Detection of transient intermediates in the metal-dependent nonoxidative decarboxylation catalyzed by α-amino-β-carboxymuconate-ε- semialdehyde decarboxylase

Article abstract of DOI:10.1021/ja073648l

The first description of a metalloenzyme-mediated nonoxidative decarboxylation reaction was that catalyzed by α-amino-β-carboxymuconate-ε-semialdehyde decarboxylase (ACMSD) (Li, T.; Walker, A. L.; Iwaki, H.; Hasegawa, Y.; Liu, A. J. Am. Chem. Soc. 2005, 127, 12282-12290 and Liu, A.; Zhang, H. Biochemistry 2006, 45, 10407-10411). Here, we report a rapid kinetic study of the ACMSD reaction using the cobalt(II)-reconstituted enzyme. Two previously undetected intermediates of the reaction are characterized. These are proposed to be the aminomuconate-ε-semialdehyde (AMS) product of the decarboxylation in complex with the enzyme and the free AMS after release from ACMSD. It was possible to determine the rates of formation of the enzyme-substrate complex (2.4 × 10⁶ s^-1 M^-1), release of the AMS product (8.8 s^-1), and spontaneous cyclization of AMS to picolinic acid (0.05 s^-1), which is the final product of the reaction. Reconstructed spectra of these reaction intermediates and a kinetic mechanism for the overall reaction are presented. Copyright

Full text of DOI:10.1021/ja073648l

Published on Web 07/11/2007
Detection of Transient Intermediates in the Metal-Dependent Nonoxidative
Decarboxylation Catalyzed by r-Amino-â-Carboxymuconate-E-Semialdehyde
Decarboxylase
Tingfeng Li, John K. Ma, Jonathan P. Hosler, Victor L. Davidson,* and Aimin Liu*
Department of Biochemistry, The UniVersity of Mississippi Medical Center, Jackson, Mississippi 39216
Received May 21, 2007; E-mail: vdavidson@biochem.umsmed.edu; aliu@biochem.umsmed.edu
Scheme 1
The first description of a metalloenzyme-mediated nonoxidative
decarboxylation reaction was that catalyzed by R-amino-â-car-
boxymuconate-ꢀ-semialdehyde decarboxylase (ACMSD) (Scheme
1).1,2 ACMSD is an important enzyme in tryptophan metabolism
and nitroaromatic biodegradation.2-4 In the absence of ACMSD,
the substrate R-amino-â-carboxymuconate-ꢀ-semialdehyde (ACMS)
is nonenzymatically converted to quinolinic acid, a neurotoxin.5-7
Analysis of the crystal structure of ACMSD, along with EPR
spectroscopy of the enzyme, has revealed that the active site
contains a mononuclear metallocofactor.2,8,9 ACMSD retains activity
with several different divalent metal ions at this site, including zinc,
cobalt, iron, and manganese.2,8,9 Recent studies have shown that
several other decarboxylases, including the structurally characterized
decreased extinction coefficient. Complete formation of intermediate
II, with the loss of substrate and intermediate I, occurs within 1 s
(Figure 2B). The primary data on which Figure 2 is based are shown
in Figure S1. Conversion of intermediate II to the final colorless
product, picolinic acid, requires several additional seconds (not
shown). Since intermediate II forms quantitatively, it was possible
to calculate its ꢀ383 ) 40 000 cm-1 M-1 by comparison with that
enzyme γ-resorcylate decarboxylase, utilize a d-block metal ion to
catalyze a variety of nonoxidative decarboxylation reactions.1,10
These enzymes catalyze similar chemical reactions, share similar
overall and active site structures, and belong to the same subset of
enzymes in the amidohydrolase superfamily.11,12 Within this large
enzyme superfamily, most members catalyze hydrolytic reactions.
The newly defined ACMSD protein subgroup is unique in that its
members catalyze nonhydrolytic C-C bond cleavage.1,9 Thus far,
the catalytic mechanism of this novel type of metallocofactor-
dependent nonoxidative decarboxylation has remained largely
uncharacterized.
In this work, we conducted a pre-steady-state kinetic study of
the reaction catalyzed by ACMSD isolated from Pseudomonas
fluorescens. These studies take advantage of the fact that the
substrate, ACMS, is a chromophore (ꢀ360 nm ) 47 500 M-1 cm-1),
while the final product, picolinic acid, is transparent at 300 nm or
above.13 ACMSD may be reconstituted after purification with
different metal ions. Of these, Co(II)-reconstituted ACMSD exhibits
the highest activity and the greatest stability, which facilitated
kinetic analysis of this enzyme.2 Transient kinetic experiments were
conducted in 25 mM HEPES buffer, pH 7.0, 5% glycerol at 25 °C
using an OLIS rapid scanning stopped-flow spectrophotometer.
Pseudo-first-order conditions were maintained using 1 µM ACMS
as the limiting reagent, while the concentration of Co(II)-
reconstituted ACMSD was varied from 7.5 to 45 µM. Protein
concentration was initially estimated using the Coomassie Plus
(Pierce) assay, and then the precise amount of metal-reconstituted
ACMSD was determined by inductively coupled plasma optical
emission spectroscopy (ICP-OES).
of ACMS (ꢀ360 ) 47 500 cm-1 M-1). It was not possible to
accurately determine this value for intermediate I because it does
not form quantitatively, as it is dependent upon the relative rates
of formation and decay and thus will vary with [ACMSD] used.
The chemical nature of intermediate I is proposed to be
R-aminomuconate-ꢀ-semialdehyde (AMS), a previously proposed
but undetected decarboxylation intermediate of the reaction14 in
complex with ACMSD (Scheme 2). The observed rate for the
conversion of ACMS to intermediate I exhibited a linear depen-
dence on the concentration of ACMSD (Figure 1). The observation
of a linear rather than hyperbolic concentration dependence indicates
that the formation of the enzyme-substrate complex is slow relative
to the catalytic step. The value of k1 in Scheme 2 determined from
the slope of the line is 2.4 × 106 s-1 M-1. The second kinetic phase
is concentration independent and therefore a unimolecular reaction.
This reaction, which exhibits a rate of 8.8 ( 1.0 s-1, is proposed
to be the release of AMS from the enzyme-product complex (k3
in Scheme 2). This is assumed to be a virtually irreversible reaction.
This would be consistent with the similar spectral properties of
intermediates I and II, with I being AMS bound to the enzyme
and II being free AMS. The rate for the disappearance of
intermediate II (k4) is 0.05 ( 0.02 s-1 and is also independent of
The ACMSD reaction was clearly multiphasic, exhibiting a rapid
rate which was dependent on ACMSD concentration and two slower
rates which were independent of ACMSD concentration (Figure
1). A species (intermediate I) with an absorption maximum red-
shifted relative to that of the substrate is rapidly formed concomitant
with the loss of ACMS absorbance centered at 360 nm (Figure 2).
This species is then converted in a slower reaction to a species
(intermediate II) with a similar absorption maximum but a
Figure 1. Concentration dependence of the observed rates for the formation
of intermediate I (red, 9), conversion of intermediate I to intermediate II
(green, O), and disappearance of intermediate II (blue, 0).
9
9278
J. AM. CHEM. SOC. 2007, 129, 9278-9279
10.1021/ja073648l CCC: $37.00 © 2007 American Chemical Society

C O M M U N I C A T I O N S
Scheme 2
ACMSD concentration, consistent with it describing the spontane-
ous cyclization of released AMS to picolinic acid. Since a hyper-
bolic dependence of the reaction rate with enzyme concentration
was not observed, it was not possible to determine k2 (the rate
constant for the catalytic conversion of ACMS to AMS). It can be
concluded that the rate of the catalytic conversion of ACMS to
AMS is rapid relative to formation of the enzyme-substrate
complex (i.e., k2 > k1[ACMSD] + k-1) and release of AMS from
the enzyme-product complex (i.e., k2 > k3).
spectra identical to those shown in Figure 2 was observed (data
not shown). This suggests that the nature of the metal at the active
site does not affect the chemical reaction mechanism of ACMSD.
Unfortunately, instability of Mn(II)-ACMSD, along with incom-
plete reconstitution of the metal, prevented accurate determination
of the concentration of active enzyme so that a kinetic analysis
like that described here for Co(II)-ACSMD was not possible.
The reaction catalyzed by ACMSD is biologically significant
because this enzyme of the tryptophan degradation pathway converts
a catabolic intermediate to a benign metabolite, thus preventing
the accumulation of a neurotoxic compound. The conversion of
ACMS substrate to picolinic acid requires both a decarboxylation
and a cyclization reaction. These results show that the enzyme-
catalyzed decarboxylation occurs first to generate an unstable AMS
intermediate, which then undergoes a relatively slow release from
the enzyme and a much slower cyclization to yield the final product.
Previous steady-state kinetic studies indicated that the identity of
the bound metal of ACMSD affected enzyme stability and Km but
had relatively little effect on kcat.2 The latter effect may be explained
by the observation that the rate-limiting step is not the metal-
dependent decarboxylation but the subsequent slower reaction. This
information provides new insight into our understanding of the
chemical and kinetic reaction mechanisms of this recently charac-
terized metal-ion-mediated nonoxidative decarboxylation.
The rate of product release (k3 ) 8.8 s-1) is approximately the
same as kcat for the steady-state reaction of Co(II)-ACMSD with
ACMS, which was measured under the same reaction conditions,
as 7.3 s-1.2 It should be noted that in those steady-state studies the
reaction was monitored by the initial rate of the loss of absorbance
at 360 nm. As can be seen in Figure 2, the reaction monitored at
this wavelength describes the steady-state conversion of ACMS to
AMS, not to the colorless picolinic acid. This explains why the
observed steady-state kcat was similar to k3 rather than the slower
k4.
Similar experiments were performed using Mn(II)-reconstituted
ACMSD, and the accumulation of intermediates with absorption
Acknowledgment. This was supported by NIH Grants GM56824
(J.P.H.), GM041574 (V.L.D.), and GM069618 (subaward to A.L.,
T. P. Begley, PI).
Supporting Information Available: Figure S1 displays the original
time courses of spectral scans from which Figure 2 is derived. This
material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Liu, A.; Zhang, H. Biochemistry 2006, 45, 10407-10411.
(2) Li, T.; Walker, A. L.; Iwaki, H.; Hasegawa, Y.; Liu, A. J. Am. Chem.
Soc. 2005, 127, 12282-12290.
(3) Hasegawa, Y.; Muraki, T.; Tokuyama, T.; Iwaki, H.; Tatsuno, M.; Lau,
P. C. FEMS Microbiol. Lett. 2000, 190, 185-190.
(4) Fukuoka, S.; Ishiguro, K.; Yanagihara, K.; Tanabe, A.; Egashira, Y.;
Sanada, H.; Shibata, K. J. Biol. Chem. 2002, 277, 35162-35167.
(5) Schwarcz, R. Curr. Opin. Pharmacol. 2004, 4, 12-17.
(6) Reinhard, J. F., Jr. Ann. N.Y. Acad. Sci. 2004, 1035, 335-349.
(7) Stone, T. W.; Darlington, L. G. Nat. ReV. Drug DiscoVery 2002, 1, 609-
620.
(8) Martynowski, D.; Eyobo, Y.; Li, T.; Yang, K.; Liu, A.; Zhang, H.
Biochemistry 2006, 45, 10412-10421.
(9) Li, T.; Iwaki, H.; Fu, R.; Hasegawa, Y.; Zhang, H.; Liu, A. Biochemistry
2006, 45, 6628-6634.
(10) Goto, M.; Hayashi, H.; Miyahara, I.; Hirotsu, K.; Yoshida, M.; Oikawa,
T. J. Biol. Chem. 2006, 281, 34365-34373.
(11) Holm, L.; Sander, C. Proteins 1997, 28, 72-82.
(12) Seibert, C. M.; Raushel, F. M. Biochemistry 2005, 44, 6383-6391.
(13) Koontz, W. A.; Shiman, R. J. Biol. Chem. 1976, 251, 368-377.
(14) Ichiyama, A.; Nakamura, S.; Kawai, H.; Honjo, T.; Nishizuka, Y.;
Hayaishi, O.; Senoh, S. J. Biol. Chem. 1965, 240, 740-749.
(15) Colabroy, K. L.; Begley, T. P. J. Am. Chem. Soc. 2005, 127, 840-841.
Figure 2. Results of the reaction of 1 µM ACMS with 12 µM ACMSD.
Reconstructed spectra (A) and changes in concentration with time (B) of
the kinetically distinguishable species fit to an A f B f C model with the
OLIS GlobalWorks software. The plot of the residuals for the fit of data is
show as an inset at the top.
JA073648L
9
J. AM. CHEM. SOC. VOL. 129, NO. 30, 2007 9279