Reverse micelles as a catalyst for the nucleophilic aromatic substitution between glutathione and 2,4-dinitrochlorobenzene

Article abstract of DOI:10.1039/a903011e

The nucleophilic aromatic substitution (S_NAr) between GSH and 2.4-dinitrochlorobenzene was studied in reverse micellar systems composed of limited amounts of water, a surfactant with a polar head and a nonpolar tail, and the organic solvent 2,2

Full text of DOI:10.1039/a903011e

View Article Online / Journal Homepage / Table of Contents for this issue
Reverse micelles as a catalyst for the nucleophilic aromatic
substitution between glutathione and 2,4-dinitrochlorobenzene
Jong-Yan Liou, Ter-Mei Huang and Gu-Gang Chang*
Department of Biochemistry, National Defense Medical Centre, PO Box 90048, Taipei 100,
Taiwan, Republic of China. E-mail: ggchang@ndmcl.ndmctsgh.edu.tw; Fax: ϩ886 22365 5746
Received (in Cambridge, UK) 15th April 1999, Accepted 28th July 1999
The nucleophilic aromatic substitution (SNAr) between GSH and 2,4-dinitrochlorobenzene was studied in reverse
micellar systems composed of limited amounts of water, a surfactant with a polar head and a nonpolar tail, and
the organic solvent 2,2,4-trimethylpentane. When the surfactant was positively charged and contained an aromatic
ring in the polar head, the second-order rate constant was increased by approximately two orders of magnitude as
compared to that in aqueous solution. The rate enhancement could be attributed to the stabilization of the negatively
charged Meisenheimer σ-complex by the positively charged polar head and the weak aromatic ring’s electric
quadrupole interactions of the surfactant. The reaction rate in reverse micelles composed of neutral polar head
groups (Triton X-100) was increased by 3-fold, which may be explained by the interactions of the hydroxy groups
of Triton molecules with the π-system of the Meisenheimer complex. An inverse relationship between the molar
concentration [H2O]/[surfactant] ratio, which reflects the inclusive volume of the reverse micellar particle, and the rate
enhancement was observed for positively charged or hydroxy-containing reverse micelles, but opposite results were
obtained with negatively charged reverse micelles. These reverse micellar systems thus mimic the active site of a
detoxification enzyme, glutathione transferase, in which stabilization of the Meisenheimer complex by a positively
charged arginine residue, on-edge quadrupole interactions of aromatic amino acids, and the hydroxy group of
tyrosine or threonine have been proposed in the enzyme-catalysed SNAr conjugation.
Introduction
Glutathione transferase (GST,† EC 2.5.1.18) catalyses the
conjugation of glutathione (GSH) with various xenobiotic
or endogenous electrophilic compounds, constituting a major
detoxification mechanism of biological systems and is respon-
sible for the development of drug resistance in cancer
chemotherapy as well as for pesticide and insecticide resistance
in agriculture.1 The GST-catalysed nucleophilic aromatic sub-
stitution (SNAr) between GSH (a -γ-Glu-Cys-Gly tripeptide)
and 2,4-dinitrochlorobenzene (CDNB) involves a general-base
catalysed ionisation of the enzyme-bound GSH in which the
cysteinyl -SH group is ionised to yield a better nucleophile gluta-
thiolate anion (GSϪ). This anion then attacks the ipso carbon
bearing the chlorine atom in CDNB, forming a Meisenheimer
σ-complex. Leaving of the chloride ion completes the reaction
(Scheme 1, reactions I and II).2 The formation of the S-(2,4-
dinitrophenyl)glutathione conjugate renders the less water sol-
uble CDNB into a more water soluble product which is then
ready for excretion from the cells by a membranous ATP-
dependent glutathione-conjugate pump system.3 The rate-
limiting step has been proposed at the σ-bond formation in the
Scheme 1 Proposed chemical mechanism for the base-catalysed
nucleophilic aromatic substitution of GSH and CDNB or GSH and
TNB. (I) Ionisation of glutathione catalysed by a general base (step a).
(II) Nucleophilic attack of the glutathiolate anion at the ipso carbon
of CDNB forming the Meisenheimer complex (step b) and leaving of
chloride ion forming the S-(DNP)GS product (step c). (III) Nucleo-
philic attack of the glutathiolate anion on TNB forming the transition
state analogue S-(TNP)GS (step d).
† Abbreviations used. GST, glutathione transferase; GSH, reduced
glutathione; CDNB, 2,4-dinitrochlorobenzene (1-chloro-2,4-dinitro-
benzene); TNB, 1,3,5-trinitrobenzene; S-(DNP)GS, S-(2,4-dinitro-
phenyl)glutathione; S-(TNP)GS, S-(2,4,6-trinitrophenyl)glutathione;
SNAr, nucleophilic aromatic substitution; AOT (Aerosol-OT), sodium
bis(2-ethylhexyl) sulfosuccinate (C20H37O7SNa); BDAC, benzyldodecyl-
bis(2-hydroxyethyl)ammonium chloride (C23H42NO2Cl); BTAC, benzyl-
dimethyltetradecylammonium chloride (C23H42NCl); CBAC, cetyl-
benzyldimethylammonium chloride (C25H46NCl); CPB, cetylpyridinium
bromide (C21H38NBr); CTAB, cetyltrimethylammonium bromide
(C19H42NBr); DEAB, dodecylethyldimethylammonium bromide (C16-
H36NBr); DTAB, dodecyltrimethylammonium bromide (C15H34NBr);
EDAB, ethylhexadecyldimethylammonium bromide (C20H44NBr);
OTAB, octadecyltrimethylammonium bromide (C21H46NBr); TDAB,
tetradecyltrimethylammonium bromide (C17H38NBr); Triton X-100,
tert-octylphenoxypolyethoxyethanol (C32–34H58–62O10–11).
Meisenheimer complex.4 Some of the factors contributing to
the transition-state stabilization have been proposed, including
the role of conserved positively charged amino acid residues in
the active site.4,5
The non-enzymatic reaction between GSH and CDNB has a
substantially measurable rate, which needs to be corrected in
the assay of GST-catalysed reaction. On the other hand, this
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176
This journal is © The Royal Society of Chemistry 1999
2171

View Article Online
non-enzymatic reaction provides one of the few reactions suit-
able for comparison with the enzymatic-catalysed reaction that
is extremely useful in characterizing the fundamental role of the
enzyme in catalysis and thus is helpful in elucidating the GST
mechanism. We have previously isolated and characterized the
octopus hepatopancreatic GST.6 The octopus GST is a sigma-
class GST,7,8 which has been recruited as S-crystallin in octopus
lens.9 The detailed kinetic analysis of the reaction catalysed by
this enzyme was performed by steady-state kinetics.10,11 Our
results indicate that octopus GST conforms to a steady-state
random Bi-Bi kinetic mechanism similar to other classes of
GST.
Non-enzymatic reaction between GSH and CDNB in reverse
micellar systems
CDNB is the generally used substrate for the enzymatic assay
of GST activity. However, GSH reacts with CDNB non-
enzymatically at alkaline pH. Under the near neutral assay
conditions (pH 6.5) used in the enzymatic assay described
above, the minor non-enzymatic conjugation was corrected
using a double beam spectrophotometer that allows all reaction
components except the enzyme to be placed in the reference
cell. Thus, the recorder tracing will represent only the enzyme-
catalysed conjugation rate.
In the reverse micellar systems, the non-enzymatic reaction
between GSH and CDNB was examined at 25 ЊC in Bis-Tris–
HCl buffer (6 mM, pH 6.5). The final total volume, not the less
definable volume entrapped in reverse micelles,15 was used for
measurement of concentrations. The CDNB and other less
water-soluble substrates were prepared as dimethyl sulfoxide
solutions. The final organic solvent’s concentration in the assay
mixture was kept below 1% and did not affect the results. An
injection method was employed in mixing the two substrate
solutions.15
Since CDNB is less soluble in water and was dissolved in
organic solvent before mixing with GSH, we have also
characterized the kinetic behavior of the enzyme in a plasma
membrane mimicking reverse micellar system composed of
water–surfactant–2,2,4-trimethylpentane which produced
a
macro-homogeneous transparent solution.11 During those
studies, we found that the non-enzymatic SNAr reaction of the
GSH–CDNB system was enhanced in reverse micelles com-
posed of positively charged surfactant but not in the negatively
charged reverse micelles.12
The second-order rate constant for the non-enzymatic
reaction between GSH and CDNB in reverse micellar systems
was estimated by the following equation (eqn. (1)) in which
Furthermore, we previously found that GST may have some
affinity with the membranous structure.11 We postulated that
association of the cytoplasmic GST with the membranous
pumping system is beneficial for the cells to efficiently excrete
toxic xenobiotics.11 These results prompted us to systematically
examine the charge and other effects of the polar heads on the
SNAr reaction rate and to critically evaluate the application of
using these reverse micelles as a model system in elucidating the
reaction mechanism of GST.
In this article, the non-enzymatic conjugation between GSH
and CDNB in aqueous solution and in various water–
surfactant–2,2,4-trimethylpentane reverse micellar systems is
examined. Our results reveal that reverse micelles are ideal
systems to access the various factors that contribute to the
rate enhancement of the GST-catalysed SNAr reaction.
1
= k2 × t ϩ 1/[A]
(1)
[A]o Ϫ [S-(DNP)GS]
[GSH] = [CDNB],16 where [Ao] is the initial substrate concen-
tration and k2 denotes the second-order rate constant. To sim-
plify the calculation further, we used 1 mM concentration for
both GSH and CDNB. In this particular case, eqn. (1) can be
simplified to eqn. (2).
1
= k2 × t ϩ 1
(2)
1 Ϫ [S-(DNP)GS]
A plot of 1/{1 Ϫ [S-(DNP)GS]} versus time (t) will give a
straight line which intercepts the vertical axis at 1.0 and whose
slope denotes the second-order rate constant.
Experimental
Materials
GSH and CDNB were obtained from Sigma-Aldrich (St. Louis
MO, USA). 2,2,4-Trimethylpentane was obtained from Merck
(Darmstadt, Germany). The purity of AOT was previously
examined.13 Other surfactants were from Sigma-Aldrich or
Merck and were used without further purification. Distilled
water further purified through a MilliQ system (Millipore,
Bedford, USA) was used throughout this work.
Results
Rate enhancement of nucleophilic aromatic substitution in
reverse micelles
We have previously suggested that a reverse micellar system
composed of positively charged polar head surfactant is an
exemplary model system to study the GST-catalysed SNAr
reaction.12 To further characterize this model system, we have
systematically examined the conjugation between GSH and
CDNB in reverse micelles composed of various surfactants.
The reactions were conveniently monitored with a UV-VIS
spectrophotometer by following the formation of S-(DNP)GS
which absorbs at 340 nm. The reaction rate measurements
in all surfactant reverse micelles followed a linear plot with
respect to the CDNB concentration and the reaction time.
Some typical plots of the data according to eqn. (2) are shown
in Fig. 1. Table 1 lists the detergent structures and the second-
order rate constants for the SNAr reaction in those reverse
micelles. We have chosen surfactants with positively charged
polar heads and nonpolar tails of different chain lengths. How-
ever, no obvious difference was observed for the reaction rate of
surfactants with different nonpolar tails. The same organic
solvent 2,2,4-trimethylpentane was used in all reverse micellar
systems. The partitioning of water insoluble CDNB between
the aqueous and organic phases and/or the interface does not
seem to account for the rate differences. The clear conclusion
that can be drawn from the data shown in Table 1 is that the
Enzyme purification
Digestive gland GST from octopus was purified once by GSH-
Sepharose 4B affinity chromatography as described previously.6
The dimeric enzyme was judged to be apparently homogeneous
by SDS/PAGE with subunit Mr 24 000. Protein concentration
was determined by the protein-dye binding method.14
Enzyme assay in aqueous system
Assay for GST activity was carried out at 25 ЊC. The reaction
mixture (1 ml) contained 89 mM potassium phosphate buffer,
pH 6.5, 1 mM each of GSH and CDNB, 0.89 mM EDTA and
an appropriate amount of the enzyme. The reaction rate was
colinear with the GST concentration up to 3.69 µg assayϪ1 11
.
The formation of S-(DNP)GS was continuously monitored at
340 nm. One unit of enzyme activity is defined as an initial rate
of 1 µmol S-(DNP)GS formed per minute under the assay
conditions using a molar absorption coefficient of 9.6 × 103
M
Ϫ1 cmϪ1 for the conjugate.
2172
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176

View Article Online
Table 1 Second-order rate constants of the nucleophilic aromatic substitution reaction between GSH and CDNB in reverse micelles composed of
different surfactants. The [H2O]/[surfactant] molar ratio was held constant at 8.3 for all reverse micellar systems in Bis-Tris–HCl buffer (6 mM, pH
6.5). The ratios of second-order rate constants in reverse micelles (k2(rm)) to that in aqueous solution (k2(H O)) are compared in the last column
2
Second-order
rate constant/
M
Surfactant
Chemical structure of surfactant
Ϫ1 sϪ1
k2(rm)/k2(H O)
2
CBAC
BTAC
0.209
0.200
91
87
BDAC
EDAB
0.124
0.109
54
47
CTAB
0.103
45
DEAB
CPB
0.101
0.96
44
42
TDAB
DTAB
0.095
0.092
41
40
OTAB
0.089
39
TRITON X-100
None
0.007
3
1
0.0023
SNAr reaction rate in positively charged polar head reverse
micelles is enhanced up to approximately two orders of magni-
tude when compared to the reaction rate in aqueous solution.
The CBAC and BTAC have the fastest reaction rates. Besides
the positive charge, the weak electric quadrupole of the benzyl
group might provide an on-edge interaction with the Meisen-
heimer complex that accounts for the approximately 2-fold rate
enhancement.
The surfactant with a hydroxy polar head (Triton X-100)
enhanced the rate by 3-fold. However, BDAC, which possesses
positively charged, benzyl and hydroxy groups, did not provide
a faster rate than those surfactants with only positive charge
and benzyl group (CBAC and BTAC). It is possible that the
hydroxy groups of BDAC shield the more important positive
charge interactions. The basicities of the amino functions from
which the positively charged head groups are derived seems
to have no effect on the rate constant. Thus values are similar
in reversed micelles derived from CPB and CTAB which
contain pyridinium and quaternary ammonium head groups
respectively.
different sizes.17 The degree of freedom of GSH in the reverse
micelles is also restricted in small reverse micelles and thus
decreases the entropy of the system. Fig. 2 shows the results of
four reverse micellar systems. If the reverse micelles are com-
posed of negatively charged polar heads (AOT), the reaction
rate decreased with decreasing water content for the smaller
reverse micelles (Fig. 2A). On the other hand, the rate in reverse
micelles composed of positively charged (CTAB and CBAC) or
hydroxy (Triton X-100) polar heads increased for the smaller
reverse micelles (Fig. 2B–D). CBAC had a positive charge and
an extra benzyl group, and gave the fastest rate enhancement
among the surfactants we examined. These results clearly
substantiate our hypothesis that the rate enhancement in the
positively charged polar head reverse micelles is due to the
ionic interactions between the polar heads and the negatively
charged Meisenheimer complex.12 In other words, the reaction
is facilitated by stabilizing the charged Meisenheimer complex
and these reverse micelles provide an excellent model system
for studying the SNAr reaction catalysed by GST.
Discussion
Effect of particle size of the reverse micellar system on the rate
enhancement of the nucleophilic aromatic substitution
Analogy between reverse micelles and glutathione transferase
To further access the charge effect, we prepared reverse micelles
under various hydration degrees of the system by adjusting the
water content of the system, thus producing reverse micelles of
In this article, we provide evidence indicating that the rate
enhancement of SNAr in reverse micelles is due to the stabiliz-
ation of the negatively charged Meisenheimer σ-complex by the
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176
2173

View Article Online
positively charged polar head and the weak aromatic ring’s
electric quadrupole interactions of the surfactant. The reaction
rate enhancement in reverse micelles composed of a neutral
polar head (Triton X-100) may be explained by the interactions
of the hydroxy groups of Triton molecules with the π-system
of the Meisenheimer complex. We found that this reverse
micellar catalysed rate enhancement is highly analogous to the
rate acceleration of the same reaction catalysed by the detoxifi-
cation enzyme GST. In GST, a highly conserved positively
charged arginine residue (Arg-13) has been proposed to stabil-
ize the negatively charged Meisenheimer complex.18–23 The
functional importance of the corresponding Arg-14 in
pi-GST22 and Arg-15 in alpha-GST23 have been demonstrated
by site-specific mutagenesis.
Here we propose that these reverse micellar systems mimic
the active site region of a GST molecule and the quater-
nary amine may play an identical role to Arg-13 in GST. In
sigma-GST, a Phe-106, located beside the dinitrophenyl moiety
of the S-(DNP)GS molecule, may play the role of the weak
aromatic ring’s electric quadrupole interactions.24,25 Although
the dinitrophenyl moiety of the product seems to have little
direct interaction with the enzyme other than a few van der
Waals interactions with Phe-106,8 the nitro groups of
S-(DNP)GS are nestled against the electropositive edge of the
Phe-106 aromatic ring in sigma-GST.26 We thus conclude that,
in mimicking the active site of sigma-GST, CBAC is a better
surfactant to choose than the CTAB that we used previously.12
The benzyl group mimics the Phe-106, whereas Tyr-7 is
mimicked by benzyl and hydroxy groups. Scheme 2 shows a
schematic model for the CBAC reverse micelles.
The above hypothesis is further supported by the fact that the
recruited cephalopods lens protein S-crystallin has very little
endogenous GST activity and can be regarded as a natural
mutant of sigma-GST. Based on homology computer modeling
data, we have suggested that the small enzymatic activity of
S-crystallin is because of its losing the charge stabilization of
the Meisenheimer complex.26 An asparagine residue at position
101 in sigma-GST has been changed, in S-crystallin, to an
aspartate residue, which is only 3.71 Å apart from Arg-13. A
Coulomb charge–charge interaction would diminish the posi-
tively charged environment provided by Arg-13 in the active
centre. This argument is strengthened by the comparison of
surface electrical potential between S-crystallin and sigma-
GST.26 A marked feature was noticed in the active centre
region, where the overall electric potential is positively charged.
Fig. 1 Effect of surfactants on the rate enhancement of the conju-
gation between GSH and CDNB. GSH (1 mM) and CDNB (1 mM) in
Bis-Tris–HCl buffer (6 mM, pH 6.5) were injected into the reverse
micellar system containing 200 mM surfactant: CBAC (ᮀ), CTAB (᭹),
Triton X-100 (᭿), water (no surfactant) (᭺). The data were fitted to
eqn. (2). The slopes of these lines denote the second-order rate
constants.
Fig. 2 Effects of [H2O]/[surfactant] ratio on the conjugation between GSH and CDNB. GSH (1 mM) and CDNB (1 mM) in Bis-Tris–HCl buffer
(6 mM, pH 6.5). The surfactant concentration was maintained constant at 200 mM. The [H2O]/[surfactant] ratio was adjusted by varying the water
amount, thus varying the dimensions of the reverse micellar particles. The surfactant used (A) AOT, (B) Triton X-100, (C) CTAB, and (D) CBAC.
2174
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176

View Article Online
Scheme 3 Thermodynamic cycle for the non-enzymatic and enzymatic
reaction of GSH and CDNB. KGSH, KCDNB, and KT (Ki,S-(TNP)GS) repre-
sent dissociation constants for GSH, CDNB and the transition-state
‡
with the enzyme, respectively. Kn and Ke‡ represent the dissociation
constants of the transition state in non-enzymatic and enzymatic reac-
tions, respectively. kn denotes the rate for non-enzymatic reaction and
ke denotes the rate for enzymatic reaction.
For the acceleration of an enzymatic reaction rate, when
making a comparison with the non-enzymatic reaction rate
constant, the second order rate constant (kcat/Km) of the enzyme
reaction should be used.30 Detoxification enzymes such as GST
are characterized as sluggish, with broad substrate specificity,
and essentially unidirectional catalysts capable of reacting with
a broad spectrum of xenobiotics that cells might encounter.1
Since the GSH concentration in cells is in the range of 3–10
mM,1 which is an order magnitude greater than the Km,GSH
value, the GST thus can be regarded as saturated with GSH
under the in vivo conditions. Therefore, the kcat/Km,CDNB is a
more appropriate kinetic parameter to be compared with the
non-enzymatic second-order rate constants. In the case of
octopus GST, the kcat is 213 sϪ1 and a catalytic efficiency of
2.8 × 105 MϪ1 sϪ1 in the kcat/Km,CDNB term is achieved by the
binding of substrates.10,11 As compared with the non-enzymatic
Scheme 2 Schematic model showing the Meisenheimer complex
entrapped in a CBAC-reverse micelle. The quaternary amine polar
heads are represented by positive charges. The benzyl groups are
represented by Φ and are drawn adjacent to the water phase.
However, quite different surface electrical potentials are
observed around Arg-13 between sigma-GST and S-crystal-
lin.26 Arginine-13 has been proposed to be the major residue
responsible for the positively charged environment in the active
site region.22,23 Our model system provides further evidence
indicating that the positively charged active centre is essential
for the SNAr reaction to proceed in GST.
second-order rate constant (0.0023 MϪ1
s
Ϫ1) (Table 1), a
rate-acceleration factor (ke/kn) for the enzymatic reaction was
estimated to be 12 × 107, a value which is still a thousand
fold faster than that estimated from the reverse micellar model
system (3.8 × 104). Since the reaction rate is increased in smaller
reverse micelles containing positive charges or hydroxy groups,
and the essential protein groups in the enzyme active centre are
more closely in contact with the substrate or transition-state,
if the inclusive volume factor is taken into consideration,
the non-enzymatic reaction rate enhancement may actually
approach the enzymatic reaction rate acceleration.
Contribution of various factors in the rate enhancement of the
GST-catalysed SNAr reaction between GSH and CDNB
The above results provide a basis to quantitatively evaluate the
contribution of various factors to the rate enhancement in the
GST-catalysed SNAr reaction. The experimental results from
our earlier study indicated that ionisation of GSH does not
contribute to the rate-acceleration of the non-enzymatic reac-
tion.12 A similar conclusion was also drawn for the enzymatic
reaction.27 Furthermore, the detergent does not affect the dis-
sociation of GSH, however, the Meisenheimer complex is
stabilized and provides a 91-fold acceleration of the reaction
rate.
Proposed reverse micelles as GST model
Our results strongly suggest that electrostatic interaction is the
key factor for the GST-catalysed SNAr reaction rate enhance-
ment. This conclusion conforms to the theoretical derivation
for enzyme catalysis by Warshel et al.,31 who predicted that the
role of an enzyme molecule in the catalysed reaction is to pro-
vide pre-oriented dipoles. These dipoles are polarized to stabil-
ize the transition state charge distribution, thus reducing the
activation energy for the chemical reaction. The charge stabil-
ization effect does account for a large part of the enzyme
acceleration rate for the SNAr reaction between GSH and
CDNB. The water–CBAC–2,2,4-trimethylpentane reverse
micellar system thus mimics the active site of the sigma-GST
(or other classes of GST) and provides an excellent model to
assess the charge effects in the nucleophilic aromatic substitu-
tion (Scheme 4).
Our reverse micellar model may be considered as a refine-
ment of the oil-lake model proposed by Jakoby for GST.32 In
that cartoon model, the lipophilic compounds are accepted by
the membranous oily lake. GSH is on an island to accom-
modate the tripeptide. Whereas in our model (Scheme 4), the
reverse micellar particles constitute the active site islands to
hold the hydrophilic substrate GSH. This model suggests that
GST is an efficient method of detoxification, capable of hand-
ling the huge numbers of water insoluble toxic compounds that
the cells might encounter, however, it would be inefficient for
To estimate the degree to which the tight binding of
transition-state contributes to the rate-enhancement of
GST-catalysed reaction, S-(TNP)GS was generated in situ and
was assumed to be a transition-state analog (see Scheme 1).
This compound was found to be an un-competitive inhibitor
for octopus GST with respect to GSH with KI,S-(TNP)GS value
of 1.89 0.17 µM according to the method of Clark and
Sinclair.28 The Km,GSH and Km,CDNB values as determined by
initial-velocity studies were 0.344 0.004 mM and
0.769 0.007 mM, respectively.10,11 According to the thermo-
dynamic relationships of these kinetic parameters29 (Scheme 3),
the rate enhancement due to tight binding of S-(TNP)GS with
the enzyme can be expressed as ke/kn = Ke‡/Kn‡ = Km,GSH
ؒ
Km,CDNB/Ki,S-(TNP)GS. An effective concentration of approxi-
mately 140 mM was obtained.
If the effects of charge stabilization for the Coulomb and
weak aromatic quadrupole electric interactions and the hydroxy
group interactions are to be considered, the rate enhancement
should be increased at least by another 273-fold (91 × 3).
Assuming that all these effects are multiplicative, i.e., their con-
tributions to the activation free energy are additive, an overall
value of approximately 3.8 × 104-fold (140 × 273) was achieved.
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176
2175

View Article Online
5 X. Ji, R. N. Armstrong and G. L. Gilliland, Biochemistry, 1993, 32,
12949; L. Prade, R. Huber, T. H. Manoharan, W. E. Fahl and
W. Reuter, Structure, 1997, 5, 1287.
6 S. S. Tang, C. C. Lin and G. G. Chang, J. Protein Chem., 1994, 13,
609.
7 T. M. Buetler and D. L. Eaton, Environ. Carcinogen. Ecotoxicol.
Rev., 1992, C10, 181.
8 X. Ji, E. C. von Rosenvinge, W. W. Johnson, S. I. Tomarev,
J. Piatigorsky, R. N. Armstrong and G. L. Gilliland, Biochemistry,
1995, 34, 5317.
9 S. I. Tomarev, R. D. Zinovieva and J. Piatigorsky, J. Biol. Chem.,
1991, 266, 24226.
10 S. S. Tang and G. G. Chang, Biochem. J., 1995, 309, 347.
11 S. S. Tang and G. G. Chang, Biochem. J., 1996, 315, 599.
12 S. S. Tang and G. G. Chang, J. Org. Chem., 1995, 60, 6183.
13 G. G. Chang and S. L. Shiao, Eur. J. Biochem., 1994, 220, 861.
14 M. M. Bradford, Anal. Biochem., 1976, 72, 248.
15 P. L. Luisi and B. Steinmann-Hofmann, Methods Enzymol., 1987,
136, 188.
16 A. Fersht, Enzyme Structure and Mechanism, 2nd edn., Freeman,
New York, 1985, p. 182.
17 M. Waks, Proteins, 1986, 1, 4; P. L. Luisi and L. J. Magid, CRC Crit.
Rev. Biochem., 1986, 20, 409; R. Bru, A. Sanchez-Ferrer and
F. Garcia-Carmona, Biochem. J., 1995, 310, 721; M. T. Gómez-
Puyon and A. Gómez-Puyon, Crit. Rev. Biochem. Mol. Biol., 1998,
33, 53.
18 S. Liu, P. Zhang, X. Ji, W. W. Johnson, G. L. Gilliland and R. N.
Armstrong, J. Biol. Chem., 1992, 267, 4296.
Scheme
4
Schematic model for the water–cetylbenzyldimethyl-
ammonium chloride–2,2,4-trimethylpentane reverse micelles. The
entrapment of the hydrophilic substrate glutathione (GSH) in the water
pool (gray area) of the reverse micelles mimics the active centre of the
glutathione transferase. The various hydrophobic substrates, e.g., 2,4-
dinitrochlorobenzene (CDNB), were dissolved in the organic solvent
2,2,4-trimethylpentane reflecting the broad substrate specificities of the
enzyme. This model suggests that GST is not an efficient detoxification
enzyme. Proximity and orientation might not be the important factors
in the GST-catalysed reaction. Stabilization of the negatively charged
Meisenheimer complex in a positively charged environment (Scheme 2)
may contribute to the major factor in the rate enhancement.
19 C. Xia, D. J. Meyer, H. Chen, P. Reinemer, R. Huber and
B. Ketterer, Biochem. J., 1993, 293, 357.
20 M. Orozco, C. Vega, A. Parraga, I. Garcia-Saez, M. Coll, S. Walsh,
T. J. Mantle and F. Javier Luque, Proteins, 1997, 28, 530.
21 A. J. Oakley, M. Lo Bello, A. Battistoni, G. Ricci, J. Rossjohn,
H. O. Villar and M. W. Parker, J. Mol. Biol., 1997, 274, 84.
22 M. Widersten, R. H. Kolm, R. Björnestedt and B. Mannervik,
Biochem. J., 1992, 285, 377.
23 R. Björnestedt, G. Stenberg, M. Widersten, P. G. Board, I. Sinning,
T. A. Jones and B. Mannervik, J. Mol. Biol., 1995, 247, 765.
24 D. A. Dougherty, Science, 1996, 271, 163.
25 G. Xiao, S. Liu, X. Ji, W. W. Johnson, J. Chen, J. F. Parsons,
W. J. Stevens, G. L. Gilliland and R. N. Armstrong, Biochemistry,
1996, 35, 4753; E. C. Dietze, C. Ibarra, M. J. Dabrowski, A. Bird
and W. M. Atkin, Biochemistry, 1996, 35, 11938.
catalysis. This low catalytic rate is compensated for by the pres-
ence of large amounts of the enzyme in cells.1,32 GST was esti-
mated to constitute as much as 5% of the total soluble protein
in cells.1,32 Our model provides a quantitative evaluation of the
previous oil-lake model.
26 C. C. Chuang, S. H. Wu, S. H. Chiou and G. G. Chang, Biophys. J.,
1999, 76, 679.
Acknowledgements
This work was supported by the Frontier Science Project
(Grant NSC 87-2312-B016-008) from the National Science
Council, Republic of China. This paper is derived from the
thesis presented by J. Y. Liou in partial fulfillment of the
requirements for an M. S. degree (Biochemistry), National
Defense Medical Center, Taipei.
27 H. W. Dirr, P. Reinemer and R. Huber, Eur. J. Biochem., 1994, 220,
645; R. N. Armstrong, Adv. Enzymol. Relat. Areas Mol. Biol., 1994,
69, 1; I. Sinning, G. J. Kleywegt, S. W. Cowan, P. Reinemer,
H. W. Dirr, R. Huber, G. L. Gilliland, R. N. Armstrong, X. Ji,
P. G. Board, B. Olin, B. Mannervik and T. A. Jones, J. Mol. Biol.,
1993, 232, 192; M. C. J. Wilce, P. G. Board, S. C. Feil and
M. W. Parker, EMBO J., 1995, 14, 2133.
28 A. G. Clark and M. Sinclair, Biochem. Pharmacol., 1988, 37, 259.
29 R. Wolfenden and L. Frick, in Enzyme Mechanism, M. I. Page and
A. Williams, Eds., Royal Society of Chemistry, London, United
Kingdom, 1987, p. 97; J. Kraut, Science, 1988, 242, 533; J. Kyte,
Mechanism in Protein Chemistry, Garland Publishing, New York,
1995, p. 199; A. S. Mildvan, Proteins, 1997, 29, 401.
30 R. H. Abeles, P. A. Frey and W. P. Jencks, Biochemistry, Jones and
Bartlett Publishers, Boston, 1992, p. 124.
References
1 For some reviews, see W. B. Jakoby and D. M. Ziegler, J. Biol.
Chem., 1990, 265, 20715; D. J. Waxman, Cancer Res., 1990, 50, 6449;
T. H. Rushmore and C. B. Pickett, J. Biol. Chem., 1993, 268, 11475;
J. D. Hayes and D. J. Pulford, CRC Crit. Rev. Biochem. Mol. Biol.,
1995, 30, 445; R. N. Armstrong, Chem. Res. Toxicol., 1997, 10, 2.
2 G. F. Graminski, Y. Kubo and R. N. Armstrong, Biochemistry, 1989,
28, 3562; W. M. Atkins, R. W. Wang, A. W. Bird, D. J. Newton and
A. Y. H. Lu, J. Biol. Chem., 1993, 268, 19188; M. R. Crampton and
S. D. Lord, J. Chem. Soc., Perkin Trans. 2, 1997, 369.
31 A. Warshel, F. Sussman and J.-K. Hwang, J. Mol. Biol., 1988, 201,
139; A. Warshel, J. Biol. Chem., 1998, 273, 27035.
32 W. B. Jakoby, in Enzyme Mechanism, M. I. Page and A. Williams,
Eds., Royal Society of Chemistry, London, United Kingdom, 1987,
p. 468.
3 T. Ishikawa, Trends Biochem. Sci., 1992, 17, 463.
4 G. F. Graminski, P. H. Zhang, M. A. Sesay, H. L. Ammon and
R. N. Armstrong, Biochemistry, 1989, 28, 6252.
Paper 9/03011E
2176
J. Chem. Soc., Perkin Trans. 2, 1999, 2171–2176