zinc center of the E. coli AP.
However, in the E. coli enzyme active site is close to
the surface and the ester group of the substrate is
solvent exposed, accounting for the lack of substrate
specificity of this enzyme. In contrast, the active site in
PLAP lies at the bottom of a large cleft,14 providing a
means for discrimination of both substrates and of
inhibitors. The phosphonothioic acids 2S-4S have 2.5
to 4.7-fold lower Ki values than their oxygen analogues.
However, this trend is reversed for the smallest (1) and
bulkiest (6) pendant groups, where the oxygen analogue
is the superior inhibitor. The compounds 7 and 8
incorporate a second sulfur atom with the potential to
coordinate to the dinuclear metal center. The ap-
proximately isosteric compounds 7 and 2S exhibit very
similar Ki values indicating the second sulfur atom does
not contribute to binding.
Yer sin ia P TP a se. The only compound to show ap-
preciable inhibition of the Yersinia PTPase is 6S. The
naphthyl ring system has previously been shown to be
superior to phenyl in phosphonic acid-derived inhibitors
of another PTPase, PTP-1B.29 The 200 µM Ki of 6S
stands in contrast to the absence of any observed
inhibition of the oxygen analogue 6O at 5 mM concen-
tration. Thus, in the case of the Yersinia PTPase the
combination of the naphthyl ring and the thiophospho-
nate moiety is much more effective than either func-
tionality alone. The difference in Ki between 6S and 6O
is also interesting in light of the similar Km values
reported for p-nitrophenyl phosphorothioate (2.7 ( 0.6
mM) and p-nitrophenyl phosphate (2.6 ( 0.1 mM).9
Inorganic phosphate is a weaker inhibitor of Yersinia
PTPase (Ki ) 15 ( 1 mM) than inorganic thiophosphate
(6.4 ( 0.3 mM).9
Ser in e/Th r eon in e P h osp h a ta ses. The phosphatas-
es PP2C and lambda are members of the serine/
threonine phosphoprotein phosphatase family, which
share the utilization of a dinuclear metal center for
catalysis. Unlike the alkaline phosphatases, evidence
indicates that the metal centers in these enzymes
catalyze the direct phosphoryl transfer to water with
no phosphoenzyme intermediate.3
Su m m a r y
Phosphonothioic acids have second pKa values lower
than those of phosphonic acids and very similar to those
of phosphate esters. In general, only moderate differ-
ences are observed between the inhibitory properties of
phosphonothioic acids and phosphonic acids toward the
phosphatases tested in this study. One notable exception
is the observation that the combination of the phospho-
nothioic acid and a naphthyl moiety (6S) yielded a much
better inhibitor of the Yersinia PTPase than any other
compound tested. It was also noted that the carboxy-
naphthyl moiety (6S and 6O) conveys fairly good inhibi-
tory properties toward PP2C, which has not been
previously reported. The phosphonothioic acid 6S was
superior to the phosphonic acid 6O. It was also found
that the phosphonothioic acids show markedly better
solubility in the presence of manganese ion than the
phosphonic acids, a useful property when working with
metallophosphatases that require metal ions supplied
in the buffer for full activity.
The serine/threonine phosphoprotein phosphatases
fall into two structurally distinct gene families, desig-
nated as PPP and PPM. X-ray structures of the PPP
subfamily members PP1,30,31 calcineurin,32,33 and lamb-
da34 reveal the same secondary structural characteris-
tics providing ligands for the dinuclear metal complex.
The PPM subfamily has a different secondary structure
and no significant sequence homology with the PPP
phosphatases. PP2C is regarded as its defining member,
and members of the PP2C subfamily show strong
dependence on Mg2+ or Mn2+ for activity. The X-ray
structure of PP2C shows an active site containing
several carboxylates that serve as metal-binding resi-
Exp er im en ta l Section
All chemicals and solvents were purchased from commercial
sources and used without purification. All melting points are
uncorrected. All reactions with phosphorus compounds were
carried out under a nitrogen atmosphere. Elemental analyses
were performed by Atlantic Microlabs. All analyses agreed
within (0.4% of the calculated values. 1H NMR (400 MHz),
13C NMR (100 MHz), and 31P NMR (162 MHz) spectra were
recorded in D2O solutions using 3-(trimethylsilyl)propionic-
1
2,2,3,3-d4 acid sodium salt as an internal standard for H and
13C NMR and 85% phosphoric acid as an external standard
for 31P NMR.

Phosphatase Inhibitors
J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17 3707
The phosphatases were commercial products with the
exception of lambda, which was provided by the laboratory of
Frank Rusnak (Mayo Clinic and Foundation), and PP2C,
which was overexpressed and purified as previously de-
scribed.36 BL21DE3 strain of E. coli were transformed with
the expression plasmid pCW-PP2C R and grown in 2xYT
medium containing 140 µg /mL ampicillin. The supernatant
obtained after centrifugation of the cells lysed by French press
was precipitated between 30% and 55% saturation with (NH4)2-
SO4 . The pellet was resuspended in buffer and loaded onto a
Q-Sepharose anion exchange column; fractions showing high
activity with pNPP (p-nitrophenyl phosphate) were pooled and
loaded on a Phenyl Sepharose hydrophobic column. Fractions
were analyzed for purity by SDS-gel electrophoresis, pooled,
concentrated to ∼40 mg/mL, and stored at -20° C in Tris-
HCl, pH 7.0, DTT, and glycerol.
Mea su r em en t of In h ibition Con sta n ts. Inhibition ex-
periments were conducted using p-nitrophenyl phosphate as
the substrate at 25 °C. Rates were measured with a range of
substrate and inhibitor concentrations. Lineweaver-Burke
plots were used to assess the type of inhibition. The Ki values
in Table 3 were obtained by fitting the data to eq 1, using
Microcal Origin 5.0 software.
25.5 mmol) was dissolved in acetone (50 mL), and excess
aniline was added to precipitate the white 6-phosphonohex-
anoic acid anilinium salt (6.13 g, 83%). Mp 141-142 °C (2-
1
butanone-ethanol). Anal. (C12H20NO5P) C, H, N. H NMR δ
7.30-7.55 (m, 5H), 2.37 (t, J ) 7.6 Hz, 2H), 1.32-1.67 (m, 8H).
13C NMR (D2O) δ 182.7 (s, CdO), 133.0 (s, CHaromatic), 130.5
(s, CHaromatic), 124.7 (s, CHaromatic), 37.1 (s, CH2), 32.6 (d, J )
17.5 Hz, CH2), 30.5 (d, J ) 132.8 Hz, P-CH2), 27.0 (s, CH2),
25.6 (d, J ) 4.6 Hz, CH2). 31P NMR δ 27.5.
7-P h osp h on om et h yln a p h t h a len e-1-ca r b oxylic Acid
(6O), Tr im eth yla m m on iu m Sa lt. To a suspension of sodium
hydride (95%, 3.88 g, 0.150 mol)) in tetrahydrofuran (100 mL)
was added (foaming) a solution of diethyl phosphite (94%,
21.52 g, 0.146 mol) in tetrahydrofuran (50 mL) at 0 °C. After
the mixture was stirred for 1 h, a solution of 7-bromomethyl-
naphthalene-1-carboxylic acid tert-butyl ester20 (44.8 g, 0.139
mol) in tetrahydrofuran (100 mL) was added. The mixture was
left to slowly reach room temperature and stirred overnight.
Methanol (100 mL) was added, and solvents were removed by
rotary evaporation. The residue was dissolved in dichlo-
romethane (250 mL), washed with sodium hydrogen carbonate
solution and brine, and then dried with anhydrous magnesium
sulfate. Solvent was evaporated under reduced pressure to give
crude 7-(diethoxyphosphorylmethyl)naphthalene-1-carboxylic
acid tert-butyl ester as an orange oil (45.4 g, 86%). This
material (6.0 g, 15.9 mmol) was refluxed with concentrated
hydrochloric acid (150 mL) overnight, evaporated under
reduced pressure, dissolved in water (100 mL), and evaporated
to dryness (this procedure was repeated three times). The
residue was dissolved in ethanol (150 mL) and saturated with
gaseous trimethylamine. Then diethyl ether (800 mL) was
added, and the precipitate was filtered off, washed with ether,
and dried to give the trimethylammonium salt of 7-phospho-
nomethylnaphthalene-1-carboxylic acid as white powder (3.77
g, 73%). Mp 245-250 °C (ethanol-diethyl ether, with decom-
[S]
[I]
v
)
(1)
Vmax
Km 1 +
+ [S]
(
)
Ki
Conditions for inhibition studies were: PLAP and E. coli
AP: 250 mM TRIS buffer, pH ) 9.0, 1 mM ZnCl2 and 1 mM
MgCl2. For YOP: 100 mM succinic acid buffer, pH ) 6.6. For
PP2C: 100 mM TRIS buffer, pH ) 7.5, 5 mM MnCl2 and 1
mM DTT. For lambda phosphatase: 100 mM MOPS buffer,
pH ) 7.3, 1 mM MnCl2 and 1 mM DTT.
Deter m in a tion of p Ka Va lu es. Samples were dissolved
in water to obtain ∼0.01 M solutions. These solutions were
titrated with 0.1 M sodium hydroxide as pH was monitored
during the titration using a glass electrode at 25 °C. The pKa
values were calculated directly from plots of titrant volume
vs pH.
Syn th eses of In h ibitor s. Compounds 1O and 2O and 3O
were commercial products. Salts of the acids 2S, 3S, 4S, 5S,
and 6S were synthesized according our previously published
method.20 The bis(cyclohexylammonium) salt of methylphospho-
nothioic acid (1S) was obtained by hydrolysis of the commercial
dichloride in aqueous KOH. After titration to pH 9.0 with
cyclohexylamine, and the bis(cyclohexylammonium) salt of 1S
was recrystallized from aqueous ethanol. 1H NMR δ 3.20-3.30
(m, 2H), 2.0-2.1 (m, 4H), 1.8-1.9 (m, 4H), 1.68-1.75 (m, 2H),
1.56 (d, J ) 14.1 Hz, 3H), 1.35-1.49 (m, 8H), 1.20-1.31 (m,
2H). 31P NMR δ 57.4.
6-P h osp h on oh exa n oic Acid (4O), An ilin iu m Sa lt. A
solution of 6-bromohexanoic acid (20.0 g, 0.103 mol) and
p-toluenesulfonic acid (2 g) in methanol (200 mL) was refluxed
overnight. The residue after evaporation of solvent was dis-
solved in benzene (100 mL), washed with water, with a sodium
hydrogen carbonate solution, and again with water, and then
dried with anhydrous magnesium sulfate. Solvent was re-
moved by rotary evaporation to give the methyl ester of
6-bromohexanoic acid (19.77 g, 92%) as a colorless oil. This
material (19.77 g, 95 mmol) was mixed with triethyl phosphite
(29.1 g, 0.175 mol) and heated to 160 °C for 48 h. The reaction
mixture was then cooled, dissolved in water (300 mL), and
heated to boiling for several minutes. After cooling, the mixture
was extracted with benzene (4 × 50 mL). The benzene extracts
were washed with water, dried with anhydrous magnesium
sulfate, and evaporated to give 6-(diethoxyphosphoryl)hexanoic
acid methyl ester (21.26 g, 84%) as a colorless oil. This material
(21.26 g, 80 mmol) was refluxed overnight with concentrated
hydrochloric acid (100 mL). The reaction mixture was then
evaporated under reduced pressure, dissolved in water (100
mL), and then evaporated to dryness (this procedure was
repeated three times) to give crude 6-phosphonohexanoic acid
as a sticky white mass (15.71 g, 100%). This material (5 g,
1
position). Anal. (C15H20NO5P) C, H, N. H NMR δ 8.22-8.26
(m, 1H), 8.02 (d, J ) 7.6 Hz, 1H), 7.93 (d, J ) 9.2 Hz, 1H),
7.82 (d, J ) 7.1 Hz, 1H), 7.48-7.60 (m, 2H), 3.23 (d, J ) 20.9
Hz, 2H), 2.86 (s, 9H). 13C NMR (D2O) δ 177.0 (s, CdO), 137.7
(d, J ) 9.2 Hz, Cnaphthyl), 134.9 (d, J ) 2.3 Hz, Cnaphthyl), 134.7
(d, J ) 1.5 Hz, CHnaphthyl), 133.1 (s, Cnaphthyl), 132.9 (d, J ) 3.1
Hz Cnaphthyl), 131.5 (d, J ) 6.1 Hz CHphenyl), 131.4 (d, J ) 3.8
Hz, CHnaphthyl), 131.2 (s, CHnaphthyl), 128.1 (d, J ) 7.6 Hz,
CHnaphthyl), 127.4 (d, J ) 1.5 Hz, CHnaphthyl), 47.6 (s, N-CH3),
39.5 (d, J ) 127.4 Hz, P-CH2). 31P NMR δ 17.2.
Meth ylsu lfan ylm eth ylph osph on oth ioic Acid (7), An ilin -
iu m Sa lt. Commercially available methylsulfanylmethylphos-
phonic acid diethyl ester (19.89 g, 0.1 mol) and sodium
ethanethiolate (8.41 g, 0.1 mol; prepared in situ prior to
reaction from sodium and ethanethiol) were refluxed for 24 h
in ethanol (250 mL). Solvent was then removed by rotary
evaporation, and the remaining yellowish precipitate was
washed with diethyl ether and dried in vacuo to yield the crude
sodium salt of methylsulfanylmethylphosphonic acid mono-
ethyl ester (17.39 g, 91%). This salt was suspended in tet-
rahydrofuran (250 mL) and a catalytic amount of DMF (0.1
mL) and cooled using an ice bath, and oxalyl chloride (13.21
g, 0.104 mol) was added (foaming). The resulting mixture was
stirred at 0 °C for 1 h and then for 3 h at room temperature.
At this time, a precipitate was filtered off, and the solvent was
removed by rotary evaporation to give the crude methylsul-
fanylmethylphosphonic acid monoethyl ester monochloride as
a yellow oil (13.02 g, 76%). This material (13.02 g, 0.069 mol)
was added at 0 °C to a solution of ethanethiol (4.29 g, 0.069
mol) and 1,8-diazabicyclo[5.4.0]undec-7-ene (10.51 g, 0.069
mol) in benzene (250 mL). The resulting suspension was
stirred for 1 h, and then the ice bath was removed and stirring
was continued overnight. The mixture was washed carefully
first with diluted (1:9) hydrochloric acid and then with a
saturated solution of sodium bicarbonate and finally dried with
anhydrous magnesium sulfate. The solvent was removed by
rotary evaporation to give crude methylsulfanylmethylphospho-
nothioic acid O,S-diethyl ester as a yellow oil (8.27 g, 56%).
The O,S-diethyl ester was converted to the free phosphono-

3708 J ournal of Medicinal Chemistry, 2003, Vol. 46, No. 17
Swierczek et al.
thioic acid according to our previously reported method20 (2.1
equiv of iodotrimethylsilane in dichloromethane followed by
hydrolysis) and precipitated as the white anilinium salt (8.53
g, 88%) after washing with ether. Mp 135.5-137 °C (with
decomposition). Anal. (C8H14NO2PS2) C, H, N. 1H NMR δ 7.27-
7.54 (m, 5H), 2.90 (d, J ) 10.2 Hz, 2H), 2.24 (s, 3H). 13C NMR
(D2O) δ 132.9 (s, CHaromatic), 128.8 (s, CHaromatic), 123.6 (s,
CHaromatic), 41.7 (d, J ) 102.2 Hz, P-CH2), 19.7 (d, J ) 4.6 Hz,
CH3). 31P NMR δ 62.0.
P h en ylsu lfan ylm eth ylph osph on oth ioic Acid (8), An ilin -
iu m Sa lt. This was prepared from commercially available
phenylsulfanylmethylphosphonic acid diethyl ester (27.99 g,
0.108 mol) following the same procedures described above for
7. The yields of crude intermediates were as follows: sodium
salt of phenylsulfanylmethylphosphonic acid monoethyl ester
(26.03 g, 95%), phenylsulfanylmethylphosphonic acid mono-
ethyl ester monochloride (22.29 g, 87%), phenylsulfanylmeth-
ylphosphonothioic acid O,S-diethyl ester (14.52 g, 59%). The
yield on the final step was 94%, yielding 15.53 g. Mp 165-
166 °C (with decomposition). Anal. (C13H16NO2PS2) C, H, N.
1H NMR δ 7.24-7.51 (m, 5H), 3.41 (d, J ) 11.7 Hz, 2H). 13C
NMR (D2O) δ 132.9 (s, CHaromatic), 132.2 (s, CHaromatic), 131.4
(s, CHaromatic), 129.3 (s, CHaromatic), 129.2 (s, CHaromatic), 123.8
(s, CHaromatic), 41.6 (d, J ) 99.9 Hz, P-CH2). 31P NMR δ 58.0.
(15) Kim, E. E.; Wyckoff, H. W. Reaction mechanism of alkaline
phosphatase based on crystal structures. J . Mol. Biol. 1991, 218,
449-464.
(16) Zhang, Z.-Y. Protein-tyrosine phosphatases: biological function,
structural characteristics, and mechanism of catalysis. CRC Crit.
Rev. Biochem. Mol. Biol. 1998, 33, 1-52.
(17) Kabachnik, M. I.; Ioffe, S. T.; Mastryukova, T. A. Tautomerism
in aprotic media. Zh. Obsch. Khim. 1960, 30, 2763-2767.
(18) Almasi, L.; Fenesan, I.; Biro, V. Heteroorganic substances. LII.
Amidothiophosphonic acids and their derivatives. J . Prakt.
Chemie 1979, 321, 913-920.
(19) Kabachnik, M. I.; Mastryukova, T. A.; Shipov, A. E.; Melent’eva,
T. A. Application of the Hammett equation to the theory of
tautomeric equilibrium. Thiono-thiolo tautomerism of thiophos-
phoric compounds. Doklady Akad. Nauk S.S.S.R. 1959, 124,
1061-1064.
(20) Swierczek, K.; Peters, J .; Hengge, A. C. A convenient synthesis
of phosphonothioic acids. Tetrahedron 2003, 59, 595-599.
(21) Savignac, P.; Lavielle, G. Monodealcoylation d’esters phospho-
riques et phosphoniques par les thiolates et thiophenates. Bull.
Soc. Chim. Fr. 1978, 1506-1508.
(22) Saady, M.; Lebeau, L.; Mioskowski, C. Convenient “One-Pot”
Synthesis of Chlorophosphonates, Chlorophosphates and Chlo-
rophosphoramides from the Corresponding Benzyl Esters. Tet-
rahedron Lett. 1995, 36, 4785-4786.
(23) Malachowski, W. P.; Coward, J . K. The Chemistry of Phos-
phapeptides: Formation of Functionalized Phosphonochloridates
under Mild Conditions and Their Reaction with Alcohols and
Amines. J . Org. Chem. 1994, 59, 7616-7624.
(24) Horner, L.; Gerhard, J . Der Oxidative Abbau von Thiophosphin-
Thiophosphon- und Thiophosphorsaureestern mit Hypochlorit.
Phosphorus Sulfur 1985, 22, 13-21.
Ack n ow led gm en t. This work was supported by
NIH grant GM47297 to A.C.H. and by an ARA award
to J .W.P. The authors gratefully acknowledge the use
of equipment in the Shimadzu Analytical Sciences
Laboratory at Utah State University, created with
equipment donated by Shimadzu Scientific Instruments,
Inc. (Columbia, MD).
(25) Crofts, P. C.; Kosolapoff, G. M. Preparation and Determination
of Apparent Dissociation Constants of Some Alkylphosphonic
and Dialkylphosphinic Acids. J . Am. Chem. Soc. 1975, 75, 3379-
3383.
(26) Freedman, L. D.; Doak, G. O. The Preparation And Properties
Of Phosphonic Acids. Chem. Rev. 1957, 57, 479-523.
(27) Catrina, I. E.; Hengge, A. C. Comparisons of Phosphorothioate
and Phosphate Monoester Transfer Reactions: Activation Pa-
rameters, Solvent Effects, and the Effect of Metal Ions. J . Am.
Chem. Soc. 1999, 121, 2156-2163.
(28) Myers, J . K.; Antonelli, S. M.; Widlanski, T. W. Motifs for
metallophosphatase inhibition. J . Am. Chem. Soc. 1997, 119,
3163-3164.
(29) Kole, H. K.; Smyth, M. S.; Russ, P. L.; Burke, T. R. Phosphonate
inhibitors of protein-tyrosine and serine/threonine phosphatases.
Biochem. J . 1995, 311, 1025-1031.
(30) Goldberg, J .; Huang, H. B.; Kwon, Y. G.; Greengard, P.; Nairn,
A. C.; Kuriyan, J . Three-dimensional structure of the catalytic
subunit of protein serine/threonine phosphatase-1. Nature 1995,
376, 745-753.
(31) Egloff, M.; Cohen, P. T.; Reinemer, P.; Barford, D. Crystal
structure of the catalytic subunit of human protein phosphatase
1 and its complex with tungstate. J . Mol. Biol 1995, 254, 942-
959.
(32) Griffith, J . P.; Kim, J . L.; Kim, E. E.; Sintchak, M. D.; Thomson,
J . A.; Fitzgibbon, M. J .; Fleming, M. A.; Caron, P. R.; Hsiao, K.;
Navia, M. A. Cell 1995, 82, 507-522.
(33) Kissinger, C. R.; Parge, H. E.; Knighton, D. R.; Lewis, C. T.;
Pelletier, L. A.; Tempczyk, A.; Kalish, V. J .; Tucker, K. D.;
Showalter, R. E.; Moomaw, E. W. Crystal structures of human
calcineurin and the human FKBP12-FK506-calcineurin com-
plex. Nature 1995, 378, 641-644.
(34) Voegtli, W. C.; White, D. J .; Reiter, N. J .; Rusnak, F.; Rosenz-
weig, A. C. Structure of the bacteriophage lambda Ser/Thr
protein phosphatase with sulfate ion bound in two coordination
modes. Biochemistry 2000, 39, 15365-15374.
Refer en ces
(1) Kennelly, P. Protein Phosphatases-A Phylogenetic Perspective.
Chem. Rev. 2001, 101, 2291-2312.
(2) Saito, H. Histidine Phosphorylation and Two-Component Signal-
ing in Eukaryotic Cells. Chem. Rev. 2001, 101, 2497-2510.
(3) J ackson, M. D.; Denu, J . M. Molecular reactions of protein
phosphatasessinsights from structure and chemistry. Chem.
Rev. 2001, 101, 2313-2340.
(4) J ohnson, L. N.; Lewis, R. J . Structural Basis for Control by
Phosphorylation. Chem. Rev. 2001, 101, 2209-2242.
(5) Adams, J . A. Kinetic and Catalytic Mechanisms of Protein
Kinases. Chem. Rev. 2001, 101, 2271-2290.
(6) Smyth, M. S.; Ford, H. F.; Burke, T. R. A general method for
the preparation of benzylic R,R-difluorophosphonic acids; non-
hydrolyzable mimetics of phosphotyrosine. Tetrahedron Lett.
1992, 33, 4137-4140.
(7) Chen, L.; Wu, L.; Otaka, A.; Smyth, M. S.; Roller, P. P.; Burke,
T. R., J r.; den Hertog, J .; Zhang, Z. Y. Why is phosphonodifluo-
romethyl phenylalanine a more potent inhibitory moiety than
phosphonomethyl phenylalanine toward protein-tyrosine phos-
phatases? Biochem. Biophys. Res. Commun. 1995, 216, 976-
984.
(8) Gerlt, J . A. Phosphate Ester Hydrolysis. The Enzymes, 3rd ed.;
Academic Press: New York, 1992; Vol. 20, pp 95-139.
(9) Zhang, Y.-L.; Hollfelder, F.; Gordon, S. J .; Chen, L.; Keng, Y.-
F.; Wu, L.; Herschlag, D.; Zhang, Z.-Y. Impaired transition state
complementarity in the hydrolysis of O-arylphosphorothioates
by protein-tyrosine phosphatases. Biochemistry 1999, 38, 12111-
12123.
(10) Breslow, R.; Katz, I. Relative reactivities of p-nitrophenyl
phosphate and phosphorothioate toward alkaline phosphatase
and in aqueous hydrolysis. J . Am. Chem. Soc. 1968, 90, 7376-
7377.
(11) Hollfelder, F.; Herschlag, D. The nature of the transition state
for enzyme-catalyzed phosphoryl transfer. Hydrolysis of O-aryl
phosphorothioates by alkaline phosphatase. Biochemistry 1995,
34, 12255-12264.
(12) Frey, P. A.; Sammons, D. Bond order and charge localization in
nucleoside phosphorothioates. Science 1985, 228, 541-545.
(13) Liang, C.; Allen, L. C. Sulfur does not form double bonds in
phosphorothioate anions. J . Am. Chem. Soc. 1987, 109, 6449-
6453.
(14) Le Du, M. H.; Stigbrand, T.; Taussig, M. J .; Menez, A.; Stura,
E. A. Crystal structure of alkaline phosphatase from human
placenta at 1.8 Å resolution. Implication for a substrate specific-
ity. J . Biol. Chem. 2001, 276, 9158-9165.
(35) Das, A. K.; Helps, N. R.; Cohen, P. T.; Barford, D. Crystal
structure of the protein serine/threonine phosphatase 2C at 2 Å
resolution. EMBO J . 1996, 15, 6798-6809.
(36) Fjeld, C. C.; Denu, J . M. Kinetic analysis of human serine/
threonine protein phosphatase 2CR. J . Biol. Chem. 1999, 274,
20336-20343.
(37) Rusnak, F.; Yu, L.; Todorovic, S.; Mertz, P. Interaction of
bacteriophage lambda protein phosphatase with Mn(II): evi-
dence for the formation of a [Mn(II)]2 cluster. Biochemistry 1999,
38, 6943-6952.
J M030106F