Synthesis of new carboxylesterase inhibitors and evaluation of potency and water solubility

Article abstract of DOI:10.1021/tx015508+

Carboxylesterases are essential enzymes in the hydrolysis and detoxification of numerous pharmaceuticals and pesticides. They are vital in mediating organophosphate toxicity and in activating many prodrugs such as the chemotherapeutic agent CPT-11. It is therefore important to study the catalytic mechanism responsible for carboxylesterase-induced hydrolysis, which can be accomplished through the use of potent and selective inhibitors. Trifluoromethyl ketone (TFK)-containing compounds are the most potent esterase inhibitors described to date. The inclusion of a thioether moiety β to the carbonyl further increased TFK inhibitor potency. In this study, we have synthesized the sulfone analogues of a series of aliphatic and aromatic substituted thioether TFKs to evaluate their potency and solubility properties. This structural change shifted the keto/hydrate equilibrium from <9% hydrate to >95% hydrate, forming almost exclusively the gem-diol. These new compounds were evaluated for their inhibition of carboxylesterase activity in three different systems, rat liver microsomes, commercial porcine esterase, and juvenile hormone esterase in cabbage looper (Trichoplusia ni) hemolymph. The most potent inhibitor of rat liver carboxylesterase activity was 1,1,1-trifluoro-3-(decane-1-sulfonyl)-propan-2,2-diol, which inhibited 50% of the enzyme activity (IC₅₀) at 6.3 ± 1.3 nM and was 18-fold more potent than its thioether analogue. However, the sulfone derivatives were consistently poorer inhibitors of porcine carboxylesterase activity and juvenile hormone esterase activity, with IC₅₀ values ranging from low micromolar to millimolar. The compound 1,1,1-trifluoro-3-(octane-1-sulfonyl)-propan-2,2-diol was shown to have a 10-fold greater water solubility than its thioether analogue, 1,1,1-trifluoro-3-octylsulfanyl-propan-2-one (OTFP). These novel compounds provide further evidence of the differences between esterase orthologs, suggesting that additional development of esterase inhibitors may ultimately provide a battery of ortholog and/or isoform selective inhibitors analogous to those available for other complex enzyme families with overlapping substrate specificity.

Full text of DOI:10.1021/tx015508+

DECEMBER 2001
VOLUME 14, NUMBER 12
© Copyright 2001 by the American Chemical Society
Articles
Syn th esis of New Ca r boxylester a se In h ibitor s a n d
Eva lu a tion of P oten cy a n d Wa ter Solu bility
Craig E. Wheelock, Tonya F. Severson, and Bruce D. Hammock*
Department of Entomology and Cancer Research Center, University of California at Davis,
One Shields Avenue, Davis, California 95616
Received April 6, 2001
Carboxylesterases are essential enzymes in the hydrolysis and detoxification of numerous
pharmaceuticals and pesticides. They are vital in mediating organophosphate toxicity and in
activating many prodrugs such as the chemotherapeutic agent CPT-11. It is therefore important
to study the catalytic mechanism responsible for carboxylesterase-induced hydrolysis, which
can be accomplished through the use of potent and selective inhibitors. Trifluoromethyl ketone
(TFK)-containing compounds are the most potent esterase inhibitors described to date. The
inclusion of a thioether moiety â to the carbonyl further increased TFK inhibitor potency. In
this study, we have synthesized the sulfone analogues of a series of aliphatic and aromatic
substituted thioether TFKs to evaluate their potency and solubility properties. This structural
change shifted the keto/hydrate equilibrium from <9% hydrate to >95% hydrate, forming
almost exclusively the gem-diol. These new compounds were evaluated for their inhibition of
carboxylesterase activity in three different systems, rat liver microsomes, commercial porcine
esterase, and juvenile hormone esterase in cabbage looper (Trichoplusia ni) hemolymph. The
most potent inhibitor of rat liver carboxylesterase activity was 1,1,1-trifluoro-3-(decane-1-
sulfonyl)-propan-2,2-diol, which inhibited 50% of the enzyme activity (IC50) at 6.3 ( 1.3 nM
and was 18-fold more potent than its thioether analogue. However, the sulfone derivatives
were consistently poorer inhibitors of porcine carboxylesterase activity and juvenile hormone
esterase activity, with IC50 values ranging from low micromolar to millimolar. The compound
1,1,1-trifluoro-3-(octane-1-sulfonyl)-propan-2,2-diol was shown to have a 10-fold greater water
solubility than its thioether analogue, 1,1,1-trifluoro-3-octylsulfanyl-propan-2-one (OTFP). These
novel compounds provide further evidence of the differences between esterase orthologs,
suggesting that additional development of esterase inhibitors may ultimately provide a battery
of ortholog and/or isoform selective inhibitors analogous to those available for other complex
enzyme families with overlapping substrate specificity.
Each member of this group of enzymes contains a similar
core comprised of an R/â-sheet of eight â-sheets connected
by R-helices. The R/â-hydrolase family was first described
by Ollis et al. and covers a broad range of enzymes that
share similarities in secondary and tertiary structures
(1). These enzymes are related by divergent evolution and
In tr od u ction
Esterases are R/â-fold hydrolase enzymes that catalyze
the hydrolysis of both endogenous and exogenous esters.
* To whom correspondence should be addressed. Phone: (530)
752-7519. Fax: (530) 752-1537. E-mail: bdhammock@ucdavis.edu.
10.1021/tx015508+ CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/27/2001

1564 Chem. Res. Toxicol., Vol. 14, No. 12, 2001
Wheelock et al.
include lipases, esterases, cholinesterases, haloalkane
dehalogenases, and epoxide hydrolases (2). The sequence
identities of esterases of the R/â-hydrolase family vary
widely, but their mechanisms are all centered around a
catalytic triad of which the most common set is Ser-His-
Glu(Asp) (3). Recent work has shown the potential for a
second serine as a fourth catalytic residue (4). Together,
this group of enzymes plays an essential role in the
activation, metabolism, and detoxification of numerous
compounds of biological importance.
Carboxylesterases (EC 3.1.1.1) are a subset of the R/â-
hydrolase family and have very broad substrate selectiv-
ity. They are B-esterases under the nomenclature of
Aldridge by virtue of their inhibition by organophos-
phates (5) and are characterized by their catalysis of the
hydrolysis of a carboxylester to the corresponding alcohol
and acid (6). Carboxylesterases are important in the
metabolism of many exogenous compounds including
pesticides and pharmaceuticals, but their endogenous
role is not well understood (7). Carboxylesterase-medi-
ated hydrolysis is used in the design of many drugs and
pharmacophores. For example, the hydrolysis of ester-
containing prodrugs has been employed in the develop-
ment of the chemotherapeutic agents CPT-11 (8, 9) and
10-hydroxycamptothecin fatty acid esters (10), and in the
activation of the blood cholesterol drug lovastatin (11).
In addition, carboxylesterases are of importance in the
transesterification of meperidine (Demerol) and meth-
ylphenidate (Ritalin) (12), and are essential in the
hydrolysis of cocaine and heroin (13, 14), as well as
aspirin (15, 16).
tion is well understood, and they appear to have little to
no exogenous function (28). J HEs hydrolyze insect juve-
nile hormone (J H), which in turn regulates molting in
some insects (29). J HE is a key enzyme in the develop-
ment of many common crop pests and is a target of
biological pesticides (30). Molting states of insects can
be disrupted through J HE inhibition, resulting in alter-
ation of larval feeding habits (31). The selectivity of J HE
for the methyl ester of J H is rare among esterases (29).
In particular, the J HEs studied are ineffective at hydro-
lyzing p-nitrophenyl acetate (PNPA), one of the most
ubiquitous esterase substrates used for experimental
determination of esterlytic activity (27).
Several recent esterase reviews have been conducted,
all of which discuss the inhibition of esterases by tri-
fluoromethyl ketones (TFK), a group of extremely potent
transition state analogue inhibitors (1, 2, 7, 32). Polarized
ketones undergo hydration, achieving equilibrium be-
tween their keto and hydrate forms (30). This hydration
shifts the geometry from trigonal planar to tetrahedral,
thus mimicking the transition state of the bound sub-
strate (33) and forming a stable analogue (34). The
binding of an enzyme to the ketone moiety also results
in a tetrahedral geometry. However, due to the lack of a
cleavable bond, the enzyme is inhibited, resulting in the
potency of this group of compounds (30). TFKs have been
extensively studied for their esterase inhibition proper-
ties and have been widely used to probe the catalytic
mechanism of numerous members of the R/â-fold hydro-
lase family (2, 30).
Roe et al. (35) reported the synthesis of a new TFK
derivative of increased potency and water solubility. This
new compound reputedly had a 3-fold increase in J HE
inhibition potency over 1,1,1-trifluoro-3-octylsulfanyl-
propan-2-one (OTFP, 2a ) in the lepidopteran cabbage
looper (Trichoplusia ni) (35). Roe et al. achieved this
increase in inhibition potency by oxidizing the sulfur
atom â to the carbonyl of the TFK to the corresponding
sulfone. This chemical change also served to drive the
gem-diol/ketone equilibrium to favor the gem-diol. The
combination of these two effects should drastically in-
crease the water solubility of the compound. It therefore
appeared to be an anomaly, a compound of increased
potency and increased water solubility. The most common
potent inhibitors of J HE are long-chain aliphatic TFK
derivatives that are hydrophobic compounds with log P
values on the order of 3-5 (30). Therefore to investigate
this phenomenon, we synthesized a series of sulfone
analogues from a set of thioether-containing TFKs and
evaluated their inhibition potency against juvenile hor-
mone esterase activity in Trichoplusia ni hemolymph to
compare with Roe et al. (35). We then investigated the
ability of these new compounds to inhibit esterlytic
activity in rat liver microsomal preparations and crude
commercial porcine liver esterase. Last, we assessed the
water solubility of the new sulfone compounds versus
their thioether analogues.
Carboxylesterases are also important in the hydrolysis
and subsequent detoxification of pyrethroid (17), car-
bamate (18) and organophosphate (19) insecticides. Car-
boxylesterases detoxify carbamates and organophos-
phates by hydrolyzing ester-containing bonds and by
acting as an inhibitor sink, thus preventing these com-
pounds from inhibiting acetylcholinesterase (20). In
mammals, the carboxylesterase-catalyzed hydrolysis of
pyrethroids is one of the main routes of detoxification of
these compounds, along with oxidase activity, and is
responsible for their relatively low mammalian toxicity
(21). In addition, inhibition of carboxylesterases has been
shown to synergize both pyrethroid and organophosphate
toxicity (20, 22). One of the major strategies of pesticide
development exploits the fact that mammals contain
higher levels of carboxylesterases compared to insects
and are thus able to more quickly detoxify and excrete
the compounds (21). Carboxylesterases are also of im-
portance in pest resistance management, with pyrethroid
(23), carbamate (24) and organophosphate (25, 26) resis-
tant strains of many insects showing elevated levels of
carboxylesterase.
J uvenile hormone esterases (J HEs)1 are insect car-
boxylesterases that are highly substrate selective (27).
Unlike other carboxylesterases, their endogenous func-
1
Abbreviations: BSA, bovine serum albumin; DMF, dimethylfor-
mamide; DMP, Dess-Martin periodinane; ESI-MS, electrospray ion-
ization MS; FID, flame ionization detector; IC50, 50% inhibition
Exp er im en ta l P r oced u r es
concentration; J H, juvenile hormone; J HE, juvenile hormone esterase;
m-CBA, m-chlorobenzoic acid; m-CPBA, m-chloroperoxybenzoic acid;
OTFP, 1,1,1-trifluoro-3-octylsulfanyl-propan-2-one; OTFPdOHSO2, 1,1,1-
trifluoro-3-(octane-1-sulfonyl)-propan-2,2-diol; OTFPmOH, 1,1,1-tri-
fluoro-3-octylsulfanyl-propan-2-ol; OTFPmOHSO2, 1,1,1-trifluoro-3-
(octane-1-sulfonyl)-propan-2-ol; PNPA, p-nitrophenyl acetate; PTU,
phenylthiourea; QSAR, quantitative structure-activity relationship;
sat, saturated; TFK, trifluoromethyl ketone; Trichoplusia ni, T. ni.
CAUTION! The hydroxyiodinane oxide precursor of the Dess-
Martin periodinane has been reported to be explosive on impact
or in conditions over 200 °C (36, 37). We found that Dess-Martin
periodinane purchased from T. C. I. America Inc. was stable but
that synthesis of the compound always resulted in at least a trace
amount of the precursor oxide, which was indeed explosive.

Synthesis and Evaluation of Carboxylesterase Inhibitors
Chem. Res. Toxicol., Vol. 14, No. 12, 2001 1565
Ta ble 1. P h ysica l a n d Str u ctu r a l Da ta for All Syn th esized Com p ou n d s
melting
point (°C)
IR
compd
GC-MS m/z (relative intensity)
1H NMR
19F NMR
(neat, cm-1)
1b
85-87
261 (39%, [M-OH]+);
δ 0.90 (t, 3H, J ) 6.45 Hz),
1.32 (br m, 4H),
1.46 (m, 2H),
1.86 (m, 2H),
3.33 (m, 2H),
-87.31 (s, 97%)
-79.21 (s, 3%)
1123.7,
1189.5,
1310.4,
2868.2,
2953.5,
3443.3,
3489.1
1133.2,
1179.4,
1303.9,
2861.0,
2927.7,
3439.4,
3523.9
1124.3,
1189.0,
1309.9,
2860.0,
2926.7,
3435.9,
3491.8
(C9H17F3O4S)
177 (3%, [M-C6H13O]+);
85 (63%, [M-C3H4F3O4S]+);
69 (76%, M-C4H8F3O4S]+);
55 (100%, [M-C5H10F3O4S]+)
3.39 (s, 2H),
4.90 (br s, 1.5H, D2O exchangeable)
δ 0.88 (t, 3H, J ) 6.45 Hz),
1.29 (br m, 8H),
1.45 (m, 2H),
1.87 (m, 2H),
3.33 (m, 2H),
3.39 (s, 2H),
4.93 (br s, 1.5H, D2O exchangeable)
δ 0.88 (t, 3H, J ) 6.45 Hz),
1.27 (br m, 12H),
1.46 (m, 2H),
2b
87-88
289 (11%, [M-OH]+);
-87.29 (s, 97%)
-79.21 (s, 3%)
(C11H21F3O4S)
177 (6%, [M-C8H17O]+);
113 (2%, [M-C3H2F3O3S]+);
69 (100%, [M-C6H12F3O4S]+);
55 (58%, [M-C7H14F3O4S]+)
3b
waxy solid 317 (10%, [M-OH]+);
177 (2%, [M-C10H21O]+);
-87.31 (s, 96%)
-79.21 (s, 4%)
(C13H25F3O4S)
205 (5%, [M-C3H4F3O2]+);
111 (17%, [M-C5H10F3O4S]+);
97 (44%, [M-C6H12F3O4S]+);
83 (97%, [M-C7H14F3O4S]+);
69 (82%, [M-C8H16F3O4S]+);
55 (100%, [M-C9H18F3O4S]+)
waxy solid 345 (21%, [M-OH]+);
177 (2%, [M-C12H25O]+);
1.87 (m, 2H),
3.34 (m, 2H),
3.39 (s, 2H),
4.88 (br s, 1.5H, D2O exchangeable)
4b
δ 0.88 (t, 3H, J ) 6.45 Hz),
1.26 (br m, 16H),
1.43 (m, 2H),
1.87 (m, 2H),
3.33 (m, 2H),
-87.31 (s, 95%)
-79.21 (s, 5%)
1133.2,
1179.4,
1310.4,
2854.4,
2924.6,
3461.0,
3497.9
(C15H29F3O4S)
233 (4%, [M-C3H4F3O2]+);
125 (12%, [M-C6H12F3O4S]+);
111 (60%, [M-C7H14F3O4S]+);
97 (96%, [M-C8H16F3O4S]+);
83 (84%, [M-C9H18F3O4S]+);
69 (100%, [M-C10H20F3O4S]+);
55 (42%, [M-C11H22F3O4S]+)
3.39 (s, 2H),
4.88 (br s, 1.5H, D2O exchangeable)
5b
96-98
105 (100%, [M-C3H5F3O4S]+);
91 (5%, [M-C4H7F3O4S]+);
77 (16%, [M-C5H9F3O4S]+)
δ 3.18 (m, 2H),
3.30 (s, 2H),
3.64 (m, 2H),
4.85 (br s, 1.5H, D2O exchangeable),
7.32 (m, 5H)
-87.29 (s, 95%)
-79.21 (s, 5%)
1121.4,
1188.7,
3430.1,
3503.9
(C11H13F3O4S)
Sch em e 1. Syn th esis Sch em e for Su lfon e
Ch em ica ls. Reagents were purchased from Aldrich Chemical
Co. (Milwaukee, WI) unless otherwise noted and were used
without further purification. Initial purity of all synthesized
compounds was assessed by 10 cm F254 silica TLC plates (EM
Science; Gibbstown, NJ ) using either a chloroform/ethanol (95:
5) or hexane/ethyl acetate (70:30) mixture and visualized with
phosphomolybdic acid and heat. Purity was assayed, in part,
by the number of spots responding to phosphomolybdic acid and
by a lack of UV sensitive spots at 254 nm, confirming removal
of UV-active impurities. To further gauge purity, dilutions were
analyzed on a Hewlett-Packard 5890A GC (Palo Alto, CA)
equipped with a flame ionization detector (FID) and a 15 m ×
0.25 mm i.d., 0.25 µm DB-5 column (J &W Scientific; Folsom,
CA). The temperature program was initiated at 100 °C for a 1
min hold and then ramped at 20 °C/min to 300 °C and held for
5 min. All compounds were of a purity of >97% by GC and NMR
and were one spot on TLC. Due to the hygroscopic nature of
these compounds, it was not possible to acquire accurate
elemental analyses. This problem was also reported by Nair et
al. (38). Melting point were determined with a Thomas-Hoover
apparatus (A. H. Thomas Co., Philadelphia, PA) and are
uncorrected.
Der iva tives of Tr iflu or om eth yl Keton es (TF K)a
Structural support was provided by 1H NMR, 13C NMR, 19F
NMR (Varian 300 MHz; Palo Alto, CA), IR (Mattson Galaxy
Series FTIR 3000, Madison, WI), and a Saturn 4D Ion Trap
mass spectrometer coupled to a Cx3400 GC (Varian; Sugarland,
Texas) and is listed in Table 1. Electrospray mass spectral data
was collected on a Quatro BQ (Fisons; Altracham, England).
Ch em ica l Syn th esis Of Su lfon e Der iva tives, Meth od A.
Two different synthetic routes were used to synthesize the
sulfone derivatives and are outlined in Scheme 1. Method A is
a new synthetic procedure that reduces the synthesis scheme
of Roe et al. (35) to two steps. The synthesis of 1,1,1-trifluoro-
3-(octane-1-sulfonyl)-propan-2,2-diol (OTFPdOHSO2, 2b) is given
as an example. First, 1,1,1-trifluoro-3-octylsulfanyl-propan-2-
a
Method A is a novel procedure developed for this paper.
Method B is based upon the work of Roe et al. (35). Compounds
are numbered in two sets, all thioether TFKs are referred to as
“thioether”, and all sulfone TFK derivatives are “sulfone”. DMP
refers to the Dess-Martin periodinane (36).
one (OTFP, 2a ) was synthesized with methods of Ashour et al.
(39). Briefly, 3-bromo-1,1,1-trifluoroacetone (75 mmol, 7.8 mL)
was added to a stirring solution of octylthiol (75 mmol, 13 mL)
in CCl4 (10 mL). The reaction was allowed to proceed under a

1566 Chem. Res. Toxicol., Vol. 14, No. 12, 2001
Wheelock et al.
gentle stream of N2 and was monitored by GC/FID. The reaction
was considered to be complete at 24 h (75% completion) and 25
mL of diethyl ether was added. The entire mixture was then
washed with 2 × 25 mL of aqueous sodium bicarbonate
(saturated, sat), and the aqueous phase was extracted with 2 ×
25 mL of diethyl ether. The organic phases were collected, dried
over magnesium sulfate, and filtered, and the solvent was
stripped by rotary vacuum evaporation. The crude mixture was
vacuum distilled (bp 85 °C, 1.5 Torr) to yield the purified product
(>97% purity by GC): 1H NMR (CDCl3) δ 0.87 (t, 3H, J ) 6.45
Hz), 1.26 (br m, 10H), 1.57 (m, 2H), 2.50 (t, 2H, J ) 7.04 Hz),
3.48 (s, 2H); 19F NMR (CDCl3) δ -76.22 (s, 91%), -85.83 (s,
90%); IR (neat) cm-1 1156.5, 1204.5, 1750.3, 2861.0, 2929.1;
single spot by TLC (hexanes:ethyl acetate, 70:30).
55.28, 66.33 (q, J CF ) 32 Hz); 123.98 (q, J CF ) 282 Hz); 19F NMR
(CDCl3) δ -79.92 (d, J HF ) 8.4 Hz); single spot by TLC
(chloroform:ethanol, 95:5); ESI-MS m/z [M - H]- (C11H20F3O3S)
) 289.1, observed 289.2.
Compound 2b was formed by the addition of OTFPmOHSO2
(0.79 mmol, 0.229 g) to a stirring solution of the Dess-Martin
periodinane (0.16 M; T. C. I. America, Portland, OR) in dichlo-
romethane, which was further stirred for 1 h followed by the
addition of 20 mL of diethyl ether. The mixture was then
extracted with 60 mL of sodium thiosulfate (0.26 M) in sodium
bicarbonate (sat). The organic fraction was washed with 2 × 25
mL of sodium bicarbonate (sat) and the combined aqueous
fractions were extracted with 2 × 50 mL of diethyl ether, dried
over magnesium sulfate, filtered, and stripped. The purified
product was recrystallized from 1-chlorobutane to give white
needlelike crystals (0.28 mmol, 0.084 g, 36% yield). Physical data
were identical as for compound 2b synthesized via Method A.
Compound 2b was formed by adding 2a (1.12 mmol, 0.286 g)
to a stirring solution of m-chloroperoxybenzoic acid (m-CPBA,
70-75%, 1.4 g) in dichloromethane. After stirring for 48 h, the
reaction was worked up using methods adapted from Camps et
al. (40). Sodium fluoride (23.8 mmol, 1.0 g), which had been
activated by heating at 100 °C for 24 h under vacuum (85 Torr),
was added to the reaction mixture followed by Celite to create
a slurry, which was then filtered through a Celite cake to remove
the m-chlorobenzoic acid (m-CBA). It was also possible to use
extensive sodium bicarbonate (sat) washes to remove the
m-CBA. The mixture was then washed with 2 × 50 mL of
aqueous sodium bicarbonate (sat) and the aqueous phases were
extracted with 2 × 50 mL of diethyl ether. The reaction wash
was monitored by GC/FID for the presence of m-CBA following
the reaction and additional sodium bicarbonate (sat) washes
were used as necessary to extract any residual acid. Once all
acid had been removed, the reaction mixture was dried with
magnesium sulfate, filtered, and stripped. The pure product was
recrystallized from 1-chlorobutane to give white needlelike
crystals (1.05 mmol, 0.320 g, 94% yield): mp 87-88 °C; 1H NMR
(CDCl3) δ 0.88 (t, 3H, J ) 6.45 Hz), 1.29 (br m, 10H), 1.87 (m,
2H), 3.33 (m, 2H), 3.39 (s, 2H), 4.79 (br s, 1.5H, D2O exchange-
able); 13C NMR (CDCl3) δ 13.96, 21.72, 22.53, 28.24, 28.41, 28.88,
31.64, 52.57, 56.27, 91.88 (q, J CF ) 33 Hz), 121.78 (q, J CF ) 286
Hz); 19F NMR (CDCl3) δ -79.21 (s, 3%), -87.29 (s, 97%); IR
(neat) cm-1 1133.2, 1179.4, 1303.9, 2861.0, 2927.7, 3439.4,
3523.9; single spot by TLC (chloroform:ethanol, 95:5); Electro-
spray ionization MS (ESI-MS) m/z calculated for [M - H-H2O]-
(C11H18F3O3S) ) 287.1, observed 287.4.
J u ven ile Hor m on e Ester a se Activity. J HE assays were
conducted with J HE from Trichoplusia ni (T. ni). Enzyme
preparations were collected from larvae of T. ni (Entopath;
Easton, PA), which had reached day two of the fifth instar.
Larvae were placed on ice to facilitate handling. After 5 min on
ice, a single proleg was cut to allow hemolymph to well from
the body, which was then harvested as described in Roe et al.
(35). Briefly, hemolymph was collected into a capillary tube and
transferred into microfuge tubes containing ∼0.5 mmol of
phenylthiourea (PTU) on ice. The hemolymph was then centri-
fuged for 5 min at 1000g and the supernatant was collected and
transferred into a microfuge tube containing 0.1 M phosphate
buffer (pH 7.4) and 0.01% PTU for a final 1:1 dilution of
hemolymph in buffer. The diluted hemolymph had a specific
activity of 1.0 nmol of juvenile hormone III/min/mg of protein
and was stored at -80 °C until use.
J HE activity was measured with a partition assay using
3[H]juvenile hormone III (3[H]J H III) (42). Substrate was
prepared by mixing 80 µL of radiolabeled J H III (10-20
Ci/mmol, 0.1 mCi/mL, >97% pure; New England Nuclear,
Boston, MA) with 20 µL of unlabeled J H III (25 mM; Sigma
Chemical Co., St Louis, MO) in hexane. The solvents were
removed by heating to ∼35 °C under a gentle stream of N2,
then 1 mL of ethanol was added to give a final J H III
concentration of 0.5 mM. Hemolymph preparations were diluted
300-fold in sodium phosphate buffer (pH 7.4, 0.1 M, 0.1 mg/mL
bovine serum albumin, BSA) and incubated with inhibitors for
10 min at 30 °C. All inhibitor stock solutions were prepared in
dimethylformamide (DMF) and the total volume of DMF added
to the assay never exceeded 1%, which control studies showed
had little effect upon J HE activity. The substrate J H III was
then added and vortexed for a final concentration of 5 µM, and
the solution was incubated for 15 min at 30 °C. Following
incubation, the mixture was extracted with 50 µL of a solution
of methanol:water:ammonium hydroxide (10:9:1) and 250 µL of
isooctane, followed by vigorous vortexing for 10 s to form an
emulsion. Samples were centrifuged for 5 min at 1000g and the
radioactivity in 50 µL of the aqueous phase was measured by a
Wallac 1409 liquid scintillation counter (Wallac; Turku, Finland)
according to Hammock and Sparks (42).
Ch em ica l Syn th esis of Su lfon e Der iva tives, Meth od B.
The second synthetic method was based upon procedures of Roe
et al. (35). It was however necessary to significantly alter those
procedures as published. The synthesis of compound 2b is given
as
a sample synthesis for Method B. Compound 2a was
synthesized as described above. 1,1,1-Trifluoro-3-octylsulfanyl-
propan-2-ol (OTFPmOH) was formed from methods adapted
from Linderman and Graves (41). Lithium aluminum hydride
(1.85 mmol, 3.7 mL) was added to a stirred solution of 2a (4.0
mmol, 1.02 g) in diethyl ether at 0 °C, followed by stirring for 1
h after which the reaction was quenched with 10 mL of aqueous
sulfuric acid (5%). The mixture was washed with 2 × 20 mL of
aqueous sodium bicarbonate (sat), followed by extraction with
2 × 20 mL of diethyl ether. The organic phase was dried over
magnesium sulfate, filtered, and stripped to give 90% yield by
GC/FID (3.7 mmol, 0.933 g). 1H NMR (CDCl3) δ 0.87 (t, 3H, J
) 6.45 Hz), 1.27 (br m, 12H), 1.62 (m, 2H), 2.58 (m, 2H), 3.09
(br s, 1H, D2O exchangeable), 4.03 (m, 1H); single spot by TLC
(hexanes:ethyl acetate, 70:30).
Ma m m a lia n Ca r boxylester a se Activity. Two sources of
mammalian carboxylesterase were used to screen activity, rat
liver microsomal preparations, and crude porcine liver carboxyl-
esterase. Assays were conducted in 96-well microtiter styrene
flat bottom plates (Dynex Technologies, Inc.; Chantilly, VA) at
30 °C using methods of Huang et al. (43). Inhibitor stock
solutions were prepared in DMF, and concentrations were
chosen such that DMF levels never exceeded 1%. Control studies
showed very little effect of DMF upon enzyme activity at these
levels. However, all samples were corrected for any potential
solvent effects. Enzyme was incubated with the inhibitor for 5
min and then substrate (p-nitrophenyl acetate, PNPA) was
added for a final concentration of 0.5 mM. After substrate
addition, the plate was read kinetically for 2 min with optical
density monitored at 405 nm for the appearance of the p-
1,1,1-Trifluoro-3-(octane-1-sulfonyl)-propan-2-ol
(OTFP-
mOHSO2) was formed by adding OTFPmOH (3.7 mmol, 0.933
g) to a stirred solution of m-CPBA (70-75%, 2.6 g) in dichlo-
romethane. After stirring for 48 h, the product was worked up
as for the 2b synthesis in Method A. The pure product was
recrystallized from 1-chlorobutane to give white needlelike
crystals (2.33 mmol, 0.677 g, 65% yield): mp 80-81 °C; 1H NMR
(CDCl3) δ 0.87 (t, 3H, J ) 6.45 Hz), 1.27 (br m, 10H), 1.84 (m,
2H), 3.16 (m, 2H), 3.36 (m, 2H), 3.99 (s, 1H), 4.64 (s, 1H);13
NMR (CDCl3) δ 14.00, 21.68, 22.53, 28.29, 28.92, 31.64, 52.20,
C

Synthesis and Evaluation of Carboxylesterase Inhibitors
Chem. Res. Toxicol., Vol. 14, No. 12, 2001 1567
nitrophenolate anion hydrolysis product (Spectramax 200; Mo-
lecular Devices, Sunnyvale, CA).
TFK-containing compounds have been synthesized, and
the structure-activity relationship for this group of
compounds has been extensively explored (30). It was
found that placing a sulfur atom â to the carbonyl greatly
increased the potency of the inhibitor. Linus Pauling
proposed that sulfur could act as an electronic mimic of
unsaturation (47). This theory was originally applied to
J HE where it was thought that the sulfur would mimic
the double bond present in the natural substrate juvenile
hormone, thus creating an inhibitor that better mimicked
the substrate (48). Roe et al. further explored this
chemistry by oxidizing the sulfur to a sulfone (35). We
therefore decided to use this chemistry in the develop-
ment of new esterase inhibitors to study other carboxy-
lesterases in addition to J HE. The inclusion of a sulfone
moiety was intriguing to us, as it appeared to offer both
increased potency and water solubility.
(A) Syn th esis. A series of TFK sulfone analogues was
synthesized for evaluation as inhibitors of carboxy-
lesterase activity. A single example of this compound
class was previously reported by Roe et al. (35). We report
here the same synthesis in two steps with significantly
increased yield and the elimination of a potentially
hazardous reagent (Method A, Scheme 1). Due to the
inability to efficiently remove the m-CBA acid after direct
oxidation of the parent thioether TFK with m-CPBA, Roe
et al. abandoned this reaction for a considerably more
difficult reaction scheme (Method B, Scheme 1). However,
we found that by using methods of Camps et al. (40), we
were able to remove all of the m-CBA. The cleanup step
following the oxidation was the key to successful direct
oxidation of the thioether to the sulfone. Camps et al.
achieved the removal of m-CBA by precipitating the acid
with sodium fluoride. This technique followed by exten-
sive washing with sodium bicarbonate (sat) served to
remove all acid from the system. Sodium carbonate (sat)
was not any more effective at m-CBA removal and
attempts to use a stronger base (such as sodium hydrox-
ide) resulted in decomposition of the molecule. By em-
ploying the methods of Camps et al., we successfully
eliminated all m-CBA from the direct oxidation allowing
for the isolation of the product by recrystallization in high
yield (94%). We then successfully applied this simplified
procedure to four additional compounds.
To ensure that the product identity was consistent
through this modified procedure, we also repeated the
synthesis of OTFPdOHSO2 (2b) using methods of Roe et
al. (35) (Method B in Scheme 1). Overall yields for this
route of synthesis were generally low due to the large
number of steps required. Roe et al. reported a 57%
overall yield from 2a to 2b, and we achieved a 22% yield.
In addition, we found it necessary to modify several
points in the procedure in order to successfully complete
the reaction. Again, it was necessary to perform a cleanup
step to remove all m-CBA following oxidation of OTFP-
mOH to OTFPmOHSO2. Roe et al. reported that they
were able to wash away all of the acid byproduct with
sodium bicarbonate (sat). We found that washing alone
was not sufficient to remove all of the acid, unless an
extremely large number of washes (>10) was used.
Precipitation of m-CBA with sodium fluoride was a much
more successful method. We also had to alter the DMP
reaction conditions. In particular, Roe et al. used a
reaction time of 3 h to oxidize OTFPmOHSO2 to OTFP-
dOHSO2. We found that 3 h was too long and that
reaction times greater than 1 h resulted in decomposition
Rat liver microsomal carboxylesterase activity assays were
performed with solubilized rat liver microsomes prepared by
methods adapted from Huang et al. (43, 44). Eight-week-old
Fisher 344 rats were sacrificed by carbon dioxide asphyxiation,
and livers were excised and perfused with a 1% KCl solution to
remove all blood, which contains high levels of serum esterases
(7). Livers were then pooled, homogenized on ice with a Polytron
homogenizer (Brinkmann Instruments; Westbury, NY), and
centrifuged at 10000g for 20 min at 4 °C. Supernatant fractions
were further centrifuged at 100000g for 60 min. The resulting
pellet was resuspended in a sodium phosphate buffer (pH 7.4,
0.1 M) to give a final protein concentration of 33.5 mg/mL and
frozen at -80 °C. Before use in enzyme assays, microsomes were
solubilized with 1% n-octyl â-D-glucopyranoside (Sigma Chemi-
cal Co.) and diluted with sodium phosphate buffer (pH 7.4, 0.1
M) to a final protein concentration of 4.2 mg/mL. Samples were
then incubated for1 h, followed by centrifugation at 100000g
for 1 h, and the supernatant was aliquoted and frozen at -80
°C. For assays, liver microsomal preparations were diluted to a
final protein concentration of 0.8 mg/mL in 0.1 M sodium
phosphate buffer (pH 7.4) with a specific activity of 67 µmol of
p-nitrophenyl acetate/min/mg protein. Assays were conducted
with 32 mg of protein in a total volume of 200 µL. Protein
concentrations were determined using methods of Bradford with
BSA as a standard (45).
Crude porcine liver carboxylesterase (EC 3.1.1.1) was pur-
chased from Sigma Chemical Co. (catalog no. E-3019). Assays
were conducted in a similar format as for rat liver carboxy-
lesterase activity with PNPA as the substrate. Working solu-
tions of enzyme were made-up at 0.02 mg of total protein/mL
in sodium phosphate buffer (pH 7.4, 0.1 M) for an activity of
0.3 units/mL (one unit will hydrolyze 1.0 µmol of ethyl butyrate
to butyric acid and ethanol at pH 8.0 at 25 °C). Assays were
conducted with 0.8 mg of protein in a total volume of 200 µL.
En zym e In h ibition Deter m in a tion . The concentration at
which 50% of the enzyme activity was inhibited (IC50) was
determined identically for all three enzymes. IC50s were calcu-
lated from a minimum of five datum points with at least two
points above and below the IC50 in the linear region of the curve.
All assays were run in at least triplicate, with each replicate
consisting of an average of 3 individual determinations.
Solu bility Deter m in a tion . Solubility was determined using
methods of Nellaiah et al. (46), which determine the solubility
of the compound based upon its turbidity. Compounds were
dissolved in DMF and 10 µL of solution was added to a cuvette
containing 1 mL of a sodium phosphate buffer (pH 7.4, 0.1 M)
at 25 °C. DMF concentrations therefore never exceeded 1% of
the total volume. The absorbance was read at 700 nm on a Cary
100 Bio Spectrophotometer (Varian, Walnut Creek, CA). The
concentration of inhibitor that resulted in a significant increase
(>4-fold) in absorbance was interpreted as exceeding the
solubility of the compound. Solubilities are therefore reported
as a range between the concentration that exceeded solubility
and the point immediately preceding it.
Resu lts a n d Discu ssion
Trifluoromethyl ketone-containing compounds are very
potent inhibitors of a large number of esterases from a
wide variety of sources across different phyla (7). It is
generally thought that the trifluoromethyl moiety en-
hances the activity of the inhibitor by further polarizing
the ketone, thereby enhancing nucleophilic attack by the
nucleophilic hydroxyl group of the serine residue in the
active site of the enzyme (30). The extreme potency of
TFKs is attributed to their ability to act as transition-
state analogues. They mimic the structure of the transi-
tion-state complex through a theorized tetrahedral ge-
ometry in the bound inhibitor (29). A number of different

1568 Chem. Res. Toxicol., Vol. 14, No. 12, 2001
Ta ble 2. Com p ou n d s Assa yed for Ester a se In h ibition Rep or ted a s IC50 (10-9 M)a
Wheelock et al.
juvenile hormone esterase activity IC50
rat carboxylesterase activity IC50
ratio
porcine carboxylesterase activity IC50
ratio
ratio
TFKdOHSO2/
TFKdOHSO2/
TFK
TFKdOHSO2/
R
TFK
TFKdOHSO2
TFK
TFK
TFKdOHSO2
TFK
TFKdOHSO2
TFK
hexyl
octyl
decyl
dodecyl
phenethyl
31a
1.6 ( 0.1
20a
354 ( 20
147 ( 89
119 ( 23
191 ( 9.0
1780 ( 570
11
92
6
54.6 ( 2.2 1290 ( 90
37.2 ( 2.1 24.5 ( 3.7
23
0.66
0.05
0.06
9
54.3 ( 6.3 39 000 ( 3900
36.8 ( 2.4 5640 ( 540
135 ( 8.0 1420 ( 53.0
719
152
11
4
52
115 ( 10
164 ( 50
6.3 ( 1.3
15a
13
10.3 ( 2.6
162 ( 17 1500 ( 100
119 ( 16
473 ( 26
NDb
75.3 ( 5.6 3390 ( 210
a
b
IC50 values taken from Szekacs et al. (30). Not determined. Compounds are composed of two different schemes as shown below,
trifluoromethyl ketone (TFK) or dihydroxysulfone trifluoromethyl ketone (TFKdOHSO2), with a range of substituents (R).
of the molecule. It was necessary for Roe et al. to
synthesize the DMP, since it was not commercially
available at the time of their work. It is therefore possible
that the DMP used by Roe et al. was of lower purity than
that now available through T. C. I. America, which could
account for the difference in reaction times. However, we
successfully carried out the reaction with commercial as
well as synthetic DMP.
observation further points to the dynamic nature of the
keto/hydrate equilibrium.
(B) En zym e In h ibition . We evaluated the inhibitory
potency of these new compounds, reported as the IC50,
in three different enzyme systems. J HE activity was
examined in order to compare results with data reported
by Roe et al. (35), and rat and porcine carboxylesterases
were used to examine inhibition potency in mammalian
systems. Rat liver carboxylesterases have been exten-
sively studied by several research groups (7, 19) and
crude porcine liver carboxylesterase is a good benchmark
esterase due to its commercial availability (Sigma Chemi-
cal Co.). All enzyme activity data are presented in Table
2. Results are also expressed as the ratio of the IC50 of
the sulfone analogue to the IC50 of the thioether com-
pound for each enzyme. These ratios are not kinetic
values and are only valid for comparisons among a given
data set, but are useful for comparing relative differences
in potency between sulfone and thioether compounds. It
is also important to note that inhibitor potency can vary
with the solvent used for the assay.
The oxidation of the thioether in compound 2a to the
sulfone affected the hydration state of the ketone. The
IR of 2a showed stretching at 1750.3 cm-1, typical for a
polarized ketone, and there was no stretching in the
3000-3500 cm-1 range where a hydroxyl group would
be expected to absorb. However, in compound 2b, the IR
showed no stretching in the ketone window and 2 distinct
broad bands at 3439.4 and 3523.9 cm-1. It is possible that
the sulfone is responsible for shifting the gem-diol/ketone
equilibrium such that the majority of the compound is
hydrated. By 19F NMR, this equilibrium was estimated
to be >95% gem-diol in compound 2b as compared to <9%
gem-diol in compound 2a (Table 1). This phenomenon
could have a large effect upon inhibitor potency by
favoring one hydration state. If the ketone is indeed the
active form of the inhibitor, it will be necessary for the
gem-diol compound to dehydrate back to the ketone
before it binds to the nucleophilic serine residue. This
dehydration step could occur either before the inhibitor
enters the active site or after and could be the rate-
limiting step of inhibition (35).
It is very likely that the ratio and sensitivity of
individual enzyme isoforms varies greatly between the
different species in this study, thus providing for different
biochemical responses for each species. It is therefore
important not to compare individual IC50 values for given
compounds from each species. Instead only overall trends
in rank potency should be compared. It is also important
to remember that all enzyme preparations used in this
study were crude mixtures potentially consisting of
multiple esterase isoforms. We therefore refer to rat and
porcine carboxylesterase as carboxylesterase activity due
to the presence of a mixture of isozymes in the enzyme
preparations. However, several laboratories demon-
strated that T. ni hemolymph contains only one enzyme
(J HE) that is responsible for J H hydrolysis (49, 50).
Additionally, J H is thought to be the natural substrate
for J HE and is not significantly hydrolyzed by other
enzymes in T. ni hemolymph (50), therefore, making it a
good substrate for J HE activity assays. This exceptional
specificity is thought to be due to the R,â-unsaturation
of the methyl ester rendering it stable to both base
hydrolysis and the action of most esterases (51). The
crude porcine carboxylesterase purchased from Sigma is
partially purified (39) as opposed to the rat liver microso-
mal preparation, which undoubtedly consist of multiple
All analytical work conducted supported the structures
as reported (Table 1). Of particular interest are the
melting point data. Compounds 1b, 2b, and 5b had sharp
melting points, but compounds 3b and 4b were waxy
solids with wide melting ranges. These compounds,
however, showed >97% purity by GC/FID, TLC, 1H NMR,
and 19F NMR and did not produce UV sensitive spots on
F254 silica plates (thus indicating the absence of the
m-CBA impurity). In addition, their mass spectral frag-
mentation patterns were consistent with other com-
pounds in this series. It was therefore concluded that the
longer chain compounds did not solidify at room temper-
ature like the shorter chain compounds. This phenom-
enon could be due to the increased lipid-like character
of these longer chain compounds. Also of interest was that
the gem-diol compounds dehydrated during electrospray
ionization mass spectrometry (ESI-MS) analysis. This

Synthesis and Evaluation of Carboxylesterase Inhibitors
Chem. Res. Toxicol., Vol. 14, No. 12, 2001 1569
Ta ble 3. Ca lcu la ted log P Va lu es for Syn th esized
enzymes and several different isoforms of carboxy-
lesterase (7). The substrate PNPA, used for the mam-
malian carboxylesterase studies, is a ubiquitous nonse-
lective substrate that is a very broad indicator of esterlytic
activity. However, it is an excellent substrate for measur-
ing carboxylesterase activity because all carboxylesteras-
es identified to date are capable of hydrolyzing PNPA
(52). In addition, Ashour and Hammock reported that
PNPA is a good substrate for distinguishing between
some protease and carboxylesterase activity (39), as
carboxylesterases are 105 times more efficient at hydro-
lyzing PNPA than proteases such as chymotrypsin (53).
Kao et al. (54) reported that Type A esterases, which
hydrolyze organophosphates, did not hydrolyze PNPA,
and Wallace et al. (19) found that inhibiting carboxy-
lesterase activity in rat microsomes with paraoxon
eliminated 100% of PNPA hydrolysis. PNPA was there-
fore chosen as an appropriate substrate for all mam-
malian carboxylesterase activity assays.
All sulfone compounds were consistently poorer inhibi-
tors of J HE than their thioether analogues. Roe et al.
reported that compound 2b was three times more potent
than 2a against T. ni J HE (35). Our results showed an
opposite effect with compound 2b being almost 100-fold
less potent than 2a . The discrepancies in IC50 results
could be due to differences in insect strains, enzyme
preparation, solvent used for inhibitor dilutions, and
assay conditions between the two laboratories. However,
it was not possible to verify the enzyme preparation
specifics from Roe et al. based upon published informa-
tion. Compound 3b was the most potent sulfone inhibitor
of J HE with an IC50 of 119 ( 23 nM and had a sulfone to
thioether IC50 ratio of six, indicating that the thioether
compound (3a ) was six times more potent than 3b. This
compound was still a relatively weak inhibitor compared
to the most potent inhibitor of J HE, compound 2a with
an IC50 of 1.6 ( 0.1 nM. This value agrees well with
previously published results (27) and further demon-
strates that 2a is one of the most potent J HE inhibitors
synthesized to date.
The results with rat liver carboxylesterase activity
were remarkably different from those of J HE, with three
of the sulfone analogues (compounds 2b, 3b, and 4b)
being substantially more potent than their thioether
analogues (compounds 2a , 3a , and 4a ). The decyl sulfone
derivative, 3b, proved to be the most potent mammalian
inhibitor of all compounds tested with a sulfone to
thioether IC50 ratio of 0.05 and an IC50 of 6.3 ( 1.3 nM
(Table 2). Compound 3b is 15-fold more potent than its
thioether analogue and 6-fold more potent than the
previous best inhibitor of rat carboxylesterase activity,
compound 2a with an IC50 of 37.2 ( 2.1 nM. Previous
studies with TFK inhibitors have shown that long-chain
aliphatic groups are the most potent carboxylesterase
inhibitors (55). This trend holds true with the sulfone
derivatives, with maximum inhibition potency reached
with 10 carbons. The hexyl compound 1b provided a very
interesting result. The IC50 ratio of sulfone analogue to
thioether compound was 23, the largest for any of the
rat carboxylesterase activity inhibitors. The addition of
a sulfone appears to interfere with the binding of the
shorter-chain inhibitors. It is possible that the increased
water solubility of 1b overcomes the hydrophobicity of
the shorter chain hexyl compound, whereas the longer
chain inhibitors still have high enough log P values to
effectively bind to the enzyme. The estimated log P value
Com p ou n d s
no.
ClogPa
no.
ClogPa
1a
1b
2a
2b
3a
3b
3.46
1.17
4.51
2.23
5.57
3.28
4a
4b
5a
5b
6.63
4.34
2.91
0.62
4.35
J H IIIb
a
ClogP indicates log P value calculated using the ClogP
b
program developed by Leo et al. (58). J H III ) juvenile hor-
mone III.
for 1b is 1.17 versus 3.46 for 1a , whereas the values for
3b and 3a are 3.28 and 5.57, respectively (Table 3). It is
possible that there is a minimum threshold log P value
for efficient binding of the inhibitor. Upon the basis of
our inhibition results, this value is probably between 2
and 3. Therefore, it appears that compounds with longer
aliphatic chains have hydrophobicities large enough to
overcome the repulsion effects arising from the addition
of the sulfone moiety. This result is in agreement with
quantitative structure-activity relationship (QSAR) stud-
ies conducted by Sze´ka´cs et al. that determined that
lipophilicity was the most important descriptor, being
responsible for over 60% of the biological activity of the
TFK inhibitors (30).
The crude porcine carboxylesterase exhibited a dra-
matically different response than that of the rat liver
microsomal preparation for the sulfone derivatives, which
were consistently worse inhibitors than the thioether
analogues. Compound 1b was particularly bad with a
sulfone/thioether IC50 ratio of 719, the highest of any of
the compounds and enzymes tested. However, the trends
do mirror those seen for rat carboxylesterase activity,
with potency increasing with the length of the aliphatic
chain. Maximal inhibition for porcine carboxylesterase
activity was achieved with a dodecyl group (4b), which
had a sulfone to thioether IC50 ratio of four. The differ-
ences in activity between rat and porcine carboxy-
lesterase activity probably result from differences in the
physiochemical environment within the active site of the
enzyme, with the porcine enzyme possibly requiring a
more hydrophobic inhibitor. Further studies could com-
pare the steric and electronic fields within the active sites
of the two enzymes to explain the variations in activity.
From our results, we can say that the enzyme pocket
appears to have a hydrophobic character that favors
inhibitors and substrates with high log P values. It is
also interesting that the results for the thioether com-
pounds for the rat and porcine enzymes are remarkably
similar. All of the aliphatic compounds have essentially
the same IC50 value within the accuracy of the assays.
This result also points to the importance of log P in the
inhibition of these enzymes. Compounds with a given log
P value inhibit carboxylesterases from different mam-
malian species with nearly identical potency.
The reasons for the sulfone analogues being consis-
tently less potent inhibitors of T. ni J HE and porcine
carboxylesterase activity than their thioether analogues
are unclear. Without structural evidence, such as a
crystal structure of both enzymes, it is difficult to
compare catalytic mechanisms. There is no structural
information available for T. ni J HE, but Ward et al.
showed that the J HE of Heliothis virescens contains a
catalytic triad comparable to other esterases (Ser-Arg-

1570 Chem. Res. Toxicol., Vol. 14, No. 12, 2001
Wheelock et al.
Glu) (56). It is possible that the sulfone is engaging in
unfavorable secondary interactions within the J HE active
site of T. ni. J HE is highly selective for J H-like molecules,
and even though its catalytic triad is similar in composi-
tion to that of the rat carboxylesterase (Ser-His-Glu) (7),
the overall sequences show very little homology. Given
the selectivity of J HE for methyl esters, it would be
expected that any bulky substituent that interfered with
the placement of the substrate or inhibitor within the
correct distance of the nucleophilic serine could decrease
activity. Thus, it is possible that the relatively large bulk
of the sulfone, compared to the thioether, sterically
hinders the binding of the inhibitor to the active residue,
resulting in lower potency. However, Linderman et al.
showed that compounds containing a methyl group either
R or R′ to the thioether were still very potent inhibitors
of J HE (29). Therefore, the reduction in potency of the
sulfone compounds is probably electronically mediated.
There could potentially be electronic repulsion resulting
from the high electron density of the oxygens of the
sulfone moiety. An additional possibility is that the active
site of J HE is very hydrophobic, resulting in unfavorable
interactions with compounds having low log P values. It
would be necessary to further analyze the position of
these inhibitors within the active site of J HE to ac-
curately predict their potential interactions with all
residues in the active site.
F igu r e 1. Proposed mechanism for intramolecular hydrogen
bond formation for sulfone and thioether trifluoromethyl ketones
(TFKs). Graphic shows the dynamic equilibrium between the
ketone and hydrate forms of the TFKs. The ketone is considered
to be the active form of the inhibitor, which undergoes nucleo-
philic attack by the hydroxy moiety of the esterase (E) catalytic
serine residue (Ser-OH). Hydrogen bonds are shown as a dashed
line. Crystallographic data showed the formation of a hydrogen
bond between the sulfur and hydrated ketone in 1,1,1-trifluoro-
3-phenethylsulfanyl-propan-2-one (57).
There are several possibilities that may explain the
increased potency of the sulfone analogue in the rat liver
microsomal preparation. First, it is possible that the
sulfone is hydrogen bonding with a hydroxy group on the
gem-diol. Although thioethers normally form weak hy-
drogen bonds, crystallographic analysis of 1,1,1-trifluoro-
3-phenethylsulfanyl-propan-2-one showed that the struc-
ture was hydrated and that a tight hydrogen bond formed
between the sulfur in the thioether and one of the
hydroxyl groups in the gem-diol, forming a five-membered
ring (57). It is probable that similar intramolecular
hydrogen bonding is also occurring in the sulfone com-
pound. It is therefore possible that the six-membered ring
of the sulfone compound is energetically favored over that
of the five-membered thioether compound, thus account-
ing for the difference in inhibitor potencies. Following
nucleophilic attack by serine, the enzyme/inhibitor com-
plex would resemble the transition state of the enzyme
with its natural substrate, forming a tetrahedral inter-
mediate. Formation of a six-membered ring would sta-
bilize this intermediate, thus preventing cleavage of the
oxygen-carbon bond between the serine and the inhibitor
(Figure 1). This mechanism would result in a more potent
inhibitor. Also possible is that the sulfone, given its
electron-withdrawing nature, increases the positive po-
larization of the carbonyl carbon and/or shifts the gem-
diol/ketone equilibrium toward the diol, thereby stabi-
lizing the product of a nucleophilic attack on the ketone
form of the sulfone TFKs.
F igu r e 2. Solubility determination of 1,1,1-trifluoro-3-(octane-
1-sulfonyl)-propan-2,2-diol (OTFPdOHSO2, 2b ) and 1,1,1-tri-
fluoro-3-octylsulfanyl-propan-2-one (OTFP, 2a ). Compounds
were dissolved in dimethylformamide and 10 µL of solution was
added to 1 mL of sodium phosphate buffer (pH 7.4, 0.1 M) in a
cuvette at 25 °C. The turbidity of the solution was then
measured by reading the absorbance at 700 nm. Compound
solubility was determined by a significant increase in absorption
(>4-fold) over baseline values (n ) 3).
the regeneration of catalytically active enzyme (27, 55).
Enzyme eluted from an affinity gel is often still inhibited
and remains so even after extensive dialysis. The use of
more potent inhibitors with increased water solubility
could greatly assist in the diffusion of the inhibitors out
of the active site of the enzyme and through the dialysis
membrane, regenerating catalytically active enzyme. It
may is also be possible to use higher concentrations of
weaker inhibitors to elute the enzymes from the affinity
column.
Carboxylesterases are important in the hydrolysis and
metabolism of numerous ester-containing compounds.
Given their role in the activation of prodrugs, metabolism
of pharmaceuticals and xenobiotics, and detoxification of
pesticides, it is essential that we understand the mech-
anism by which these enzymes function. Toward this end
we have synthesized four new inhibitors that will pro-
vide tools to explore key differences in the catalytic
mechanism of different carboxylesterases. These new
inhibitors have ortholog specific potency and could prove
useful for distinguishing specific esterase isoforms. These
compounds could also assist in the affinity purification
of these essential enzymes. Further research should
address the mechanism behind the variation in potency
of these different compounds. Analysis of the key phys-
iochemical parameters responsible for carboxylesterase
activity will greatly increase our understanding of these
enzymes.
Solubility studies with 2a and 2b demonstrated that
the oxidation of the thioether to the sulfone increases
water solubility. The water solubility of 2a was between
0.2 and 0.3 mM whereas the solubility of 2b was between
1.5 and 2.0 mM. Compound 2b is therefore approximately
10 times more soluble under the conditions examined
(Figure 2). An inhibitor that is more potent and water-
soluble would be very useful for applications in affinity
chromatography. One difficulty often encountered in
affinity purification of esterases with TFK inhibitors is

Synthesis and Evaluation of Carboxylesterase Inhibitors
Chem. Res. Toxicol., Vol. 14, No. 12, 2001 1571
(17) Casida, J . E., Gammon, D. W., Glickman, A. H., and Lawrence,
L. J . (1983) Mechanisms of selective action of pyrethroid insec-
ticides. Annu. Rev. Pharmacol. Toxicol. 23, 413-438.
(18) Gupta, R. C., and Dettbarn, W. D. (1993) Role of carboxylesterases
in the prevention and potentiation of N-methylcarbamate toxicity.
Chem.-Biol. Interact. 87, 295-303.
(19) Wallace, T. J ., Ghosh, S., and Grogan, W. M. (1999) Molecular
cloning and expression of rat lung carboxylesterase and its
potential role in the detoxification of organophosphorus com-
pounds. Am. J . Respir. Cell Mol. Biol. 20, 1201-1208.
(20) Maxwell, D. M. (1992) The specificity of carboxylesterase protec-
tion against the toxicity of organophosphate compounds. Toxicol.
Appl. Pharmacol. 114, 306-312.
Ack n ow led gm en t. The authors gratefully acknowl-
edge Dr. Christophe Morisseau and Dr. Donald Stoutamire
from the Department of Entomology at the University
of California Davis for many useful discussions. We also
thank Dr. Yoshiaki Nakagawa and Dr. Masahiro Mi-
yashita from the Graduate School of Agriculture at Kyoto
University for performing the log P calculations. C.E.W.
was supported by a UC TSR&TP Lead Campus Program
in Ecotoxicology Fellowship and a UC TSR&TP Graduate
Fellowship. T.F.S. was supported by USDA Competitive
Research Grant 2001-35302-09919 and a UC Davis
Biotechnology Training Grant Fellowship. This work was
supported in part by NIEHS Grant R37 ES02710, NIEHS
Superfund Grant P42 ES04699, and NIEHS Center for
Environmental Health Sciences Grant P30 ES05707.
(21) Glickman, A. H., and Casida, J . E. (1982) Species and structural
variations affecting pyrethroid neurotoxicity. Neurobeha. Toxicol.
Teratol. 4, 793-799.
(22) Casida, J . E. (1980) Pyrethrum flowers and pyrethroid insecti-
cides. Environ. Health Perspect. 34, 189-202.
(23) Byrne, F. J ., Gorman, K. J ., Cahill, M., Denholm, I., and
Devonshire, A. L. (2000) The role of B-type esterases in conferring
insecticide resistance in the tobacco whitefly, Bemisia tabaci
(Genn). Pest Manage. Sci. 56, 867-874.
(24) Foster, S. P., and Devonshire, A. L. (1999) Field-simulator study
of insecticide resistance conferred by esterase-, MACE- and kdr-
based mechanisms in the peach-potato aphid, Myzus persicae
(Sulzer). Pestic. Sci. 55, 810-814.
(25) Harold, J . A., and Ottea, J . A. (2000) Characterization of esterases
associated with profenofos resistance in the tobacco budworm,
Heliothis virescens (F.). Arch. Insect Biochem. Physiol. 45, 47-
59.
(26) Conyers, C. M., MacNicoll, A. D., and Price, N. R. (1998)
Purification and characterisation of an esterase involved in
resistance to organophosphorus insecticides in the saw-toothed
grain beetle, Oryzaephilus surinamensis (Coleoptera: Silvanidae).
Insect Biochem. Mol. Biol. 28, 435-448.
(27) Shiotsuki, T., Huang, T. L., Uematsu, T., Bonning, B. C., Ward,
V. K., and Hammock, B. D. (1994) J uvenile hormone esterase
purified by affinity chromatography with 8-mercapto-1,1,1-tri-
fluoro-2-octanone as a rationally designed ligand. Protein Expres-
sion Purif. 5.
(28) Abdel-Aal, Y. A. I., and Hammock, B. D. (1985) Apparent multiple
catalytic sites involved in the ester hydrolysis of juvenile hor-
mones by the hemolymph and by an affinity-purified esterase
from Manduca sexta J ohannson (Lepidoptera: Sphingidae). Arch.
Biochem. Biophys. 243, 206-219.
(29) Linderman, R. J ., Leazer, J ., Venkatesh, K., and Roe, R. M. (1987)
The inhibition of insect juvenile hormone esterase by trifluoro-
methylketones: steric parameters at the active site. Pestic.
Biochem. Physiol. 29, 266-277.
(30) Sze´ka´cs, A., Borda´s, B., and Hammock, B. D. (1992) Transition
state analogue enzyme inhibitors: Structure-activity relation-
ships of trifluoromethyl ketones. In Rational Approaches to
Structure, Activity, and Ecotoxicology of Agrochemicals (Draber,
W., and Fujita, T., Eds.) pp 219-249, CRC Press, Boca Raton,
FL.
(31) Abdel-Aal, Y. A. I., and Hammock, B. D. (1986) Transition state
analogs as ligands for affinity purification of juvenile hormone
esterase. Science 233, 1073-1075.
(32) Sone, T., and Wang, C. Y. (1997) Microsomal amidases and
carboxylesterases. In Biotransformation (Guengerich, F., Eds.) pp
265-281, Elsevier Science, Oxford.
(33) Wolfenden, R. (1969) Transition state analogues for enzyme
catalysis. Nature 223, 704-705.
(34) Lienhard, G. E. (1973) Enzymatic catalysis and transition-state
theory. Science 180, 149-154.
(35) Roe, R. M., Anspaugh, D. D., Venkatesh, K., Linderman, R. J .,
and Graves, D. M. (1997) A novel geminal diol as a highly specific
and stable in vivo inhibitor of insect juvenile hormone esterase.
Arch. Insect Biochem. Physiol. 36, 165-179.
(36) Dess, D. B., and Martin, J . C. (1983) Readily accessible 12-I-5
oxidant for the conversion of primary and secondary alcohols to
aldehydes and ketones. J . Org. Chem. 48, 4155-4156.
(37) Ireland, R. E., and Liu, L. (1993) An improved procedure for the
preparation of the Dess-Martin periodinane. J . Org. Chem. 58,
2899.
Refer en ces
(1) Ollis, D. L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F.,
Franken, S. M., Harel, M., Remington, S. J ., Silman, I., Schrag,
J ., Sussman, J . L., Verschueren, K. H. G., and Goldman, A. (1992)
The R/â hydrolase fold. Protein Eng. 5, 197-211.
(2) Quinn, D. M. (1999) Ester hydrolysis. In Enzymes, Enzyme
Mechanisms, Proteins, and Aspects of NO Chemistry (Poulter, C.,
Eds.) pp 101-137, Elsevier Science, Oxford.
(3) Quinn, D. M. (1997) Esterases of the R/â hydrolase fold family.
In Biotransformation (Guengerich, F., Eds.) pp 243-264, Elsevier
Science, Oxford.
(4) Thomas, B. A., Church, W. B., Lane, T. R., and Hammock, B. D.
(1999) Homology model of juvenile hormone esterase from the crop
pest, Heliothis virescens. Proteins 34, 184-196.
(5) Aldridge, W. N. (1953) Serum esterases 1. Two types of esterase
(A and B) hydrolysing p-nitrophenyl acetate, propionate and
butyrate, and a method for their determination. Biochem. J . 53,
110-117.
(6) Aldridge, W. N. (1953) Serum esterases 2. An enzyme hydrolysing
diethyl p-nitrophenyl phosphate (E600) and its identity with the
A-esterase of mammalian sera. Biochem. J . 53, 117-124.
(7) Satoh, T., and Hosokawa, M. (1998) The mammalian carboxy-
lesterases: From molecules to functions. Annu. Rev. Pharmacol.
Toxicol. 38, 257-288.
(8) Senter, P. D., Marquardt, H., Thomas, B. A., Hammock, B. D.,
Frank, I. S., and Svensson, H. P. (1996) The role of rat serum
carboxylesterase in the activation of paclitaxel and camptothecin
prodrugs. Cancer Res. 56, 1471-1474.
(9) Kojima, A., Hackett, N. R., Ohwada, A., and Crystal, R. G. (1998)
In vivo human carboxylesterase cDNA gene transfer to activate
the prodrug CPT-11 for local treatment of solid tumors. J . Clin.
Invest. 101, 1789-1796.
(10) Takayama, H., Watanabe, A., Hosokawa, M., Chiba, K., Satoh,
T., and Aimi, N. (1998) Synthesis of a new class of camptothecin
derivatives, the long-chain fatty acid esters of 10-hydroxycamp-
tothecin, as
a potent prodrug candidate, and their in vitro
metabolic conversion by carboxylesterases. Bioorg. Med. Chem.
Lett. 8, 415-418.
(11) Tang, B. K., and Kalow, W. (1995) Variable activation of lovastatin
by hydrolytic enzymes in human plasma and liver. J . Clin.
Pharmacol. 47, 449-451.
(12) Bourland, J . A., Martin, D. K., and Mayersohn, M. (1997)
Carboxylesterase-mediated transesterification of meperidine (De-
merol) and methylphenidate (Ritalin) in the presence of [2H6]
ethanol: preliminary in vitro findings using a rat liver prepara-
tion. J . Pharm. Sci. 86, 1494-1496.
(13) Pindel, E. V., Kedishvili, N. Y., Abraham, T. L., Brzezinski, M.
R., Zhang, J ., Dean, R. A., and Borson, W. F. (1997) Purification
and cloning of a broad substrate specificity human liver carboxy-
lesterase that catalyzes the hydrolysis of cocaine and heroin. J .
Biol. Chem. 272, 14769-14775.
(14) Brzezinski, M. R., Spink, B. J ., Dean, R. A., Berkman, C. E.,
Cashman, J . R., and Bosron, W. F. (1997) Human liver carboxy-
lesterase hCE-1: binding specificity for cocaine, heroin, and their
metabolites and analogs. Drug Metab. Dispos. 25, 1089-1096.
(15) Kim, D. H., Yang, Y. S., and J akoby, W. B. (1990) Aspirin
hydrolyzing esterases from rat liver cytosol. Biochem. Pharmacol.
40, 481-487.
(16) Williams, F. M., Mutch, E. M., Nicholson, E., Wynne, H., Wright,
P., Lambert, D., and Rawlins, M. D. (1989) Human liver and
plasma aspirin esterase. J . Pharm. Pharmacol. 41, 407-409.
(38) Nair, H. K., Lee, K., and Quinn, D. M. (1993) m-(N,N,N-Tri-
methylammonio)trifluoroacetophenone: A femtomolar inhibitor
of acetylcholinesterase. J . Am. Chem. Soc. 115, 9939-9941.
(39) Ashour, M. B. A., and Hammock, B. D. (1987) Substituted
trifluoroketones as potent, selective inhibitors of mammalian
carboxylesterases. Biochem. Pharmacol. 36, 1869-1879.

1572 Chem. Res. Toxicol., Vol. 14, No. 12, 2001
Wheelock et al.
(40) Camps, F., Coll, J ., Messeguer, A., and Pericas, M. A. (1981)
Improved oxidation procedure with aromatic peroxyacids. Tetra-
hedron Lett. 22, 3895-3896.
(41) Linderman, R. J ., and Graves, D. M. (1989) Oxidation of fluoro-
alkyl-substituted carbinols by the Dess-Martin reagent. J . Org.
Chem. 54, 661-668.
the cabbage looper, Trichoplusia Ni. J . Insect Physiol. 25, 551-
560.
(51) Hammock, B. (1981) Metabolism and environmental fate of
juvenoids. In Regulation of insect development and behavior (F,
S., Zabza, A., J J , M., and B, C., Eds.) pp 841-852, Wroclaw
Technical University Press, Wroclaw, Poland.
(42) Hammock, B. D., and Sparks, T. C. (1977) A rapid assay for insect
juvenile hormone esterase activity. Anal. Biochem. 82, 573-579.
(43) Huang, T. L., Shiotsuki, T., Uematsu, T., Borhan, B., Li, Q. X.,
and Hammock, B. D. (1996) Structure-activity relationships for
substrates and inhibitors of mammalian liver microsomal car-
boxylesterases. Pharm. Res. 13, 1495-1500.
(44) Huang, T. L., Villalobos, S. A., and Hammock, B. D. (1993) Effect
of hepatotoxic doses of paracetamol and carbon tetrachloride on
the serum and hepatic carboxylesterase activity in mice. J .
Pharm. Pharmacol. 45, 458-465.
(45) Bradford, M. B. (1976) A rapid and sensitive technique for the
quantification of microgram quantities of protein utilizing the
principles of dye-binding. Anal. Biochem. 72, 248-254.
(46) Nellaiah, H., Morisseau, C., Archelas, A., Furstoss, R., and
Baratti, J . C. (1996) Enantioselective hydrolysis of p-nitrostyrene
oxide by an epoxide hydrolase from Aspergillus niger. Biotechnol.
Bioeng. 49, 70-77.
(47) Pauling, L. (1960) The nature of the chemical bond, Cornell
University Press, Ithaca, NY.
(48) Hammock, B. D., Wing, K. D., McLaughlin, J ., Lovell, V. M., and
Sparks, T. C. (1982) Trifluoromethylketones as possible transition
state analog inhibitors of juvenile hormone esterase. Pestic.
Biochem. Physiol. 17, 76-88.
(49) Rudnicka, M., and Hammock, B. (1981) Approaches to the
purification of the juvenile hormone esterase from the cabbage
looper, Trichoplusia Ni. Insect Biochem. 11, 437-444.
(50) Sparks, T., and Hammock, B. (1979) Induction and regulation
of juvenile hormone esterases during the last larval instar of
(52) Mentlein, R., Heiland, S., and Heymann, E. (1980) Simultaneous
purification and comparative characterization of six serine hy-
drolases from rat liver microsomes. Arch. Biochem. Biophys. 200,
547-559.
(53) Horgan, D., Webb, E., and Zerner, B. (1966) Determinations of
the normality of pig liver carboxylesterase solutions. Biochem.
Biophys. Res. Commun. 23, 23-28.
(54) Kao, L., Motoyama, N., and Dauterman, W. (1985) Multiple forms
of esterases in mouse, rat, and rabbit liver, and their role in
hydrolysis of organophosphate and pyrethroid insecticides. Pestic.
Biochem. Physiol. 23, 66-73.
(55) Shiotsuki, T., Huang, T. L., and Hammock, B. D. (1994) Affinity
purification of mouse liver carboxylesterases with trifluoromethyl
ketones as ligands. J . Pestic. Sci. 19, 321-324.
(56) Ward, V. K., Bonning, B. C., Huang, T., Shiotsuki, T., Griffeth,
V. N., and Hammock, B. D. (1992) Analysis of the catalytic
mechanism of juvenile hormone esterase by site-directed mu-
tagenesis. Int. J . Biochem. 24, 1933-1941.
(57) Olmstead, M. M., Musker, W. K., and Hammock, B. D. (1987)
Structure of the hydrate form of a â-thiotrifluoromethyl ketone,
a potent esterase inhibitor. Acta Crystallogr., Sect. C 43, 1726-
1728.
(58) Leo, A. J . (1993) Calculating log Poct from structures. Chem. Rev.
93, 1281-1306.
TX015508+