Synthesis and Kinetic Evaluation of Inhibitors of the Phosphatidylinositol-Specific Phospholipase C from Bacillus cereus

Article abstract of DOI:10.1021/jo960850q

Substrate analogues of phosphatidylinositol (1) were synthesized and evaluated as potential inhibitors of the bacterial phosphatidylinositol-specific phospholipase C (PI-PLC) from Bacillus cereus.The chiral analogues of the water-soluble phospholipid substrate 5 were designed to probe the effects of varying the inositol C-2 hydroxyl group, which generally believed to serve as the nucleophile in the first step of the hydrolysis of phosphatidylinositols by PI-PLC.In the analogues 6-9, the C-2 hydroxyl group on the inositol ring of the phosphatidylinositol derivatives was rationally altered in several ways.Inversion of the stereochemistry at C-2 of the inositol ring led to the scyllo derivative 6.The inositol C-2 hydroxy group was replaced with inversion by a fluorine to produce the scyllo-fluoro inositol 7 and with a hydrogen atom to furnish the 2-deoxy compound 8.The C-2 hydroxyl group was O-methylated to prepare the methoxy derivative 9.The natural inositol configuration at C-2 was retained in the nonhydrolyzable phosphorodithioate analogue 10.The inhibition of PI-PLC by each of these analogues was then analyzed in a continuous assay using D-myo-inositol 1-(4-nitrophenyl phosphate) (25) as a chromogenic substrate.The kinetic parameters for each of these phosphatidylinositol derivatives were determined, and each was found to be a competitive inhibitor with K_i's as follows: 6, 0.2 mM; 10, 0.6 mM; 8, 2.6 mM; 9, 6.6 mM; and 7, 8.8 mM.This study further establishes that the hydrolysis of phosphatidylinositol analogues by bacterial PI-PLC requires not only the presence of a C-2 hydroxyl group on the inositol ring, but the stereochemistry at this position must also correspond to the natural myo-configuration.For future inhibitor design, it is perhaps noteworthy that the best inhibitors 6 and 10 each possess a hydroxyl group at the 2-position.Several of the inhibitors identified in this study are now being used to obtain crystallographic information for an enzyme-inhibitor complex to gain further insights regarding the mechanism of hydrolysis of phosphatidylinositides by this PI-PLC.

Full text of DOI:10.1021/jo960850q

8016
J . Org. Chem. 1996, 61, 8016-8023
Syn th esis a n d Kin etic Eva lu a tion of In h ibitor s of th e
P h osp h a tid ylin ositol-Sp ecific P h osp h olip a se C fr om Ba cillu s
cer eu s
Stephen F. Martin* and Allan S. Wagman
Department of Chemistry and Biochemistry, The University of Texas, Austin, Texas 78712
Received May 8, 1996X
Substrate analogues of phosphatidylinositol (1) were synthesized and evaluated as potential
inhibitors of the bacterial phosphatidylinositol-specific phospholipase C (PI-PLC) from Bacillus
cereus. The chiral analogues of the water-soluble phospholipid substrate 5 were designed to probe
the effects of varying the inositol C-2 hydroxyl group, which is generally believed to serve as the
nucleophile in the first step of the hydrolysis of phosphatidylinositols by PI-PLC. In the analogues
6-9, the C-2 hydroxyl group on the inositol ring of the phosphatidylinositol derivatives was
rationally altered in several ways. Inversion of the stereochemistry at C-2 of the inositol ring led
to the scyllo derivative 6. The inositol C-2 hydroxy group was replaced with inversion by a fluorine
to produce the scyllo-fluoro inositol 7 and with a hydrogen atom to furnish the 2-deoxy compound
8. The C-2 hydroxyl group was O-methylated to prepare the methoxy derivative 9. The natural
inositol configuration at C-2 was retained in the nonhydrolyzable phosphorodithioate analogue 10.
The inhibition of PI-PLC by each of these analogues was then analyzed in a continuous assay using
D-myo-inositol 1-(4-nitrophenyl phosphate) (25) as a chromogenic substrate. The kinetic parameters
for each of these phosphatidylinositol derivatives were determined, and each was found to be a
competitive inhibitor with Ki’s as follows: 6, 0.2 mM; 10, 0.6 mM; 8, 2.6 mM; 9, 6.6 mM; and 7, 8.8
mM. This study further establishes that the hydrolysis of phosphatidylinositol analogues by
bacterial PI-PLC requires not only the presence of a C-2 hydroxyl group on the inositol ring, but
the stereochemistry at this position must also correspond to the natural myo-configuration. For
future inhibitor design, it is perhaps noteworthy that the best inhibitors 6 and 10 each possess a
hydroxyl group at the C-2 position. Several of the inhibitors identified in this study are now being
used to obtain crystallographic information for an enzyme-inhibitor complex to gain further insights
regarding the mechanism of hydrolysis of phosphatidylinositides by this PI-PLC.
Sch em e 1
In tr od u ction
Unraveling the intricate regulatory pathways of cel-
lular signal transduction that involves the receptor-
mediated turnover of inositol phospholipids by mamma-
lian enzymes of the phosphatidylinositol-specific phos-
pholipase C (PI-PLC) family has recently been the subject
of unrivaled interest.1,2 The intense attention accorded
this class of enzymes was stimulated by the exciting
discovery that when certain extracellular signaling mol-
ecules including hormones, neurotransmitters, and some
growth factors bind to their G-protein linked surface
receptors, a calcium-dependent PI-PLC is activated. The
consequent hydrolysis of phosphatidylinositol 4,5-bis-
phosphate (PIP2) (1, R2 ) PO32-) by this enzyme produces
1,2-diacylglycerol (DAG) (2) and inositol 1,4,5-trisphos-
phate (IP3) (3, R2 ) PO32-) (Scheme 1). Both of these
products function as intracellular second messengers that
are integral to the regulation and stimulation of cellular
metabolism, secretion, and proliferation. For example,
IP3 effects the release of calcium ions from the endoplas-
mic reticulum thereby activating numerous cytosolic
enzymes, whereas DAG stimulates protein kinase C
(PKC), a calcium- and phosphatidylserine-dependent
enzyme that phosphorylates serine and threonine resi-
dues of the target enzymes.3 Mammalian PI-PLC’s also
hydrolyze the unphosphorylated phosphatidylinositol 1
(R2 ) H) to give 2 together with mixtures of the inositol
1-phosphate (3) (R2 ) H) and cyclic inositol phosphate
(IcP) (4) (R2 ) H). On the other hand, hydrolysis of 1
(R2 ) H) by bacterial PI-PLC produces only 2 and 4 (R2
) H).
X Abstract published in Advance ACS Abstracts, November 1, 1996.
(1) (a) Hokin, L. E. Ann. Rev. Biochem. 1985, 54, 205-235. (b)
Nishizuka, Y. Science 1992, 258, 607-614. (c) Rhee, S.-G.; Choi, K. D.
J . Biol. Chem. 1992, 267, 12393-12396. (d) Cockcroft, S.; Thomas, G.
M. H. Biochem. J . 1992, 288, 1-14. (e) Berridge, M. J . Nature 1993,
361, 315-325. (f) Bruzik, K. S.; Tsai, M.-D. BioMed. Chem. 1994, 2,
49-72. (g) Potter, B. V. L.; Lampe, D. Angew. Chem., Int. Ed. Engl.
1995, 34, 1933-1972.
(2) (a) Exton, J . H. J . Biol. Chem. 1990, 265, 1-4. (b) Billah, M. M.;
Anthes, J . C. Biochem. J . 1990, 269, 281-291. (c) Liscovitch, M. TIBS
1992, 17, 393-399. (d) Cockcroft, S. Biochem. Biophys. Acta 1992, 1113,
135-160. (e) Lassegue, B.; Alexander, R. W.; Clark, M.; Akers, M.;
Griendling, K. K. Biochem. J . 1993, 292, 509-517.
(3) For a review, see: Newton, A. C. J . Biol. Chem. 1995, 270,
28495-28498.
S0022-3263(96)00850-X CCC: $12.00 © 1996 American Chemical Society

Inhibitors of the Phospholipase C from B. cereus
J . Org. Chem., Vol. 61, No. 23, 1996 8017
Despite the biological importance of PI-PLC’s, mecha-
nistic information and three-dimensional structural de-
tails that provide clues as to how these enzymes catalyze
the hydrolysis of the phosphodiester bond of phospholip-
ids are only recently becoming available. The fact that
the cyclic phosphate 4 (R2 ) H) is produced upon
hydrolysis of 1 (R2 ) H) by mammalian and bacterial PI-
PLC suggests the C-2 hydroxyl group participates in the
hydrolysis.4 Supporting this hypothesis are the recent
findings that modified phosphatidylinositol analogues
lacking a free hydroxyl function at C-2 are inhibitors, not
substrates, of PI-PLC.5 The PI-PLC from Bacillus cereus
has been cloned and over-expressed in Escherichia coli,
and the crystal structure of this enzyme has been solved
for the native enzyme and for the enzyme complexed with
myo-inositol.6 The recently reported X-ray structure of
a catalytically-active deletion variant of the rat phos-
phatidylinositol-specific PLC-δ1 isozyme complexed with
calcium ion and inositol trisphosphate provides further
evidence for nucleophilic attack on phosphorus by the C-2
hydroxyl group and suggests mechanistic roles for several
active site residues.7
synthesis of modified phosphatidylinositides in which the
hydroxyl function at C-2 of the inositol ring is replaced,
altered, or removed appeared to be the most direct route
to developing competitive inhibitors of the enzyme. To
test this hypothesis, we envisioned that the series of
phosphatidylinositol derivatives 6-10, which are pat-
terned after the modified substrate 5, would be interest-
ing candidates as potential competitive inhibitors of
bacterial PI-PLC. Each of these compounds contains
n-hexanoyl side chains on the sn-glycerol moiety to confer
water solubility so that the biological evaluation of these
substrate analogues could be conducted below the critical
micelle concentrations (CMC) of substrates and inhibi-
tors, thereby simplifying analysis of the kinetic data. The
reported CMC of 5 is 12.3 mM in pure water and 9.0 mM
in Tris (tris(hydroxymethyl)aminomethane)acetate buffer
at pH 7.5.5c
We have been engaged in a general program directed
toward unveiling the mechanistic details in the hydroly-
sis of phosphatidylcholine by the PLC from B. cereus
(PLCBc).8 As a logical extension of this work, we became
interested in the design and synthesis of novel inhibitors
of PI-PLC to identify some structural features that are
important for substrate specificity, binding, and catalysis.
Ultimately, we hoped to gather details that would further
illuminate the mechanistic features of PI-PLC-catalyzed
hydrolysis of the phosphatidylinositols. As the first step
in this endeavor, we focused upon the design, synthesis,
and kinetic evaluation of a series of analogues of 1 (R2 )
H) that would be inhibitors of PI-PLC. Only one struc-
tural feature would be varied at a time so that the
importance of each change could be reasonably quanti-
fied. X-ray crystallographic studies would then be un-
dertaken with selected enzyme-inhibitor complexes in
order to help identify the important active site amino acid
residues and elucidate their roles in binding and cataly-
sis. These studies would also provide insights into the
function of the inositol C-2 hydroxyl group in enzyme/
substrate recognition and enzymatic hydrolysis.
We envisioned that the phosphatidylinositol targets
5-10 could be prepared by applying the general phos-
phite coupling techniques that had been developed in our
laboratories for the synthesis of a diverse array of
phosphodiesters. In each of these substrate analogues,
the D-myo-inositol enantiomer was used since it had been
determined that phospholipids incorporating L-myo-inosi-
tol are processed at only 0.2% the rate of those of D
configuration.9 Analogues 6-9 incorporated changes of
the C-2 hydroxyl group of the myo-inositol moiety to help
elucidate what alterations in the inhibitor would lead to
optimal binding at the enzyme active site. The stereo-
chemistry at C-2 in 6 was inverted to test the importance
of stereochemistry at this position. In 7-9, the hydroxyl
group at C-2 was replaced so the nucleophile that has
been proposed to initiate hydrolysis of the phosphodiester
linkage is no longer available. Compound 10 maintains
the natural integrity of inositol ring, but in place of the
phosphate function, it possesses a phosphorodithioate
group, which we anticipated would stabilize the phos-
phodiester linkage toward enzymatic hydrolysis.10 While
we were engaged in this effort, there were several reports
of phosphatidylinositol analogues in which the C-2 hy-
droxyl group was removed or methylated; these com-
pounds were found to be weak inhibitors, but no Ki’s were
determined.5a-c
Resu lts a n d Discu ssion
The weight of current experimental evidence suggests
that the C-2 hydroxyl in phosphatidylinositol 1 (R2 ) H)
is the nucleophile in the first step of the hydrolysis of 1
by bacterial PI-PLC (Scheme 1).4-7 Thus, the design and
(4) (a) Lin, G.; Bennett, C. F.; Tsai, M.-D. Biochem. 1990, 29, 2747-
2757. (b) Bruzik, K. S.; Morocho, A. M.; J hon, D.-Y.; Rhee, S. G.; Tsai,
M.-D. Biochem. 1992, 31, 5183-5193.
(5) (a) Seitz, S. P.; Kaltenbach, R. F., III; Vreekamp, R. H.;
Calabrese, J . C.; Perrella, F. W. Bioorg. Med. Chem. Lett. 1992, 2, 171-
174. (b) Garigapati, V. R.; Roberts, M. F. Tetrahedron Lett. 1993, 34,
5579-5582. (c) Lewis, K. A.; Garigapati, V. R.; Zhou, C.; Roberts, M.
F. Biochemistry 1993, 32, 8836-8841. See also: (d) Morris, J . C.; Ping-
Sheng, L.; Zhai, H.-X.; Shen, T.-Y.; Mensa-Wilmot, K. J . Biol. Chem.
1996, 271, 15468-15477.
(6) (a) Koke, J . A.; Yang, M.; Henner, D. J .; Volwerk, J . J .; Griffith,
O. H. Protein Expression Purif. 1992, 2, 51-58. (b) Heinz, D. W.; Ryan,
M.; Bullock, T. L.; Griffith, O. H. EMBO J . 1995, 3855-3863.
(7) Essen, L.-O.; Perisic, O.; Cheung, R.; Katan, M.; Williams, R. L.
Nature 1996, 380, 595-602.
(8) (a) Hansen, S.; Hough, E.; Svensson, L. A.; Wong, Y.-L.; Martin,
S. F. J . Mol. Biol. 1993, 234, 179-187. (b) Martin, S. F.; Wong, Y.-L.;
Wagman, A. S. J . Org. Chem. 1994, 59, 4821-4831. (c) Martin, S. F.;
Spaller, M.; Hergenrother, P. Biochemistry, in press.
Syn th esis of Mod ified P h osp h a tid ylin ositols. The
phosphatidylinositol derivatives 5-8 and 10 are known
compounds that were prepared using a one-pot, chloro-
phosphite coupling protocol that had been developed in
the our group.11 In this procedure, which is summarized
in Scheme 2, a chlorophosphite is sequentially treated
(9) (a) Shashidar, M. S.; Volwerk, J . J .; Griffith, O. H.; Keana, J . F.
W. Chem. Phys. Lipids 1991, 60, 101-110. (b) Leigh, A. J .; Volwerk,
J . J .; Griffith, H. O.; Keana, J . F. W. Biochemistry 1992, 31, 8978-
8983.
(10) (a) Lin, G.; Bennett, C. F.; Tsai, M.-D. Biochemistry 1990 29,
2747-2757. (b) Marshall, W. S.; Caruthers, M. H. Science 1993 259,
1564-1570.

8018 J . Org. Chem., Vol. 61, No. 23, 1996
Martin and Wagman
Sch em e 2a
Sch em e 3a
a
Key: (a) R3OPCl2 (1 equiv); R2OH (1 equiv); (b) [O] or S8; (c)
deprotect.
with two different nucleophiles to give the phosphotri-
ester 12 that is oxidized to give the phosphate triester
13. The desired phosphodiester 14 is then generated by
deprotection of the phosphate. In many cases the chlo-
rophosphite of choice is the commercially available meth-
yl dichlorophosphite, but the 2-(trimethylsilyl)ethyl dichlo-
rophosphite has been used as an alternative coupling
reagent because the (trimethylsilyl)ethyl group may be
removed under mildly acidic conditions.12 A diverse
array of phospholipid analogues with a variety of fatty
acid side chains and different head groups are readily
available by application of this method.
a
The synthesis of 9 by the same procedure used to
prepare 5-8 required a modified protecting group strat-
egy to manipulate the C-1 and C-2 hydroxyl groups of
3,4,5,6-tetra-O-benzyl-myo-inositol (15) (Scheme 3). Re-
action of 15 with dibutyltin oxide generated an interme-
diate stannylene ketal that was treated in situ with
p-methoxybenzyl chloride in the presence of CsF to give
16 in 89% yield.13,14 The selective alkylation of 1,2-diols
mediated by a dialkyltin oxide is a common procedure
in carbohydrate chemistry.15 O-Methylation of 16 fol-
lowed by removal of the p-methoxybenzyl protecting
group with dichlorodicyanobenzoquinone (DDQ) gave 17
in 89% overall yield. The phosphate triester 18 was then
prepared in 66% yield by coupling the protected inositol
17 with the diacylglycerol 19 using 2-(trimethylsilyl)ethyl
dichlorophosphite followed by oxidation according to our
standard protocol.11c Global deprotection of 18 by cata-
lytic hydrogenolysis of the benzyl groups and removal of
the trimethylsilylethyl group with aqueous HF gave 9
in 50% yield.
Key: (a) n-Bu2SnO, C6H6; ∆; p-MeOC6H4CH2Cl; CsF, DMF;
(b) NaH, THF; MeI; (c) DDQ, H2O, CH2Cl2; (d) TMSCH2CH2OPCl2,
i-Pr2NEt, THF, -78 °C; 19, -78 f 20 °C; t-BuOOH, CH2Cl2, 0
°C; (e) H2 (500 psi), 20% Pd(OH)2/C, EtOH, rt; (f) aqueous HF,
MeCN/THF (2:1), rt.
Sch em e 4a
As part of a general program directed toward develop-
ing methods for the syntheses of isosteric phosphate
replacements, we invented a facile method to prepare
phosphorodithioates using the phosphatitylating reagent
2-chloro-1,3,2-dithiaphospholane (20).11c,16 This protocol
was applied to the synthesis of 10 (Scheme 4). Thus,
(11) (a) Martin, S. F.; J osey, J . A. Tetrahedron Lett. 1988, 29, 3631-
3634. (b) Martin, S. F.; Wagman, A. S. J . Org. Chem. 1993, 58, 5897-
5899. (c) Martin, S. F.; J osey, J . A.; Wong, Y.-L.; Dean, D. W. J . Org.
Chem. 1994, 59, 4805-4820.
a
Key: (a) 19, EtN(i-Pr)2, CH3CN, -38 °C f rt; S8, CS2, rt; (b)
23, DBU, MeCN, rt; (c) aqueous HF, MeCN, rt.
(12) (a) Mautz, D. S.; Davisson, V. J .; Poulter, C. D. Tetrahedron
Lett. 1989, 30, 7333-7336. (b) Sawabe, A.; Filla, S. A.; Masamune, S.
Tetrahedron Lett. 1992, 33, 7685-7686.
reaction of 20 with 1,2-di-n-hexanoyl-sn-glycerol (19)
followed by sulfurization with elemental sulfur gave the
2-alkoxy-2-thio-1,3,2-dithiaphospholane 21 in 84% yield.
Reaction of 21 with the protected inositol 23 in the
presence of 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) and
elution of the intermediate DBU salt through an Am-
berlyst-15 ion exchange column (basic form) gave the
protected phosphorodithioate 22 in 85% yield.11c Acid-
(13) For a related example, see: Gigg, J .; Gigg, R.; Payne, S.; Conant,
R. J . Chem. Soc., Perkin Trans. 1 1987, 423426.
(14) Nagashima, N.; Ohno, M. Chem. Lett. 1987, 141-144.
(15) (a) Nashed, M. A.; Anderson, L. Tetrahedron Lett. 1976, 39,
3503-3506. (b) Nashed, M. A.; Anderson, L. Carbohydr. Res. 1977,
56, 419-422. (c) Kong, X; Grindley, T. B. Can. J . Chem. 1994, 72,
2396-2404 and 2405-2415.
(16) Boudjebel, H.; Goncalves, H.; Mathis, F. Bull. Soc. Chim. Fr.
1974, 5, 1671-1674.

Inhibitors of the Phospholipase C from B. cereus
J . Org. Chem., Vol. 61, No. 23, 1996 8019
Sch em e 5a
catalyzed deprotection of the inositol moiety of 22 fur-
nished 10 in 93% yield.
Biologica l Eva lu a tion of 6-10 a s In h ibitor s of P I-
P LC. The hydrolysis of phosphatidylinositol 1 (R2 ) H)
by bacterial PI-PLC produces diacylglycerol 2 and the
cyclic 1,2-phosphoinositol 4 (R2 ) H) (Scheme 1). Because
no acid is produced by this enzymatic reaction, a pH
titrimetric assay similar to that used for PLCBc cannot
be employed,11c,17 and alternate tactics have been devel-
oped to follow the hydrolysis of phosphatidylinositols by
PI-PLC.1f,18 After considering the available options, we
decided to examine the inositol alkylthiophosphate 2419
and the nitrophenyl inositolphosphate 259 as a potential
water-soluble substrates to develop a continuous assay
to measure the rates in the PI-PLC catalyzed hydrolysis
of phosphoinositides in the presence of 6-10. Cleavage
of 24 by PI-PLC in the presence of 4,4′-dipyridyl disulfide
produces pyridyl-4-thiol, the production of which would
be monitored by UV spectrometry, and inositol-1,2-cyclic
phosphate (26). Alternatively, hydrolysis of 25 by PI-
PLC forms 26 and p-nitrophenol, whose formation may
also be detected by UV.
a
Key: (a) pyridine, rt; H2O; rt; (b) AcOH/H2O (1:4), rt.
mM depending upon buffer;9 the Vmax values of 25 and
phosphatidylinositol were reported to be of the same
order of magnitude. In exploratory experiments, we
discovered that the rates of hydrolysis of 25 at low
concentrations were easily measurable when using mod-
erate (1-5 mM) concentrations of inhibitors having Ki’s
in the low millimolar range.
Compound 25 was prepared by the route outlined in
Scheme 5, which is similar to the published procedure.9
The protected myo-inositol 23 was coupled to com-
mercially available 4-nitrophenyl phosphorodichloridate
(27) to deliver the protected phosphate 28, which was
immediately deprotected to give 25 in 57% overall yield.
Since the pure phosphate 25 hydrolyzes slowly in the
presence of water or in aqueous base, it was dissolved in
water and extracted with ether to remove traces of
p-nitrophenol. The aqueous solution was then carefully
freeze-dried to a glass, evaporated to dryness with
toluene, dried in vacuo, and stored at -20 °C under
argon. In this state, 25 could be kept for several months
without decomposition. Upon dissolution of 25 in water
at neutral or slightly acidic pH, no free p-nitrophenol was
detected by UV spectrometry at 396 nm.
The rate of release of p-nitrophenol upon hydrolysis of
25 by PI-PLC at pH ) 7.0 (100 mM N-(2-hydroxyethyl)-
piperazine-N′-2-ethanesulfonic acid (Hepes) buffer) and
25 °C was followed by the change in the UV absorbance
at 396 nm (see Experimental Section). The assays were
performed at pH 7.0 as the best compromise for the
stability of 25 to hydrolysis, the molar extinction coef-
ficient of p-nitrophenol and the activity of the enzyme.
The progression curves for the enzymatic hydrolysis of
25 were nearly linear for the first 60-70 s of the reaction
for all substrate concentrations, [S], ranging from 1 to
25 mM so the initial rates, υo, were determined by
extrapolating a tangent to the curve back to the time at
which the enzyme was added. The initial velocity data
were then fit with a nonlinear least-squares algorithm,
and the curved line was calculated from the Michaelis-
Menten equation, υo ) Vmax[S](Km + [S])-1 (Figure 1). The
values for the kinetic parameters Km and Vmax for 25 that
gave a reasonable fit to the experimental data were
estimated to be 34 ((5) mM and 3800 ((500) µmol/min‚
mg, respectively.
At the outset of our work, the range of concentrations
of 25 used in our assay was based upon the reported Km
of 5 mM (Vmax ) 650 µmol/min‚mg) for 25; this value had
been determined in the same buffer system at concentra-
tions of 25 ranging from 0.1 to 7.7 mM.9b Examination
of Figure 1 reveals that even at 25 mM, the concentration
of 25 is significantly below saturation for the enzyme.
Because accurate kinetic parameters are most reliably
obtained at substrate concentrations varying from 0.5 to
Preliminary attempts to develop a kinetic assay using
24 as a substrate for PI-PLC proved problematic owing
to a significant background rate for the formation of
pyridyl-4-thiol by the nonenzymatic, chemical hydrolysis
of 4,4′-dipyridyl disulfide. Higher homologues of 24 are
better substrates for the enzyme, but they have much
lower CMC’s and therefore were not examined. On the
other hand, a continuous colorimetric assay using 25 as
the substrate was simple to implement (vide supra). In
the absence of PI-PLC, there was negligible hydrolysis
of 25 under the assay conditions, so the change in UV
absorbance over time was directly related to the amount
of p-nitrophenol that was released by the enzymatic
reaction. The only drawback to this assay is that the
substrate 25 is significantly different from the natural
phosphatidylinositol substrate in that p-nitrophenol is
the leaving group rather than a hydrophobic diacylglyc-
erol group. The natural phosphatidylinositol substrate
has a Km of 1-2 mM, whereas 25 has been reported to
exhibit a relatively high Km that varies from 5 to >15
(17) For a facile chromogenic assay for PLCBc, see: Hergenrother,
P. J .; Spaller, M. R.; Haas, M.; Martin, S. F. Anal. Biochem. 1995,
229, 313-316.
(18) Reynolds, L. J .; Washburn, W. N.; Deems, R. A.; Dennis, E. A.
Methods Enzymol. 1991, 197, 3-31.
(19) (a) Hendrickson, E. K.; J ohnson, J . L.; Hendrickson, H. S.
BioMed. Chem. Lett. 1991, 1, 615-618. (b) Hendrickson, H. S.;
Hendrickson, E. K.; J ohnson, J . L.; Khan, T. H.; Chial, H. J .
Biochemistry 1992, 31, 12169-12172.

8020 J . Org. Chem., Vol. 61, No. 23, 1996
Martin and Wagman
F igu r e 2. Proposed first step in catalytic mechanism for
hydrolysis of phosphatidylinositol by PI-PLC.
mM. It should be noted that because these assays were
not run at saturating substrate concentrations, there will
be a tendency to underestimate the Km and the apparent
Km’s. Consequently, these values for the Ki’s should be
viewed as an indication of the relative binding affinities
within this series of inhibitors rather than well-deter-
mined quantities.
F igu r e 1. Velocity versus substrate concentration plot for the
enzymatic hydrolysis of 25 by recombinant bacterial PI-PLC.
The curved line is calculated from the Michaelis-Menten
equation, υo ) Vmax[S](Km + [S])-1
.
These studies provide useful information that is rel-
evant to the mechanism of the enzymatic hydrolysis of
phosphatidylinositols by bacterial PI-PLC and to the
future design of competitive inhibitors of this and related
enzymes. Replacement of the axial hydroxyl group at C-2
of the inositol ring in compounds 7-9 leads to phosphati-
dylinositol derivatives that are inhibitors rather than
substrates. This observation supports the mechanism for
the hydrolysis of phospholipids by bacterial PI-PLC that
was first proposed by Tsai,4b,10a who suggested that the
inositol C-2 hydroxyl group attacks phosphorus in the
first step of the reaction. Because 6 is an inhibitor rather
than a substrate of PI-PLC, the enzymatic reaction is
clearly sensitive to the precise orientation of the C-2
hydroxyl group as well as its presence. More specific
details regarding this mechanism have recently emerged
from the X-ray structural work of Heinz,6b who has
suggested that this enzymatic reaction is catalyzed by
His-32 and His-82, which serve as the general base and
general acid, respectively (Figure 2). Site-directed mu-
tagenesis has shown that both histidines are critical for
catalytic activity. In this mechanistic model, Arg 69
activates the phosphate moiety for nucleophilic attack
by neutralizing charge, and the developing charge on the
imidazole of His-32 is neutralized by proton transfer to
Asp-274. This type of relay mechanism is similar to those
proposed for DNAse I21 and mammalian PLC-δ1.7
The two most potent inhibitors 6 and 10 each possess
a hydroxyl group at the C-2 position of the inositol ring
that can participate as a hydrogen bond donor, whereas
the weaker inhibitors 7-9 do not. In this context, it
should be recognized that differences in the hydrogen-
binding interactions at the active site are not the only
factors affecting the Ki’s, and solvation-desolvation
effects may also play important roles. The fluoro scyllo
analogue 7 was the worst inhibitor of all substrate
analogues tested, but the efficacy this inositol replace-
ment for inhibitor design should not yet be ruled out as
2-deoxy-2-fluoro-scyllo-inositol 1-O-dodecylphosphonate
was the most potent of several phosphonate inhibitors
of the bacterial PI-PLC from B. cereus.5d Substitution
of the phosphodiester linkage in 5 with a phospho-
rodithioate group gave the nonhydrolyzable analogue 10,
5 Km, all of the determinations of Vmax and Km are
presumably subject to significant experimental error.
Indeed, the measured Km of 25 could be even higher if
saturating substrate concentrations were used.
However, several factors conspired to limit the range
of concentrations of 25 used in the subsequent parts of
this study to 1-25 mM. First, the labor associated with
synthesizing the quantities of 25 and 6-10 that would
be required to evaluate each inhibitor at concentrations
of 25 ranging from 15 to g150 mM seemed prohibitive.
Furthermore, at higher concentrations of substrate and
inhibitor, there loomed the danger of forming micelles
or mixed micelles, and the analysis of the data obtained
in the resulting heterogeneous system would not be
amenable to simple treatment.
The water-soluble substrate analogues 6-10 were
evaluated as inhibitors of bacterial PI-PLC in competition
experiments in which enzyme was added to a stirred
solution of substrate 25 and inhibitor under the standard
assay conditions. The inhibitor concentration needed for
the kinetic analysis was determined by adding different
amounts of inhibitor to 2.5 mM solutions of 25, and the
inhibitor concentration [I] that gave a 30-50% rate
reduction was used. The raw rate data were then fit
using a nonlinear least-squares fit of the hyperbolic data
of [S] vs υo. The curved line for each inhibitor assay was
calculated from the Michaelis-Menten equation (υo
V
)
max[S](Km + [S])-1) giving directly the apparent Km (Km′).
The values obtained for Vmax for the enzymatic hydrolysis
of 25 by PI-PLC in the presence of each inhibitor 6-10
were similar, but the measured apparent Km’s were
always greater than for the substrate 25. Although this
behavior suggests that each inhibitor is competitive,20
small components of noncompetitive inhibition would not
be detected under the constraints of our assays. The
initial velocities were used to generate double-reciprocal
plots that also indicated that each of the compounds
6-10 was a competitive inhibitor. Using the expression
Km′ ) Km(1 + [I]/Ki), the Ki’s were calculated as follows:
6, 0.2 mM; 10, 0.6 mM; 8, 2.6 mM; 9, 6.6 mM; and 7, 8.8
(20) For
a review of the concepts and the derivations for the
equations for the different types of inhibition. see: (a) Segel, I. H. In
Enzyme Kinetics; J ohn Wiley & Sons: New York, 1975; pp 1-467. (b)
Cornish-Bowden, A. In Fundamentals of Enzyme Kinetics; Butter-
worths, Inc.: Boston, 1979; pp 1-230.
(21) Suck, D.; Oefner, C. Nature 1986, 321, 620-625.

Inhibitors of the Phospholipase C from B. cereus
J . Org. Chem., Vol. 61, No. 23, 1996 8021
enzyme used in the kinetic assays was determined by the
Bradford method.23 The enzymatic hydrolysis reactions were
carried out in disposable 0.5 mL methyl methacrylate cuvettes
available from Fisher Scientific. The progression of the
enzymatic reactions was monitored and recorded using a
Hewlett-Packard 8452A UV diode array spectrophotometer.
Compounds 5-8 and 10 were prepared according to literature
so this replacement should be more broadly examined for
the development of novel inhibitors. Because sulfur is
less electronegative than oxygen, the ionic interaction of
a phosphorodithioate moiety with a positive charge on
the enzyme, which might be provided by either an
arginine residue or a calcium ion, will be less than for a
phosphate group. This reduced interaction not only
decreases the susceptibility of phosphorous to nucleo-
philic attack, but it also suggests that inhibitors of PI-
PLC that contain the phosphorodithioate moiety may
bind less tightly than the corresponding phosphate
analogues; this hypothesis remains to be tested. The
importance of the hydroxyl groups at C-3, C-4, and C-5
in the design of inhibitors of bacterial PI-PLC must also
be evaluated.
A number of mechanistic issues remain unresolved.
The X-ray structure of PI-PLC with myo-inositol reveals
only limited information regarding the specific roles of
the various active site residues in binding and catalysis,
and a structure of PI-PLC complexed with an inhibitor
such as 6-10 is essential to developing more complete
insights regarding the mechanism of hydrolysis of inositol
phosphates by bacterial PI-PLC. Details regarding the
kinetic mechanism are also lacking. Toward this end, it
is necessary to determine the slow step and the order of
product release in the enzymatic cycle. The nature of
product inhibition by the enzyme must also be clarified.
Resolution of these questions and the development of new
kinetic assays for PI-PLC are in progress, and the results
of these and related investigations will be reported in due
course.
1
procedures and were determined to be g95% pure by H, 13C,
and 31P NMR.11
D-1-O-(4-Met h oxyb en zyl)-3,4,5,6-t et r a -O-b en zyl-m yo-
in ositol (16). A suspension of tetrabenzyl-myo-inositol (15)24
(9.29 g, 17.0 mmol) and dibutyltin oxide (4.48 g, 17.0 mmol)
in C6H6 (100 mL) was heated overnight at reflux using a
Dean-Stark trap to remove water. After cooling, the solvent
was removed under reduced pressure. The yellow residual oil
was dissolved in DMF (60 mL) containing CsF (5.24 g, 34.5
mmol) and 4-methoxybenzyl chloride (2.3 mL, 2.66 g, 17.0
mmol). The reaction was stirred for 8 h, and the volatiles were
removed in vacuo leaving a white crystalline solid, which was
partially dissolved in EtOAc (50 mL). The solid was removed
by filtration, and the filtrate was evaporated to give a residue
that was purified by flash chromatography eluting with
hexane/EtOAc (7:3) to deliver 9.95 g (89%) of 16 as a white
solid: mp 109-111 °C; 1H NMR δ 7.65-7.28 (comp, 22 H),
6.97 (d, J ) 7.0 Hz, 2 H), 5.17-4.92 (comp, 6 H), 4.88-4.71
(comp, 4 H), 4.19 (app t, J ) 2.4 Hz, 1 H), 4.18 (m, 2 H), 3.85
(s, 3 H), 3.63 (app t, J ) 7.8 Hz, 1 H), 3.53 (app t, J ) 2.4 Hz,
1 H), 3.50 (app t, J ) 2.4 Hz, 1 H), 2.86 (br s, 1 H); 13C NMR
δ 159.3, 138.8, 138.7, 138.6, 137.9, 129.9, 129.4, 128.4, 128.3,
127.9, 127.8, 127.5, 113.8, 83.1, 81.1, 79.7, 79.4, 75.8, 72.6, 72.3,
67.4, 55.2; IR 3578 cm-1; mass spectrum (CI(+)) m/ z 659.3001
(C42H42O7 + H requires 659.3009), 630 (base).
D-1-O-(4-Met h oxyb en zyl)-2-O-m et h yl-3,4,5,6-t et r a -O-
ben zyl-m yo-in ositol. A solution of 16 (0.69 g, 1.04 mmol) in
THF (2 mL) was added to a 60% NaH oil dispersion (0.16 g,
4.0 mmol) in THF (10 mL). After 30 min, freshly distilled
methyl iodide (0.5 mL, 1.1 g, 8.0 mmol) was added, and the
mixture was stirred for 24 h at 40 °C. Water (10 mL) was
cautiously added, and the reaction was then diluted with Et2O
(100 mL). The layers were separated, and the organic layer
was washed with brine (1 × 30 mL), dried (MgSO4), and
evaporated. The residue was purified by flash chromatography
eluting with hexane/EtOAc (7:3) to provide 0.79 g (ca 100%)
Exp er im en ta l Section
Gen er a l P r oced u r es. Unless otherwise noted, all starting
materials were obtained from commercial suppliers and were
used without further purification. Reagent-grade solvents
were dried according to established protocols by distillation
under inert atmosphere from an appropriate drying agent
immediately prior to use. Thus, tetrahydrofuran (THF) was
distilled from the potassium/benzophenone ketal, and aceto-
nitrile (MeCN) and benzene (C6H6) were distilled from CaH.
Diisopropylamine and pyridine (Py) were distilled from calcium
hydride and stored over 4 Å molecular sieves under argon.
Ethanol (EtOH) was distilled from magnesium ethoxide and
stored over 3 Å molecular sieves. Reactions involving air- and/
or moisture-sensitive reagents were executed under an atmo-
sphere of dry argon, and the glassware was flame dried under
vacuum. Melting points and boiling points are uncorrected.
Infrared (IR) spectra were recorded as solutions in CHCl3. All
spectra are reported in wavenumbers (cm-1) and are refer-
enced to the 1601 cm-1 absorption of a polystyrene film.
Proton (1H) (250 MHz), carbon (13C) (63 MHz), and phosphorus
(31P) (120 or 146 MHz) nuclear magnetic resonance (NMR)
spectra were obtained as solutions in deuteriochloroform
(CDCl3) unless otherwise indicated. 1H and 13C chemical shifts
are reported in parts per million (ppm, δ) downfield relative
to tetramethylsilane (TMS), which was referenced to the
solvent, whereas 31P chemical shifts are reported in parts per
million (ppm, δ) downfield relative to external 80% phosphoric
acid. Coupling constants are reported in hertz (Hz). Spectral
splitting patterns are designated as follows: s, singlet; d,
doublet; t, triplet; q, quartet; m, multiplet; comp, complex
multiplet; and br, broad. Flash chromatography was per-
formed according to the established protocol with Merck silica
gel 60 (230-400 ASTM).22
1
of the product as a white solid: mp 151-153 °C; H NMR δ
7.50-7.36 (comp, 22 H), 6.99 (d, J ) 6.7 Hz, 2 H), 5.11-4.94
(comp, 6 H), 4.89-4.71 (comp, 4 H), 4.19-4.12 (m, 2 H), 3.90
(s, 3 H), 3.84 (app t, J ) 2.4 Hz, 1 H), 3.81 (s, 3 H), 3.60 (app
t, J ) 9.2 Hz, 1 H), 3.49 (app t, J ) 2.4 Hz, 1 H), 3.45 (app t,
J ) 2.4 Hz, 1 H); 13C NMR δ 159.4, 139.1, 139.0, 138.4, 130.5,
129.5, 128.5, 128.4, 128.1, 127.9, 127.6, 113.9, 83.7, 81.8, 80.8,
80.5, 77.8, 77.3, 76.8, 76.0, 73.0, 72.7, 61.4, 55.3; mass
spectrum (CI(+)) m/ z 673.3165 (C43H44O7 + H requires
673.3165), 553 (base).
D-2-O-Meth yl-3,4,5,6-tetr a -O-ben zyl-m yo-in ositol (17).
2,3-Dichloro-5,6-dicyano-1,4-benzoquinone (DDQ) (0.39 g, 1.7
mmol) was added to a stirred solution of the product of the
preceding experiment (0.79 g, 1.10 mmol) and water (0.5 mL)
in CH2Cl2 (20 mL). The reaction was stirred for 30 min at rt
and then diluted with CH2Cl2 (100 mL). The mixture was
washed with saturated aqueous NaHCO3 (1 × 30 mL) and
brine (1 × 30 mL), dried (MgSO4), and concentrated under
reduced pressure. The residue was purified by flash chroma-
tography eluting with hexane/EtOAc (7:3) to provide 0.79 g
(ca. 100%) of 17 as a white solid: mp 133-135 °C; 1H NMR δ
7.38-7.21 (comp, 20 H), 4.96-4.67 (comp, 8 H), 4.97 (app t, J
) 9.5 Hz, 1 H), 3.84-3.70 (comp, 2 H), 3.65 (s, 3 H), 3.51-
3.38 (comp, 3 H), 2.23 (br s, 1 H); 13C NMR δ 138.7, 138.6,
138.5, 138.0, 128.4, 128.3, 128.2, 127.9, 127.8, 127.7, 127.4,
83.4, 82.0, 81.8, 81.0, 79.0, 75.8, 75.6, 75.5, 72.9, 72.2, 61.3;
mass spectrum (CI(+)) m/ z 555.2744 (base) (C35H38O6 + H
requires 555.2747), 463.
Recombinant bacterial PI-PLC was obtained from Professor
O. H. Griffith (University of Oregon). The concentration of
(23) Bradford, M. M. Anal. Biochem. 1975,72, 248-254.
(24) Bruzik, K. S.; Salamonczyk, G. M. Carbohydr. Res. 1989, 195,
67-73.
(22) Still, W. C.; Kahn, M.; Mitra, A. J . Org. Chem. 1978, 43, 2923-
2924.

8022 J . Org. Chem., Vol. 61, No. 23, 1996
Martin and Wagman
O-(1,2-Dih exan oyl-sn -3-glycer yl) O-(D-2-O-m eth yl-3,4,5,6-
tetr a -O-ben zyl-m yo-in ositolyl) O-[2-(tr im eth ylsilyl)eth yl]
p h osp h a te (18). A solution of the protected inositol 17 (150
mg, 0.28 mmol) in THF (1 mL), was added dropwise to a
vigorously stirred solution of 2-(trimethylsilyl)ethyl dichloro-
phosphite (67 mg, 0.31 mmol) and N,N-diisopropylethylamine
(60 µL, 44 mg, 0.78) in dry, deoxygenated THF (2 mL) at -78
°C. After stirring for 1 h, a solution of diacylglycerol 19 (97
mg, 0.34 mmol) in THF (1 mL) was slowly added, and the
resulting suspension was stirred for 2 h at -78 °C. The cooling
bath was removed, and stirring was continued for an ad-
ditional 1 h at rt. The reaction was filtered through a plug of
celite (1 g), which was subsequently washed with dry THF (2
mL). The filtrate and washings were combined and cooled to
0 °C, and a solution of 3 M tert-butyl hydroperoxide in
isopentane (0.3 mL, 1.0 mmol) was added with stirring. After
0.5 h, the excess oxidizing agent was destroyed by adding
trimethyl phosphite (2 mL). When the mixture no longer gave
a positive indication on acidic starch/iodine paper, the solvents
were removed under reduced pressure, and the residue was
dried in vacuo. The crude phosphate triester was then purified
by flash chromatography eluting with hexanes/EtOAc (2:1) to
give 0.18 g (66%) of 18 as inseparable mixture of diastereo-
isomers and a colorless oil: 1H NMR δ 7.42-7.18 (comp, 20
H), 5.11-5.09 (m, 1 H), 4.99-4.70 (comp, 8 H), 4.31-3.90
(comp, 10 H), 3.71 (s, 3 H), 3.51 (t, J ) 7.0 Hz, 2 H), 2.42-
2.23 (m, 4 H), 1.73-1.52 (m, 4 H), 1.41-1.22 (m, 8 H), 1.15 (t,
J ) 7.0 Hz, 1.1 H), 1.05 (t, J ) 7.0 Hz, 0.9 H), 0.96-0.85 (m,
6 H), 0.06 (s, 5 H), 0.01 (s, 4 H); 13C NMR δ 172.8, 172.4, 138.5,
138.3, 137.8, 128.2, 128.1, 127.7, 127.5, 127.4, 127.3, 127.3,
82.8, 81.1, 80.2, 79.8, 79.7, 78.3, 78.2, 77.9, 77.2, 75.6), 75.2,
72.7, 69.2, 69.1, 67.6, 66.8, 66.7, 66.6, 65.2, 65.1, 64.7, 33.8,
33.7, 31.0, 30.9, 24.3, 24.2, 22.0, 19.4, 19.3, 13.7, -1.7, -1.8;
IR 1736 cm-1; mass spectrum (CI(+)) m/ z 1005.4939 (base)
(C55H77O13PSi + H requires 1005.4949), 977, 289, 271.
1,2-Dih exa n oyl-sn -3-glycer yl-3-p h osp h o-D-2-O-m eth yl-
m yo-in ositol (9). To a solution of 18 (0.18 g, 0.18 mmol) in
absolute EtOH (10 mL) was added 20% Pd(OH)2/C (0.15 g).
The resulting suspension was stirred under an atmosphere of
H2 (500 psi) for 14 h. The catalyst was removed by filtration
through a plug of celite, and the solvent was removed under
reduced pressure and dried in vacuo. The residue was
dissolved in MeCN/THF (2:1) (2.8 mL), and 2.8 N aqueous HF
in THF (195 µL) was added. After stirring at rt for 1 h, the
volatiles were removed under reduced pressure, and the
residue was purified by flash chromatography eluting with
CHCl3/MeOH/H2O (6.6:3.0:0.4) to obtained 49 mg (50%) of 9
as a colorless glass: 1H NMR (CD3OD) δ 5.31-5.19 (m, 1 H),
4.49-4.39 (m, 1 H), 4.25-3.88 (comp, 4 H), 3.90-3.81 (m, 2
H), 3.81-3.68 (m, 1 H), 3.68-3.57 (m, 3 H), 3.57-3.51 (m, 1
H), 3.51-3.38 (m, 1 H), 3.23-3.15 (m, 1 H), 2.43-2.26 (m, 4
H), 1.71-1.49 (m, 4 H), 1.43-1.20 (m, 8 H), 0.90 (t, J ) 6.6
Hz, 6 H); 13C NMR (CD3OD) δ 175.0, 174.7, 83.3, 78.1, 76.2,
74.3, 73.5, 73.1, 71.9, 65.0, 63.7, 62.3, 35.0, 34.8, 32.3, 25.6,
23.3, 14.3); 31P NMR (CD3OD) δ 0.41; IR 3338, 1733, 1449,
1230 cm-1; mass spectrum (FAB(-)) m/ z 543.2208 (base)
(C22H41O13P - H requires 543.2207).
refrigerated): 1H NMR δ 3.78-3.51 (comp, 4 H); 13C NMR δ
42.8 (d, J CP ) 7.8 Hz); 31P NMR δ 168.9; IR 3008, 1417, 1281,
1077, 988 cm-1; mass spectrum (CI(+)) m/ z 157.9181 (C2H4-
PS2 requires 157.9181), 125, 123 (base).
2-O-(1,2-Dih exa n oyl-sn -3-glycer yl)-2-th io-1,3,2-d ith io-
p h osp h ola n e (21). 2-Chloro-1,3,2-dithiophospholane (20)
(0.79 g, 5.00 mmol) in MeCN (1 mL) was added dropwise to a
stirred solution of 1,2-di-n-hexanoyl-sn-3-glycerol (19) (1.44 g,
5.00 mmol) and Hu¨nig’s base (0.96 mL, 5.50 mmol) in dry,
degassed MeCN (2 mL) at -38 °C. After stirring for 2 h at
-38 °C, the reaction mixture was warmed to rt and stirred
for an additional 1 h. A solution of S8 (0.48 g, 15.00 mmol)
dissolved in CS2 (5 mL) was added, and the resultant light
yellow heterogeneous mixture was stirred vigorously for 6 h.
The reaction mixture was concentrated under reduced pres-
sure, EtOAc (5 mL) was added, and the mixture was filtered
through a plug of glasswool. The filtrate was concentrated
under reduced pressure, and the residue was purified by flash
chromatography eluting with hexanes/EtOAc (7:3) to deliver
1.87 g (84%) of 21 as a viscous oil: 1H NMR δ 5.26-5.18 (m,
1 H), 4.31-4.09 (comp, 4 H), 3.70-3.52 (comp, 4 H), 2.29 (t, J
) 7.3 Hz, 2 H), 2.27 (t, J ) 7.3 Hz, 2 H), 1.64-1.51 (comp, 4
H), 1.31-1.20 (comp, 8 H), 0.84 (t, J ) 6.7 Hz, 6 H); 13C NMR
δ 173.2, 172.7, 69.1 (d, J CP ) 10.1 Hz), 65.4 (d, J CP ) 8.6 Hz),
61.7, 41.5 (d, J CP ) 5.3 Hz), 34.1, 33.9, 31.2, 24.4, 22.2, 13.9;
31P NMR δ 124.0; IR 1737, 1466, 1229, 1165, 1020 cm-1; mass
spectrum (CI(+)) m/ z 443.1140 (C17H31O5PS3 + H requires
443.1150), 271 (base).
O-(1,2-Dih exa n oyl-sn -3-glycer yl) O-(2,3:5,6-Bis-O-(1-
m et h ylet h ylid en e)-4-O-(ter t-b u t yld im et h ylsilyl)-D-m yo-
in osit olyl) P h osp h or od it h ioa t e (22). 1,8-Diazo[5.4.0]-
bicycloundec-7-ene (DBU) (0.19 mL, 1.25 mmol) was added to
a solution of the dithiophospholane 21 (0.63 g, 1.40 mmol) and
the protected D-myo-inositol 23 (0.48 g, 1.25 mmol) in MeCN
(50 mL), and the reaction was stirred for 1 h at rt. The solvent
was removed under reduced pressure, and the crude product
was purified by flash chromatography eluting first with MeOH/
CHCl3 (1:19) and then Me2CO/CHCl3/H2O (6.7: 3.2: 0.1) to
provide 0.80 g (85%) of the DBU salt of 22. The free
phosphorodithioate was isolated by ion exchange chromatog-
raphy using an Amberlyst column (basic form), which was
prepared by washing the hydrated resin (ca. 10g) with etha-
nolic NaOH and then absolute EtOH. The DBU salt of 22 was
eluted through the basic resin using absolute EtOH. The
eluant was acidified with dilute aq. HCl, and the solvents were
removed in vacuo to yield 0.76 g (95%) 22 as a light yellow
glass: 1H NMR δ 5.26-5.18 (m, 1 H), 4.92-4.81 (m, 1 H), 4.72
(t, J ) 4.5 Hz, 1 H), 4.40-4.03 (comp, 4 H), 3.93-3.84 (m, 2
H), 3.71 (dd, J ) 9.8, 6.3 Hz, 1 H), 3.25 (t, J ) 9.8 Hz, 1 H),
2.23 (t, J ) 7.4 Hz, 2 H), 2.21 (t, J ) 7.4 Hz, 2 H), 1.68-1.51
(comp, 4 H), 1.36-1.19 (comp, 8 H), 1.32 (s, 3 H), 1.23 (s, 6
H), 1.20 (s, 3 H), 0.84 (s, 9 H), 0.83 (t, J ) 6.9 Hz, 6 H), 0.03
(s, 6 H); 13C NMR δ 173.3, 172.9, 111.7, 109.6, 79.9, 82.5, 79.1,
77.3, 75.8, 72.9 (d, J CP ) 8.5 Hz), 70.2 (d, J CP ) 8.6 Hz), 66.2
(d, J CP )7.7 Hz), 63.1, 34.2, 34.0, 31.1, 28.2, 26.8, 24.4, 22.2,
18.1, 13.8, -4.6, -4.8; 31P NMR δ 115.6; IR 1731, 1456, 1170
cm-1; mass spectrum (CI(+)) m/ z 755.3092 (C33H59O11PSi2 +
H requires 755.3084), 469, 271 (base), 173.
2-Ch lor o-1,3,2-dith ioph osph olan e (20).16 1,2-Ethanedithi-
ol (85 mL, 1 mol) was added dropwise over 1.5 h with stirring
to a solution of freshly distilled PCl3 (90 mL, 1.0 mol) in
degassed C6H6 (350 mL) at 50 °C (oil bath). After the addition
was complete, the solution was heated for 2 h at which time
the reaction flask was fitted for a simple distillation. The
solvent was removed by distillation at atmospheric pressure
(head temp ∼ 61 °C), and the product was then distilled at
0.1 mm Hg (head temp, 48 °C). (NOTE: It is important to
seal all joints with grease to exclude oxygen and to distill the
product slowly (∼5-6 h) to avoid an explosion! Signs of
imminent danger include a darkening to a deep orange or red,
and a thickening of the material in the distillation pot.) The
crude distillation was stopped when the oil in the distillation
pot becomes viscous. The crude product was then purified by
fractional distillation using a 6 in. vigreux column (∼0.1
mmHg, head temp 75 °C), and the vacuum is quenched with
a stream of dry argon to yield 142.7 g (90%) of 31 as clear
colorless liquid (which must be stored under argon and
O-(1,2-Dih exa n oyl-sn -3-glycer yl) O-(2,3:5,6-Bis-O-(1-
m et h ylet h ylid en e)-4-O-(ter t-b u t yld im et h ylsilyl)-D-m yo-
in ositolyl) p h osp h or od ith ioa te (10). A mixture of the
protected inositol dithioate 22 (0.40 g, 0.53 mmol) and aq. 2.5
M HF (0.6 mL, 0.74 mmol) in MeCN (10 mL) was stirred for
2 h at rt, whereupon the solvents were removed in vacuo. The
residue was suspended in CHCl3 (10 mL), and the insoluble
solids were removed by vacuum filtration. The solvent was
removed under reduced pressure, and the residue was purified
by flash chromatography eluting with CHCl3/MeOH/H2O (7:
3: 0.5) to provide 0.28 g (93%) of 10 as a yellow glass: 1H
NMR (DMSO-d6) δ 5.17-5.04 (br s, 1 H), 4.67-4.63 (br s, 1
H), 4.52-4.49 (br s, 1 H), 4.46-4.42 (br s, 1 H), 4.39-4.35 (br
s, 1 H), 4.31-3.90 (comp, 5 H), 3.56-3.46 (m, 1 H), 3.42-3.31
(m, 2 H), 3.13-3.04 (m, 1 H), 3.00-2.89 (m, 1 H), 2.34-2.21
(comp, 4 H), 1.60-1.44 (comp, 4 H), 1.37-1.15 (comp, 8 H),
0.83 (t, J ) 6.9 Hz, 6 H); 13C NMR (CD3OD) δ 175.1, 174.1,
88.9 (d, J CP ) 6.0 Hz), 76.0, 73.9, 73.2, 73.1 (d, J CP ) 3.6 Hz),

Inhibitors of the Phospholipase C from B. cereus
J . Org. Chem., Vol. 61, No. 23, 1996 8023
72.7 (d, J CP ) 7.5 Hz), 71.7 (d, J CP ) 7.6 Hz), 64.9 (d, J CP )5.6
Hz), 63.8, 35.1, 34.9, 32.3, 25.6, 23.4, 14.3; 31P NMR δ 116.7;
IR 3251, 1737, 1466, 1164 cm-1; mass spectrum (CI(+)) m/ z
561.1600 (base) (C21H37O11PS2 + H requires 561.1593), 257.
m yo-In ositol 1-(4-Nitr op h en ylp h osp h a te) (25). A mix-
ture of 4-nitrophenyl phosphorodichloridate (1.56 g, 6.00 mmol)
and 23 (0.75 g, 2.00 mmol) in dry pyridine (10 mL) was stirred
overnight at rt. Water (0.5 mL) was added, and the mixture
was stirred for 1 h at rt. The solvents were removed in vacuo,
and the residual oil was dissolved in CH3CN/THF (2:1) (30
mL) containing 2.45 N aqueous HF (2 mL). After stirring for
2 h at rt, the volatiles were removed under reduced pressure,
and the crude product was purified by flash chromatography
eluting with CHCl3/MeOH/H2O (7:3:0.5) to provide 0.45 g (57%)
of 25 as a yellow glass. Any traces of 4-nitrophenol were
removed by extracting an aqueous solution of 25 with ether.
The aqueous layer was then freeze-dried to yield 25 as a stable
glass, the physical properties of which were identical to those
reported.9
Sta n d a r d Sp ectr op h otom etr ic Assa y. The activity of
recombinant bacterial PI-PLC (B. cereus) was assayed spec-
trophotometrically at 396 nm by measuring the rate of
formation of 4-nitrophenol that was produced by the enzyme
catalyzed hydrolysis of the chromogenic substrate 25.9 Stock
solutions of substrate 25 were obtained by dissolving a weighed
amount of dry 25 in the required amount of acidic water (pH
4.5-5) to give 1000 mM and 100 mM solutions. The concen-
trations of these stock solutions were then accurately deter-
mined by adding a small aliquot of the 25 solution to 0.01 N
aq. NaOH and measuring the absorbance at 396 nm after
complete hydrolysis (30 min).9a The stock solution of recom-
binant PI-PLC was prepared by diluting a small aliquot of
enzyme in Tris buffer (20 mM, pH 8.6) in the required amount
of 1% bovine serum albumin (BSA) at pH 7-8 to give a 5.25
µg/mL solution. Inhibitor stock solutions were made by
dissolving a weighed amount of dry inhibitor with the required
volume of water. During the kinetic evaluations the stock
solutions were kept on ice. In a typical assay, the required
amount of 25 (5-25 µL) and inhibitor stock solution (10 µL)
were mixed in a cuvette containing the proper volume of 100
mM Hepes/1 mM ethylenediamine tetraacetic acid (EDTA)
buffer at pH 7.0. The mixture was stirred in a constant
temperature bath at 25 °C. After a background measurement
was recorded by the spectrophotometer, the reaction was
initiated by the addition of 0.01 µg of PI-PLC (10 µL) with
stirring. The rate of release of 4-nitrophenol was followed
continuously at 396 nm for 2 min. PI-PLC activity was
assayed at [S]’s of 25.0, 10.0, 5.0, 2.5, and 1.0 mM in both the
presence and in the absence of inhibitors 6-10; only one [I]
was assayed for each inhibitor. The assays conducted with
10 mM and 25 mM substrate concentrations were performed
at least in duplicate, whereas those executed at lower substrate
concentrations were done at least in triplicate. The specific
activity of PI-PLC was determined by calculating the rate of
product released over time by using the molar extinction
coefficient of 4-nitrophenol, which is 7700 M-1 cm-1 at 396 nm
at pH 7.0.9a The initial velocity was determined by fitting a
tangent anchored at time zero to the initial linear region of
the rate progression curve. This tangent line was calculated
using the kinetics software provided by Hewlett-Packard that
also provides the standard deviation of the data points in the
fitted curve. The initial velocity data from each enzyme assay
were examined both by non-linear least squares analysis and
by extrapolation of the double reciprocal plot of velocity versus
substrate concentration. The hyperbolic data in the plot of
velocity versus concentration were fit with a non-linear least
squares algorithm, and the curved line was calculated from
the Michaelis-Menten equation υo ) Vmax[S](Km + [S])-1 using
the program KaleidaGraph. The values for Km and Vmax for
the substrate 25 were estimated directly from the hyperbolic
data. For the assays in which the hydrolysis of 25 was
monitored in the presence of the inhibitors 6-10, the apparent
Km’s (Km′) were determined by the same method assuming
competitive inhibition for each compound.20 The initial veloci-
ties were used to generate double-reciprocal plots from which
the type of inhibition of each of the compounds 6-10 was also
determined to be competitive. The Ki’s were calculated using
the expression Km′ ) Km(1 + [I]/Ki).
Ack n ow led gm en t. We thank the National Insti-
tutes of Health and the Robert A. Welch Foundation for
their generous support of this research. We also thank
Mr. Daniel W. Dean for preparing compounds 6-8, for
the synthesis of a number of other intermediates, and
for assistance in developing the assay. We are espe-
cially grateful to Professor O. Hayes Griffith (University
of Oregon) for a generous gift of recombinant PI-PLC
and to Dr. Dirk W. Heinz (Universita¨t Freiburg) for
helpful discussions.
1
Su p p or tin g In for m a tion Ava ila ble: Copies of H NMR
spectra of all new compounds (9 pages). This material is
contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
J O960850Q