The reversible enolization and hydration of pyruvate: Possible roles of keto, enol, and hydrated pyruvate in lactate dehydrogenase catalysis

Article abstract of DOI:10.1139/v99-071

The reversible enolization and hydration of pyruvic acid and pyruvate anion were monitored using spectrophotometric methods at several temperatures. Widely varying values for the equilibrium constant for the enolization of pyruvic acid and pyruvate ion appear in the literature. To accurately determine the position of equilibrium for the enolization reaction, we have developed a method that gives consistent results in which purified samples of sodium pyruvate are first "titrated" with triiodide ion to remove any triiodide-scavenging impurities such as those resulting from aldol condensation reactions. After reequilibration to allow the regeneration of enol pyruvate, the addition of small quantities of triiodide result in an initial burst in the decrease of absorbance at 353 nm, followed by the much slower zero-order decrease due to the formation of new enol pyvuvate molecules. The absorbance change during the burst phase of the reaction is proportional to the enol concentration plus that of any triiodide-scavenging impurity which may be present in the original pyruvate solution. Thus, as the quantity of triiodide used in the pretreatment stage of the experiments is increased, these burst absorbance changes, ΔA, decrease until a constant value of ΔA is reached. Accordingly, this final ΔA value is proportional to enol pyruvate (or enol pyruvic acid) in the absence of triiodide-scavenging impurity, allowing the accurate and reproducible determinations of K_enol. The equilibrium constants for both pyruvate and pyruvic acid are relatively temperature insensitive and, typically, K_enol (pyruvate anion) = 2.6 × 10^-5 and K_enol (pyruvic acid) = 7.8 × 10^-5 at 25.0°C. The zero-order phase of the reaction of triiodide ion may be used to calculate rate constants for enolization. The hydration and dehydration of pyruvic acid were followed directly by following absorbance changes in the peak at 340 nm due to the keto group. The thermodynamic and kinetic results reported in this paper are used to help determine whether the observed "substrate" inhibition of the lactate dehydrogenase catalyzed reduction of pyruvate is actually caused by keto, hydrated, or enol pyruvate.

Full text of DOI:10.1139/v99-071

1108
The reversible enolization and hydration of
pyruvate: possible roles of keto, enol, and
hydrated pyruvate in lactate dehydrogenase
catalysis
A. Esposito, A. Lukas, J.E. Meany, and Y. Pocker
Abstract: The reversible enolization and hydration of pyruvic acid and pyruvate anion were monitored using
spectrophotometric methods at several temperatures. Widely varying values for the equilibrium constant for the
enolization of pyruvic acid and pyruvate ion appear in the literature. To accurately determine the position of
equilibrium for the enolization reaction, we have developed a method that gives consistent results in which purified
samples of sodium pyruvate are first “titrated” with triiodide ion to remove any triiodide-scavenging impurities such as
those resulting from aldol condensation reactions. After reequilibration to allow the regeneration of enol pyruvate, the
addition of small quantities of triiodide result in an initial burst in the decrease of absorbance at 353 nm, followed by
the much slower zero-order decrease due to the formation of new enol pyvuvate molecules. The absorbance change
during the burst phase of the reaction is proportional to the enol concentration plus that of any triiodide-scavenging
impurity which may be present in the original pyruvate solution. Thus, as the quantity of triiodide used in the
pretreatment stage of the experiments is increased, these burst absorbance changes, ∆A, decrease until a constant value
of ∆A is reached. Accordingly, this final ∆A value is proportional to enol pyruvate (or enol pyruvic acid) in the
absence of triiodide-scavenging impurity, allowing the accurate and reproducible determinations of Kenol. The
equilibrium constants for both pyruvate and pyruvic acid are relatively temperature insensitive and, typically, Kenol
(pyruvate anion) = 2.6 × 10–5 and Kenol (pyruvic acid) = 7.8 × 10–5 at 25.0°C. The zero-order phase of the reaction of
triiodide ion may be used to calculate rate constants for enolization. The hydration and dehydration of pyruvic acid
were followed directly by following absorbance changes in the peak at 340 nm due to the keto group. The
thermodynamic and kinetic results reported in this paper are used to help determine whether the observed “substrate”
inhibition of the lactate dehydrogenase catalyzed reduction of pyruvate is actually caused by keto, hydrated, or enol
pyruvate.
Key words: pyruvate, enolization, hydration, lactate dehydrogenase. 1117
Résumé : Opérant à plusieurs températures et faisant appel à des méthodes spectrophotométriques, on a examiné
l’énolisation réversible et l’hydratation de l’acide pyruvique et de l’anion pyruvate. Des valeurs très variables
apparaissent dans la littérature pour la constante d’équilibre pour l’énolisation de l’acide pyruvique et de l’ion
pyruvate. Dans le but de déterminer la position de l’équilibre pour la réaction d’énolisaton d’une façon précise, on a
développé une méthode qui donne des résultats reproductibles; elle implique l’utilisation d’échantillons purifiés de
pyruvate de sodium « prétitrés » avec l’ion triiodure pour enlever toutes les impuretés susceptibles de capter le
triiodure, telles que celles résultant de réactions de condensation aldolique. Après une rééquilibration pour permettre
une régénération de l’énol du pyruvate, l’addition de faibles quantités de triiodure provoque une diminution initiale
rapide de l’absorbance à 353 nm, suivie d’une diminution beaucoup plus lente d’ordre zéro causée par la formation de
nouvelles molécules d’énol du pyruvate. Le changement rapide d’absorbance dans la phase initiale de la réaction est
proportionnel à la concentration d’énol plus celle de toute impureté susceptible de capter le triiodure qui peut encore
être présente dans la solution originale de pyruvate. Donc, si l’on augmente la quantité de triiodure utilisée dans le
stage de prétraitement des expériences, ces changements rapides d’absorbance, ∆A, diminuent jusqu’à l’atteinte d’une
valeur de ∆A constante. Il en résulte que cette valeur finale de ∆A est proportionnelle à l’énol du pyruvate (ou à l’énol
de l’acide pyruvique) en l’absence d’impureté susceptible de capter le triiodure et qu’elle permet d’obtenir des valeurs
précises et reproductibles pour les déterminations de Kénol. Les constantes d’équilibre tant pour le pyruvate que pour
l’acide pyruvique sont relativement peu sensibles à la température et les valeurs types de Kénol (anion pyruvate) = 2,6
× 10–5 et Kénol (acide pyruvique) = 7,8 × 10–5, à 25°C. On peut utiliser la phase d’ordre zéro de la réaction du
Received October 5, 1998.
This paper is dedicated to A. Jerry Kresge in recognition of his many contributions to physical organic chemistry.
A. Esposito, A. Lukas, and J.E. Meany.1 Department of Chemistry, Seattle University, Seattle, WA 98122, U.S.A.
Y. Pocker.1 Department of Chemistry, University of Washington, Seattle, WA 98195, U.S.A.
1Authors to whom correspondence may be addressed. J.E.M.: Telephone: (206) 296-5942; Y.P.: Telephone: (206) 543-1672.
Can. J. Chem. 77: 1108–1117 (1999)
© 1999 NRC Canada

Esposito et al.
1109
triiodure pour calculer les constantes de vitesse d’énolisation. Les réactions d’hydratation et de déshydratation de
l’acide pyruvique sont suivies directement par les changements d’absorbance dans le pic à 340 nm dû au groupe
cétonique. Les résultats thermodynamiques rapportés dans ce travail ont été utilisés pour déterminer si l’inhibition de
« substrat » observé lors de la réduction du pyruvate catalysée par la déshydrogénase du lactate est de fait cause par
le groupe cétonique, son hydrate ou l’énol du pyruvate.
Mots clés : pyruvate; énolisation; hydratation; déshydrogénase du lactate.
[Traduit par la Rédaction]
Introduction
Esposito et al.
condensation product α-keto-γ-valerolactone-γ-carboxylic
acid (6c). One may also envisage the bimolecular aldol con-
densation of pyruvic acid, followed by dehydration, to form
an α,β-conjugated diacid. Accordingly, apparent Kenol values
determined by the Meyer method (5) are typically larger
than the actual value. In the present work, we have circum-
vented this problem by utilizing purified pyruvate solutions
that have been pretreated with triiodide ion to “titrate” out
Pyruvate is an important intermediate that is linked to
many biochemical processes regulated by enzymes. In the
glycolytic pathway, for example, the anaerobic regeneration
of nicotinamide adenine dinucleotide (NAD+) from its re-
duced form (NADH) is facilitated by the reduction of
pyruvate to lactate in the presence of lactate dehydrogenase
LDH (EC1.1.1.27). Under physiological conditions,
pyruvate exists as an equilibrated mixture of enol, keto, and
–
the I3 scavenging impurity. After the reequilibration be-
tween enol and keto forms we utilize a modification of the
Meyer (5) method to determine Kenol. On the other hand, be-
cause the burst of triiodide uptake is relatively rapid, the fi-
nite time required for the complete mixing of the reaction
components renders it difficult to accurately extrapolate
spectrophotometric absorbances back to the kinetic zero.
And since the burst absorbance changes, ∆A, typically are
relatively small, the uncertainty of the absorbance at the ki-
netic zero can result in large deviations of the apparent Kenol
from the actual values. In this report, we describe our solu-
tion to this problem as well.
Investigators have also attempted to approximate Kenol by
independently determining the rates of enolization of keto
pyruvate and ketonization of enol pyruvate using the rela-
tionship Kenol = kenol / kketo (1g, 1h, 2b). Again, using this
method, uncertainties may arise because of differences in the
experimental conditions under which the two reactions are
independently carried out (7a–7c, 8a, 8b).
It has been demonstrated that keto pyruvate serves as the
preferential substrate of LDH (9, 10). This enzyme is one of
many oxidoreductase enzymes that is subject to “substrate”
inhibition (11). Although substrate inhibition for LDH is
more pronounced for the heart than for the muscle
isoenzymes (12), the physiological importance of this mode
of inhibition in warm-blooded animals is controversial (13a–
13d). Somero (14), however, has shown that the substrate in-
hibition of the enzyme in certain fishes becomes increas-
ingly more pronounced at lower temperatures and has
suggested that this phenomenon serves an important role in
biological adaptation.
An earlier report by Coulson and Rabin (15a) suggested
that enol pyruvate is a potent inhibitor of LDH and is
responsible for the observed substrate inhibition of the en-
zyme. That original report was quoted widely in biochemis-
try textbooks (15b) and other references (15c) for a number
of years. Although the original suggestion (15a) was based
on reasonable experimental results, an alternative interpreta-
tion of these results is indicated by more recent, detailed
studies of the kinetics and thermodynamics of the various
nonenzymatic reactions of pyruvic acid and pyruvate anion
in aqueous solution. Indeed, studies have since been con-
ducted to determine the possible role of hydrated pyruvate
hydrated forms (1a–1f).
Equilibrium constants have been determined in previous
work for the reversible hydrations of pyruvic acid and
pyruvate anion using pmr (1a) and spectrophotometric meth-
ods (1f) and the various data obtained are in relatively good
agreement when due consideration is taken of the fact that
the stoichiometry of the hydration of pyruvic acid involves
three water molecules (1a). Because of the electron-
withdrawing effect of the carboxyl group, pyruvic acid is
considerably more hydrated in aqueous solutions than is the
pyruvate ion. The hydrations of pyruvic and pyruvate ion are
both known to be exothermic (1a, 1d, 1f).
The results of kinetic studies demonstrate that the revers-
ible hydration of pyruvate anion is a general-acid, general-
base-catalyzed process, which includes a small catalytic
term for general-base catalysis by the pyruvate anion itself
(2a, 2b, 2c). The kinetic studies also suggest that the sponta-
neous hydration of pyruvate anion involves the contribution
of a kinetic term representing an intramolecular general-
base-catalyzed path involving the neighboring carboxylate
group (2b).
Although there is little ambiguity associated with the de-
termination of kinetic parameters for the enolization of
pyruvate and pyruvic acid (2b, 3, 4a, 4b), reported equilib-
rium constants for these reversible processes vary (1g, 1h,
2b, 4b). In the Meyer method (5) the burst of triiodide up-
take, presumably by enol pyruvate (or enol pyruvic acid), is
spectrophotometrically measured at 353 nm. This method
generally gives unreliable values of Kenol because solutions
of pyruvic acid or pyruvate contain a triiodide-scavenging
impurity. For this reason, the classic protocol for the Meyer
method (5) often gives rise to inaccurate determinations of
the enol content of carbonyl compounds in solution (6a, 6b).
The existence of an impurity formed in samples of pyruvic
acid was reported many years ago and identified as the aldol
© 1999 NRC Canada

1110
Can. J. Chem. Vol. 77, 1999
–
–
I3
as well as enol pyruvate in the observed substrate inhibition
of the enzyme (16, 17).
of the association constant for the reaction I2 + I– » I3 , K
= [I3 ]/[I2][I–] (20a, 20b). The initiation of all kinetic runs
–
was accomplished with the rapid mixing of the final reaction
components by means of a small stirring paddle attached to
the needle tip of the Hamilton microlitre syringe (21).
These reactions are known to take place in two phases
(2b, 3, 4): a rapid burst phase in which enol pyruvate (and
any unsaturated impurities inherent in the original pyruvate
It was first proposed by Fromm (18) that the substrate in-
hibition of the enzyme is the result of the formation of an
abortive ternary complex with the stoichiometric composi-
tion of LDH, pyruvate, and NAD+. Two modes of LDH inhi-
bition by pyruvate have been proposed (17, 19) involving,
respectively, (i) the rapidly reversible formation of a ternary
complex (LDH·NAD+·pyruvate), and (ii) the slowly revers-
ible formation of a binary complex between LDH and a co-
valently bound adduct of NAD+ and pyruvate resulting in
the formation of LDH·NAD+–pyruvate. Both hydrated
pyruvate and enol pyruvate have been reported to form ter-
nary abortive complexes with LDH and NAD+ (16, 17).
The kinetic and thermodynamic parameters associated
with the reversible hydration and enolization of pyruvate
may be used to assess the likelihood of each of the different
forms of pyruvate serving as potential participants either as
enzyme substrates or as enzyme inhibitors. Such data pro-
vide a basis to determine whether there exists a high enough
concentration of a given form of pyruvate to explain a hypo-
thetical kinetic scheme or whether a given form of pyruvate
may be formed rapidly enough to be consistent with known
kinetic parameters. Over the years, two of the present au-
thors have carried out detailed investigations of the revers-
ible hydration and enolization of pyruvic acid and pyruvate
anion (1a, 2a–2c, 3a, 3b, 9, 16). In the present paper, we
provide selected thermodynamic and kinetic data for these
reactions and show how these data can help address two
questions: (i) which chemical form of pyruvate (enol, keto,
or hydrate) is involved in the substrate inhibition of LDH?
(ii) is the substrate inhibition recognized by Somero at lower
temperatures a property of the enzyme alone, or is it also de-
pendent on variations in the position of equilibria between
enol, keto, and hydrated pyruvate as a function of tempera-
ture?
–
sample) reacts directly with the added I3 , followed by the
–
zero-order reaction of I3 with enol pyruvate as it is formed
in the rate-determining step. In attempting to employ the
Meyer method (5) in which the burst phase absorbance
change, ∆A, is used to determine equilibrium concentrations
of enol pyruvate, [enol] = ∆A/ε{1 + 1/(KI [I–])}, it has been
–
found by us (2b) and others (4a) that the3values obtained are
usually considerably larger than those determined through
–
other methods, suggesting the existence of an I3 scavenging
impurity in the pyruvate samples. Even using recrystallized
sodium pyruvate, the method described above gives rise to
an equilibrium constant for enolization, Kenol = 1 × 10–4,
which is considerably larger than that reported previously
(1g, 1h, 2b).
We found, however, that the Meyer method could be
adapted to allow the determination of valid values of Kenol
–
by removing or inactivating the I3 scavenging impurity. Ac-
cordingly, we pretreated the buffered pyruvate solutions with
–
small quantities of I3 to react with and, therefore, remove
the impurity. After the pyruvate in solution reacted com-
–
pletely with the added I3 (note the pyruvate concentration
was always much larger than that of iodine) it was allowed
to reequilibrate with its enol form. The time allowed for
reequilibration was always greater than 10 half-lives for the
known rate of reequilibration between keto and enol
pyruvate (2b). After reequilibration had taken place, we in-
–
jected 0.05 mL additional I3 to initiate the burst uptake of
–
I3 by the equilibrium concentration of enol pyruvate. The
burst was again followed by the zero-order production of
new enol pyruvate. Thus, we were able to essentially titrate
–
out the I3 scavenging impurity by varying the amount of
Experimental section
triiodide used in pretreatment.
Materials used to prepare buffer solutions, sodium pyruvate,
pyruvic acid, and all other reagents were in their commer-
cially available reagent-grade forms. Pyruvic acid was distilled
through a Vigreux column in an atmosphere of nitrogen gas,
and sodium pyruvate was recrystallized using methanol–water
mixtures. All solutions were prepared from distilled, dei-
onized water. Spectrophotometric measurements were made
using either a Shimadzu UV-2101PC or a Varian Cary 210
spectrophotometer. Both instruments gave rise to essentially
identical kinetic and thermodynamic parameters. The tem-
perature of the cell compartments was regulated and maintained
constant by a VWR 1160 thermostated bath and circulator.
The enolization of pyruvate anion was monitored by
following its iodination at 353 nm as described in earlier
work (2b, 3, 4). Typically, 0.10 M sodium pyruvate solutions
were prepared directly prior to use in 0.04 M phosphate
buffers in which the ionic strength was adjusted to 0.35 M
by the addition of appropriate quantities of sodium chloride.
Most of the reactions were initiated by the addition of
0.05 mL of 2.05 × 10–3 M I2 in 1.0 M KI to the 3 mL reac-
tion cuvettes containing the buffered solutions of pyruvate.
For each temperature studied, we used the appropriate value
Because the time required for complete mixing of the re-
action mixtures (ca. 3–5 s) interferes with monitoring the
initial stages of the burst phase, it is not possible to accu-
rately extrapolate absorbances of the burst phase back to the
–
kinetic zero. Thus, the actual absorbance of the I3 solution
used to initiate the run after pretreatment must be be deter-
mined independently, using experimental conditions that
simulate those used in the actual iodination experiments.
Even when using 0.1 M pyruvate, the absorbance differ-
ences, ∆A, used to calculate the equilibrium concentration of
enol pyruvate are relatively small (typically from ca. 0.06 to
–
0.07), so that the precise determination of A is crucial. In
these determinations, account must be takenIo3 f the residual
–
iodide ion left over from the original I3 pretreatment since
–
the added I– increases the concentration of I3 , thereby in-
creasing the spectrophotometric absorbance. In Table 1, we
exemplify the absorbancies observed from solutions at vari-
ous initial concentrations of I– and I2 at 25°C. It will be
–
noted that our values of A
agreement with those calculated, A (calcd), on the basis of
in Table 1 are in excellent
(obsd)
I3
–
I3
–
published values of KI (20a, 20b).
3
© 1999 NRC Canada

Esposito et al.
Table 1. Absorbance of I3 solutions at 353 nm and 25.0°C.
1111
–
[I–] (M)
0.0082
1.68
1.45
0.0164
3.36
3.10
0.0263
3.36
3.19
0.0360
3.36
3.23
0.0455
3.36
3.27
0.0549
3.36
3.28
0.0639
3.36
3.29
0.0323a
3.30
3.17
0.0476a
3.29
3.20
105[I2]0 (M) b
[I3 ] (M) c,d
–
AI – (obsd) e
0.363
0.778
0.796
0.813
0.821
0.825
0.828
0.794
0.807
3
εobsd (M–1 cm–1)
25 030
0.364
25 100
0.778
24 960
0.801
25 170
0.811
25 110
0.821
25 150
0.823
25 170
0.826
25 020
0.796
25 220
0.804
AI – (calc) f
3
–
–
aThe data in the last two columns refer to absorbances of I3 in the presence of 0.01 M pyruvate anion prior to enolization and in the absence of I3
scavenging impurity. See text for experimental protocol.
bFormal concentration.
–
cAssuming: KI = [I3 ]/[I2][I–] = 774 at 25.0°C. References 20a and 20b give comparable values of KI
−
over two overlapping temperature ranges.
–
3
3
d[I3 ] = {774(3.36 × 10–5)[I–]}/(1 + 774[I–]).
–
eEach value represents the average of 10 determinations. The average standard deviation for these values was 0.0025 absorbance units.
fUsing the average ε = 25 100 M–1 cm–1 calculated from the data in this table.
–
Our observed values of the absorbance of I3 in the ab-
sence of pyruvate,
In earlier papers, we (1a) and others (1b–1h) reported that
–
A
(obsd), were compared with the extrap-
aqueous solutions of pyruvic acid below pH 2.2 consist pre-
dominantly of the hydrated form, while at higher values of
pH the pyruvate anion is hydrated to a much lesser extent.
We also demonstrated from pmr experiments the third-power
dependence of the ratio of 2,2-dihydroxypropionic to
pyruvic acid on the water concentration (1a), [hy-
drate]/[keto] = Keq[H2O]3 (3a), a finding that has not always
been taken into consideration in reports from other laborato-
ries. Our spectrophotometric determinations in the present
work were carried out using 0.098 M pyruvic acid and the
olated linear portioIn3 of our iodination runs, A0,extrap to
determine ∆A for the direct and rapid reaction of triiodide
with enol pyruvate. According to Davies and Gwynne (20a),
–
KI is independent of ionic strength up to 1 M, and our own
ob3servations are in agreement with this finding. Nonethe-
–
less, the question arises whether the value of A in the ab-
I3
sence of pyruvate is the same as that in the presence of the
0.098 M pyruvate solutions used in the iodination runs. To
determine this, we performed a number of very simple ex-
periments. We first added triiodide to the buffered solution
of pyruvate and when the zero-order portion of the reaction
was complete, we immediately injected 0.05 mL of the
triiodide solution into the same reaction mixture. This latter
injection gives rise to no burst reaction whatsoever, showing
values of χhyd = [hydrate]/([hydrate] + [keto]) and Khyd
=
[hydrate]/[keto] given in this paper were determined at that
concentration.
The largely different fractions of hydration of pyruvic acid
and pyruvate anion have been utilized in a kinetic method de-
signed to follow the dehydration of 2,2-dihydroxypropionate
in aqueous buffers (2a, 2b). Rate constants reported in the
present work were determined from kinetic runs that were
initiated by the injection of an acidic solution of pyruvic
acid (with a relatively high fraction of hydrated pyruvic
acid) into a neutral buffered solution along with sufficient
sodium hydroxide to immediately neutralize the added acid.
The essentially instantaneous deprotonation of 2,2-dihydrox-
ypropionic acid resulted in its dehydration to keto pyruvate,
a reaction that was monitored directly by the increase in
absorbance at 325nm (λmax = 325 nm, ε = 20 M–1 cm–1).
Typically, 0.10 mL of an acidic solution of 0.23 M pyruvic
acid (pH 1.45) was injected into 3.0 mL of the buffer solu-
tion. First-order rate constants, kdehyd, were determined from
the linear plots of ln(A∞ – At) vs. t.
–
that the I3 scavenging impurity has been deactivated and
that no new enol has yet had the time to form. The linear
extrapolation of absorbance to the endpoint of the first reac-
tion can be made with certainty and gives yet another means
–
of determining A . It will be noted, then, that this value
was determined iIn3 the presence of “pure” pyruvate, but in
the absence of triiodide-scavenging material either in the
form of impurity or as enol pyruvate. The last two columns
–
of data in Table 1 exemplify values of A determined in this
way. These data demonstrate that A is not only insensitive
to ionic strength but also independent of the presence or ab-
I3
–
I3
sence of keto pyruvate anion.
For the determination of the kinetic data for the eno-
lization of pyruvate anion and pyruvic acid, reaction velocities
were calculated from the linear portions of the spectrophoto-
–
–
metric traces at 353 nm, venol = ∆A/ε∆t{1 + 1/(K [I ])} and
Since the dehydration of 2,2-dihydroxypropionate is sen-
sitive to catalysis both by acids and bases (2b, 22)
I3
the corresponding observed pseudo-zero-order rate constants,
kenol = venol/[keto pyruvate]. The reversible enolization of
pyruvate is subject to general-base catalysis (2b, 2c) so that
the observed rate constants are actually a sum of kinetic
terms:
+
kdehyd = k0 + kH O [H3O ] + kOH– [OH–]
+
[3]
3
+ kHA[HA] + kA– [A–]
[2]
kenol = k0 + kB[B] + kOH-[OH–]
where HA and A– refer, respectively, to the acid and conju-
gate base of the buffer system employed. At the approxi-
mately neutral pH’s used in the present work, the kinetic term
in hydronium ion is negligible (2b). The rate of equilibration
between keto and enol forms, kequil,hyd = khyd + kdehyd, may be
calculated from kequil,hyd = kdehyd(1 + Khyd), where Khyd is the
equilibrium constant for hydration, Khyd = [hydrate]/{keto].
where B is the basic buffer component and k0 is the “sponta-
neous” rate usually taken to be k0 = kH O [H2O]. Rate con-
stants for the equilibration between enol and keto forms,
kequil, enol = kenol + kketo, were calculated using the equilib-
rium constant for the reaction, Kenol = [enol]/[keto], and the
2
half-lives for the equilibrations from τ1/2 = 0.693/kequil,enol
.
© 1999 NRC Canada

1112
Can. J. Chem. Vol. 77, 1999
Table 2. Half-lives, t1/2, for the equilibration of the reversible enolization and hydration of pyruvate anion and pyruvic acid.
Enolizationa,b
Hydrationa,c
τ1/2 (s)
T (°C)
Solution
pH
τ1/2 (s)
Pyruvate anion
5
25
25
25
0.01 M Phosphate buffer
0.04 M Phosphate buffer
0.20 M Phosphate buffer
0.20 M Imidazole buffer
6.9
6.9
7.0
7.0
100d
2.8f
0.6f
0.6f
120d,e
22f
8f
15f
Pyruvic acid
0
11
15
20
25
0.1 M HCl
0.1 M HCl
0.1 M HCl
0.1 M HCl
0.1 M HCl
1.15
1.15
1.15
1.15
1.15
570g
140f
110g
70g
11h
40g
ca. 1e
akequil = kf + kr; Keq = kf/kr, where Keq values for enolization are taken from Table 2 of the present work and for hydration from ref. 1a.
bWhere t1/2 = 0.693/kobsd and kobsd = k0 + kB[B].
cWhere t1/2 = 0.693/kobsd and kobsd = k0 + k
[H3O+] + kOH-[OH–] + kB[B] + kBH+ [BH+].
H 3O+
dPresent work.
eThis value is slightly dependent on pyruvate ion concentration (see ref. 2b) That reported here applies to 1 × 10–4 pyruvate so that it ccorresponds
more closely to the concentration ranges used for its enzymatic reactions with LDH.
fReference 2b.
gReferences 3a, 3b.
hReference 2a.
Again, half-lives (τ1/2) were calculated from τ1/2
0.693/kequil,hyd
=
and for the correspondinng equilibration for pyruvic acid in
0.1 M HCl.
.
Finally, the hydration of pyruvic acid in aqueous hydro-
chloric acid solutions (pH <2.2) was initiated by injecting
0.01 mL of neat pyruvic acid, or the equivalent amount di-
luted with acetonitrile, into 3 mL of hydrochloric acid solu-
tions of varying concentrations. For the hydration of pyruvic
In developing our experimental protocol for the Meyer
method (5), we found that optimal conditions for determin-
ing accurate and reproducible values of Kenol utilized a con-
centration of pyruvate around 0.1 M (actually 0.098 M). The
resultant equilibrium concentration of enol pyruvate allowed
an easily measured burst of triiodide uptake, ∆A ≈ 0.06
absorbance units, while at the same time the background
absorbance due to keto pyruvate at 353 nm was maintained
low enough to allow the zeroing of the instrument. However,
since the burst-phase absorbance change was relatively
small, determination of the actual absorbance of triiodide at
the kinetic zero for the enolization of pyruvic acid and
pyruvate anion is crucial to the accurate determination of
+
+
acid (2a), khyd = k0 + kH O [H3O ], kequil,hyd = khyd(1 + Khyd)/Khyd
.
3
Results and discussion
In the present investigation, we observed that the magni-
tude of the “burst” region resulting from the intial reaction
of triiodide in solutions of pyruvate and pyruvic acid some-
times increased markedly on standing, implying that a
triiodide-scavenging impurity forms over time. The rate of
formation of this impurity appears to depend on the pH of
K
enol. Since it is not possible to directly observe the original
absorbance due to triiodide anion, it must be determined as
described in the Experimental section. The close comparison
of the data in the last two columns of Table 1 with the rest
the solution. In neutral phosphate buffers, the initial triiodide
–
–
of the data in the table demonstrates that A in the absence
–
uptake = ∆A/ε{1 + 1/(KI [I ])} remains constant for up to
I3
several days, while under3more acidic conditions it increases
much more rapidly. For example, at pH 1.3, triiodide uptake
in pyruvic acid solutions (unpretreated with triiodide ion)
varied as follows: 1.1 × 10–5 (1 h), 1.5 × 10–5 (2 h), 3.3 ×
10–5 (4 h), 6.0 × 10–5 (8 h), 9.2 × 10–5 (24 h). Consequently,
when using the Meyer method to make valid determinations
of Kenol, one must be aware of a finite “window of opportu-
nity” for the previously described pretreatment incubation
studies using triiodide. Depending on the pH and tempera-
ture for a given series of such determinations, the limits of
this “window” used in the present work were such that the
experiments were always carried out after 10 half-lives for
the reequilibration between keto and enol pyruvate, but be-
of pyruvate is the same as that in the presence of 0.1 M so-
dium pyruvate. Not only is this an important finding that
–
A
confirms the validity of the value of
we have used in
I3
–
–
determining Kenol = ∆A/ε{1 + 1/(KI [I ])}, but it also sug-
3
gests the lack of formation of a complex from the reaction of
–
hydrated pyruvate and I3 as proposed earlier by Albery and
Bell (4a).
Figure 1 shows the results of experiments in which the
magnitude of the burst phase, ∆A, was measured as a func-
–
tion of the amount of I3 used in the pretreatment of
pyruvate solutions. The experiments were conducted as fol-
lows. Solutions (3 mL) of buffered pyruvate (0.098 M) were
pretreated with various quantities of triiodide; this phase of
the iodination was allowed to go to completion and then in-
cubated for at least 10 half-lives to allow the reequilibration
between enol and keto pyruvate. After that time, 0.05 mL of
a solution of 2.05 × 10–3 M I2 in 0.10 M KI was injected
–
fore the time where a measurable quantity of I3 scavenging
impurity was known to reform in the pyruvate solutions.
Table 2 lists the half-lives for the equilibration of the
enolization of pyruvate anion in selected buffer solutions
© 1999 NRC Canada

Esposito et al.
1113
Fig. 1. Burst as a function of the total iodine concentration used in pretreating solutions of pyruvate anion.
into the incubated mixture. The plot illustrated in the figure
indicates that pretreatment of the buffered pyruvate solution
with triiodide effectively removes the I3 scavenging impu-
rity until, at approximately a total iodine concentration of
1.5 × 10–5 M, a constant value for the “burst” is reached.
Table 3 lists values of the fraction of enolization deter-
mined as described above for pyruvate and pyruvic acid at
various temperatures. It is important to note from the data in
We note that Meyer method determinations of Kenol re-
ported earlier (1g, 1h) were made using approximately 1/10
the concentration of pyruvate used in the present work.
Since no mention was made in those reports concerning the
use of a longer pathlength reaction cell, we assume the path-
length was the standard 1 cm. Using the value of Kenol and
the extinction coefficient ε = 26 000 cm–1 M–1 used in those
earlier calculations, the absorbance change resulting from
the reaction of triiodide with the equilibrium concentration
of enol pyruvate would be approximately 0.004 units, a value
that would be difficult to measure reproducibly. Finally, we
were surprised at the relatively low value of Kenol reported,
given the fact that no precautions were apparently taken to
rid the original pyruvate solution of triiodide-scavenging material.
Burgner and Ray (4b), on the other hand, reported the de-
termination of Kenol = 4.1 × 10–6 for pyruvate anion using
the Meyer method at a sodium pyruvate concentration of 1.0
M. We were unable to reproduce their experiments since, at
353 nm, a 1.0 M solution of pyruvate across a pathlength of
1 cm has an absorbance much greater than 10 units. Even
using their reported pathlength of 0.4375 cm would result in
an absorbance due to keto pyruvate of around 5 units, mak-
ing it virtually impossible to obtain accurate values of ∆A
using the instrumentation they reportedly used. Again, since
the work was presumably carried out using unpurified keto
pyruvate and was untreated with the necessary initial aliquot
–
the table that the calculated fractions of the enol forms, χenol
,
remain relatively constant in the temperature range 11–25°C
both for pyruvate anion and for pyruvic acid and, for this
reason, the evaluation and interpretation of thermodynamic
parameters would not be meaningful. In fact, the corre-
sponding values of Kenol = χenol/(1 – χhydration) are even more
similar over the pH range studied. It will be noted that the
enolization of pyruvate anion is apparently slightly endo-
thermic in contrast to the exothermic hydration of keto
pyruvate. While the variation in Kenol for pyruvic acid is
within the limits of experimental error, the corresponding
hydration of pyruvic acid is clearly exothermic.
For pyruvate anion, at 25°C, χenol = [enol]/([enol] + [keto])
= 2.5 × 10–5 (Table 3). It will be noted that this value is
somewhat larger than that reported by other workers (1g, 1h)
using the Meyer method. Actually, our attempts at estimat-
ing, via direct extrapolation, the absorbance of the burst
phase at the kinetic zero gave a value which was very simi-
lar to that in the literature cited above (1g, 1h). However, we
observed that the mixing time interrupted the spectrophoto-
metric monitoring of the decrease in absorbance of triiodide
to the extent that the extrapolated triiodide absorbance at the
kinetic zero was impossible to measure accurately and was,
–
of I3 to rid the original pyruvate solution of triiodide-
scavenging impurity, the reliability of these data must be
questioned.
We found values of Kenol for pyruvic acid to be about 3
times larger than those for pyruvate anion, perhaps partially
due to the electron-withdrawing inductive effect of the car-
boxyl group of pyruvic acid. We must note, however, that
–
in fact, consistently lower than the actual values of A (obsd)
I3
given in Table 1.
© 1999 NRC Canada

1114
Can. J. Chem. Vol. 77, 1999
Table 3. Equilibria for the enolization and hydration of pyruvate anion and pyruvic acid.
Pyruvate anion
T (°C)
11
20
25
a,b
χhyd
0.11
0.069
0.054
a,b
Khyd
0.12
0.074
0.057
χenol
(1.6 ± 0.2) × 10–5
(1.9 ± 0.2) × 10–5
(1.7 ± 0.2) × 10–5
(2.0 ± 0.2) × 10–5
(2.5 ± 0.2) × 10–5
(2.6 ± 0.2) × 10–5
b,c
Kenol
Pyruvic acid
T (°C)
11
15
25
a,b
χhyd
0.68
0.65
0.58
a,b
Khyd
2.1
1.9
1.4
χenol
(2.2 ± 0.2) × 10–5
(6.9 ± 0.5) × 10–5
(2.4 ± 0.2) × 10–5
(6.8 ± 0.5) × 10–5
(2.9 ± 0.2) × 10–5
(6.9 ± 0.5) × 10–5
b,c
Kenol
aThese apparent values of χhyd and Khyd for the hydration of pyruvic acid vary with the third power of the water concentration
(see ref. 1a). The values of Khyd and χ given in this table correspond to a total pyruvate or pyruvic acid concentration of 0.098 M.
bPortions of these data were taken from earlier work (see ref. 1a).
cValues take into consideration the fraction of hydration, so that Kenol = [enol pyruvate]/[keto pyruvate].
there is not always close correlation between inductive ef-
fects of groups attached to the carbonyl of ketones and the
magnitude of their values of Kenol. For example, although the
The method of determining Kenol for carbonyl compounds
from the ratio of rate constants for ketonization to the corre-
sponding quantities for enolization obviates the difficulty as-
sociated with minute levels of triiodide-scavenging impurity
in solutions of the carbonyl compound to be studied. How-
ever, in the case of the pyruvic acid – pyruvate anion sys-
tem, there can exist uncertainties in the calculation of Kenol
from the ratio of rates of enolization to ketonization. In our
own analysis in earlier work (2b), for example, we approxi-
mated Kenol = kenol/kketo = 8.6 × 10–5 based on accurate and
reproducible values for kenol but, at the same time, using val-
ues for kketo (7a) that necessitated approximation to compen-
sate for the different experimental conditions (e.g., ionic
strength, buffer composition, temperature) under which the
two reactions were independently carried out. We are also
concerned that, in general, the validity of kenol/kketo ratios
may be weakened when the determination of values of kketo
require complex or ambiguous kinetic methods (1g, 1h, 7a,
7b, 8a, 8b) involving reaction intermediates. For example,
when attempting to determine kketo for pyruvic acid by intro-
ducing solutions rich in enol content into aqueous acidic so-
lutions, the ketonization of pyruvic acid enol would be
paralleled by the even more rapid hydration of keto pyruvic
acid, thereby complicating the interpretation of the ensuing
decreasing absorbances.
–
inductive effects exerted by -CH3 and -CO2 are similar,
Kenol for acetone (4.7 × 10–9) (23a, 23b) is almost four or-
ders of magnitude lower than that for keto pyruvate (2.5 ×
10–5). It is conceivable that the enol form of the pyruvate an-
ion is stabilized, to some extent, via intramolecular hydro-
gen bonding between the enol hydroxy and the negatively
charged carboxylate oxygens. Thus any increase in Kenol for
pyruvic acid in comparison to pyruvate anion, caused by the
electron-withdrawing inductive effect of -CO2H, may be
partially offset by such a mode of stabilization of the enol
pyruvate anion. Unfortunately, because of the small changes
in values of Kenol as a function of temperature in the range
studied, it is not possible to calculate reliable values for the
entropies of the respective enolizations of pyruvate anion
and pyruvic acid.
On the other hand, the data in Table 3 indicate a very
large difference in the extent to which pyruvic acid and
pyruvate anion undergo hydration. It will be noted that these
acyl addition reactions involve bond making between the
nucleophile and the electrophilic carbonyl carbon. Accord-
ingly, one would expect that electronic inductive effects
caused by groups directly attached to the carbonyl group
would be of greater importance for hydration than for the
corresponding enolization reaction. Indeed, α-halogen sub-
stituents are known to greatly increase fractions of hydration
of carbonyls. For example, at 25°C, whereas the fraction of
hydration for pyruvate anion is ca. 0.06, that for
fluoropyruvate anion is about 0.85 (9), a value that is even
greater than that for the neutral pyruvic acid molecule.
The comparison of catalytic ratios for the ketonization and
enolization of the pyruvic acid – pyruvate anion system may
present additional difficulties in which the analysis of data
requires certain assumptions. For example, values of Kenol
=
+
H3 O+ ,keto
kOH– , enol /kOH
and Kenol, pyruvic acid = k
,enol /k
– , keto
H3 O
were evaluated in earlier work (1g) It must first be noted
–
that reported values (1g, 2b) of kOH ,enol vary by as much as
60%. Furthermore, it was observed that kketo is nonlinear for
both pyruvate anion with increasing hydroxide concentra-
tion, and for pyruvic acid with increasing hydronium ion.
Thus, although the kinetic experiments themselves may be
For pyruvic acid, a recent report (1g) gives the value Kenol
= 6.76 × 10–4 using the Meyer method. This larger value, al-
most 10 times larger than the value determined in the pres-
ent investigation (Table 3), may be due to the formation of
the triiodide-scavenging impurity that we observed to occur
in acidic solutions of pyruvic acid as described earlier.
carried out carefully, calculated values of kOH ,keto and
-
kH O ,keto at higher concentrations of OH– and H3O+, respec-
+
3
© 1999 NRC Canada

Esposito et al.
1115
tively, depend on the appropriateness of the theoretical
kinetic model used in its calculation. For these reasons, the
use of spontaneous rate constants for ketonization and
in solutions of pyruvic acid at pH 1.3. Furthermore, the data
in Table 2 demonstrate that at 0°C the half-life for
enolization is around 10 min and, accordingly, after even 1 h
the equilibrium between keto and enol pyruvate would have
been virtually complete. Thus all of the enzyme assays car-
ried out by Coulson and Rabin (15a) actually involved the
same fraction of enolized pyruvate and, therefore, any varia-
tions in LDH inhibition in these experiments were unrelated
to the content of enol pyruvate in solution. In fact, the ther-
modynamic data in Table 3 indicate that the pyruvic acid
solutions at pH 1.0 at 0°C utilized in that study contained
mostly hydrated pyruvate.
enolization may give less ambiguous values of Kenol
=
k0,enol/k0,keto, provided, of course, that valid determinations of
the spontaneous rate constants are made under identical
reaction conditions. For pyruvic acid, k0,enol values that are
in close agreement (when corrected for fractions of
hydration) have been reported at 25°C (3a, 3b, 4a), k0,enol
=
1.2 × 10–6 s, and when this average value is compared to
k0,keto for pyruvic acid from the work of Chiang et al. (1g),
k0,keto = 1.06 × 10–2 s, it follows that Kenol = k0,enol/k0,keto
=
1.2 × 10–6 s/1.06 × 10–2 s = 1.1 × 10–4, which is in closer
agreement with the value we report in Table 3 (6.9 × 10–5, at
25°C), allowing for differences in experimental conditions.
The kinetic and thermodynamic parameters associated
with the reversible hydration and enolization of pyruvate re-
ported in this paper may be used to access the likelihood of
the keto, hydrated, and enol forms of pyruvate to serve ei-
ther as enzyme substrates or as enzyme inhibitors. Spe-
cifically, the availability of these data allow a closer look at
the experimental data determined by Coulson and Rabin
(15a).
In that report, it was concluded that the substrate inhibi-
tion observed in the LDH reduction of pyruvate by NADH
was actually due to a substance that can be removed from
pyruvate solutions by column chromatography. It was sug-
gested that the enzyme-inhibiting substance was enol
pyruvate and not 2,2-dihydroxypropionate (hydrated pyruvate)
because, at that time, it was believed that hydrated pyruvate
could not be observed at neutral values of pH. However, ac-
cording to the thermodynamic data provided in Table 3 it is
now known that there exists over 2000 times more hydrated
pyruvate than enol pyruvate at neutral pH at room tempera-
ture.
Buffer, ionic strength, and pH effects on the magnitude of
inhibition constants have been observed for LDH (16) and
for other enzymes (24a–24f). For example, whereas at neu-
tral pH a solution of 1 × 10–3 M keto pyruvate results in
27% substrate inhibition in 0.1 M phosphate buffer, the
same concentration of keto pyruvate causes only 9% sub-
strate inhibition in 0.1 M diethylmalonate buffer(16).
Schwert and Winer (25) have also observed greater substrate
inhibition in phosphate buffers than in other buffers. There-
fore we suggest that the differences Coulson observed in the
degree of substrate inhibition arising from assays in 0.2 M
imidazole and 0.2 M phosphate buffers were due to specific
interactions between the buffer components and the en-
zymes.
Somero (14a, 14b) has demonstrated that the substrate in-
hibition observed for muscle LDH isolated from the fish
Gillichtghys mirabilis is much stronger at colder tempera-
tures than at warmer temperatures. He suggested that this
phenomenon is a property of the enzyme itself. In a more re-
cent article (26), it is reported that differences in enzyme–
substrate binding and thermal stabilities of Sphyraena LDH-
A from various barracuda species are actually due to amino
acid residues outside the active site.
It is also instructive to interpret other of Coulson and
Rabin’s results (15a) in light of the kinetic data presented in
Table 2. Kinetic runs were carried out in the earlier study
(15a) using chromatographically purified solutions of
pyruvic acid to oxidize NADH in the presence of LDH at
room temperature. Prior to the introduction of NADH and
LDH to initiate the kinetic run, the pyruvate was incubated
in 0.2 M buffer solutions for 2 min. It was noted that while
substrate inhibition was not observed in 0.2 M imidazole
buffer, rather strong inhibition was noted in 0.2 M phosphate
buffer. Because of the lack of reliable kinetic data at that
time, it was tentatively concluded that the purified pyruvate
enolized rapidly in the phosphate buffer, thereby forming the
actual substance that inhibited LDH. It was argued that the
enolization was so slow in imidazole buffer that no such in-
hibition was observed. The data in Table 2, however, show
that the half-life for the equilibration between keto and enol
pyruvate is less than a second in either 0.2 M phosphate or
0.2 M imidazole buffer. Consequently, identical (equilib-
rium) concentrations of enol pyruvate would have prevailed
under the experimental conditions in the article cited (15a).
It was also stated in that article that purified pyruvic acid so-
lutions were stable at pH 1.0 for several hours at 0°C. How-
ever, according to the results we now report at the beginning
of this section, we find that measurable quantities of a
triiodide-scavenging impurity begin to form after about 1 h
However, we have been able to show that for trout muscle
LDH, the increased sensitivity of substrate inhibition with
decreasing temperatures is due, at least in part, to the equi-
librium shift that increases the fraction of hydrated pyruvate
at lower temperatures (27a–27c). Similar results were ob-
tained using purified beef muscle and beef heart isoenzymes
of LDH (16, 27a–27c) in which substrate inhibition was ob-
served to be more severe in the case of the latter. In a typical
set of experiments with beef heart LDH (16), small quanti-
ties of preincubated solutions of pyruvate (pyruvic acid)
were added to reaction mixtures containing NADH and LDH
to initiate kinetic runs in phosphate buffer at neutral pH at
5.0°C. The preincubated solutions of pyruvate (pyruvic acid)
were as follows: (i) prepared in phosphate buffer and
preincubated at 41°C (χhydrate = 0.054); (ii) prepared in phos-
phate buffer at 5.0°C (χhydrate = 0.136); (iii) prepared in HCl,
pH 4.5, at 41°C, yielding an aqueous pyruvic acid solution
(χhydrate = 0.412); (iv) prepared in HCl, pH 4.9, at 5.0°C,
yielding an aqueous pyruvic acid solution (χhydrate = 0.801).
The data in Table 4 show that increasing the initial fraction
of hydrated pyruvate in the assay results in increased sub-
strate inhibition in the enzymatic reactions. Since the four
experiments illustrated in Table 4 represent kinetic runs at
5.0°C, the kinetic differences cannot be a property of the en-
zyme alone but must, in fact, be related to the different ini-
tial compositions of the pyruvate solutions themselves (keto
© 1999 NRC Canada

1116
Can. J. Chem. Vol. 77, 1999
Table 4. Lactate dehydrogenase catalysis reduction of pyruvate at various fractions of keto, hydrated, and enol
pyruvate.a
% “Substrate” inhibition
at 5 × 10–4 M pyruvate
χ keto,initial
χ hydrate,initial
χ enol,initial
T (°C)
pH
41
5.0
41
6.8
6.8
0.45
0.49
0.973
0.864
0.543
0.137
0.027
0.136
0.457
0.863
3 × 10–5
1 × 10–5
4 × 10–5
2 × 10–5
15
30
80
5.0
-100
aRuns initiated by injection of preincubated pyruvic acid solutions at various temperatures and values of pH. Kinetic runs in pH
6.8 phosphate buffers µ = 0.1 M (see ref. 16).
Fig. 2. Percent (%) inhibition of beef heart LDH as a function of initial fractions of hydration of pyruvate. All kinetic runs were
carried out in phosphate buffer at 5.0°C.
vs. hydrated). On the other hand, the entries in Table 4 dem-
onstrate that there is no relationship between the fraction of
available enol pyruvate and the degree of substrate inhibi-
tion. Figure 2 shows that the initial fraction of hydrated
pyruvate during the enzymatic reaction is directly related to
the severity of substrate inhibition. Finally, we note that the
kinetic experiments illustrated in Table 4 were purposely
conducted at 5.0°C, where the half-life for the hydration–de-
hydration reequilibration (Table 2) is 100 s. Since the “ini-
tial” velocities used to generate the data in Table 4 were
taken 15 s following the kinetic zero, the reequilibration
between ketopyruvate and hydrated pyruvate was less than
1/10 complete when the progress of the reactions were first
recorded.
tures of 2,2-dihydroxypropanoate (hydrated pyruvate) and 2-
hydroxypropanoate (lactate), one would expect that the en-
zyme·NAD+ complex would, indeed, accommodate hydrated
pyruvate. Related experiments, involving the incubation of
LDH with pyruvate solutions containing varying fractions of
hydrated pyruvate (16) also have suggested that a more stable
inhibitory complex is formed, again with the same stoichio-
metry as the rapidly formed ternary complex. In light of the
kinetic and thermodynamic analysis provided in the present
paper, we again suggest that hydrated pyruvate is involved in
the formation of the more stable ternary complex as well.
References
Pyruvate inhibition of LDH as observed from initial veloc-
ity studies such as those represented in Table 4 has been at-
tributed to the reversible formation of a relatively loosely
bound abortive ternary complex (17) composed of LDH,
NAD+, and some form of pyruvate, which we suggest is hy-
drated pyruvate. Given the similarity of the shapes and struc-
1. (a) Y. Pocker, J.E. Meany, B.J. Nist, and C. Zadorojny. J.
Phys. Chem. 73, 2879 (1969); (b) H. Strehlow. Z.
Elektrochem. 66, 392 (1962); (c) M. Eigen, K. Kustin, and H.
Strehlow. Z. Phys. Chem. 31, 140 (1962); (d) V. Gold, G. Soc-
rates, and M.R. Crampton. J. Chem. Soc. 5888 (1964); (e) V.S.
Griffiths and G. Socrates. Trans. Faraday Soc. 673 (1967);
© 1999 NRC Canada

Esposito et al.
(f) M. Becker. Ber. Bunsen-Ges. Phys. Chem. 68, 669 (1964);
1117
Strategies of biochemical adaptation. W.B. Saunders Co., Phil-
adelphia. 1973. p. 58.
(g) Y. Chiang, A.J. Kresge, and P. Pruszynski. J. Am. Chem.
Soc. 114, 3103 (1992); (h) A.J. Kresge. Pure Appl. Chem. 63,
213 (1991).
15. (a) C.J. Coulson and B.R. Rabin. FEBS Lett. 3, 333 (1969);
(b) A.L. Lehniger. Biochemistry, 2nd ed. Worth Publishers,
New York. 1975. p. 484; (c) M.G. Rossman and N.O. Kaplan.
Structure function in lactate dehydrogenase. Proc. Natl. Acad.
Sci. (U.S.A.), 70, 1970 (1973).
16. M.D.Waddington and J.E. Meany. Arch. Biochem. Biophys.
211, 447 (1981).
17. J.W. Burgner II and W.J. Ray. Biochemistry, 17, 1654 (1978);
J.W. Burgner, G.R. Ainslie, W.W. Cleland, and W.J. Ray. Bio-
chemistry, 17, 1646 (1978).
18. H.J. Fromm. Biochim. Biophys. Acta, 52, 199 (1961).
19. W.E. Trommer, H. Huth, and H.R. Wenzel. Biochim. Biophys.
Acta, 547, 49 (1979).
20. (a) M. Davies and E. Gwynne. J. Am. Chem. Soc. 74, 2748
(1952); (b) G. Jones and B. Kaplan. J. Am. Chem. Soc. 50,
1845 (1928).
2. (a) Y. Pocker and J.E. Meany. J. Phys. Chem. 74, 1486 (1970);
(b) J. Damitio, G. Smith, J.E. Meany, and Y. Pocker. J. Am.
Chem. Soc. 114, 3081 (1992); (c) T. Germay, J.E. Meany, and
Y. Pocker. Joint 45th Northwest/Rocky Mountain Regional
Meeting, American Chemical Society. 1990. Abstr. No. 255.
3. (a) J.E. Meany. J. Phys. Chem. 73, 3421 (1969); (b) M. Hegazi
and J.E. Meany. J. Phys. Chem. 76, 3121 (1972).
4. (a) W.J. Albery and R.P. Bell. Trans. Faraday Soc. 61, 1194
(1965); (b) J.W. Burgner II and W.J. Ray. Biochemistry, 13,
4229 (1974).
5. K. Meyer. Justus Liebigs Ann. Chem. 380, 212 (1911).
6. (a) Y. Chiang and A.J. Kresge. Science (Washington, D.C.),
253, 395 (1991); (b) A.J. Kresge. Chem. Rev. 275 (1996);
(c) L. Wolff. Ann. Chem. 317, 8 (1901).
7. (a) D.J. Kuo, L. O’Connell, and I.A. Rose. J. Am. Chem. Soc.
101, 5025 (1979); (b) M. Fisher and P. Warneck. Ber. Bunsen-
Ges. Phys. Chem. 95, 523 (1991); (c) J. Peliska and H.
O’Leary. J. Am. Chem. Soc. 113, 1841 (1991).
8. (a) B.A. Miller and D.L. Leussing. J. Am. Chem. Soc. 107,
7146 (1985); (b) A. Schellenberger and G. Hubner. Chem. Ber.
98, 1938 (1965).
9. R. Tienharra and J.E. Meany. Biochemistry, 12, 2067 (1973).
10. B.J. Vennesland. J. Biol. Chem. 212, 787 (1955).
11. I.H. Segel. Enzyme kinetics. Wiley, New York. 1975; (b) M.
Dixon and E.C. Webb. Enzymes. 2nd ed. Academic Press,
New York. 1964.
12. (a) P.G.W. Plagemann, K.F. Gregory, and F. Wrobiewski.
Biochem. Z. 334, 37 (1961); (b) R. Stambaugh and P. Post. J.
Biol. Chem. 241, 1462 (1966); (c) A. Pesce, T.P. Fondy, F.E.
Stolzenbach, D. Castillo, and N.O. Kaplan. J. Biol. Chem. 242,
2151 (1967).
21. Y. Pocker, J.E. Meany, and R.C. Jones. J. Am. Chem. Soc.
104, 4885 (1982).
22. G.T. Spyridis and J.E. Meany. J. Chem. Educ. 65, 461 (1988).
23. (a) Y. Chiang, A.J. Kresge, and N.P. Schepp. J. Am. Chem.
Soc. 111, 3977 (1989); (b) J.E. Meany and J. Spaulding. Un-
published work.
24. (a) J.M. Wilson, Q. Tanko, M.M. Wenland, J.E. Meany, J.F.
Nedved, and Y. Pocker. Physiol. Chem. Phys. Med. NMR, 30,
149 (1998); (b) Y. Pocker and J.T. Stone. Biochemistry, 6, 668
(1967); (c) Biochemistry, 7, 2936 (1968); (d) Biochemistry, 7,
3021 (1968); (e) Y. Pocker, N. Janjic, and C.H. Miao. In Zinc
enzymes. Edited by I. Bertini, C. Luchinat, W. Maret, and M.
Zeppezauer. Birkhauser, Boston. 1986. p. 341. (f) Y. Pocker
and C.H. Miao. Biochemistry, 26, 8481 (1987).
25. G.W. Schwert and A.D. Winer. Enzymes, 7, 127 (1963).
26. L.Z. Holland, N.M. McFall, and G.N. Somero. Biochemistry,
36, 3207 (1997).
13. (a) A.L. Latner, S.S. Siddiqui, and A.W. Skillen. Science
(Washington, D.C.), 154, 527 (1966); (b) R.D. Cahn, N.O.
Kaplan, L. Levine, and E. Zwilling. Science (Washington,
D.C.), 136, 962 (1962); (c) D.M. Dawson, T.L. Goodfriend,
and N.O. Kaplan. Science (Washington, D.C.), 143, 929
(1964); (d) K.L. Manchester. Biochem. Educ. 22, 9 (1994).
14. (a) G.N. Somero. Comp. Biochem. Physiol. B: Biochem. Mol.
Biol. 44B, 205 (1973); (b) P.W. Hochachka and G.N. Somero.
27. (a) D.H. Robinson, J. Ju, and J.E.Meany. 40th Northwest Re-
gional Meeting, American Chemical Society. 1985. Abstr. No.
255; (b) J. Ju and J.E. Meany. 41st Northwest Regional Meet-
ing, American Chemical Society.1986. Abstr. No. 195.
(c) 42nd Northwest Regional Meeting, American Chemical
Society. 1987. Abstr. No. 101.
© 1999 NRC Canada