(Received for publication, September 19, 1996, and in revised form, November 6, 1996)
From the Department of Chemistry, University of
British Columbia, Vancouver, British Columbia V6T 1Z1, Canada
and ¶ Wake Forest University Medical Center,
Winston-Salem, North Carolina 27157
UDP-glucuronic acid is used by many pathogenic
bacteria in the construction of an antiphagocytic capsule that is
required for virulence. The enzyme UDP-glucose dehydrogenase catalyzes the NAD+-dependent 2-fold oxidation of
UDP-glucose and provides a source of the acid. In the present study the
recombinant dehydrogenase from group A streptococci has been purified
and found to be active as a monomer. The enzyme contains no
chromophoric cofactors, and its activity is unaffected by the presence
of EDTA or carbonyl-trapping reagents. Initial velocity and product
inhibition kinetic patterns are consistent with a bi-uni-uni-bi
ping-pong mechanism in which UDP-glucose is bound first and
UDP-glucuronate is released last. UDP-xylose was found to be a
competitive inhibitor (Ki, 2.7 µM) of
the enzyme. The enzyme is irreversibly inactivated by uridine
5-diphosphate-chloroacetol due to the alkylation of an active site
cysteine thiol. The apparent second order rate constant for the
inhibition (ki/Ki) was found to
be 2 × 103 mM
1
min
1. Incubation with the truncated compound,
chloroacetol phosphate, resulted in no detectable inactivation when
tested under comparable conditions. This supports the notion that
uridine 5
-diphosphate-chloroacetol is bound in the place of
UDP-glucose and is not simply acting as a nonspecific alkylating
agent.
The enzyme UDP-glucose dehydrogenase
(UDPGDH)1 catalyzes the
NAD+-dependent oxidation of UDP-glucose to
UDP-glucuronic acid (Fig. 1). It belongs to a small
group of dehydrogenases that are able to carry out the 2-fold oxidation
of an alcohol to an acid without the release of an aldehyde
intermediate (1). Much of the work to date has focused on the
properties and mechanism of the beef liver enzyme, and relatively
little is known about the enzyme purified from bacterial sources
(2-4). In many strains of bacteria that act as human pathogens, UDPGDH
provides the UDP-glucuronic acid required for the construction of an
antiphagocytic capsular polysaccharide. It is well established that the
formation of the capsule is required for virulence (5, 6), and it is
thought that the capsule enables the bacteria to evade the host's
immune system (7, 8). Group A and C streptococci are mammalian pathogens that use UDPGDH in the synthesis of a capsule composed of
hyaluronic acid (a polysaccharide consisting of alternating glucuronic
acid and N-acetylglucosamine residues) (9, 10). Many of the
known strains of Streptococcus pneumoniae also use UDP-glucuronic acid in the construction of their polysaccharide capsule
(11), and it has recently been shown that UDPGDH is required for
capsule production in S. pneumoniae type 3 (12). The
encapsulated Escherichia coli K5 is also known to use the enzyme for a similar purpose (2, 13). The hasB gene that encodes for UDPGDH in group A streptococci (Streptococcus
pyogenes) has been cloned and overexpressed in Escherichia
coli (14). Its gene product shares 57% sequence identity and 74%
sequence similarity with the S. pneumoniae enzyme (15, 16).
These properties make the HasB UDPGDH an attractive target for further
mechanistic studies and inhibitor design.
The mechanism used by the bovine liver dehydrogenase is thought to involve an initial oxidation of the C-6 hydroxyl of UDP-glucose to form an aldehyde intermediate and NADH (Fig. 1) (17-21). The aldehyde is tightly bound to the enzyme and is inaccessible to carbonyl-trapping reagents (17, 18). It has been suggested that the aldehyde is bound via an imine linkage with a lysine residue (19), however this is somewhat at odds with solvent oxygen isotope incorporation studies (20). The second oxidation is initiated by the addition of an active site cysteine thiol to the aldehyde to form a thiohemiacetal intermediate (21). A hydride transfer from the intermediate to NAD+ then takes place to produce an enzyme-bound thioester intermediate and a second molecule of NADH. In a final irreversible step the thioester is hydrolyzed to give the product UDP-glucuronic acid. The formation of a thioester intermediate in the second oxidation step is precedented by studies on glyceraldehyde-3-phosphate dehydrogenase and several other aldehyde dehydrogenases (22-25).
The existence of an active site cysteine residue has been well
documented in the case of UDP-glucose dehydrogenase from beef liver.
Thiol modification studies using iodoacetate (26), iodoacetamide (26),
5-[[(iodoacetamido)ethyl]amino]naphthalene-1-sulfonic acid (27),
iodoacetamidofluorescein (28), 5,5-dithiobis-(2-nitrobenzoate) (21,
29), and 6,6
-dithionicotinate (30) all indicate that a cysteine is
required for catalysis. The observation that the cysteine is protected
from modification by the addition of UDP-glucose or UDP-xylose suggests
that it is located in the active site (26, 29). Proteolytic digestion
and sequencing studies on the labeled enzyme have indicated that
Cys-275 supplies the reactive thiol (28, 31). The use of relatively
nonspecific thiol labeling reagents can, however, lead to misleading
results in certain cases (see Ref. 24 and references therein). The
enzyme histidinol dehydrogenase also catalyzes the
NAD+-dependent 2-fold oxidation of an alcohol
to an acid (without the release of aldehyde) and was generally thought
to use a mechanism similar to that of UDP-glucose dehydrogenase (1).
Treatment of the enzyme with 7-chloro-4-nitro-2,1,3-benzoxadiazole
resulted in the covalent labeling of an active site cysteine and the
loss of enzyme activity (32). This seemed to indicate that Cys-116 supplied a thiol that participated in the second oxidation step. Subsequent mutagenesis experiments indicated, however, that this enzyme
did not require a cysteine for catalysis and therefore does not follow
the glyceraldehyde-3-phosphate dehydrogenase paradigm (33, 34). A way
in which one can maximize the chances that a labeling reagent will
target a catalytically important group is to incorporate structural
elements of the natural substrate into its design. In the case of
UDP-glucose dehydrogenase the diphosphate nucleotide group provides a
convenient handle that can be used to orient a reactive moiety within
the active site. The reagent uridine 5
-diphosphate chloroacetol (UDC)
was first synthesized by Flentke and Frey (35) and positions an
electrophilic carbon atom three bond lengths away from the
-phosphate of a UDP group (Fig. 1, inset). It is
therefore a useful tool in probing the active site of UDP sugar
nucleotide-modifying enzymes.
In this article we report the purification and initial characterization
of the recombinant HasB UDPGDH. The streptococcal enzyme is notably
different from the mammalian enzyme in that it is a monomer (as opposed
to a hexamer). The initial velocity and product inhibition patterns are
shown to be consistent with a bi-uni-uni-bi ping-pong mechanism (36) in
which UDP-glucose is bound first and UDP-glucuronic acid is released
last. We have also demonstrated that UDC is an active site-directed,
irreversible inhibitor of the enzyme and covalently modifies a cysteine
thiol. The apparent second order rate constant for the inhibition
(ki/Ki) was found to be 2 × 103 mM1 min
1.
Incubation with the truncated compound, chloroacetol phosphate, resulted in no detectable inactivation when tested under comparable conditions and concentrations.
The hasB
gene product was overexpressed in E. coli JM109 (DE3,
pGAC147) as described elsewhere (14). The cells (1.9 g) were pelleted
3 h after induction with
isopropyl-1-thio--D-galactopyranoside. They were
resuspended in buffer A (5 ml; 50 mM Trien-HCl buffer, pH
8.7, containing glycerol (10%, v/v), dithiothreitol (2 mM), pepstatin (1 mg/liter), aprotinin (1 mg/liter), and
phenylmethanesulfonyl fluoride (1.5 mM)). Cells were lysed
by two passes through a French pressure cell at 20,000 p.s.i. Following
ultracentrifugation at 30,000 rpm (60,000 × g) for
1 h, the resulting supernatant was loaded onto a column (15 ml) of
diethylaminoethylcellulose that had been preequilibrated with buffer A. The column was washed with buffer A (50 ml) and then eluted with buffer
A containing 200 mM NaCl (50 ml). Active fractions were
pooled and concentrated (to 6.7 ml) using Centriprep-10 concentrators
and dialyzed against 1 liter of buffer B (50 mM Trien-HCl
buffer, pH 8.7, containing glycerol (10%, v/v) and dithiothreitol (2 mM)) for 48 h. The resulting solution was frozen in
liquid nitrogen and stored at
75 °C. The protein solution was
quickly thawed and applied (approximately 27 mg of total protein in
each injection) to a Waters AP-1 Protein-Pak Q column (10 × 100 mm) that had been equilibrated with buffer B (at 22 °C). The column
was washed with buffer B (10 ml) and eluted with a linear gradient of
NaCl (0-120 mM, over 35 min at 1.3 ml/min) in buffer B. The active fractions were pooled, frozen in liquid nitrogen, and stored
at
75 °C. Unless noted otherwise, all manipulations were performed
at 4 °C. Protein concentrations were determined by the method of
Bradford (37), using bovine serum albumin as a standard. A unit of
enzyme is defined as the amount of enzyme required to produce 2 µmol
of NADH/min at 30 °C.
The subunit molecular weight was determined on a Perkin-Elmer Sciex API300 electrospray mass spectrometer. The native molecular weight was estimated using gel filtration chromatography on a Protein-Pak 300SW column (Waters). The column was run in potassium phosphate buffer (100 mM, pH 7.0) at a flow rate of 1 ml/min. Blue dextran (2,000 kDa) was used to mark the void elution time, and bovine serum albumin (66 kDa), ovalbumin (43 kDa), and cytochrome c (12.4 kDa) were used as molecular mass standards.
UV Spectrum of the DehydrogenaseA sample of UDPGDH (0.4 ml of a 2.5 mg/ml solution) purified as described above was dialyzed twice for 24 h each time against potassium phosphate buffer (1 liter, 10 mM, pH 8.7, containing 2 mM dithiothreitol). To obtain a clean spectrum of the enzyme it was necessary to prevent the aerobic oxidation of dithiothreitol by carrying out the dialysis under an atmosphere of argon. The buffer was degassed and thoroughly sparged with argon prior to the addition of solid dithiothreitol. Following dialysis, particulate matter was removed from the sample by centrifugation at 10,000 rpm for 15 min. A UV spectrum of the enzyme (final concentration, 1.15 mg/ml or 25 µM) was recorded at 30 °C. A second spectrum was recorded following the addition of excess UDP-glucose (final concentration, 500 µM). A final spectrum was recorded following the addition of both UDP-glucose (500 µM) and 1 eq of NAD+ (25 µM).
Enzyme KineticsUDP-Glucose dehydrogenase activity was
assayed using a modification of a previously described procedure (4).
Rates were determined by following the reduction of NAD+ at
340 nm using a Varian Cary 3E UV-visible spectrophotometer. All assays
were performed at 30 °C in 50 mM Trien-HCl buffer, pH
8.7, containing 2 mM dithiothreitol (1 ml total volume).
Initial velocities were measured during the first 40 s after
initiation with UDPGDH (2.0 × 102 units), and
slopes were calculated using a least squares analysis with Cary 3 software version 3.0. NADH concentrations were determined using
A340 with
= 6,220 M
1. The compounds NAD+, NADH,
UDP-glucose, UDP-glucuronic acid, and UDP-xylose were all purchased
from Sigma.
UDC was prepared according to the
procedure of Flentke and Frey (35), and chloroacetol phosphate was
prepared as described by Hartman (38). The concentration of UDC was
determined using A260 with = 8,900 M
1 (35). Inactivation kinetics were measured
in the absence of dithiothreitol to prevent the reaction of the
inhibitor with added thiols. Inactivation incubations were therefore
performed under an atmosphere of argon, and aliquots were removed by
syringe. The buffer was degassed and sparged with argon prior to use.
In the absence of inhibitor the enzyme was shown to lose negligible amounts of activity under these conditions. The inactivations were
carried out at 30 °C in 50 mM Trien-HCl buffer, pH 8.7, containing either 5.0 mM UDP-glucuronic acid or 75 µM UDP-xylose and the inhibitor (0.5 ml total volume).
Timed aliquots (50 µl) were removed and diluted in the assay buffer
(900 µl containing 500 µM UDP-glucose) prior to
analysis. The initial velocities were obtained following initiation
with NAD+ (500 µM). First order rate
constants describing the loss of enzyme activity were obtained using
the program Grafit.
The sample used to determine the mass of the inhibited enzyme was prepared in the following manner. A sample of UDPGDH (1.2 mg/ml in the assay buffer) was incubated at 37 °C for 30 min in the presence of UDC (2.0 mM). Any remaining inhibitor was quenched by the addition of excess dithiothreitol (20 mM) prior to analysis.
To determine the nature of the tagged residue a sample of UDPGDH (2 mg/ml in 40 µl of the assay buffer) was inactivated by incubation at 30 °C for 30 min in the presence of UDC (250 µM). A second sample was incubated under identical conditions in the absence of UDC. Both samples were denatured by the addition of saturated urea (40 µl prepared in the assay buffer lacking dithiothreitol). Following 1 h of incubation at room temperature, iodoacetate was added to each sample (to a final concentration of 1 mM). The solutions were incubated in the dark for an additional 1 h and then quenched by the addition of excess dithiothreitol (to a final concentration of 5 mM) prior to mass spectral analysis.
The dehydrogenase was purified to homogeneity (as judged by SDS-polyacrylamide gel electrophoresis) by ion exchange chromatography. The purification gave 35 mg of protein/500 ml of culture, which accounts for 73% of the activity in the crude extract (Table I). A specific activity of 1.5 units/mg was measured. The mass of the protein was 45,489 ± 4 Da (versus 45,487 Da calculated) as determined by electrospray mass spectrometry.
|
The enzyme was found to fairly stable when stored in the presence of 2 mM dithiothreitol. Incubation for 1.6 h at 30 °C (50 mM Trien-HCl, pH 8.7) resulted in the loss of 40% of the activity. Similar incubations in the absence of any reduced thiol resulted in the loss of 75% of the activity. When stored at 5 °C (in the presence of dithiothreitol) the enzyme was found to be quite stable (greater than 90% of the activity remained after 24 h). The enzyme was also stable to repeated rapid freeze-thaw cycles when kept in a buffer containing 10% glycerol and dithiothreitol.
Gel filtration chromatography indicated that the enzyme has a mass of
approximately 52 kDa when compared with known standards (Fig.
2). This indicates that the enzyme is active as a
monomer.
Effect of Additives on the Enzymic Reaction
As mentioned previously, the activity of the dehydrogenase was dependent on the presence of reduced thiols such as dithiothreitol. The activity was essentially unaffected by the addition of 5 mM EDTA, indicating that exchangeable divalent cations were not required.
To investigate whether the putative aldehyde intermediate is released or is accessible to solvent, the stoichiometry of the reaction was determined in the presence of the aldehyde-trapping reagents semicarbazide and hydroxylamine. The ratio of NADH produced/UDP-glucose added was found to be constant (about 1.9:1) over a range of 0-50 mM of each of the additives.
UV Spectrum of the Purified DehydrogenaseThe UV spectrum of
the dehydrogenase (Fig. 3, trace 1) did not
show any significant absorptions past 300 nm that might be indicative
of a bound chromophore such as NADH. The addition of excess UDP-glucose
to the cuvette did not cause any changes in this area of the spectrum
(Fig. 3, trace 2). This indicates that bound
NAD+ is not present, since the equilibrium lies strongly in
favor of NADH and UDP-glucuronic acid (17). When one equivalent of NAD+ was added to the same cuvette, absorbance due to the
NADH chromophore was clearly observed (340 nm maximum; Fig. 3,
trace 3). These studies suggest that only substoichiometric
amounts, if any, of NAD+ or NADH are present in the enzyme
sample.
Initial Velocity Kinetic Studies
The initial velocity kinetic
patterns with UDP-glucose as the variable substrate are shown in Fig.
4, and those with NAD+ as the variable
substrate are shown in Fig. 5. In both cases linear
intersecting patterns are observed, indicating that the binding of
UDP-glucose and the first NAD+ are reversibly connected
(39). It also suggests that the binding of the first NAD+
and the second NAD+ are not reversibly connected (a
reversible connection should result in nonlinear initial velocity
patterns) (39). This indicates that a ping-pong mechanism is operative
in which the release of the first NADH occurs before the binding of the
second NAD+. A replot of the intercepts against the
reciprocal of the concentration of changing fixed substrate (Figs. 4
and 5, right graphs) was extrapolated to the x
axis to reveal the Km values for each substrate. For
UDP-glucose Km = 20 µM, and for
NAD+ Km = 60 µM. The
y-intercept of these replots also provided the
kcat value of 1.7 s1 (based on
UDP-glucuronic acid production).
Product Inhibition Studies
Product inhibition studies were performed by including changing fixed concentrations of either UDP-glucuronic acid or NADH in the assay buffer and varying the concentration of either UDP-glucose or NAD+ in the presence of saturating concentrations of the other substrate (200 µM UDP-glucose or 500 µM NAD+).
When UDP-glucuronic acid was the inhibitor and UDP-glucose was the
variable substrate, a double reciprocal plot of the initial velocity
data revealed a pattern consistent with competitive inhibition (Fig.
6). This indicates that the two species are interacting with the same enzyme form and is consistent with a kinetic mechanism in
which UDP-glucose binds first and UDP-glucuronic acid is released last
(39). A replot of the slopes against the UDP-glucuronic acid
concentration (Fig. 6, right graph) gave a
Ki of 200 µM for UDP-glucuronic acid.
The apparent small differences in Vmax seen in
Fig. 6 could potentially be caused by weak substrate inhibition. When
UDP-glucuronic acid was the inhibitor and NAD+ was the
variable substrate (with saturating levels of UDP-glucose), no
inhibition was observed. This is entirely consistent with the observation of competitive inhibition between UDP-glucuronic acid and
UDP-glucose. Since the two UDP sugars bind to the same form of the
enzyme, saturation with UDP-glucose will prevent the binding of
UDP-glucuronic acid.
When NADH was the inhibitor and UDP-glucose was the variable substrate
(in the presence of saturating NAD+), a parallel pattern of
lines resulted (Fig. 7). This apparent uncompetitive
inhibition is seen if the two species are interacting with different
enzyme forms that are not reversibly connected (39). It is consistent
with a mechanism in which one or both of the NAD+ molecules
bind to the enzyme after UDP-glucose binds and before the first NADH is
released. When NADH was the inhibitor and NAD+ was the
variable substrate, an apparent noncompetitive pattern was observed
(Fig. 8). This pattern indicates that at least one NADH
molecule and one NAD+ molecule interact with different
enzyme forms, and the points of interaction are not separated by any
effectively irreversible steps, such as the binding of UDP-glucose
(when present at saturating levels) or the release of UDP-glucuronic
acid (when absent from the solution).
Inhibition by UDP-Xylose
Inhibition by UDP-xylose was
investigated by including changing fixed concentrations of UDP-xylose
in the assay buffer and varying the concentration of UDP-glucose in the
presence of saturating concentrations of NAD+ (500 µM). A double reciprocal plot of the data showed a
pattern consistent with competitive inhibition (Fig. 9).
A replot of the slopes against the UDP-glucuronic acid concentration
(Fig. 9, right graph) gave a Ki of 2.7 µM for UDP-xylose.
Irreversible Inhibition by UDC
Incubation of UDPGDH with 0.20 mM solution of UDC resulted in the rapid inactivation of the enzyme. Dialysis of the inactivated enzyme against inhibitor-free buffer for 48 h did not result in the restoration of any measurable activity, indicating that the process is irreversible. An electrospray mass spectrum of the inactivated enzyme showed that the mass of the protein increased from 45,493 ± 4 to 45,958 ± 4 Da on inactivation. This is consistent with the covalent attachment of one molecule of inhibitor (molecular weight, 495 g/mol) to the enzyme with an accompanying loss of a chlorine atom. The presence of known competitive inhibitors decreased the rate of inactivation, suggesting that the inhibitor is active site-directed (see the following section).
An attempt was made to measure the kinetics of inactivation from the rate of decrease of residual activity as a function of incubation time at various inhibitor concentrations. This proved to be problematic, since the inhibitor was effective at very low concentrations, and the enzymatic assay proved to be insufficiently sensitive to determine accurate rates with small amounts of enzyme. Instead the inactivation rates were measured in the presence of a competitive inhibitor that permitted the use of higher concentrations of both inactivator and enzyme. As long as the inhibition is caused by an initial reversible binding event followed by irreversible inactivation, and the initial binding is competitive with that of a known competitive inhibitor, then the rate constant for inactivation (kobs) will be lowered according to Equation 1:
![]() |
(Eq. 1) |
To demonstrate that UDC was acting as a specific irreversible inhibitor
of the enzyme, as opposed to a relatively nonspecific alkylating agent,
the inactivation by chloroacetol phosphate (38) was briefly
investigated. Chloroacetol phosphate contains the chemically reactive
-haloketone moiety required for covalent modification to occur yet
lacks the terminal UMP functionality that is likely important in active
site binding. When the dehydrogenase was incubated in the presence of
17 µM UDC and 5 mM UDP-glucuronic acid, 50%
of the activity was lost within 1 minute (see Fig. 10). Analogous
incubations with 17 µM chloroacetol phosphate, however, showed no detectable inactivator-dependent loss of activity
over a 10-min period (2-3% of the activity is lost even in the
absence of inactivator). When the concentration of chloroacetol
phosphate was increased to 2 mM (with 5 mM
UDP-glucuronic acid) only 25% of the activity was lost over a 10-min
incubation. Incubation with 2 mM chloroacetol phosphate in
the absence of any competitive inhibitor resulted in inactivation with
a half-life of approximately 1 min. These results show that the
terminal UMP functionality is required for the potency of UDC and is
consistent with a specific binding event occurring prior to alkylation.
The nature of the tagged residue was investigated in the following manner. A sample of the dehydrogenase was denatured in 5 M urea and then treated with excess iodoacetate. An electrospray mass spectrum showed that the mass of the protein increased from 45,493 ± 4 to 45,606 ± 4 Da, consistent with the attachment of two acetate units (expected difference, 116 daltons). A similar analysis using an enzyme previously inactivated by UDC showed that the mass of the protein-inhibitor adduct increased from 45,958 ± 4 to 46,009 ± 4 Da, consistent with the attachment of only 1 acetate unit (expected difference, 58 daltons). The enzyme contains only two cysteine residues, and in a denatured state both will be readily alkylated by iodoacetate (as indicated by the mass spectral data). The observation that prior inactivation with UDC results in the removal of one of the sites of acetate attachment strongly indicates that the inhibitor acts by alkylating an active site cysteine residue.
A well documented characteristic of the hexameric bovine liver UDP-glucose dehydrogenase is that it displays half-of-the-sites reactivity (26, 30, 40, 41). The enzyme binds only three molecules of UDP-glucose at saturation, and only three of its six active site cysteine thiols are readily accessible to alkylation by reagents such as iodoacetic acid. Studies on partially denatured samples of the beef liver enzyme led researchers to speculate that two active sites, one on each subunit, were involved in the overall reaction (40). They suggested that the first oxidation produces an aldehyde intermediate, which is covalently bound via an imine linkage with a lysine residue. The intermediate is then transferred to a cysteine thiol in the active site of an adjacent subunit where the second oxidation takes place. Resonance energy transfer experiments were found to be at odds with this notion, since they indicated that the six active sites of the hexamer were spatially remote (27). The observation made in this article that the streptococcal enzyme is monomeric is significant in that it rules out any mechanisms invoking the participation of more than one subunit. The sequence similarities (31% positional identity) between the streptococcal enzyme and the bovine liver enzyme suggest that they share a common ancestry and likely use a common mechanism (31).
The only other bacterial UDP-glucose dehydrogenase that has been purified is that from E. coli (2-4). Many of the properties of the streptococcal enzyme are similar to that of the E. coli enzyme, with the notable exceptions that the E. coli enzyme is reported to be active as a dimer and that the UDP-glucuronic acid inhibition patterns display positive cooperativity. The observation that the dehydrogenase activity is dependent on the presence of reduced thiols and is unaffected by treatment with EDTA or carbonyl trapping reagents is consistent with studies on the enzyme obtained from other sources (2, 4, 12, 17). The UV spectrum of the enzyme showed no significant absorptions past 300 nm that would indicate the presence of a chromophoric cofactor. Spectra taken in the presence of added UDP-glucose and NAD+ indicate that the purified enzyme does not contain stoichiometric amounts of bound NAD+ or NADH.
The kinetic mechanism of the bovine liver UDP-glucose dehydrogenase and
the similar enzyme histidinol dehydrogenase (which also catalyzes the
2-fold oxidation of an alcohol to an acid without release of an
aldehyde intermediate) have been investigated previously (42-45). In
both cases the results indicated that a bi-uni-uni-bi ping-pong
mechanism was followed in which the alcohol was bound first and the
acid was released last (Fig. 11). The results obtained here for the streptococcal enzyme were also consistent with this mechanism. It was, however, impossible to rule out an ordered Ter-Ter mechanism (36) in which UDP-glucose binds first and UDP-glucuronic acid is released last. The linearity of the initial velocity plots (Figs. 4 and 5) suggests that the NAD+
binding events are separated by an effectively irreversible step (dissociation of NADH at zero concentration) and supports the former
mechanism. Conversely, the linearity of the NADH inhibition data (with
NAD+ as the variable substrate; Fig. 8) does not suggest
that a reversible link is established between the binding of the two
NAD+ molecules at significant NADH concentrations
(indirectly supporting the latter mechanism). To discern between the
two mechanisms, parabolic effects would have to be observed in one of
these two plots. These effects can be hard to see, as the curvature may be slight (42, 43). The ordered Ter-Ter mechanism requires that the enzyme binds both NAD+ molecules before a product
is released. It seems unlikely that this mechanism is operative, since
the enzyme is monomeric, and inspection of the protein sequence reveals
only one obvious NAD+ binding site (14).
The inhibitor UDC was first used in studies on UDP-galactose 4-epimerase in an effort to identify an active site general base (35). The reagent did not covalently label the protein in the conventional sense but instead was shown to act as a suicide substrate. The inactivation was caused by an initial enolization of the chloroacetol moiety followed by the alkylation of the nicotinamide ring of a bound NAD+ cofactor. The resulting chromophoric adduct remained tightly bound by the enzyme, resulting in an essentially irreversible inactivation. The inhibition of UDP-glucose dehydrogenase by UDC is clearly taking place in a different manner. The inactivation proceeds in the absence of bound NAD+, and the electrospray mass spectrum indicates that the mass of the protein has increased by the mass of the inhibitor less one chlorine atom. This is consistent with a mechanism in which the alkylation of an enzyme nucleophile is taking place. It should be noted that a related compound, p-bromoacetamidophenyl uridyl pyrophosphate, has been reported to be a potent inhibitor of UDP-glucose dehydrogenase; however, few details concerning the nature of the inhibition have been published (46, 47).
The observation that the inactivation by UDC is slowed by the addition of known competitive inhibitors indicates that the process is active site-directed. Furthermore, the agreement in the values of ki/Ki obtained in the presence of two different competitive inhibitors indicate that the initial binding of UDC is competitive with that of UDP-glucose. The assumption that the UDP group of UDC occupies the normal UDP binding site is supported by the observation that the truncated compound, chloroacetol phosphate, is a markedly less potent inhibitor. The electrophilic carbon of bound UDC should therefore rest in a position comparable with that of the aldehyde during the course of the normal reaction pathway. The iodoacetate-labeling experiments clearly indicate that one thiol has been modified during the inactivation process and support the notion that a cysteine is in the proper vicinity to participate in covalent catalysis. When the protein sequence of the streptococcal enzyme is compared with that of the other known UDPGDHs, Cys-260 is strictly conserved, whereas Cys-162 clearly is not (14-16, 31, 48). This observation, combined with the previous identification (28, 31) of the iodoacetate-labeled tryptic fragment from the bovine liver enzyme (containing the corresponding conserved residue Cys-275), strongly indicates that Cys-260 is the active site thiol. Further studies involving site-directed mutagenesis and proteolytic digestion techniques will be undertaken to investigate this further.
We thank Shouming He for collecting the electrospray mass spectral data.