Kinetics of the reaction catalyzed by inositol dehydrogenase from Bacillus subtilis and inhibition by fluorinated substrate analogs

Article abstract of DOI:10.1139/V06-033

Inositol dehydrogenase (EC 1.1.1.18) from Bacillus subtilis catalyzes the oxidation of myo-inositol to scyllo-inosose by transfer of the equatorial hydride of the substrate to NAD⁺. This is a key enzyme in the metabolism of myo-inositol, a primary carbon source for soil bacteria. In light of our recent discovery that the enzyme has a broad substrate spectrum while maintaining high stereoselectivity, we seek a more thorough understanding of the enzyme and its active site. We have examined the kinetics of the recombinant enzyme, and synthesized fluorinated substrate analogues as competitive inhibitors. We have evaluated all rate constants in the ordered, sequential Bi Bi mechanism. No steady-state kinetic isotope effect is observed using myo-[2-²H]-inositol, indicating that the chemical step of the reaction is not rate-limiting. We have synthesized the substrate analogs 2-deoxy-2-fluoro-myo-inositol, its equatorial analog 1-deoxy-1-fluoro-scyllo- inositol, the gem-difluorinated analog 1-deoxy-1,1-difluoro-scyllo-inositol, and the sugar analog α-D-glucosyl fluoride. Of these, 1-deoxy-1-fluoro- scyllo-inositol showed no inhibition, while all others tested had K_i values comparable to the K_m values of the analogous substrates myo-inositol and α-D-glucose.

Full text of DOI:10.1139/V06-033

522
Kinetics of the reaction catalyzed by inositol
dehydrogenase from Bacillus subtilis and
inhibition by fluorinated substrate analogs1
Richard Daniellou, Hongyan Zheng, and David R.J. Palmer
Abstract: Inositol dehydrogenase (EC 1.1.1.18) from Bacillus subtilis catalyzes the oxidation of myo-inositol to scyllo-
inosose by transfer of the equatorial hydride of the substrate to NAD+. This is a key enzyme in the metabolism of
myo-inositol, a primary carbon source for soil bacteria. In light of our recent discovery that the enzyme has a broad
substrate spectrum while maintaining high stereoselectivity, we seek a more thorough understanding of the enzyme and
its active site. We have examined the kinetics of the recombinant enzyme, and synthesized fluorinated substrate ana-
logues as competitive inhibitors. We have evaluated all rate constants in the ordered, sequential Bi Bi mechanism. No
steady-state kinetic isotope effect is observed using myo-[2-2H]-inositol, indicating that the chemical step of the reac-
tion is not rate-limiting. We have synthesized the substrate analogs 2-deoxy-2-fluoro-myo-inositol, its equatorial analog
1-deoxy-1-fluoro-scyllo-inositol, the gem-difluorinated analog 1-deoxy-1,1-difluoro-scyllo-inositol, and the sugar analog
α-D-glucosyl fluoride. Of these, 1-deoxy-1-fluoro-scyllo-inositol showed no inhibition, while all others tested had Ki
values comparable to the Km values of the analogous substrates myo-inositol and α-D-glucose.
Key words: inositol dehydrogenase, enzyme mechanism, kinetics, competitive inhibitor, substrate analogue.
Résumé : La déshydrogénase de l’inositol (EC 1.1.1.18) du Bacillus subtilis catalyse l’oxydation du myo-inositol en
scyllo-inosose par le transfert d’un hydrure équatorial du substrat vers la NAD+. C’est un enzyme clé dans le métabo-
lisme du myo-inositol, une source de carbone primaire pour les bactéries du sol. Compte tenu de notre découverte ré-
cente à l’effet que l’enzyme peut agir sur un large éventail de substrats tout en maintenant une stéréosélectivité élevée,
une étude plus approfondie a été entreprise pour mieux comprendre cet enzyme et son site actif. À cette fin, on a étu-
dié la cinétique de l’enzyme recombinant et on a synthétisé des analogues fluorés de substrats comme inhibiteurs com-
pétitifs. On a évalué toutes les constantes de vitesse dans le mécanisme Bi Bi, ordonné et séquentiel. On n’a pas
observé d’effet isotopique cinétique à l’état stationnaire lorsqu’on a étudié le myo-[2-2H]-inositol; cette observation in-
dique que l’étape chimique n’est pas cinétiquement limitante. On a aussi synthétisé les analogues de substrats 2-
désoxy-2-fluoro-myo-inositol, son analogue équatorial 1-désoxy-1-fluoro-scyllo-inositol, son analogue gem-difluoré 1-
désoxy-1,1-difluoro-scyllo-inositol et le sucre analogue fluorure de α-D-glucosyle. De tous ces composés, le 1-désoxy-1-
fluoro-scyllo-inositol ne présente pas d’inhibition alors que, avec tous les autres composés évalués, les valeurs de Ki
sont comparables aux valeurs de Km des substrats analogues du myo-inositol et du α-D-glucose.
Mots clés : inositol, déshydrogénase, mécanisme d’enzyme, cinétique, inhibition compétitive, analogue de substrat.
[Traduit par la Rédaction] Daniellou et al. 527
Introduction
been known for many years that environmental microorgan-
isms perform regioselective oxidation reactions on inositols
and related carbohydrate derivatives (1–3). The first step in
the catabolism of inositol by soil bacteria is the NAD+-
dependent oxidation of myo-inositol to scyllo-inosose by
inositol dehydrogenase (IDH, EC 1.1.1.18), shown in
Scheme 1. Recently, interest in IDH has increased because
of links between inositol catabolism, particularly IDH activ-
ity, and the competitiveness of certain beneficial soil bacte-
ria that can form symbiotic relationships with important
crops such as peas, beans, and other legumes (4–6).
Inositol is a ubiquitous natural product that is abundant in
soil. Inositol derivatives are known to occupy key roles in
the biology of bacteria, fungi, plants, and animals. It has
Received 25 August 2005. Published on the NRC Research
Press Web site at http://canjchem.nrc.ca on 21 April 2006.
This paper is dedicated to Professor Walter Szarek, whose
knowledge of and enthusiasm for his science are
inspirational.
The most detailed in vitro study of IDH catalysis has used
the enzyme from Bacillus subtilis. The native enzyme was
isolated and the kinetic mechanism described by Ramaley et
al. (7) more than 25 years ago. This work revealed several
aspects of IDH, including some tolerance of the active site to
structural variations in the substrate: α-D-glucose and α-D-
xylose were also observed to be substrates. Also demon-
R. Daniellou, H. Zheng, and D.R.J. Palmer.2 Department of
Chemistry, University of Saskatchewan, 110 Science Place,
Saskatoon, SK S7N 5C9, Canada.
1This article is part of a Special Issue dedicated to Professor
Walter A. Szarek.
2Corresponding author (e-mail: palmer@sask.usask.ca).
Can. J. Chem. 84: 522–527 (2006)
doi:10.1139/V06-033
© 2006 NRC Canada

Daniellou et al.
523
Scheme 1. The IDH-catalyzed oxidation of myo-inositol to
scyllo-inosose and α-D-glucose to D-glucono-1,5-lactone.
acteristics matched those reported before (9). Inositol
dehydrogenase was purified as described previously (8).
UV–vis absorbance was measured using a Beckman DU-640
spectrophotometer with a circulating-water-bath-controlled
temperature block. Thin layer chromatography was per-
formed on aluminium-backed plates of Silica gel 60F254
(EM Science, Gibbstown, New Jersey) using phosphomoly-
bdic acid – ethanol reagent, or a 10% solution of sulfuric
acid in ethanol, and (or) UV at 254 nm to visualize spots.
Silica gel 60 (40–63 µm) was used for flash chromatogra-
phy. NMR spectra were recorded on a Bruker 500 MHz
spectrometer. Chemical shifts are reported in ppm downfield
from tetramethylsilane. Mass spectra were recorded with a
VG 70SE mass spectrometer. NMR, mass spectrometry, and
elemental analysis facilities are part of the Saskatchewan
Structural Sciences Centre, University of Saskatchewan,
Saskatoon, Saskatchewan.
Scheme 2. Cleland diagram of an ordered, sequential Bi Bi
mechanism. For IDH, A is NAD+, B is myo-inositol, P is
scyllo-inosose, and Q is NADH.
2-Deoxy-2-fluoro-myo-inositol (1), 1-deoxy-1-fluoro-
scyllo-inositol (2), 1-deoxy-1,1-difluoro-scyllo-inositol (3)
(10–13), and α-D-glucosyl fluoride (4) (14) were synthesized
according to literature methods. All spectroscopic data
matched that reported previously.
myo-[2-2H]-Inositol
NaBD4 was added to a solution of scyllo-inosose (0.53 g,
3 mmol) in water (27 mL). After stirring overnight at room
temperature, the reaction was quenched by the addition of
acetone (5 mL). The mixture was then concentrated and the
residue was peracetylated by stirring in acetic anhydride
(20 mL) and pyridine (10 mL) at 100 °C for 24 h (15). Flash
chromatography of the residue (hexane – ethyl acetate, 1:1),
followed by deacetylation using sodium methoxide in meth-
anol at reflux for 10 min, neutralization with Dowex 50X8
(H+ form), filtration and concentration, gave myo-[2-2H]-
inositol isolated in 41% yield as a white solid. No scyllo-
strated was a sequential, ordered Bi Bi mechanism in which
NAD+ binds first and NADH is released last, as indicated in
Scheme 2. However, the authors reported only the apparent
Michaelis constant for myo-inositol at 0.5 mmol/L NAD+
(K Ampp = 18 mmol/L) rather than the true Km. No specific in-
sight into the active-site architecture was garnered.
Our laboratory has recently subcloned and expressed re-
combinant IDH from B. subtilis. We observed that IDH can
oxidize 1L-4-O-subtituted derivatives of myo-inositol, and
that the substituent can be as large as a camphorsulfonyl
group (8). Disaccharides could also act as substrates, but the
presence of a negative charge on the substituent decreased
affinity for the enzyme. This suggests a large, nonpolar
pocket adjacent to the active site, allowing stereoselective
recognition of these alternative substrates.
We seek a thorough understanding of this enzyme, and to
that end we have evaluated rate constants for each of the
steps represented in Scheme 2. In addition, we have synthe-
sized substrate mimics in which the axial hydroxyl group of
the reaction centre has been replaced with an axial and (or)
equatorial fluorine atom. Fluorine can act as a hydrogen
bond acceptor, but not a donor, and the ability of these com-
pounds to act as inhibitors will indicate the role of the
hydroxyl group in binding to the active site. Development of
a potent inhibitor could prove very useful in determining the
roles of active-site residues if a high-resolution crystal struc-
ture of inhibitor-bound enzyme can be generated.
1
inositol derivative was observed. H NMR (500 MHz, D2O)
δ: 3.15 (t, J = 9.3 Hz, 1H, H-5), 3.42 (d, J = 10.0 Hz, 2H, H-
1 and H-3), 3.51 (m, 2H, H-4 and H-6). 13C NMR
(125 MHz, D2O) δ: 71.44, 72.13 (t, JC-D = 22 Hz), 72.78,
74.73. ESI-MS : 204.06 [M + Na]+, 385.13 [2M + Na]+.
Anal. calcd. for C6H11DO6: C 39.78, H 7.23; found: C
39.99, H 6.99.
Kinetics
Forward (oxidation of myo-inositol) and reverse (reduc-
tion of scyllo-inosose) reactions were measured by continu-
ously monitoring the appearance or the disappearance of
NADH at 340 nm (ε = 6.2 (mmol/L)–1 cm–1). Initial rates
were determined using various concentrations of substrates
in 100 mmol/L Tris-HCl pH 9.0 at 25 °C. Kinetic constants
were determined using the program Leonora (16) by nonlin-
ear least-squares fitting of the data to eq. [1], which
describes an ordered sequential Bi Bi mechanism in the ab-
sence of products,
Experimental
General
Vmax[A][B]
Chemical reagents, including buffers, salts, myo-inositol,
NAD+, and NADH, were obtained from Sigma-Aldrich Can-
ada, Ltd. (Oakville, Ontario), or VWR CanLab (Mississauga,
Ontario), and were categorized as molecular biology grade
or were the highest grade available. scyllo-Inosose was syn-
thesized as previously described and its spectroscopic char-
[1]
v =
KiAKmB + KmB [A] + KmA[B] + [A][B]
where [A] is the concentration of the substrate that first
binds (NAD+ or NADH), [B] is the concentration of the sec-
ond substrate (myo-inositol or scyllo-inosose), KiA is the dis-
© 2006 NRC Canada

524
Can. J. Chem. Vol. 84, 2006
Fig. 1. Double reciprocal plot of the variation of rate with:
(A) NAD+ concentration at constant myo-inositol concentrations
(2 mmol/L (᭹), 4 mmol/L (ٗ), 7 mmol/L (᭿), 10 mmol/L (᭡),
20 mmol/L (᭺)); and (B) inositol concentration at constant
inositol concentrations (0.1 mmol/L (᭹), 0.2 mmol/L (᭺),
0.4 mmol/L (᭿), 1.0 mmol/L (ٗ)).
sociation constant for A, KmA and KmB are the respective
limiting Michaelis constants for the substrates A and B, and
Vmax is the maximum velocity when both substrates are in
saturating condition.
The concentration of each inhibitor needed to decrease the
velocity by 50% (IC50) was determined at [S] = 20 mmol/L
by fitting the dependence of v/v0 on [I], where v0 is the rate
in the absence of inhibitor. The inhibition constant (Ki) was
calculated using the Cheng and Prusoff equation (17).
IC50
[2]
Ki =
[S]
1 +
KApp
To test the validity of this method in these conditions, the
inhibition constant for 1-deoxy-1-fluoro-α-D-glucose (4) was
determined by a Dixon plot, based on eq. [3],
V
App[S]
[3]
v =
⎛
⎞
⎟
⎠
[I]
KApp 1 +
+ [S]
⎜
K
⎝
i
where S is the varied substrate, VApp is the maximal velocity
at the concentration of the fixed substrate, KApp is the sub-
strate concentration necessary for half-maximal velocity un-
der these conditions, and Ki is the inhibition constant. The
result of this analysis (66 5 mmol/L) was in good agree-
ment with the value determined using eq. [2] (58
8 mmol/L).
Results and discussion
Kinetics and mechanism
The work of Ramaley et al. (7) provided the basic facts re-
garding the IDH mechanism, including product inhibition
studies consistent with an ordered Bi Bi mechanism and a
pH–rate profile, which we have confirmed in our laboratory
(data not shown). To examine the details of the kinetic
mechanism, the initial velocities of the enzymatic reaction at
pH 9.0 were recorded at several constant concentrations of
one substrate (B) while varying the concentration of the
other substrate (A). When data were plotted in the form 1/v
against 1/A, a series of straight lines was obtained for the
forward and reverse enzymatic reactions. The lines inter-
sected to the left of the ordinate axis of the double-
reciprocal plot, in agreement with the ordered Bi Bi mecha-
nism depicted in Scheme 2 (18). The families of lines are
shown in Fig. 1. Our results, shown in Table 1, provide a
Km = 1.1 mmol/L for myo-inositol for the first time, compa-
rable to the value of 1 mmol/L for scyllo-inosose and consis-
tent with a catabolic pathway.
Some other facts become apparent from the data of Ta-
ble 1. First, although we are operating at the optimal pH for
oxidation of myo-inositol, the reverse reaction, reduction of
scyllo-inosose, is still faster. Second, the value of Kiq, repre-
senting the disassociation constant of NADH binding to free
IDH, is notably lower than the Michaelis constant of the
substrate inositol. Product inhibition thus limits the applica-
tion of IDH as a catalyst for the synthesis of inosose deriva-
tives, but this nonproductive ternary complex would yield a
wealth of information if it could be crystallized.
Table 1. Kinetic constants for the IDH-
catalyzed reaction.
Constant
Value
Kia
5 1 mmol/L
Km(NAD
0.26 0.05 mmol/L
1.1 0.1 mmol/L
13.5 2 µmol min–1 mg–1
9 1 s–1
+
)
Km(inositol)
Vforward
kforward
Kiq
50 5 µmol/L
Km(NADH)
Km(inosose)
Vreverse
36 5 µmol/L
1.0 0.1 mmol/L
19 2 µmol min–1 mg–1
13 1 s–1
kreverse
Note: Conditions for the reaction were
100 mmol/L TRIS-HCl, pH 9.0, 25 °C.
© 2006 NRC Canada

Daniellou et al.
525
Table 2. Calculated kinetic constants of IDH-catalyzed reactions.
Constant
Expression
Value
(3.6 0.4) × 104 s–1 (mol/L)–1
Vforward
k1
Km(NAD + [E0]
)
VforwardKia
k–1
k2
180 30 s–1
Km(NAD + [E0]
)
(1.4 0.2) × 104 s–1 (mol/L)–1
14 1 s–1
Vforward(k−2 + k3)
k3Km (inositol)[E0]
VforwardVreverse Kia
k–2
k3
[E0](Vforward Kia − VreverseKm(NAD +
)
)
)
19 1 s–1
VforwardVreverse Kiq
[E0](Vreverse Kiq − VforwardKm(NADH)
Vreverse(k−2 + k3)
k−2Km (inositol)[E0]
k–3
k4
(3.0 0.4) × 104 s–1 (mol/L)–1
18 1 s–1
1
Vreverse
Kiq
Km (NADH)[E0]
Vreverse
k–4
Kib
Kip
3.5 0.4 × 105 s–1 (mol/L)–1
14 2 mmol/L
Km (NADH)[E0]
k−1 + k−2
k2
k−1 + k4
k1
5.5 0.9 mmol/L
Note: Conditions for the reactions were 100 mmol/L TRIS-HCl, pH 9.0, 25 °C.
The internal consistency of the kinetic constants in Ta-
ble 1 was confirmed using the Haldane relationship for an
ordered Bi Bi kinetic mechanism. A value of 5.7 × 10–3 was
calculated for Keq. This is in good agreement with the value
of 2.5 × 10–3, which was determined experimentally at
pH 9.0 and at 25 °C for the thermodynamic equilibrium con-
stant.
values for k1–k–4 were calculated, and are summarized in
Table 2. In both directions, the breakdown of the ternary
complex (k3 and k–2) appears to constitute the rate-limiting
step, although in the forward direction, liberation of the co-
factor NADH is also slow and partially rate-limiting. This
analysis does not distinguish between the chemical step, i.e.,
the hydride transfer, and the release of the first product.
To determine whether the chemical step or the liberation
of the product is the limiting step, primary kinetic isotope
effect experiments were performed. The synthesis of myo-[2-
2H]-inositol was easily effected by reaction of scyllo-inosose
with sodium borodeuteride. One might expect a mixture of
myo- and scyllo-inositol isomers, generated by hydride at-
tack from either face of the ketone, however, only the myo-
isomer is formed. The high stereoselectivity of this reaction,
in agreement with results using related compounds (21, 22),
is in accordance with the prevalence of the steric effects in
reduction reactions with sodium borohydride; hydride attack
occurs at the face anti to the substituents on the two adjacent
carbons. A simple acetylation–deacetylation sequence was
used to remove salts from the compound. For consistency,
this process was also carried out using sodium borohydride
to make myo-inositol, and this inositol was used for direct
comparison with the deuterated analogue. myo-Inositol pre-
pared in this manner showed no difference in behaviour
from purchased myo-inositol. At [NAD+] = 0.5 mmol/L, no
kinetic isotope effect was observed on Vmax (VH/VD = 1.0)
and the isotope effect on Vmax/Km was only 1.1. These re-
sults lead us to the conclusion that the chemical reaction is
rapid relative to liberation of product, the latter being the
rate-limiting step of the enzymatic reaction.
[NADH][scyllo-inosose]
Keq
=
[NAD+][myo-inositol]
The Theorell–Chance mechanism is a special case of the
ordered Bi Bi mechanism in which the concentrations of the
central EAB and EPQ complexes are essentially zero. To
evaluate the kinetic importance of these species and discrim-
inate between the two mechanisms, we have calculated the
ratio R, as defined by Janson and Cleland (19).
⎛
⎞
Km(NAD+
⎛
⎜
⎝
⎞
⎟
⎠
Km(NADH)
1
1
)
⎜
⎜
⎝
⎟
⎟
⎠
1 −
+
+
1 −
V
Kia
V
Kiq
forward
reverse
R =
1
1
V
V
reverse
forward
The value of R can vary between 0 and 1, where the limit-
ing case R = 0 would indicate the canonical Theorell–
Chance mechanism. Using our data, we calculated R = 0.68
for IDH under these conditions, indicating that the ternary
complex has a lifetime.
By using the relationships of kinetic parameters to micro-
scopic rate constants for an ordered Bi Bi mechanism (20),
© 2006 NRC Canada

526
Can. J. Chem. Vol. 84, 2006
Chart 1.
Table 3. Inhibition constants and IC50 values for inhibitors 1–4.
Compound
IC50 (mmol/L)
Ki (mmol/L)
77 2
No inhibition
82 2
1
2
3
4
37 5
—
39 5
58 8
116 3
Note: Collected under the following conditions: 100 mmol/L TRIS-HCl,
pH 9.0, 25 °C, 20 mmol/L myo-inositol, 0.5 mmol/L NAD+.
Fig. 2. Dixon plot of variation of rate with varying concentra-
tions of inhibitor 4 at constant myo-inositol concentrations:
10 mmol/L (᭺), 20 mmol/L (᭹), 40 mmol/L (ٗ), 80 mmol/L
(᭡).
Fluorine-containing analogues
Replacement of the hydroxyl group of a substrate with a
fluorine atom can help one understand the mechanism of an
enzyme in a number of ways (23, 24). The carbon–fluorine
bond length is close to that of the carbon–oxygen bond, but
fluorine is a better leaving group than hydroxyl and cannot
act as a hydrogen bond donor. A substrate bearing a fluorine
atom might therefore behave as a pseudosubstrate, undergo-
ing the “normal” enzymatic chemistry, but with altered
reactivity; as an inhibitor, recognized by the enzyme but un-
dergoing no reaction, or as an inactivator, alkylating the en-
zyme at the active site. However, a number of studies giving
unexpected results has prompted Seebach to dub fluorinated
substrates “flustrates” (25).
We chose to replace the reaction centre of myo-inositol
with a carbon bearing a fluorine substituent since the result-
ing molecules are approximately isosteric with the substrate,
but cannot undergo the dehydrogenation reaction. The inhib-
itors envisioned were all accessible by published procedures.
2-Deoxy-2-fluoro-myo-inositol (1) matches the configuration
of the substrate, with an axial fluorine replacing the reactive
hydroxyl. 1-Deoxy-1-fluoro-scyllo-inositol (2) has all non-
hydrogen substituents in the equatorial position. The all-
equatorial stereoisomer of myo-inositol (scyllo-inositol) is
not a substrate nor an inhibitor of IDH (7). Therefore, if we
understand the recognition behaviour of the active site, 2
should show little or no inhibition. While these two inhibi-
tors test for the requirement of an axial substituent, the gem-
difluoro analog, 1-deoxy-1,1-difluoro-scyllo-inositol (3),
tests whether the presence of an equatorial substituent is del-
eterious. α-D-Glucosyl fluoride (4) is the isosteric analog of
glucose, which is also a substrate for IDH (Chart 1).
The results of the inhibition studies with 1–4 are shown in
Table 3. None of these compounds could act as substrates,
i.e., no reduction of NAD+ was observed in the absence of
myo-inositol. All experiments were performed under our
standard assay conditions, where the K Ampp for myo-inositol
is 18 mmol/L, and for α-D-glucose is 56 mmol/L. Solubility
of the fluorinated compounds in aqueous buffer prevented
very high concentrations of inhibitor from being used, but an
IC50 value could be determined. Using the method of Cheng
and Prusoff (17), a value for Ki was calculated from the
IC50. The Dixon plot generated for 4 shown in Fig. 2 is as
expected for a competitive inhibitor. As predicted, the myo-
inositol analog 1 inhibits IDH, while the scyllo-derivative 2
does not. The Ki value is about double the Km value for myo-
inositol under these conditions, suggesting that binding of
the inhibitor to the active site is comparable, although not as
favourable, as binding of substrate. Interestingly, The Ki
value for 3 is very close to that for 1, clearly indicating that
the presence of an equatorial substituent is not intrinsically
detrimental to binding in the active site. In this compound, a
carbon–fluorine atom is taking the place of the scissile
carbon–hydrogen bond (in addition to the carbon–oxygen
bond), therefore, a fluorine atom can be accommodated be-
tween the carbon atom of the inositol derivative and carbon-
4 of the nicotinamide ring of NAD+. Inhibition by the glu-
cose analog 4 suggests that binding is only slightly less fa-
vourable than substrate binding, suggesting that the fluorine
atom can replace the hydroxyl group with only a small ener-
getic penalty.
Conclusion
IDH is an enzyme that B. subtilis can use for catabolism
of not only myo-inositol, but also for several mono-
saccharides and disaccharides (7, 8). Our results are in ac-
cord with the conclusions of Ramaley et al. (7) regarding the
ordered mechanism of the enzyme, but we have now shown
that the enzyme has evolved a fast hydride transfer mecha-
nism, such that release of products limits the rate. The en-
zyme also regulates its production of inosose by product
inhibition. The use of fluorinated analogs as inhibitors
shows that an axial electronegative group is required for
binding, and that fluorine can replace hydroxyl as this axial
group. The presence of an equatorial carbon–fluorine bond
at the reaction centre can be accommodated by the enzyme
© 2006 NRC Canada

Daniellou et al.
527
9. C.A. Hansen, A.B. Dean, and J.W. Frost. J. Am. Chem. Soc.
121, 3799 (1999)
10. S.K. Chung, Y.U. Kwon, Y.T. Chang, K.H. Sohn, J.H. Shin,
K.H. Park, B.J. Hong, and I.H. Chung. Bioorg. Med. Chem. 7,
2577 (1999).
if an axial substituent is also present. We hope that these or
other inhibitors can be used to form crystals of IDH in a ter-
nary complex, giving a clear picture of the active-site archi-
tecture during catalysis.
11. G. Lowe and F. McPhee. J. Chem. Soc. Perkin Trans. 1, 1249
(1991).
12. D.A. Sawyer and B.V.L. Potter. J. Chem. Soc. Perkin Trans. 1,
923 (1992).
Acknowledgements
This work was supported by a Natural Sciences and Engi-
neering Research Council of Canada (NSERC) Discovery
grant and a New Opportunities grant from the Canadian
Foundation for Innovation. The authors thank Ken Thoms
for mass spectrometry and elemental analysis.
13. T. Desai, J. Gigg, R. Gigg, E. Martin-Zamora, and N. Schnetz.
Carbohydr. Res. 258, 135 (1994).
14. M. Hayashi, S.I. Hashimoto, and R. Noyori. Chem. Lett. 1747
(1984).
15. R. Mukherjee and E.M Axt. Phytochemistry, 23, 2682 (1984).
16. A. Cornish-Bowden. Analysis of enzyme kinetic data. Oxford
University Press, New York. 1995.
17. Y.C. Cheng and W.H. Prusoff. Biochemical Pharmacology, 22,
3099 (1973).
References
1. T. Posternak. Biochem. Prep. 2, 52 (1952).
18. I. Segel. Enzyme kinetics. Wiley Interscience, New York.
Chap. 9. 1975.
19. C.A. Janson and W.W. Cleland. J. Biol. Chem. 249, 2562
(1974).
2. G.W. Schnarr and W.A. Szarek. Can. J. Chem. 56, 1752 (1978).
3. K.-I. Yoshida, D. Aoyama, I. Ishio, T. Shibayama, and Y.J.
Fujita. J. Bacteriol. 179, 4591 (1997).
4. J. Fry, M. Wood, and P.S. Poole. Mol. Plant-Microbe Interac.
20. A. Cornish-Bowden. Fundamentals of enzyme kinetics. Port-
land Press, Ltd., London. 1995.
14, 1016 (2001).
5. G. Jiang, A.H. Krishnan, Y.M. Kim, T.J. Wacek, and H.B.
Krishnan. J. Bacteriol. 183, 2594 (2001).
6. M.P. Galbraith, S.F. Feng, J. Borneman, E.W. Triplett, F.J. de
Brujin, and S. Rossbach. Microbiology, 144, 2915 (1998).
7. R. Ramaley, Y. Fujita, and E. Freese. J. Biol. Chem. 254, 7684
(1979).
21. G. Catelani, A. Corsaro, F. D’Andrea, M. Mariani, and V.
Pistara. Bioorg. Med. Chem. Lett. 12, 3313 (2002).
22. V. Pistara, P.L. Barili, G. Catelani, A. Corsaro, F. D’Andrea,
and S. Fisichella. Tetrahedron Lett. 41, 3523 (2000).
23. R. Pongdee and H.W. Liu. Bioorg. Chem. 32, 393 (2004), and
refs. therein.
8. R. Daniellou, C.P. Phenix, P.H. Tam, M.C. Laliberte, and
D.R.J. Palmer. Org. Biomol. Chem. 3, 401 (2005).
24. S.J. Williams and S.G. Withers. Carbohydr. Res. 327, 27 (2000).
25. D. Seebach. Angew. Chem. Int. Ed. Engl. 29, 1320 (1990).
© 2006 NRC Canada