Product-like inhibitors of inositol monophosphatase

Article abstract of DOI:10.1016/S0040-4039(03)01878-1

A series of product-like inhibitors of inositol monophosphatase have been prepared and tested for activity in vitro as possible leads for treatment of bipolar disorders. Compounds possessing a 6-alkyloxy side chain were inhibitors but less efficacious tha

Full text of DOI:10.1016/S0040-4039(03)01878-1

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 7677–7679
Product-like inhibitors of inositol monophosphatase
M. Bashir-Uddin Surfraz, David J. Miller, David Gani and Rudolf K. Allemann*
Department of Chemical Sciences, The University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
Received 23 June 2003; accepted 1 August 2003
Abstract—A series of product-like inhibitors of inositol monophosphatase have been prepared and tested for activity in vitro as
possible leads for treatment of bipolar disorders. Compounds possessing a 6-alkyloxy side chain were inhibitors but less efficacious
than those possessing a 6-aminoalkyl side chain. These new structures show promise as inhibitors possessing the bioavailability
and characteristics necessary for drug development.
© 2003 Elsevier Ltd. All rights reserved.
Inositol monophosphatase (IMPase, EC 3.1.3.25) plays
a crucial role in the recycling of metabolites of the
important second-messenger molecule inositol-1,4,5-
trisphosphate. The enzyme catalyses the cleavage of the
phosphate group from various isomers of inositol
monophosphate to generate myo-inositol. This is used
to resynthesise phosphatidylinositol-4,5-bisphosphate,
the membrane-bound lipid that is the direct precursor
to the second messenger.1–4 Inhibition of IMPase in
vivo squeezes the supply of myo-inositol available for
recycling, limiting the cellular production of inositol-
1,4,5-trisphosphate and hence cell response to external
stimuli. It is believed that this is the therapeutic mode
of action of lithium ion in the treatment of bipolar
disorders.5,6 Lithium treatment unfortunately leads to
many toxic side effects and so alternative inhibitors of
IMPase are attractive therapeutic targets.6
Work performed both within our laboratories and else-
where has led to the design and synthesis of numerous
IMPase inhibitors, some with impressive potency yet to
date, none have the required bioavailability of a drug-
candidate.7,8 Most members of the current range of
inhibitors possess a phosphate monoester functionality
and it seems that the negative charge on this group
renders the molecules too hydrophilic to cross the
lipophilic barrier between the blood and the brain
where suppression of IMPase activity must occur.7 It
seemed to us, therefore that a new approach to this
problem was needed.
Figure 1. A Representation of the active site of IMPase after cleavage of the inositolꢀphosphate bond and before release of
inositol. B Proposed product-like inhibitor bound to the enzyme–phosphate complex. X=O or NH, R=alkyl or aryl.
* Corresponding author. Tel.: +44-121-4359; fax: +44-121-4446; e-mail: r.k.allemann@bham.ac.uk
0040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01878-1

7678
M. Bashir-Uddin Surfraz et al. / Tetrahedron Letters 44 (2003) 7677–7679
Investigation of the kinetics of the reactions catalysed
by IMPase has shown that after the enzyme has cleaved
-inositol-1-phosphate, the first molecule to leave the
isolated in excellent yields. The benzyl groups proved
resistant to hydrogenolysis in the presence of an amine
side-chain but after some experimentation it was found
that treatment of 2d–h with trimethylsilyl bromide in
chloroform at 50°C effected their cleavage in good
yield.13
D
active site is myo-inositol followed by inorganic phos-
phate (Fig. 1) and it has been shown that Pi exhibits
product inhibition (Ki=0.3 mM at pH 8.0).5 It has also
been demonstrated that at therapeutic concentrations,
lithium ion binds to the enzyme–Pi complex preventing
the release of Pi and therefore the binding of another
substrate molecule.5 Moreover, it is estimated that the
concentration of Pi in the brain is ꢀ3 mM and so at
any given time >90% of the enzyme must be Pi bound
in vivo.9 It was therefore reasoned that investigation of
molecules that mimic the action of lithium by binding
to the enzyme–Pi complex might lead to a new series of
IMPase inhibitors. The proposed series of inhibitors
were designed as mimics of the product myo-inositol
(Fig. 1); such inhibitors would lack the usual phosphate
group of IMPase inhibitors and so should be more
lipophilic, hopefully improving the chances of transport
across the blood-brain barrier. myo-Inositol itself dis-
plays poor product inhibition (Ki=400 mM),5 and so to
improve binding to the enzyme–Pi complex structural
features already known to enhance inhibitor binding to
IMPase were included in the design.8 The hydroxyl
groups at positions 3 and 5 on the inositol ring would
be deleted and on position 6 a lipophilic chain would
replace the hydroxyl group of inositol. IMPase pos-
sesses a lipophilic pocket bounded by Val40 and Leu42
which this structural feature seeks to exploit.10
The results of initial testing of these compounds as
inhibitors of IMPase are shown in Table 1. The assay
procedure used was a modification of the procedure of
Gee et al. using commercially available IMPase.14
Assays (50 ml) were initiated by addition of enzyme
solution (5 ml, 0.02 units) and contained 100 mM
Tris–HCl, 50 mM KCl, 3 mM MgCl2 and (2-3H)-DL
-
myo-inositol-1-phosphate
(500
dpm/nmol)
plus
inhibitor at pH 8.0. Assays were quenched after 20 min
by addition of 1 M NaOH (5 ml). The solution was then
passed down a small pre-equilibrated column of
DOWEX 1X2-400 anion exchange resin (OH form, 200
mg) to remove unreacted substrate and the filtrate plus
washings were emulsified with scintillation cocktail and
then the radioactivity of each sample was determined
by scintillation counting.
6-Propyloxy derivative 3a was the first compound
tested since its benzylated precursor had already been
prepared within our laboratories.11 It proved an ineffec-
tive inhibitor of IMPase, significantly retarding the
turnover of inositol-1-phosphate only as it approached
its saturating concentration in the assay buffer. Never-
theless, inhibition of IMPase with an IC50 of approxi-
mately 150 mM represented an improvement upon the
efficacy of myo-inositol as a product inhibitor. The
6-hexyloxy derivative 3b was a more effective inhibitor
of IMPase but also was significantly more lipophilic
The synthesis of the first series of inhibitors of this type
is illustrated in Figure 2. Epoxide 1 was used to prepare
all of the required compounds.11 Regioselective ring
opening of 1 to the 6-alkyloxy derivatives 2a–c was
achieved in good to excellent yield by heating the
epoxide with excess alcohol in 1,2-dichloroethane in the
presence of catalytic ytterbium(III) triflate.11,12 Simi-
larly, amines 2d–h were prepared in good yields by
heating 1 with the parent primary amine and ytterbium-
(III) triflate in a 3:1 mixture of toluene and THF.
1,2-Dichloroethane was not a suitable solvent in this
instance as it was reactive towards the amines under the
reaction conditions required to open the epoxide; it was
also found necessary when opening the epoxide with
amines to use 1 equivalent of ytterbium(III) triflate. To
remove the benzyl protecting groups from compounds
2a–c catalytic hydrogenolysis using hydrogen over 5%
palladium on activated carbon was effective, 3a–c being
Table 1.
Compound
Side chain
IC50 (mM)
myo-Inositol
–
400 (Ki)
3a
3b
3c
3d
3e
3f
OC3H7
OC6H13
OC4H8O-[(2-OH)C6H4]
NHC4H9
NHC6H13
]150
]10
4
0.5
0.5
4
NHC8H17
3g
3h
NH(CH2)2C6H5
NH(CH2)4C6H5
6
10
Figure 2. Reagents and conditions: (i) ROH, cat. Yb(OTf)3, DCE, reflux, 72–99%; (ii) RNH2, 3:1 toluene–THF, Tb(OTf), reflux,
76–96%; (iii) H2, 5% Pd–C, MeOH, 91% quantitative; (iv) Me3SiBr, CHCl3, 50°C, 54–88%.

M. Bashir-Uddin Surfraz et al. / Tetrahedron Letters 44 (2003) 7677–7679
7679
than 3a. It appeared to have an IC50 value of around 10
Acknowledgements
mM but it was not possible to measure this precisely as
above this concentration it too was insoluble in the assay
buffer. Compound 3c was expected to show much
promise as a product-like inhibitor of IMPase. Phospho-
rylation of the hydroxyl group at position 1 on the
inositol ring gives the most potent inhibitor of IMPase
known (IC50=40 nM),7 and it was hoped that this
compound would therefore bind effectively to the
enzyme even in the absence of the phosphate group. In
the event it proved the most effective of the 6-alkyloxy
derivatives tested in this work but still gave a disappoint-
ing IC50 of only 4 mM. Removal of the phosphate group
from position 1 of the inositol ring effectively retarded
the binding to the enzyme by six orders of magnitude.
We thank the EPSRC and the BBSRC for financial
support.
References
1. Berridge, M. J.; Irvine, R. F. Nature 1989, 341, 197–205.
2. Abdel-Latif, A. A. Pharmacol. Rev. 1986, 38, 227–272.
3. Gani, D.; Downes, C. P.; Batty, I.; Bramham, J. Biochim.
Biophys. Acta 1993, 1177, 253–269.
4. Potter, B. V. L.; Lampe, D. Angew. Chem., Int. Ed. Engl.
1995, 34, 1933–1972.
5. Leech, A. P.; Baker, G. R.; Shute, J. K.; Cohen, M. A.;
Gani, D. Eur. J. Biochem. 1993, 212, 693–704.
6. Baraban, J. M. Proc. Nat. Acad. Sci. USA 1994, 91,
5738–5739.
7. Faroux, C. M. J.; Freeman, S. J. Enzyme Inhibition 1999,
14, 97–108.
8. Schulz, J.; Wilkie, J.; Beaton, M. W.; Miller, D. J.; Gani,
D. Biochem. Soc. Trans. 1998, 26, 315–322.
9. Voet, D.; Voet, J. G. Biochemistry; Wiley: New York,
1990.
Amines 3d–h proved more interesting when tested as
inhibitors of IMPase. Of particular note is the fact that
the structure–activity relationship with respect to the
length of the side chain appears reversed when compared
with the 6-alkyloxy derivatives. Long lipophilic side
chains possessing aromatic functionality (compounds 3g
and 3h) were the least effective of the 6-aminoalkyl
derivatives and those possessing relatively short simple
alkyl chains were the most effective. Indeed, the 6-butyl-
amino derivative 3d and the 6-hexyl-amino derivative 3e
were effective inhibitors at sub-millimolar concentrations
and so represent an improvement of three orders of
magnitude upon the product inhibition of myo-inositol
itself.
10. Cole, A. G.; Wilkie, J.; Gani, D. J. Chem. Soc., Perkin
Trans. 1 1995, 2695–2707.
11. Schulz, J.; Beaton, M. W.; Gani, D. J. Chem. Soc.,
Perkin Trans. 1 2000, 943–954.
12. Beaton, M.; Gani, D. Tetrahedron Lett. 1998, 39, 8544–
8552.
13. Jung, M. E.; Lyster, M. A. J. Org. Chem. 1977, 42,
3761–3764.
14. Gee, N. S.; Ragan, C. I.; Watling, K. J.; Aspley, S.;
Jackson, R. G.; Reid, G. G.; Gani, D.; Shute, J. K.
Biochem. J. 1988, 249, 883–889.
The 6-aminoalkyl derivatives described above effectively
validate our new approach to developing inhibitors of
IMPase. Further analogues of this type are currently
being investigated and a more detailed analysis of their
mode of action will be published in due course.