Stereochemistry at phosphorus of the reaction catalyzed by myo-inositol monophosphatase

Article abstract of DOI:10.1021/jm011056m

myo-Inositol monophosphatase (IMPase), the proposed target for lithium therapy for manic depression, is an important enzyme in the biosynthesis of second messengers. Earlier studies have shown that the IMPase-catalyzed hydrolysis of myo-inositol monophosphates to inorganic phosphate and myo-inositol proceeds by direct attack of water at phosphorus. However, research groups have independently proposed either an in-line displacement (with inversion of stereochemistry at phosphorus) or an adjacent attack with a pseudorotation (with retention of stereochemistry at phosphorus). Here, the elucidation of the stereochemical pathway is presented. The IMPase-catalyzed hydrolysis of D-1-S_p-myo-inositol [¹⁷O]-thiophosphate in the presence of H₂¹⁸O gave inorganic R_p-[¹⁶O, ¹⁷O, ¹⁸O]-thiophosphate, with inversion of configuration at phosphorus. This is only consistent with an in-line displacement, and it rules out the controversial adjacent/pseudorotation mechanism. This result will assist in the design of alternative inhibitors of IMPase.

Full text of DOI:10.1021/jm011056m

J . Med. Chem. 2002, 45, 1363-1373
1363
Ster eoch em istr y a t P h osp h or u s of th e Rea ction Ca ta lyzed by m yo-In ositol
Mon op h osp h a ta se
Christine M.-J . Fauroux,† Michael Lee,‡ Paul M. Cullis,‡ Kenneth T. Douglas,† Michael G. Gore,§ and
Sally Freeman*,†
School of Pharmacy and Pharmaceutical Sciences, Manchester University, Oxford Road, M13 9PL, United Kingdom,
Department of Chemistry, University of Leicester, LE1 7RH, United Kingdom, and Division of Biochemistry and Molecular
Biology, School of Biological Sciences, University of Southampton, Southampton SO16 7PX, United Kingdom
Received October 11, 2001
myo-Inositol monophosphatase (IMPase), the proposed target for lithium therapy for manic
depression, is an important enzyme in the biosynthesis of second messengers. Earlier studies
have shown that the IMPase-catalyzed hydrolysis of myo-inositol monophosphates to inorganic
phosphate and myo-inositol proceeds by direct attack of water at phosphorus. However, research
groups have independently proposed either an in-line displacement (with inversion of
stereochemistry at phosphorus) or an adjacent attack with a pseudorotation (with retention of
stereochemistry at phosphorus). Here, the elucidation of the stereochemical pathway is
presented. The IMPase-catalyzed hydrolysis of D-1-Sp-myo-inositol [17O]-thiophosphate in the
presence of H218O gave inorganic Rp-[16O,17O,18O]-thiophosphate, with inversion of configuration
at phosphorus. This is only consistent with an in-line displacement, and it rules out the
controversial adjacent/pseudorotation mechanism. This result will assist in the design of
alternative inhibitors of IMPase.
Sch em e 1. Direct Attack by Water with Inversion
(a)9,17 or Retention (b)15,17 of Stereochemistry at
Phosphorus (R ) myo-Inositol)
In tr od u ction
myo-Inositol monophosphatase (IMPase 3.1.3.25) is
an important enzyme for mechanistic investigation due
to its critical role in the phosphatidyl inositol signaling
pathway and its inhibition by lithium used for the
treatment of manic depression.1,2 This enzyme, which
has been purified from bovine3 and human brain,4 is
responsible for supplying free myo-inositol by catalyzing
the hydrolysis of all myo-inositol monophosphates,
except the 2-isomer,5 into inorganic phosphate and myo-
inositol. The X-ray crystal structures of both the bovine
and the human enzymes (2.1-2.6 Å resolution)6-8 are
known. The enzyme is a dimer with identical 30 kDa
subunits (277 residues), with the active sites located in
large hydrophilic caverns.6,7 Two magnesium cations are
present in the active site.8,9
Early mechanistic studies attempted to address
whether the hydrolysis proceeded by direct nucleophilic
attack by water at phosphorus or via a phospho-enzyme
(P-E) intermediate10,11 involving two in-line displace-
ment reactions.12 While a P-E intermediate is typical
for nonspecific phosphatase enzymes such as alkaline
phosphatase, experiments designed to trap such an
intermediate in the case of IMPase failed.10,12 X-ray
crystallography of IMPase in the presence of D- or L-myo-
inositol 1-phosphate also argued against the formation
of a P-E intermediate, as there is no nucleophilic amino
acid side chain close to the phosphoryl group.8 In
addition, kinetic and site-directed mutagenesis studies
identified an activated water molecule rather than an
amino acid as the nucleophile.9,13 A dissociative mech-
anism via a metaphosphate intermediate has also been
discounted; compounds bearing a hydroxyethyl arm are
potent competitive enzyme inhibitors,14 whereas trans-
phosphorylation of the hydroxyethyl side chain may
have been expected for a dissociative mechanism.
However, two different associative mechanisms
(Scheme 1) with differing stereochemical outcomes have
been suggested for IMPase9,15,16 depending on the loca-
tion of water in the active site. Pollack and co-workers9
proposed an in-line displacement mechanism in which
the nucleophile attacks opposite the leaving group,
which would proceed with inversion of stereochemistry
at phosphorus (Scheme 1a). From structural and muta-
genesis studies, they proposed that the water nucleo-
phile is activated by Glu-70, Mg2+-1, and possibly Thr-
95.9 While the first metal ion activates water for
nucleophilic attack, the second metal ion, coordinated
by three aspartate residues, appears to act as a Lewis
acid, stabilizing the leaving inositol oxyanion.
In contrast, Gani and co-workers15,17 proposed a
nonin-line associative mechanism where the nucleophile
attacks the phosphorus at the same face as the leaving
group (Scheme 1b). Pseudorotation is required to place
the leaving group in an apical position, and the reaction
would proceed with retention of stereochemistry at
* To whom correspondence should be addressed. Tel.: 44 161 275
2366. Fax: 44 161 275 2396. E-mail: sally.freeman@man.ac.uk.
† Manchester University.
‡ University of Leicester.
§ University of Southampton.
10.1021/jm011056m CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/13/2002

1364 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6
Fauroux et al.
phorus.25 A small amount of phosphorylation at the 4/6-
position was also observed. Diastereomers 3 and 4
coeluted by flash chromatography but could be sepa-
rated by preparative high-performance liquid chroma-
tography (HPLC), to give pure samples of the first (δP
82.4 ppm) and second (δP 81.0 ppm) eluted diastereo-
Sch em e 2. Synthesis of D-1-Sp-myo-Inositol
[17O]-thiophosphate (6) Using (1) as a Chiral
Phosphorylating Agent
1
mers, which were fully characterized by 31P, H, and
13C nuclear magnetic resonance (NMR) spectroscopy.
To establish which diastereomer was which, diol 2
was resolved via its biscamphanate using the method
of Vacca and co-workers.26 In their original paper, the
compounds were misassigned, but a correction con-
firmed that the less polar diastereomer was D-1,2:4,5-
di-O-cyclohexylidene 3,6-biscamphanate. Deprotection
gave D-(-)-1,2:4,5-di-O-cyclohexylidene myo-inositol,
which on phosphorylation with 1 gave diastereomer 4.
By comparison with the spectral data of the two dia-
stereomers isolated by HPLC, it was established that
the first eluted diastereomer was 4 and the second
eluted 3.
Treatment of 3 with a solution of trifluoroacetic
anhydride in H217O (52.8% 17O, 46.0% 16O, and 1.2% 18O)
cleaved the P-N bond with inversion of configuration
and removed the cyclohexylidene protecting groups to
form zwitterion 5 (Scheme 2). Compounds containing
the 17O label attached to phosphorus cannot readily be
detected by 31P NMR spectroscopy; however, because of
incomplete enrichment, other isotopes were also incor-
porated into this site, which allowed 5 to be observed
at δP 57.0 ppm. Some phospho group migration (13.5
and 10.5%, respectively) to positions 6 (δP 58.6 ppm) and
4 (δP 58.9 ppm) was also observed. It is proposed that
after acetal deprotection in a minor pathway, the
adjacent equatorial hydroxy group forms a cyclic pen-
tacoordinate intermediate at phosphorus, which on
degradation cleaves to give these phospho migration
products.
Sodium in liquid ammonia22 was used to remove the
chiral scaffold by reductive cleavage of the substituted
benzyl ester group in 5 by C-O cleavage to give optically
pure D-1-Sp-myo-inositol [17O]-thiophosphate (6), δP
44.85 (Scheme 2). A peak at δP 46.6 (26%) was also
present for myo-inositol 4/6-thiophosphate. Mass spec-
trometry (MS) of D-1-Sp-myo-inositol [17O]-thiophosphate
established that the actual 17O incorporation was 30%
with the remaining isotope composition being 69% 16O
and 1% 18O.
phosphorus. The pseudorotation mechanism is unusual
and, as yet, has not been proven for any enzymatic
system.18 They suggest that the 6-OH of myo-inositol
forms a hydrogen bond with the water nucleophile that
completes the primary coordination sphere of Mg2+-2.
One of these two associative mechanisms can be ruled
out by establishing the stereochemical course of the
hydrolysis reaction catalyzed by IMPase. This can be
achieved by analyzing the chiral product inorganic
[16O,17O,18O]-thiophosphate formed from the hydrolysis
of D-1-Sp-myo-inositol [17O]-thiophosphate in the pres-
ence of H218O.19 In this study, we provide a firm
stereochemical basis from which inhibitors and potential
lead drugs can be securely designed.
Un la beled Hyd r olysis Stu d ies w ith IMP a se. Ul-
timately, the hydrolysis of 6 in H218O catalyzed by
IMPase4,27 is required; however, it was first necessary
to optimize the conditions to monitor the hydrolysis of
unlabeled substrate.
Resu lts a n d Discu ssion
Ch em istr y. The chiral phosphorylating agent (2R,4S,
5R)-2-chloro-3,4-dimethyl-5-phenyl-1,3,2-oxazaphospho-
lidin-2-one has been used by our group to prepare D-(+)-
myo-inositol 3-phosphate.20 The thio analogue (1) has
been utilized to introduce 17O and 18O labels into organic
thiophosphate analogues for stereochemical studies at
phosphorus,21-23 and here, 1 has been used to prepare
D-1-Sp-myo-inositol [17O]-thiophosphate (6) (Scheme 2).
Phosphorylation of the anion from diol (2)24 gave (D/
L)-1,2:4,5-di-O-cyclohexylidene-3-[(2S,4R,5S)-3,4-dimethyl-
5-phenyl-2-sulfide-1,3,2-oxazaphospholidinyl]-myo-inos-
itol 3 and 4 with retention of configuration at phos-
The activity of recombinant IMPase from bovine brain
was confirmed by observing the rapid hydrolysis of
D-myo-inositol 1/3-phosphate (δP 4.8 ppm) to inorganic
phosphate (δP 3.4 ppm) by 31P NMR spectroscopy. The
corresponding hydrolysis of D-myo-inositol 1/3-thiophos-
phate was significantly slower, with 10% substrate (δP
44.8 ppm) remaining after 24 h at 37 °C. The major
product was inorganic thiophosphate (PSi, δP 32.2 ppm);
however, inorganic phosphate (δP 5.1 ppm) and a
desulfurization intermediate (δP 16.6 ppm, 2-O3P-S-
S-PO32-) were also formed.

Reaction Catalyzed by myo-Inositol Monophosphatase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6 1365
Ta ble 1. Percentages of Pi (%) Observed with Time (h) from a
Solution of PSi (8.95 mM) in Tris-HCl, pH 9.00, at 25 °C with
and without Different Concentrations of IMPasea
time without 0.119 mg mL-1 0.235 mg mL-1 0.343 mg mL-1
(h) enzyme 3.97 × 10-6
M
7.82 × 10-6
M
11.43 × 10-6
M
24
48
72
4.3
11.4
15.3
4.7
8.1
16.4
8.7
17.9
19.6
14.7
26.9
34.6
a
The reactions were monitored by 31P NMR spectroscopy, and
peak areas were used.
Sch em e 3. Coupled Enzymatic Reaction with Inositol
Dehydrogenase To Monitor Hydrolysis of myo-Inositol
(Thio)phosphate (X ) O or S) Catalyzed by IMPase
F igu r e 1. Kinetics of the hydrolysis of myo-inositol 1/3-
thiophosphate [S] catalyzed by IMPase (6.7 µg ml-1) in 50.4
mM Tris-HCl buffer containing 102 mM NaCl and 3 mM
MgCl2‚6H2O, pH 9.00, at 25 °C with 5.0 mM â-NAD+ and 0.19
units mL-1 inositol dehydrogenase. Plot of initial velocity vs
[S].
The rate of desulfurization of PSi is critical for the
labeled study as loss of sulfur leads to loss of the
stereochemical information. Therefore, the chemical
stability of PSi was monitored at pD 11-12 and 8.2 by
31P NMR spectroscopy at 22 and 37 °C. In agreement
with Dittmer and Ramsay,28 desulfurization was slower
at pD 11-12 than at pD 8.2, and as anticipated, the
rate increased with increasing temperature. Therefore,
to minimize chemical desulfurization of PSi, reactions
catalyzed by IMPase were carried out at pH 9 and 25
°C. In addition to chemical instability, it is of importance
to understand the possible role of IMPase in the
desulfurization of PSi. Solutions of PSi (8.95 mM) were
incubated at 25 °C in Tris-HCl buffer (pH 9.00) in the
presence and absence of IMPase at various concentra-
tions. The reactions were monitored by 31P NMR
spectroscopy over 72 h; the percentages of Pi formed are
given in Table 1. The data, fitted with the Grafit
program29 to the Michaelis-Menten equation, showed
that IMPase catalyzes desulfurization of PSi. Therefore,
the labeled reaction had to be designed to maximize the
recovery of PSi. This result is consistent with other
enzymes,30 for example, the desulfurization of PSi is also
catalyzed by acid and alkaline phosphatases.31
IMPase),4 kcat ) 0.48 ( 0.01 s-1, and kcat/Km ) 4.42 ×
103 M-1 sec-1
.
The coupled assay was used to estimate the kinetic
constants for the IMPase-catalyzed hydrolysis of myo-
inositol 1/3-thiophosphate (Figure 1). At pH 9.00 at 25
°C, the Km value was 2.07 ( 0.54 mM (by fitting the
substrate inhibition equation using force estimates and
simple weighting; see Figure 1), the kcat value was 0.185
( 0.037 s-1, the Ksi value was 1.14 ( 0.35 mM, and the
k
cat/Km value was 89.4 M-1 sec-1. These are in the range
of those published by Baker et al.35 and Cole and Gani15
(Km ) Ksi ) 1 mM). At high concentrations of myo-
inositol 1/3-thiophosphate (>1 mM), inhibition of hy-
drolysis was observed. The turnover (kcat) of myo-inositol
1/3-thiophosphate was 3 times slower than that ob-
served for D-myo-inositol 3-phosphate. The specificity
constant (kcat/Km) for D-myo-inositol 3-phosphate was 49
times higher than that for myo-inositol 1/3-thiophos-
phate confirming that the phosphate ester is a better
substrate of IMPase than the thiophosphate ester.
Substrate inhibition was confirmed when a high con-
centration of myo-inositol 1/3-thiophosphate (24 mM, δP
(D2O) 44.8 ppm) in the presence of IMPase (0.71 mg
mL-1) at pH 8.85 and 37 °C was shown to be stable
toward hydrolysis by 1H and 31P NMR spectroscopy. The
presence of the P-S group reduces the susceptibility of
phosphorus to nucleophilic attack via an associative
mechanism.
It was difficult using 31P NMR spectroscopy to obtain
good kinetic data for the hydrolysis of myo-inositol 1/3-
thiophosphate; therefore, a UV assay was developed.
The reaction catalyzed by IMPase can be monitored
spectrophotometrically at 340 nm by coupling to inositol
dehydrogenase, which catalyzes the oxidation of myo-
inositol to scyllo-inosose (Scheme 3).32-34 The kcat, Km,
and kcat/Km values for the oxidation of myo-inositol
catalyzed by inositol dehydrogenase at pH 9.13 were
found to be 13.3 ( 2.0 s-1, 0.58 ( 0.17 mM, and 2.3 ×
104 M-1 sec-1, respectively. The coupled enzyme assay
required that the IMPase step was rate-determining.
This was readily achieved by using inositol dehydroge-
nase in excess (0.14 units mL-1), such that the oxidation
of myo-inositol to scyllo-inosose was very fast. The
observed velocity was shown to be independent of
inositol dehydrogenase concentration but proportional
to IMPase concentration. The kinetic constants for the
hydrolysis of D-myo-inositol 3-phosphate catalyzed by
IMPase at pH 9.00 at 25 °C were Km ) 0.11 ( 0.01 mM
(similar to literature values for the bovine recombinant
It was considered that substrate inhibition might
arise through a covalent interaction between IMPase
and myo-inositol 1/3-thiophosphate. The crystal struc-
ture of IMPase with D- or L-myo-inositol 1-phosphate
bound7,8 shows a disulfide bond between Cys-24 and
Cys-125, although in the absence of substrate this
disulfide bond was not observed. Electrospray mass
spectrometry (ESI-MS) was used to probe the possibility
of a covalent interaction between the cystine disulfide
bond of IMPase and the myo-inositol 1/3-thiophosphate.
The experimental mass of native, recombinant IMPase
was 29 924.4 Da, which corresponds to 276 amino acids
with loss of the N-terminal methionine. The recombi-
nant enzyme, which had been in a 20.3 mM solution of
myo-inositol 1/3-thiophosphate for 90 min, gave a mo-
lecular mass of 29 922.6 ( 3 Da, which confirms that

1366 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6
Fauroux et al.
aliquots of substrate were added every 2 h to allow the
hydrolysis to proceed to near completion between ad-
ditions. After 8 aliquots, the 31P NMR (D2O) spectrum
gave δ 32.5 (29.4%, PSi), 45.0 (58.0%, D-1-Sp-myo-inositol
[17O]-thiophosphate), and 46.7 ppm (12.6%, myo-inositol
4/6-thiophosphate). The hydrolysis was not complete as
some accumulation of substrate caused inhibition. In
the second experiment, the concentration of substrate
was halved; aliquots of D-1-Sp-myo-inositol [17O]-thio-
phosphate (12 µL, initial concentration 1.07 mM) were
added every 2 h (14 additions) to the IMPase solution
in H218O. The enzymatic reaction was monitored 90 min
after the addition of the first aliquot and shown to
contain 1.0 mM myo-inositol, with only 0.1 mM starting
material remaining. After 14 aliquots, 12.15 mM of myo-
inositol was formed with 3.25 mM D-1-Sp-myo-inositol
[17O]-thiophosphate remaining. After complete addition,
the 31P NMR (D2O) spectrum gave δ 5.5 (14.9%, Pi), 32.2
(39.7%, PSi), 44.8 (35.4%, D-1-Sp-myo-inositol [17O]-
thiophosphate), and 46.5 ppm (10.0%, myo-inositol 4/6-
thiophosphate). Hydrolysis was more complete in the
second experiment than in the first; however, 15% Pi
was formed in the longer second experiment, arising
from desulfurization of PSi.
For each experiment, inorganic [16O,17O,18O]-thio-
phosphate was separated from D-1-Sp-myo-inositol [17O]-
thiophosphate by anion exchange chromatography. By
MS, the experimental 18O incorporation during enzy-
matic hydrolysis was estimated to be 83%, with the
remaining isotope composition being 1% 17O and 16%
16O. The reactions gave a total of 23 µmol of inorganic
[16O,17O,18O]-thiophosphate (based on PSi having an ꢀ
of 4200 M-1 cm-1 at 225 nm, pH 7),36 which was
subjected to stereochemical analysis.
F igu r e 2. Plot of percentage activity in the hydrolysis of myo-
inositol 1/3-phosphate vs concentration of PSi as an inhibitor
of IMPase in 50 mM Tris-HCl buffer containing 100 mM KCl
and 2.7 mM MgCl2‚6H2O, pH 9.00, at 25 °C with 4.82 mM
â-NAD+, 0.19 units mL-1 inositol dehydrogenase, 1.42 µg mL-1
IMPase, and 0.5 mM myo-inositol 1/3-phosphate. Points are
experimental: line is theoretical assuming linear competitive
inhibition with Ki ) 0.48 mM.
myo-inositol 1/3-thiophosphate does not form a covalent
complex with IMPase.
IMPase could also be affected by product inhibition;
therefore, its activity in the presence of PSi was moni-
tored. Pi is a known competitive inhibitor of IMPase (Ki
values 0.32-0.52 mM at pH 8.0 and 8 mM at pH 6.5).3,10
The inhibition of IMPase by PSi, evaluated at 0.5 mM
myo-inositol 1/3-phosphate (Figure 2, data for pH 9.0),
was similar to that for Pi: assuming linear competitive
inhibition gave Ki ) 0.5 mM at either pH 6.5 or pH 9.0.
This result is consistent with PSi being dianionic at both
pH 6.5 and pH 9.0 (pKa1 2.05, pKa2 5.6, and pKa3 10.2).28
As product inhibition was shown not to be sensitive to
pH, the labeled experiment was conducted at pH 9.00
because PSi is chemically more stable at basic pH.
In summary, the following factors had to be consid-
ered in the experimental design for the IMPase-
catalyzed hydrolysis of D-1-Sp-myo-inositol [17O]-thio-
phosphate in H218O. Hydrolysis of myo-inositol 1/3-
thiophosphate competes with desulfurization of PSi;
therefore, the reaction should not be left longer than
necessary. Product inhibition by PSi occurs, which
prevents the reaction from proceeding to completion.
Substrate inhibition by myo-inositol 1/3-thiophosphate
occurred, requiring that its concentration should not
exceed 1-2 mM. For 25 mg of D-1-Sp-myo-inositol [17O]-
thiophosphate, this would require 100 mL of H218O,
which was prohibitively expensive. To minimize sub-
strate inhibition, a method was developed in which myo-
inositol 1/3-thiophosphate was added in several aliquots
to IMPase in 5 mL of H218O. This protocol, which
maintained a low substrate concentration, was then
used for the labeled experiment.
Arnold, Bethell, and Lowe developed a method for the
stereochemical analysis of labeled inorganic thiophos-
phate, which is shown in Scheme 4 for inorganic Rp-
[16O,17O,18O]-thiophosphate.37 Each 16O, 17O, or 18O
oxygen of the thiophosphate has an equal chance of
reacting with diphenyl phosphorochloridate and there-
fore has an equal chance of being eliminated in the
cyclization reaction. The isotopomeric cyclic phospho-
thiolate triesters derived from the chiral inorganic
[16O,17O,18O]-thiophosphate were analyzed by high-field
31P NMR spectroscopy (Figure 3). Any products contain-
ing an 17O atom bonded to phosphorus are silent in the
31P NMR spectrum. In addition, the incorporation of 17
O
and 18O into chiral inorganic thiophosphate is not 100%
because both of the commercial isotopically labeled
waters contain some of the other isotopes. A total of 12
triesters are formed, but only 8 are observable by 31P
NMR, the percentages of which are given in Scheme 4.
In this experiment, only 6 species were detected because
10 and 14 did not appear in the 31P NMR spectrum as
there was only 1.2% 18O in H217O. Compounds with only
16O, 7, and 11 (Scheme 4) give rise to the most downfield
peaks in the 31P NMR spectrum.
Ster eoch em ica l Stu d y w ith IMP a se. The hydroly-
sis of D-1-Sp-myo-inositol [17O]-thiophosphate with H218O
catalyzed by IMPase was completed in two experiments.
In the first experiment, an aliquot of D-1-Sp-myo-inositol
[17O]-thiophosphate in H218O (25 µL, initial concentra-
tion 2.14 mM) was added every 2 h (8 additions) to a
solution of IMPase in 97% H218O at 25 °C. The reaction
was monitored 1 h after the addition of the first aliquot
and shown to contain 1.14 mM of myo-inositol and 1.00
mM of substrate. To minimize substrate inhibition,
From the chiral inorganic Rp-thiophosphate, com-
pounds 8 and 12, which contain 18O-P and 18OdP,
respectively, are the major products. Comparison of the
ratios (8/9 and 12/13) of the peak heights determines if
the inorganic thiophosphate is of the R or S configura-
tion at phosphorus.

Reaction Catalyzed by myo-Inositol Monophosphatase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6 1367
Sch em e 4. Stereochemical Analysis of Chiral Inorganic [16O,17O,18O]-Thiophosphate, Showing the Cyclization Reaction
that Eliminates 17Oa,b
a
Shown are the eight triester products, which are observed by 31P NMR spectroscopy. The percent of each species given is from mass
spectral data with 17O incorporation being 69% 16O, 30% 17O, and 1% 18O and 18O incorporation being 16% 16O, 1% 17O, and 83% 18O.
Reagents: (i) 2[HN+(n-octyl)3], DMF. (ii) (PhO)2POCl, NBun3. (iii) CH2N2.
b
F igu r e 3. 31P NMR spectrum (Bruker ARX400; 162 MHz) of the sample from the configurational analysis of inorganic [16O,17O,18O]-
thiophosphate derived from the hydrolysis of D-1-Sp-myo-inositol [17O]-thiophosphate in H218O catalyzed by recombinant bovine
IMPase. (The spectrum was recorded at a digital resolution of 0.038 Hz per point and processed without resolution enhancement,
with natural line widths of 0.27 Hz.)
From the mass spectral data of both the labeled
substrate and the product, the predicted experimental
ratios for the resonances from the trans ester (7:8:9) for
inversion of configuration are 40.6:34.9:24.5, whereas
for retention of configuration the ratios would be 40.6:
24.5:34.9. The experimental ratios 7:8:9 of 40.5:33.5:26
(obtained from the peak heights in the 31P NMR
spectrum) are in excellent agreement with a complete
inversion of configuration at phosphorus for the hy-
drolysis of D-1-Sp-myo-inositol [17O]-thiophosphate to PSi
catalyzed by IMPase.
This result is entirely consistent with an in-line
nucleophilic displacement reaction (Scheme 1a). It rules
out a mechanism involving a P-E intermediate10 and
also the adjacent displacement reaction accompanied by
pseudorotation (Scheme 1b).17,38 It also argues against
a mechanism involving a free metaphosphate interme-
diate (providing that this intermediate was able to
rotate in the active site) since this would be ac-
companied by racemization of the stereochemistry at
phosphorus. These observations are consistent with the
“Merck” model for the hydrolysis mechanism, which is

1368 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6
Fauroux et al.
based on their X-ray structural work.6-9 In this mech-
anism, the nucleophilic water is coordinated to Mg2+-1
and is activated by Glu-70 and Thr-95, allowing ap-
proach in-line with the inositol leaving group.
The stereochemical result has already led Gani and
co-workers to revise their proposal regarding the crucial
6-hydroxy group in D-myo-inositol 1-phosphate: there
is reduced catalytic activity or inhibition with substrate
analogues that lack the 6-OH (e.g., D-6-O-methylinositol
1-phosphate and 6-deoxyinositol 1-phosphate).39,40 They
have now proposed that the 6-OH group hydrogen bonds
with a water molecule bound to Mg2+-2, while the
nucleophilic water is associated with Mg2+-1.
77.0 ppm) or CH3CN (δC 1.6 ppm) and benzene (δC 128.0 ppm)
in D2O. All 13C and 31P NMR spectra were 1H-decoupled.
Infrared spectra were recorded on a Philips PU 9516 infrared
spectrophotometer. MS data were recorded in the Department
of Chemistry, Manchester University. Fast atom bombardment
(FAB) and chemical ionization (CI) mass spectra were recorded
on a Fisons VG Trio 2000 mass spectrometer. FAB mass
spectra were recorded with a mNBA matrix, while CI mass
spectra were recorded using NH3 as carrier gas. ESI mass
spectra were recorded on a Micromass Platform spectrometer
using acetonitrile-water (1:1) as the mobile phase. Melting
points were determined using a Gallenkamp melting point
apparatus. Optical rotations were measured on an AA-100
polarimeter (sensitivity 1 millidegree).
The solvents used were dried by heating under reflux over
the appropriate drying agent for 1 h, followed by distillation
with tetrahydrofuran (THF) (sodium wire, benzophenone) and
triethylamine (KOH).
Con clu sion
The stereochemical course of the reaction at phos-
phorus catalyzed by IMPase has been elucidated by the
hydrolysis of D-1-Sp-myo-inositol [17O]-thiophosphate in
H218O. Analysis of the chiral inorganic [16O,17O,18O]-
thiophosphate product showed that the reaction pro-
ceeded with inversion of configuration (Scheme 1a),
which is only consistent with the in-line displacement
mechanism. It is interesting to note that fructose 1,6-
bisphosphatase utilizes two magnesium cations, is
sensitive to lithium, and also proceeds with inversion
of stereochemistry at phosphorus.39,41
IMPase has been implicated in manic depression, and
lithium inhibition is used currently in the treatment of
this disease. Lithium concentrations are difficult to
control, and new inhibitors for the treatment of manic
depression are required.2 The results of the present
study provide the solid platform of structure/mechanism
required in the design of novel inhibitors for IMPase,
especially if mechanism-based inhibitors with advan-
tages in potency and selectivity are to be contemplated.
Flash column chromatography42 was performed using chro-
matography grade silica gel 60 (35-75 µm, Prolabo). Thin-
layer chromatography (TLC) was carried out using Merck 60
F254 silica gel TLC plates, and the components were revealed
with phosphomolybdic acid reagent (PMA) in MeOH and
molybdic acid reagent (1 g of molybdic acid in 30 mL concen-
trated H2SO4, diluted with 50 mL water) or observed by 254
nm UV lamp. TLC of phosphorus compounds was performed
using Whatman K6 Silica gel on glass TLC plates. Spots were
visualized using malachite green spray reagent.43 Preparation
of the Dowex-50 (H+ form, length ) 10 cm, diameter ) 3 cm)
and the Dowex-50 (Na+ form, length ) 20 cm, diameter ) 2
cm) resin cation exchange columns was as follows: Two
columns were prepared approximately two-thirds full of
Dowex-50 (Dowex 50-X8, 20-50 Mesh). These were washed
with H2O (500 mL), 6 M HCl (500 mL), and then H2O (500
mL) to give both columns in the H+ form. One of the columns
was washed with 6 M NaOH (500 mL) and then H2O (500 mL)
to obtain Dowex-50 in the Na+ form. Anion exchange chroma-
tography was performed using a fraction collector (Pharmacia
LKB 2211-010 SuperRac). A conductivity meter (conductance
bridge, Griffin and George Ltd.) with conductivity cell Griffin
(PJ K-320-518Q, platinum black) was used to measure the
conductance (from 0.1 to 1 Ω1-) of the triethylammonium
bicarbonate (TEAB) buffer at different strengths and of the
anion exchange loading sample. Anion exchange was per-
formed with an analytical grade anion exchange resin (AG
1-X8, 200-400 mesh, formate form, BIO-RAD, total capacity
of 2.9 mequiv mL-1 resin bed).
Exp er im en ta l Section
Labeled 17O water (16O: 46.0%, 17O: 52.8%, and 18O: 1.2%)
was obtained from Cambridge Isotope Laboratories. Labeled
18O water (O: 1.78%, 17O: 0.96%, and 18O: 97.26%) was
obtained from Goss Scientific Instruments Ltd. Cyclohexy-
lammonium D-myo-inositol 1/3-phosphate (∼75%) containing
the 2-phosphate isomer (25%), malachite green carbinol,
trisodium thiophosphate dodecahydrate (97%), myo-inositol,
labeled 18O water (18O: 95%), and anhydrous solvents (dim-
ethylformamide (DMF), diethyl ether, and ethanol) were
obtained from Aldrich Chemical Co. Ammonium molybdate
[(NH4)6Mo7O24‚4H2O], sodium chloride, sodium hydroxide pel-
lets, magnesium chloride hexahydrate, disodium hydrogen
orthophosphate dodecahydrate, potassium chloride, sodium
hydrogen carbonate, tris(hydroxymethyl)methylamine, and
standard pH buffers were obtained from BDH Chemicals Ltd.
Low molecular weight protein standards, Phast gel sample
applicator, Phast gel sodium dodecyl sulfate (SDS) buffer
strips, Phast system separation, and control machine unit gels
were obtained from Pharmacia, U.K. myo-Inositol dehydroge-
nase from Enterobacter aerogenes and bovine serum albumin
(BSA, crystallized and lyophilised) were obtained from Sigma
Chemical Company.
To determine concentrations of (organic) phosphate or
thiophosphate solutions, samples were oxidized with perchloric
acid to give inorganic phosphate, which was assayed at 660
nm as a complex with ascorbic acid and ammonium molybdate.
IMPase, purified from bovine brain,44 was stored in 60%
(NH4)2SO4 containing 10 mM ethylenediaminetetraacetic acid
(EDTA). Assay conditions of pH 8.0, 37 °C, 50 mM Tris/HCl
containing KCl (100 mM), MgCl2 (7 mM), and myo-inositol
1-monophosphate (1 mM) gave a specific activity of 13 µmol
min-1 mg-1. Electrophoresis was carried out using the Phar-
macia Phast system with manual gel staining using precast
Phastgel Gradient 10-15 gels and PhastGel SDS buffer strips,
which contained 0.2 M Tricine, 0.2 M Tris, and 0.55% SDS,
pH 8.1, in a 3% agarose isoelectric focusing support. Gels were
stained with Coomassie blue stain. Protein concentration was
determined by the method of Bradford (1976)45 using BSA as
standard.
Ultrafiltration (10 YM 10 (25 mm) 136 12 DIAFLO mem-
brane, Amicon Corporation) of the enzyme was used to dilute
the storage EDTA concentration (10 mM) to 10 µM against 3
times 10 mL of Tris/HCl buffer (10 mM, pH ∼8.1) in the
presence of the antioxidant dithioerythritol (DTE) (0.5 mM)
as required. The diluted solution of protein was concentrated
to a volume of 1 mL and stored in a sterilized Eppendorf tube
in the freezer (∼ -30 °C). All metal ion solutions were prepared
in 50 mM Tris/HCl, pH 8.0. A stock solution of antioxidant
DTE (5mM) was prepared in 50 mM Tris/HCl, pH ∼8.1. The
Routine NMR spectra were recorded on a J EOL NMR-
1
EX270 spectrometer for H (270.0 MHz), 31P (109.2 MHz), and
13C (67.8 MHz) data. The 31P NMR (162 MHz) spectrum of the
sample from the stereochemical analysis was recorded on a
1
Bruker ARX400 spectrometer. H spectra were referenced to
tetramethylsilane for samples run in CDCl3 and relative to
benzene (at 7.44 pm) for samples run in D2O. 31P NMR spectra
were referenced to 85% phosphoric acid (upfield chemical shifts
are negative). 13C NMR spectra were referenced to CDCl3 (δC

Reaction Catalyzed by myo-Inositol Monophosphatase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6 1369
pH required was obtained by addition of HCl or NaOH. The
pD value was calculated from the measured pH value.46
Kinetic parameters were determined by fitting the data to
the appropriate equation by nonlinear least squares regression
analysis using Grafit29 (Erithacus Software Ltd., distributed
by Sigma Chemical Co.).
assignments made with the aid of a DEPT spectrum) δC
(ppm): 14.4 (d, J PC ) 2.5 Hz, CHCH3), 23.5, 23.6, 23.9, 24.0,
24.9, 29.7, 34.8, 36.3, 36.5, 38.1 (10 × CH2 of cyclohexylidene),
29.5 (d, J PC ) 7.3 Hz, NCH3), 59.5 (d, J PC ) 11 Hz, CHCH3),
75.2 (d, J PC ) 6.1 Hz, inositol-CH), 75.6 (inositol-CH), 75.7
(inositol-CH), 76.5 (inositol-CH), 77.8 (inositol-CH), 81.4 (inosi-
tol-CH), 82.4 (CHPh), 110.9, 113.3 (C-1 of cyclohexylidene),
126.5, 128.2 (CH of Ph, 1 overlapping), 136.2 (d, J PC ) 4.9 Hz,
C-1 of Ph). IR (Nujol): νmax 3405 (OH) cm-1; m/z (CI) 566 (M
+ H+, [12C] 12.3%), 567 (M + H+, [13C] 4%), 148 (Ph+CH-
CHMe-NMeH, 100%).
To identify each diastereomer, 1,2:4,5-di-O-cyclohexylidene-
myo-inositol 3,6-bis-camphanate and 2,3:5,6-di-O-cyclohexyl-
idene-myo-inositol 1,4-bis-camphanate were prepared26 from
(D/L)-1,2:4,5-di-O-cyclohexylidene-myo-inositol and S-(-)-cam-
phanic chloride. The less polar diastereomer was 1,2:4,5-di-
O-cyclohexylidene 3,6-bis-camphanate, which was deprotected
using KOH in absolute ethanol to give D-(-)-1,2:4,5-di-O-
cyclohexylidene myo-inositol in 99% yield. This diol was
phosphorylated using (2R,4R,5S)-(+)-2-chloro-3,4-dimethyl-5-
phenyl-1,3,2-oxazaphospholidin-2-sulfide to give 4 with all of
the spectroscopic data in agreement with those reported above.
Measurements of pH values were taken with a Corning
digital 240 pH meter and BDH Colorkey buffer standards were
used as reference solutions. Values of pH of small volumes of
buffer were determined using a Kent EIL 7005 pH meter
calibrated with standard buffers at 20 °C. Enzyme assays,
spectral scans, and absorbance readings at fixed wavelengths
were performed using either a Perkin-Elmer λ3 spectropho-
tometer thermostated to 37 °C with a Haake GH water bath
or a Cary UV/vis spectrophotometer thermostated to 25 °C
using a Cary temperature controller. Poly(methyl methacry-
late) (PMMA) plastic cuvettes or quartz cuvettes were used
for readings in the UV region at 280 and 340 nm; otherwise,
disposable polystyrene (PS) plastic cuvettes were used for
readings in the visible region. All buffer solutions were
prepared using analytical grade reagents and water, which
had been distilled, deionized, and then treated using a Milli-Q
purification system. All stock solutions (except buffer) were
stored over ice during kinetic experiments.
(1S,2R)-(2-Meth yla m in o-1-p h en ylp r op yl) 1-m yo-In osi-
tol-(Rp)-[16O,17O]-th iop h osp h a te (5). A solution of diaste-
reomer 4 (0.167 g, 0.295 mmol) in THF (0.5 mL) was added to
a solution of trifluoroacetic anhydride (0.38 mL, 2.83 mmol)
in the presence of labeled 17O water (0.5 mL, 52.8% 17O) under
argon at ∼5 °C. The mixture was stirred for 30 min, and then,
the solution was concentrated under vacuum. Water (20 mL)
and diethyl ether (30 mL) were added, and the aqueous layer
was separated and concentrated under vacuum. The reaction
mixture gave optically pure 17O-labeled zwitterion (5) (120 mg,
96%) in the presence of 16O- and 18O-labeled zwitterions, which
were observed by 31P NMR. 31P NMR (D2O) δ (ppm): 56.51 (s,
1.7%, [16O,16O]-zwitterion phosphorylated at position 3), 56.98
(s, 70.54%, [16O,16O]-zwitterion phosphorylated at position 1),
57.4 (s, 3.4%, impurities), 58.6 (s, 13.5%, [16O,16O]-zwitterion
phosphorylated at position 6), 58.9 (s, 10.5%, [16O,16O]-zwit-
terion phosphorylated at position 4). 1H NMR (D2O, referenced
to water at 4.7 ppm) δ (ppm): 0.87 (3H, d, J HH ) 6.93 Hz,
CHCH3), 2.52 (3H, s, NCH3), 3.06 (1H, t, J HH ) 9.24 Hz, H-5),
3.25-3.45 (2H, m, H-1/H-3), 3.53 (1H, dd appears as a br t,
J HH ) 9.9 Hz, J HH ) 9.57 Hz, H-2), 3.96-4.02 (2H, m, H-4/H-
6), 5.4-5.5 (1H, m, CHPh), 7.15 (5H, m, Ph). m/z (ES-): 422
(M - H+ [16O,16O], 100), 423 (M - H+ [16O,17O] or M - H+
[16O,16O,113C], 65.24), 424 (M - H+ [16O,18O] or M - H+
[16O,17O,113C], 10.97). Considering the natural 13C content, this
mass spectral data corresponds to 54% [16O,16O], 36% [16O,17O].
(D)-2,3:5,6-Di-O-cycloh exylid en e 1-[(2S,4R,5S)-3,4-Di-
m eth yl-5-ph en yl-2-su lfide-1,3,2-oxazaph osph olidin yl]myo-
in ositol (3) an d (D)-1,2:4,5-Di-O-cycloh exyliden e 3-[(2S,4R,
5S)-3,4-Dim eth yl-5-p h en yl-2-su lfid e-1,3,2-oxa za p h osp h o-
lid in yl]m yo-in ositol (4). (D/L)-1,2:4,5-Di-O-cyclohexylidene-
myo-inositol24 (2.57 g, 7.5 mmol) and 60% NaH (1.6 equiv,
0.297 g, 12.4 mmol) in dry DMF (25 mL) was stirred for 1 h at
0 °C under argon. To this reaction mixture was added dropwise
(2R,4R,5S)-(+)-2-chloro-3,4-dimethyl-5-phenyl-1,3,2-oxazaphos-
pholidin-2-sulfide (1.974 g, 7.5 mmol) in dry DMF (7.5 mL),
and the mixture was stirred for 2 h allowing the temperature
to rise to room temperature. Flash chromatography on silica
gel eluting with EtOAc-hexane (1:5, 1:3, and then 1:2) gave
one pair of diastereomers as the major product, 42.7% (1.286
g): Rf ) 0.36 (EtOAc-hexane, 1:2). 31P NMR (CDCl3) δP
1
(ppm): 81.4 (s, 39%, 4) and 83.1 (s, 61%, 3). H NMR (CDCl3)
δH (ppm): 0.83 (d, J HH ) 6.6 Hz, 3) and 0.84 (d, J HH ) 6.6 Hz,
4), (total 3H, CHCH3), 2.75 (d, J PH ) 12.9 Hz, 3) and 2.76 (d,
J PH ) 12.5 Hz, 4), (total 3H, NCH3), 4.56 (t, J HH ) 4.6 Hz, 3)
and 4.69 (t, J HH ) 4.6 Hz, 4), (total 1H, H-2). IR (Nujol): νmax
3410 (OH) cm-1
.
Diastereomers 3 and 4 were separated by HPLC on a
preparative C18 reverse phase column, eluting with CH3CN-
H2O (60:40). The diastereomer that eluted first (160 mg) with
a retention time of 34.2 min was 4. 31P NMR (d4-MeOH) δP
D-1-Sp-m yo-In ositol [17O]-Th iop h osp h a te (6). A small
piece of sodium was added to distilled liquid NH3 (5 mL) to
give a blue solution, and the labeled zwitterion (5) (122 mg,
0.29 mmol) was added. The reaction mixture was stirred for
30 min at - 78 °C, the reaction was quenched by the addition
of EtOH (5 mL), and the reaction mixture was allowed to warm
to room temperature. Water (20 mL) was added and then
Dowex-50 (H+ form) resin to adjust the pH to 5. This solution
was applied to a Dowex-50 (Na+ form) resin column and eluted
with H2O (100 mL). The eluant was concentrated on a high
vacuum rotary evaporator to give disodium D-1-Sp-myo-inositol
[17O]-thiophosphate (80 mg, 86%), which was stored under
argon in the freezer. 31P NMR (D2O) δ (ppm): 44.85 (s, 74.3%,
myo-inositol 1-[16O,16O]-thiophosphate), 46.6 (s, 25.7%, myo-
inositol 4-[16O,16O]-thiophosphate and myo-inositol 6-[16O,16O]-
thiophosphate). Compound 6 is not observed by 31P NMR
1
(ppm): 81.0 (s). H NMR (CDCl3) δH (ppm): 0.84 (3H, d, J HH
) 6.6 Hz, CHCH3), 1.2-1.8 (20H, m, 10 × CH2 of cyclohexy-
lidene), 2.53 (1H, d, J HH ) 3.0 Hz, 6-OH), 2.76 (3H, d, J PH
)
12.5 Hz, NCH3), 3.42 (1H, dd, J HH ) 10.6 Hz, J HH ) 9.6 Hz,
H-5), 3.6-3.8 (1H, m, CHCH3), 3.8-3.95 (1H, m, H-6), 4.0-
4.1 (2H, m, H-1/H-4), 4.69 (1H, t, J HH ) 4.6 Hz, H-2), 4.8-5.0
(1H, m, H-3), 5.68 (1H, dd, J HH ) 6.3 Hz, J HH ) 3.0 Hz, CHPh),
7.3-7.4 (5H, m, Ph). 13C NMR (CDCl3) δ (ppm): 13.8 (s,
C
CHCH3), 23.6, 23.7, 23.8, 24.1, 24.9, 25.1, 34.9, 36.3, 36.5, 37.9
(10 × CH2 of cyclohexylidene), 29.3 (d, J PC ) 6.1 Hz, NCH3),
59.8 (d, J PC ) 9.7 Hz, CHCH3), 75.0 (d, J PC ) 6.1 Hz, inositol-
CH), 75.2 (inositol-CH), 75.3 (d, J PC ) 6.1 Hz, inositol-CH),
76.0 (inositol-CH), 77.7 (inositol-CH), 81.3 (inositol-CH), 82.3
(CHPh), 110.7, 113.3 (C-1 of cyclohexylidene), 126.1, 128.1,
128.2 (CH of Ph), 136.1 (d, J PC ) 6.1 Hz, C-1 of Ph).
The diastereomer that eluted second (360 mg) with a
1
retention time of 35.7 min was 3. 31P NMR (d4-MeOH) δP
spectroscopy. H NMR (270.0 MHz, D2O, referenced to water
1
at 4.7 ppm) δ (ppm): 1.79 (1H, s, impurity not assigned), 3.24
(1H, t, J HH ) 8.91 Hz, H-5), 3.42-3.57 (2H, m, H-3/H-4), 3.65
(1H, dd appears as a br t, J HH ) 9.6 Hz, J HH ) 9.9 Hz, H-6),
3.99 (1H, dd appears as br t, J PH ∼ J 1/6 ∼ 10.4 Hz, H-3), 4.18
(1H, br s, H-2). m/z (ES-): 275 (M-2Na+ + H+ [16O,16O], 59.3),
276 (M-2Na+ + H+ [16O,17O] or M-2Na+ + H+ [16O,16O,113C]
27.6), 277 (M-2Na+ + H+ [16O,18O] or M-2Na+ + H+ [16O,17O,
113C] 6.1). Considering the natural 13C content, this mass
(ppm): 82.4 (s). H NMR (CDCl3) δH (ppm): 0.84 (3H, d, J HH
) 6.6 Hz, CHCH3), 1.2-1.75 (20H, m, 10 × CH2 of cyclohexy-
lidene), 2.59 (1H, d, J HH ) 3.0 Hz, 4-OH), 2.75 (3H, d, J PH
)
12.5 Hz, NCH3), 3.42 (1H, dd ∼ appears at t, J HH ∼ 10.1 Hz,
H-5), 3.6-3.8 (2H, m, CHCH3), 3.8-3.95 (1H, m, H-4), 4.0-
4.1 (2H, m, H-3/H-6), 4.56 (1H, t, J HH ) 4.6 Hz, H-2), 4.9-
5.05 (1H, m, H-1), 5.63 (1H, dd, appears as t, J HH ∼J PH ∼ 5.8
Hz, CHPh), 7.3-7.4 (5H, m, Ph). 13C NMR data (CDCl3,

1370 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6
Fauroux et al.
spectral data corresponds to 69% [16O,16O], 30% [16O,17O], and
1% [16O,18O].
Tris-HCl buffer containing 100 mM KCl and 3.3 mM MgCl2‚
6H2O, pH 9.13 (0.695-0.885 mL), to give a total volume of
1.00 mL. The mixture was incubated at 25 °C for 5 min, after
which inositol dehydrogenase (5 µL, final concentration 0.0486
units mL-1, 1.39 µg mL-1) was added. The absorbance at 340
nm was monitored with time.
Sta bility of IMP a se a t Va r iou s p H Va lu es. The stability
of IMPase was evaluated under the conditions required for the
continuous coupled enzymatic assay in the presence of a small
amount of 1,10-phenanthroline as a chelating agent. The
incubation media consisted of 50 mM Tris-HCl at different pH
values (7.06, 8.06, or 9.00), 100 mM KCl, 3 mM MgCl2‚6H2O,
20 mM â-NAD+, 0.55 mM cyclohexylammonium myo-inositol
1/3-phosphate, and IMPase (0.129 mg mL-1). These solutions
were incubated at 25 °C for 5 min, and the reaction was
initiated by addition of IMPase (30 µL, final concentration 3.81
µg mL-1). The absorbance at 340 nm was monitored for several
minutes after intervals of 30 and 75 min incubation. The
activity of IMPase was not detectably changed by incubation
at pH 7.06, 8.06, or 9.00.
Va lid a tion of th e Cou p led En zym e Assa y. The coupled
enzyme assay (Scheme 3) was arranged so that the reaction
catalyzed by IMPase was rate-determining, with inositol
dehydrogenase in excess; the observed velocity was indepen-
dent of inositol dehydrogenase and proportional to IMPase
under the optimized coupled assay conditions. To establish the
concentration of inositol dehydrogenase required that its
concentration was varied, with constant concentrations of myo-
inositol 1/3-phosphate, IMPase, and â-NAD+, until the rate of
the overall enzymatic coupled assay was constant. Assays were
completed at concentrations of myo-inositol 1/3-phosphate both
around the Km value (0.1 mM) and less than the Km value
(0.038 mM). The minimum quantity of inositol dehydrogenase
required was 0.14 units mL-1. When the concentration of
inositol dehydrogenase was doubled (0.284 units mL-1), the
rate of the enzymatic reaction remained the same for 0.1 and
0.038 mM myo-inositol 1/3-phosphate.
Disod iu m (D)-m yo-In osit ol 1/3-Th iop h osp h a t e. This
compound was prepared for kinetic studies using a method
similar to that described for the labeled material except that
a mixture of diastereomers 3 and 4 were used. 31P NMR (D2O)
δ (ppm): 44.8 (s, 95%, 1/3-phosphorylated compounds), 46.5
(s, 5%, 4/6-phosphorylated compounds). 1H NMR (D2O): δ 3.37
(1H, t, J HH ∼ 9.1 Hz, H-5), 3.55-3.7 (2H, m, H-1 and H-6),
3.78 (1H, t, J HH ) 9.7 Hz, H-4), 4.11 (1H, br t, J PH ) 10.4 Hz,
H-3), 4.31 (1H, br s, H-2). 13C NMR (D2O, referenced to
acetonitrile at 1.6 ppm): δ 70.3 (s, inositol-CH), 71.0 (s,
inositol-CH), 71.6 (d, J PC ∼ 3.7 Hz, inositol-CH), 71.8 (s, CH-
inositol), 73.9 (s, CH-inositol), 74.2 (d, J PC ∼ 4.9 Hz, CH-
inositol). IR (Nujol): νmax 3250 (OH) cm-1. m/z (ES-): 275 (M
- H, [12C], 100%), 276 (M - H, [13C], 8.1%).
Hyd r olysis of D-m yo-In ositol 1/3-Mon op h osp h a te w ith
IMP a se Mon itor ed by 31P NMR Sp ectr oscop y. A solution
of cyclohexylammonium D/L-myo-inositol 1-monophosphate
(75% approximately, 1.3 mg, 4 mM) in 50 mM Tris/HCl buffer
in D2O (0.8 mL, pD ∼ 8.5) containing 100 mM KCl and 7 mM
MgCl2‚6H2O in the presence or absence of 5mM DTE (0.1 mL)
gave 31P NMR (D2O) δ 4.8 (s, ∼77%, myo-inositol 1/3-mono-
phosphate) and 5.5 (s, ∼23%, myo-inositol 2-monophosphate).
IMPase (0.2 mL, concentration 0.54 mg mL-1, after ultrafil-
tration) was added, and the reaction was monitored by 31P
NMR spectroscopy over 40 min at 37 °C, which showed the
hydrolysis of myo-inositol 1/3-monophosphate to inorganic
phosphate: δ 3.4 (s, Pi), 5.5 (d, J PH ) 7.8 Hz, myo-inositol
2-monophosphate). In agreement with the literature, myo-
inositol 2-monophosphate is not a substrate for IMPase.5
Hyd r olysis of D-m yo-In ositol 1/3-Th iop h osp h a te w ith
IMP a se Mon itor ed by 31P NMR Sp ectr oscop y. The 31P
NMR spectrum of myo-inositol 1/3-thiophosphate (2 mg, 4.8
mM) in Tris/HCl buffer in D2O (0.8 mL, pD ∼ 8.5), KCl (100
mM), MgCl2‚6H2O (7 mM), and DTE (0.2 mL, 5mM) gave δ
(D2O) 3.4 (s, 6%, Pi), 45.1 (s, 94%). After IMPase (0.3 mL, 0.54
mg mL-1, after ultrafiltration to remove EDTA) was added,
the reaction was followed by 31P NMR spectroscopy at 37 °C,
which after 24 h, showed δ 3.4 (sharp s, Pi), 36-38 (br s, PSi),
45.1 (sharp s, myo-inositol 1/3-thiophosphate). To sharpen the
broad peak for PSi to estimate the percentage of each compo-
nent, the pD of the solution was adjusted to 11-12 with NaOH
(1 M, 3 drops), and the 31P NMR spectrum was recorded as
follows: δ 5.1 (s, 20%, Pi), 16.6 (s, 2.2%, intermediate in the
oxidative desulfurization of PSi), 32.2 (sharp, s, 66.7%, PSi),
44.8 (s, 11.1%, myo-inositol 1/3-thiophosphate).
Desu lfu r iza tion of In or ga n ic Th iop h osp h a te Mon i-
tor ed by 31P NMR Sp ectr oscop y. A solution of PSi (4.6 mg,
11.6 mM) in Tris/HCl buffer (1 mL) at pD 11-12 (adjusted
with 1 M NaOH) was left for 6 days at 22 °C after which
reaction was assessed by 31P NMR δ 5.7 (Pi, 6%), 16.6 (2-O3P-
S-S-PO32-, 5%), 32.2 (PSi, 89%).
A solution of PSi (10 mg, 25.2 mM) in Tris/HCl buffer (1
mL) at pD ∼ 8.24 was left for 6 days at 22 °C after which
reaction was assessed by 31P NMR δ 5.7 (Pi, 21%) and 32.2
(PSi, 79%).
A solution of PSi (10.7 mg, 27 mM) in Tris/HCl buffer (1
mL) at pD ∼ 8.24 was left for 3 days at 37 °C after which
reaction was assessed by 31P NMR δ 5.7 (Pi, 25%) and 32.2
(PSi, 75%).
In ositol Deh yd r ogen a se Assa y. The rate of oxidation of
myo-inositol was shown to be proportional to the concentration
of inositol dehydrogenase, which was varied at constant
concentrations of myo-inositol (0.39 M) and â-NAD+ (4.75 mM).
The observed reaction velocities (5.95 × 10-2 and 12.6 × 10-2
min-1) were proportional to the inositol dehydrogenase con-
centration (1.39 and 2.78 µg mL-1).
The rate of reaction was shown to be proportional to the
concentration of IMPase, which was varied at constant con-
centrations of myo-inositol 1/3-phosphate (0.1 mM), â-NAD+
(1.99 mM), and inositol dehydrogenase (0.14 units mL-1). The
observed velocities of the reaction (6.12 × 10-3, 1.1 × 10-2
,
1.3 × 10-2, and 1.77 × 10-2 ∆A min-1) were proportional to
the concentrations of IMPase (7.36, 13.9, 20.78, and 27.6 µg
mL-1), respectively. Under similar conditions, the velocities
(3.71 × 10-3, 5.49 × 10-3, and 7.95 × 10-3 ∆A min-1) of the
reaction were also proportional to the concentration of IMPase
(6.3, 12.55, and 18.73 µg mL-1) at a myo-inositol monophos-
phate concentration of 0.038 mM. Assays with D-myo-inositol
1/3-thiophosphate also needed a concentration of inositol
dehydrogenase of 0.14 units mL-1
.
Kin etic Stu d y of D-m yo-In ositol 3-P h osp h a te Usin g th e
In ositol Deh yd r ogen a se Cou p led Assa y. The cuvettes
contained 0.1 mL of â-NAD+ (5.22 mM), 20 µL of inositol
dehydrogenase (0.19 units mL-1), sodium D-myo-inositol 3-phos-
phate (final concentrations: 14.83, 11.92, 6.82, 3.41, 2.22, 1.02,
0.68, 0.34, 0.17, and 0.085 mM), and 50.4 mM Tris-HCl buffer
containing 102 mM NaCl and 3.37 mM MgCl2‚6H2O at pH
9.10. This mixture was incubated at 25 °C for 5 min, and the
reaction was initiated by addition of IMPase (5 µL, final
concentration 2.9 µg mL-1). The change in absorbance at 340
nm was monitored over several minutes.
Kin etic Stu d y of D-m yo-In ositol 1/3-Th iop h osp h a te
Usin g th e In ositol Deh yd r ogen a se Cou p led Assa y. The
cuvettes contained 0.1 mL of â-NAD+ (5 mM), 20 µL of inositol
dehydrogenase (0.19 units mL-1), myo-inositol 1/3-thiophos-
phate (final concentrations: 7.4, 6.56, 4.23, 2.63, 2.12, 1.27,
0.95, 0.85, 0.74, 0.64, 0.53, 0.42, 0.32, 0.21, or 0.11 mM), and
50.4 mM Tris-HCl buffer containing 102 mM NaCl and 3.4
mM MgCl2‚6H2O at pH 9.00. This mixture was incubated at
25 °C for 5 min, and then, the reaction was initiated by the
addition of IMPase (10 µL, final concentration 6.7 µg mL-1).
The change in absorbance at 340 nm was monitored over
several minutes.
To determine the effect of the concentration of myo-inositol
in the coupled assay at a constant concentration of inositol
dehydrogenase at pH 9.13, the assay cuvettes contained 0.1
mL of â-NAD+ (4.75 mM final concentration), myo-inositol
(0.039-0.784 mM from 0.01 to 0.2 mL of a stock solution of
3.92 M myo-inositol in 50 mM Tris-HCl buffer, pH 9.13), and

Reaction Catalyzed by myo-Inositol Monophosphatase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6 1371
Sta n d a r d Cu r ve of m yo-In ositol. After myo-inositol 1/3-
thiophosphate was hydrolyzed, the myo-inositol formed was
quantitated by means of a standard curve. The cuvettes
contained 0.1 mL of â-NAD+ (final concentration: 4.42 mM),
myo-inositol (0.00349-0.0349 mM from 0.01 to 0.1 mL of a
stock solution of 0.349 mM myo-inositol in Tris-HCl buffer (pH
8.90)), and 0.78-0.87 mL) Tris-HCl buffer (52.8 mM, pH 8.90)
containing 99.8 mM NaCl and 3.6 mM MgCl2‚6H2O to give a
total volume of 1.00 mL. This mixture was incubated at 25 °C
for 5 min after which time the reaction was initiated by the
addition of inositol dehydrogenase (20 µL, final concentration
0.107 92 units mL-1, 1.08 × 10-2 mg mL-1). The absorbance
at 340 nm was monitored with time until a steady state was
reached. The absorbance (0.006 23-0.0693) was proportional
to the myo-inositol concentration (0.003 49-0.0349 mM).
Stu d ies of P u ta tive In ter a ction s of IMP a se w ith High
Con cen tr a tion s of m yo-In ositol 1/3-Th iop h osp h a te by
ESI-MS. IMPase (50 µL of 0.436 mg mL-1) was added to 0.95
mL of 50 mM Tris-HCl containing 100 mM NaCl and 3.44 mM
MgCl2‚6H2O at pH 8.80. To half of this mixture (0.5 mL) was
added myo-inositol 1/3-thiophosphate (7.36 mg, 20.3 mM), and
the reaction was incubated at 37 °C for 90 min. The other half
of the mixture, containing only IMPase, was also incubated
at 37 °C for 90 min as a control. The samples were kept at 4
°C overnight, and the Tris-HCl buffer was exchanged with
ammonium acetate (5 mM) by centrifugation through an
Ultrafree-MC filter (Millipore). An aliquot (250 µL) of each
sample was kept at 4 °C, diluted with ammonium acetate (250
µL, 5 mM), and centrifuged until 50 µL remained. Ammonium
acetate was added (4 times) to return the volume to 250 µL.
The solutions were concentrated to 50 µL to give final
concentrations of IMPase of ∼0.46 mg mL-1. The two samples
were analyzed by ESI-MS.
inorganic thiophosphate from myo-inositol, inorganic phos-
phate, and myo-inositol 1/3-thiophosphate was performed by
anion exchange chromatography. Fractions 25-31 corre-
sponded to myo-inositol 1/3-thiophosphate, and fractions 38-
42 corresponded to inorganic thiophosphate.
Hyd r olysis of D-1-Sp-m yo-In ositol [17O]-Th iop h osp h a te
(6) in H218O Ca ta lyzed by IMP a se. First reaction: Tris-HCl
buffer (2 mL, 50.0 mM, pH 9.00) containing MgCl2‚6H2O (3.0
mM) and NaCl (100.4 mM) was lyophilyzed overnight. To this
was added H218O (400 µL), and lyophilyzation of the resulting
solution was repeated. H218O (1.8 mL) was added under argon,
followed by IMPase (16.34 mg previously lyophilyzed with 250
µL of H218O), and the solution was incubated at 25 °C under
argon for the duration of the experiment. D-1-Sp-myo-Inositol
[17O]-thiophosphate (10.5 mg, 0.033 mmol) was lyophilyzed
with H218O (250 µL), and the resulting solid was dissolved in
Tris-HCl buffer (pH 9.00) in H218O (210 µL) to give a 157 mM
substrate stock solution.
An aliquot of D-1-Sp-myo-inositol [17O]-thiophosphate stock
solution (25 µL, 2.14 mM) was added every 2 h (8 additions)
to the IMPase solution at 25 °C. One hour after the addition
of the first aliquot, the formation of myo-inositol was measured.
An aliquot of the enzymatic reaction mixture (0.01 mL) was
added to Tris-HCl buffer (0.87 mL, pH 9.00), â-NAD+ (0.1 mL,
4.89 mM), and inositol dehydrogenase (0.02 mL). A slope of
0.0067 ∆A min-1 at 340 nm corresponded to a concentration
of 1.14 mM of myo-inositol. The initial concentration of the
substrate was 2.14 mM; therefore, 1.00 mM starting material
remained. Aliquots were added every 2 h. After the 8 additions,
the reaction was put in the fridge to slow the hydrolysis and
then quenched by the addition of aqueous NaOH (4 drops, 1
M, pH 13.0). 31P NMR spectroscopy (109.2 MHz, D2O) gave δ
(ppm) 32.46 (s, 29.4%, inorganic thiophosphate), 45.00 (s, 58%,
myo-inositol 1-thiophosphate), 46.74 (s, 12.6%, myo-inositol
6-thiophosphate. The mixture was subjected to anion exchange
chromatography. Fractions 25-31 corresponded to D-1-Sp-myo-
inositol [17O]-thiophosphate, and fractions 37-43 corresponded
to inorganic [16O,17O,18O]-thiophosphate (observed by UV at
222 nm) and were analyzed by 31P NMR spectroscopy. Frac-
tions 37-43 were concentrated by evaporation and freeze-dried
overnight, dissolved in distilled water (3 mL), and diluted 10-
fold. An absorbance of 1.180 at 222 nm gave a concentration
of 2.8 mM of inorganic thiophosphate, which corresponds to
8.4 µmol (based on PSi having an ꢀ of 4200 M-1 cm-1 at 225
nm, pH 7).36
Second reaction: This was identical to the first reaction
except that the 160 mM stock solution of D-1-Sp-myo-inositol
[17O]-thiophosphate (10.8 mg, 0.034 mmol) in Tris-HCl buffer/
H218O (pH 9.00, 210 µL) was added in 15 × 12 µL aliquots
every 2 h to the buffer containing 14.4 mg of IMPase in H218O
at 25 °C. Ninety minutes after the addition of the first aliquot,
the inositol dehydrogenase assay showed the formation of 1.00
mM of myo-inositol. The initial concentration of the substrate
was 1.1 mM; therefore, 0.1 mM starting material remained.
Aliquots were added every 2 h to avoid accumulation of
starting material. After 28 h (14 aliquots), another inositol
dehydrogenase assay gave a myo-inositol concentration of
12.15 mM. The total concentration of added substrate was 15.4
mM; therefore, 3.25 mM of D-1-Sp-myo-inositol [17O]-thiophos-
phate remained. After all of the additions, the reaction was
put in the fridge to slow hydrolysis and stopped by the addition
of aqueous NaOH (3 drops, 1 M, pH 12.0). 31P NMR spectros-
copy (109.2 MHz, D2O) δ (ppm) gave 5.53 (s, 14.9%, inorganic
phosphate), 32.23 (s, 39.7%, inorganic thiophosphate), 44.78
(s, 35.4%, myo-inositol 1-thiophosphate), 46.51 (s, 10.0%, myo-
inositol 6-thiophosphate). Separation by anion exchange chro-
matography (as described for the first reaction) gave inorganic
[16O,17O,18O]-thiophosphate (14.8 µmol).
Desu lfu r iza tion of In or ga n ic Th iop h osp h a te Ca ta -
lyzed by IMP a se. Sodium thiophosphate tribasic dodecahy-
drate (0.020 63 g) was dissolved in Tris-HCl buffer (5 mL, 50.3
mM, pH 9.00, containing MgCl2‚6H2O (3 mM) and NaCl (100.5
mM)) to give a 10.44 mM stock solution. Into three NMR tubes
were added 0.6 mL of the above stock solution of PSi and D2O
(0.1 mL). The reactions were initiated by the addition of
IMPase to give final enzyme concentrations of 0.119, 0.235,
and 0.343 mg mL-1 corresponding to 3.97 × 10-6, 7.82 × 10-6
,
and 11.43 × 10-6 M. The samples were incubated at 25 °C,
and the 31P NMR spectra were recorded after 24, 48, and 72
h. The area of the peaks of PSi remaining and Pi formed were
analyzed by weighing the paper cut-outs of the peaks. Des-
ulfurization with time (h) is shown in Table 1.
La r ge-Sca le En zym a tic Rea ction w ith Un la beled m yo-
In ositol 1/3-Th iop h osp h a te. IMPase (10.72 mg) was added
to Tris-HCl buffer (1.42 mL, 49.8 mM, pH 9.00) containing
MgCl2‚6H2O (3 mM) and NaCl (100 mM). myo-Inositol 1/3-
thiophosphate (12.19 mg, stock solution: 476.0 mM) was
dissolved in Tris-HCl buffer (80 µL, pH 9.00). An aliquot of
myo-inositol 1/3-thiophosphate solution (10 µL, final concen-
tration: 3.33 mM) was added every 2 h (for 6 additions) to
the solution containing IMPase (7.15 mg mL-1) at 25 °C. After
the first aliquot was added, the enzymatic reaction mixture
(0.01 mL) was added to Tris-HCl buffer (0.87 mL, pH 9.00),
â-NAD+ (0.1 mL, final concentration 4.68 mM), and inositol
dehydrogenase (0.02 mL). A slope of 0.0098 ∆A min-1 at 340
nm corresponded to a concentration of 2.5 mM of myo-inositol.
The initial concentration of the substrate was 3.33 mM;
therefore, 0.83 mM of starting material remained. After 11 h
and 6 aliquots was added, the myo-inositol assay (0.002 mL
of aliquot, 0.878 mL of Tris-HCl, 0.1 mL of â-NAD+, and 0.02
mL of inositol dehydrogenase) gave a slope of 0.0096 ∆A min-1
corresponding to a concentration of 12.23 mM. The concentra-
tion of the sample was 19.98 mM myo-inositol, with 7.75 mM
starting material remaining.
Altogether, 23 µmol of inorganic [16O,17O,18O]-thiophosphate
was isolated by anion exchange chromatography by combining
both samples. This was analyzed using the method of Arnold,
Bethell, and Lowe.37
After 12 h, the reaction was stopped by the addition of
NaOH solution (4 drops, 1 M), and 31P NMR spectroscopy
(109.2 MHz, D2O) gave δ (ppm) 4.76 (s, 19.6%, Pi), 31.45 (s,
57.7%, PSi), 44.02 (s, 19.9%, myo-inositol 1/3-thiophosphate),
45.74 (s, 2.8%, myo-inositol 4/6-thiophosphate). Separation of
(S)-1-P h en yleth an ol-2-S-[16O,17O,18O]-th ioph osph ate. The
inorganic [16O,17O,18O]-thiophosphate (approximately 23 µmol)

1372 J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6
Fauroux et al.
(14) Schulz, J .; Wilkie, J .; Lightfoot, P.; Rutherford, T.; Gani, D.
Synthesis and Properties of Mechanism-based Inhibitors and
Probes for Inositol Monophosphatase derived from 6-O-(2′-
Hydroxylethyl)-(1R,2R,4R,6R)-cyclohexane-1,2,4,6-tetraol. J .
Chem. Soc., Chem. Commun. 1995, 2353-2356.
(15) Cole, A. G.; Gani, D. “Active” Conformation of the Inositol
Monophosphatase Substrate, Adenosine 2′-Phosphate: Role of
the Ribofuranosyl O-Atoms in Chelating a Second Mg2+ ion. J .
Chem. Soc., Perkin Trans. 1 1995, 2685-2694.
(16) Cole, A. G.; Wilkie, J .; Gani, D. Probes for the Position and
Mechanism Role of the Second “Catalytic” Magnesium Ion in
the Inositol Monophosphatase Reaction. J . Chem. Soc., Perkin
Trans. 1 1995, 2695-2707.
(17) Wilkie, J .; Gani, D. Comparison of Inline and Noninline As-
sociative and Dissociative Reaction Pathways for Model Reac-
tions of Phosphate Monoester Hydrolysis. J . Chem. Soc., Perkin
Trans. 2 1996, 783-787.
(18) Gani, D.; Wilkie, J . Stereochemistry, Mechanistic, and Structural
Features of Enzyme-catalysed Phosphate Monoester Hydrolysis.
J . Chem. Soc. Rev. 1995, 55-63.
(19) Part of the study has been published as a communication.
Fauroux, C. M.-J .; Lee, M.; Cullis, P. M.; Douglas, K. T.;
Freeman, S.; Gore, M. G. Inversion of Configuration during the
Hydrolysis of D-1-Sp-myo-Inositol [17O]Thiophosphate Catalyzed
by myo-Inositol Monophosphatase. J . Am. Chem. Soc. 1999, 121,
8385-8386.
(20) Spiers, I. D.; Schwalbe, C. H.; Blake, A. J .; Solomons, K. R. H.;
Freeman, S. Structure of (()-1,2,4,5-Di-O-cyclohexylidene myo-
Inositol and Synthesis of myo-Inositol 3-Phosphate via its
Phosphorylation with (2R,4R,5R)-2-Chloro-3,4-dimethyl-5-phen-
yl-1,3,2-oxazaphospholidin-2-one. Carbohydr. Res. 1997, 302,
43-51.
isolated from the two enzymatic reactions was converted to
its trioctylammonium salt and dried by evaporation of dry
DMF. The residue was dissolved in DMF (0.3 mL) and treated
with excess (S)-2-iodo-1-phenylethanol (14 mg, 56 µmol). After
ca. 26 h, the reaction was shown to be complete by 31P NMR,
and the product was isolated by ion exchange chromatography
on (diethylamino)ethyl (DEAE) Sephadex (A25) eluting with
a gradient of TEAB buffer (25-250 mM; pH 7.8). (S)-1-
Phenylethanol-2-S-[16O,17O,18O]-thiophosphate (11 µmol based
on integration of 1H NMR signals with respect to dioxan added
as an internal integration standard) was isolated and cyclized
directly without further characterization.
cis- a n d tr a n s-4-(S)-P h en yl-2-m eth oxy-2-oxo-1,3,2-th ia -
oxa p h osp h ola n e. The tributylammonium salt of (S)-1-Phen-
ylethanol-2-S-[16O,17O, 18O]-thiophosphate (11 µmol) was dried
by evaporation of dry DMF and then reacted with diphen-
ylphosphorochloridate (13.2 µmol) and tributylamine (11 µmol)
in dry DMF (0.3 mL). After 15 min, the cyclization reaction
was quenched by the addition of TEAB buffer (2 mL; 100 mM
solution, pH 7.8), and the solution was evaporated to dryness
and purified by ion exchange chromatography on DEAE
Sephadex (A25) eluting with a gradient of triethylammonium
biocarbonate buffer (5-120 mM; pH 7.8). The triethylammo-
nium salt of 4-(S)-phenyl-2,2-dioxo-1,3,2-thiaoxaphospholane
was converted directly to the pyridinium salt using Dowex 50W
(pyridinium form), and the salt was dried by coevaporation of
dry CH3CN. The residue was dissolved in CD3CN (0.5 mL) and
treated with excess distilled diazomethane solution (dried over
KOH pellets). The excess diazomethane was evaporated with
a stream of dry nitrogen, and the solution was filtered and
analyzed by high field 31P NMR spectroscopy (Figure 3).
(21) Knowles, J . R. Enzyme-Catalyzed Phosphoryl Transfer Reac-
tions. Annu. Rev. Biochem. 1980, 49, 877-919.
(22) Cullis, P. M.; Iagrossi, A.; Rous, A. J . Thiophosphoryl-Transfer
Reactions: A General Synthesis and Configurational Analysis
of O-Substituted [16O,18O]Thiophosphates. J . Am. Chem. Soc.
1986, 108, 8, 7869-7870.
(23) Frey, P. A. Chiral Phosphorothioates: Stereochemical Analysis
of Enzymatic Substitution at Phosphorus. Adv. Enzymol. Relat.
Areas Mol. Biol. 1989, 62, 119-201.
Ack n ow led gm en t. We thank Mrs. Nicola Muir for
protein purification and Dr. Gerald Griffiths for high-
field NMR spectroscopy.
Refer en ces
(24) Dreef, C. E.; Tulnman, R. J .; Lefeber, A. W. M. Synthesis of
Racemic 3-Methylphosphonate Analogues of myo-Inositol 3,4-
Bis- and 1,3,4-Trisphosphate. Tetrahedron 1991, 47, 4709-4722.
(25) Cooper, D. B.; Hall, C. R.; Harrison, J . M.; Inch, T. D. 1,3,2-
Oxazaphospholidines from (-)-Ephedrine. Intermediates for the
Stereospecific Synthesis of Optically Active Dialkyl Alkylphospho-
nothioates and -selenoates, Trialkyl Phosphoro-thioates and
-selenoates, Dialkyl Methylphosphonates, and Trialkyl Phos-
phates. J . Chem. Soc., Perkin Trans 1 1977, 1969-1980.
(26) (a) Vacca, J . P.; deSolms, S. J .; Huff, J . R.; Billington, D. C.;
Baker, R.; Kulagowski, J . J .; Mawer, I. M. The Total Synthesis
of myo-Inositol Polyphosphates. Tetrahedron 1989, 45, 5679-
5701. (b) Vacca, J . P.; deSolms, S. J .; Huff, J . R.; Billington, D.
C.; Baker, R.; Kulagowski, J . J .; Mawer, I. M. Tetrahedron 1991,
47, 907.
(27) Diehl, R. E.; Whiting, P.; Potter, J .; Gee, N.; Ragan, C. I.;
Linemeyer, D.; Schoepfer, R.; Bennet, C.; Dixon, A. F. Cloning
and Expression of Bovine Brain Inositol Monophosphatase. J .
Biol. Chem. 1990, 265, 5946-5949.
(28) Dittmer, D. C.; Ramsay, O. B. Reactivity of Thiophosphates. I.
Hydrolysis of Phosphorothioic Acid. J . Org. Chem. 1963, 28,
8-1272.
(29) Leatherbarrow, R. J . Grafit, version 3.0; Erithacus Software
Ltd.: Staines, U.K., 1992; Chapter 32, pp 243-262.
(30) Van Etten, R. L.; Saini, M. S. pH-Dependent Oxygen Exchange
of Phosphate and Hydrolysis (of Sulfur) from Thiophosphate
Catalyzed by Acid Phosphatases. Arch. Biochem. Biophys. 1982,
219, 155-162.
(31) Chlebowski, J . F.; Coleman, J . E. Mechanisms of Hydrolysis of
O-Phosphorothioates and Inorganic Thiophosphate by Escheri-
chia coli Alkaline Phosphatase. J . Biol. Chem. 1974, 249, 7192-
7202.
(32) Kwok, F.; Lo, S. C. L. Development of a Continuous Coupled
Enzymatic Assay for myo-Inositol Monophosphatase. J . Biochem.
Biophys. Methods 1994, 29, 173-178.
(33) Weissbach, A. The Enzymic Determination of myo-Inositol.
Biochim. Biophys. Acta 1958, 27, 608-611.
(1) Atack, J . R.; Broughton, H. B.; Pollack, S. J . Inositol Monophos-
phatase - A Putative Target for Li+ in the Treatment of Bipolar
Disorder. TINS 1995, 18, 343-349.
(2) Fauroux, C. M. J .; Freeman, S. Review Article: Inhibitors of
Inositol Monophosphatase. J . Enzyme Inhib. 1999, 14, 97-108.
(3) Gee, N. S.; Ragan, C. I.; Watling, K. J .; Aspley, S.; J ackson, R.
G.; Reid, G. G.; Gani, D.; Shute, J . K. The Purification and
Properties of myo-Inositol Monophosphatase from Bovine Brain.
Biochem. J . 1988, 249, 883-889.
(4) McAllister, G.; Whiting, P.; Hammond, E. A.; Knowles, M. G.;
Atack, J . R.; Bailey, F. J .; Maigetter, R.; Ragan, C. I. cDNA
Cloning of Human and Rat Brain myo-Inositol Monophos-
phatase. Biochem. J . 1992, 284, 749-754.
(5) Majerus, P. W. Inositol Phosphate Biochemistry. Annu. Rev.
Biochem. 1992, 61, 225-250.
(6) Bone, R.; Springer, J . P.; Atack, J . R. Structure of Inositol
Monophosphatase, the Putative Target of Lithium Therapy.
Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 10031-10035.
(7) Bone, R.; Frank, L.; Springer, J . P.; Pollack, S. J .; Osbourne, S.;
Atack, J . R.; Knowles, M. R.; McAllister, G.; Ragan, C. I.;
Broughton, H. B.; Baker, R.; Fletcher, S. R. Structural Analysis
of Inositol Monophosphatase Complexes with Substrates. Bio-
chemistry 1994, 33, 9460-9467.
(8) Bone, R.; Frank, L.; Springer, J . P.; Atack, J . R. Structural
Studies of Metal Binding by Inositol Monophosphatase: Evi-
dence for Two-Metal Ion Catalysis. Biochemistry 1994, 33, 9468-
9476.
(9) Pollack, S. J .; Atack, J . R.; Knowles, M. R.; McAllister, G.; Ragan,
C. I.; Baker, R.; Fletcher, S. R.; Iversen, L. L.; Broughton, H. B.
Mechanism of Inositol Monophosphatase, the Putative Target
of Lithium Therapy. Proc. Natl. Acad. Sci. U.S.A. 1994, 91,
5766-5770.
(10) Leech, A. P.; Baker, G. R.; Shute, J . K.; Cohen, M. A.; Gani, D.
Chemical and Kinetic Mechanism of the Inositol Monophos-
phatase Reaction and its Inhibition by Li+. Eur. J . Biochem.
1993, 212, 693-704.
(11) Baker, G. R.; Gani, D. Inositol Dependent Phosphate-Oxygen
Ligand Exchange Catalysed by Inositol Monophosphatase. Bioorg.
Med. Chem. Lett. 1991, 1, 193-196.
(34) Larner, J .; J ackson, W. T.; Graves, D. J .; Stamer, J . R. Inositol
Dehydrogenase from Aerobacter aerogenes. Arch. Biochem. Bio-
phys. 1956, 60, 352-363.
(35) Baker, G. R.; Billington, D. C.; Gani, D. Synthesis of D- and
L-myo-Inositol 1-Phosphorothioate, Substrates for Inositol Mono-
phosphatase. Tetrahedron 1991, 47, 3895-3908.
(12) Shute, J . K.; Baker, R.; Billington, D. C.; Gani, D. Mechanism
of the myo-Inositol Phosphatase Reaction. J . Chem. Soc., Chem.
Commun. 1988, 626-628.
(13) Ganzhorn, A. J .; Chanal, M.-C. Kinetic Studies with myo-Inositol
Monophosphatase from Bovine Brain. Biochemistry 1990, 29,
6065-6071.

Reaction Catalyzed by myo-Inositol Monophosphatase
J ournal of Medicinal Chemistry, 2002, Vol. 45, No. 6 1373
(36) Tsai, M.-D. Stereochemistry of the Hydrolysis of Adenosine 5′-
Thiophosphate Catalyzed by Venom 5′-Nucleotidase. Biochem-
istry 1980, 19, 5310-5316.
(42) Still, W. C.; Khan, M.; Mitra, A. Rapid Chromatographic
Technique for Preparative Separations with Moderate Resolu-
tion. J . Org. Chem. 1978, 43, 2923-2925.
(43) Itaya, K.; Ui, M. A New Micromethod for the Colorimetric
Determination of Inorganic Phosphate. Clin. Chim. Acta 1966,
14, 361-366.
(44) Gore, M. G.; Greasley, P. J .; Ragan, C. I. Bovine Inositol
Monophosphatase: Development of a Continuous Fluorescence
Assay of Enzyme Activity. J . Biochem. Biophys. Methods 1992,
25, 55-60.
(45) Bradford, M. M. A Rapid and Sensitive Method for the Quan-
titation of Microgram Quantities of Protein Utilizing the Prin-
ciple of Protein-Dye Binding. Anal. Biochem. 1976, 72, 248-
254.
(37) Arnold, J . R. P.; Bethell, R. C.; Lowe, G. Chiral Inorganic
[
16O,17O,18O]Thiophosphate: Synthesis and Stereochemical Analy-
sis. Bioorg. Chem. 1987, 15, 250-261.
(38) Wilkie, J .; Cole, A. G.; Gani, D. 3-Dimensional Interactions
Between Inositol Monophosphatase and its Substrates, Inhibi-
tors and Metal Ion Cofactors. J . Chem. Soc., Perkin Trans. 1
1995, 2709-2727.
(39) Atack, J . R.; Broughton, H. B.; Pollack, S. J . Structure and
Mechanism of Inositol Monophosphatase. FEBS Lett. 1995, 361,
1-7.
(40) Miller, D. J .; Beaton, M. W.; Wilkie, J .; Gani, D. The 6-OH Group
of D-Inositol 1-Phosphate Serves as an H-Bond Donor in the
Catalytic Hydrolysis of the Phosphate Ester by Inositol Mono-
phosphatase. Chem. Bio. Chem. 2000, 1, 262-271.
(41) Domanico, P. L.; Rahil, J . F.; Benkovic, S. J . Unambiguous
Stereochemical Course of Rabbit Liver Fructose Bisphosphatase
Hydrolysis. Biochemistry 1985, 24, 1623-1628.
(46) Fife, T. H.; Bruice, T. C. The Temperature Dependence of the
∆pD Correction for the Use of the Glass Electrode in D2O. J .
Phys. Chem. 1961, 65, 1079-1080.
J M011056M