Mechanism-Based Trapping of the Quinonoid Intermediate by Using the K276R Mutant of PLP-Dependent 3-Aminobenzoate Synthase PctV in the Biosynthesis of Pactamycin

Article abstract of DOI:10.1002/cbic.201500426

Mutational analysis of the pyridoxal 5′-phosphate (PLP)-dependent enzyme PctV was carried out to elucidate the multi-step reaction mechanism for the formation of 3-aminobenzoate (3-ABA) from 3-dehydroshikimate (3-DSA). Introduction of mutation K276R led to the accumulation of a quinonoid intermediate with an absorption maximum at 580 nm after the reaction of pyridoxamine 5′-phosphate (PMP) with 3-DSA. The chemical structure of this intermediate was supported by X-ray crystallographic analysis of the complex formed between the K276R mutant and the quinonoid intermediate. These results clearly show that a quinonoid intermediate is involved in the formation of 3-ABA. They also indicate that Lys276 (in the active site of PctV) plays multiple roles, including acid/base catalysis during the dehydration reaction of the quinonoid intermediate.

Full text of DOI:10.1002/cbic.201500426

Accepted Article
Title: Mechanism-based Trapping of the Quinonoid Intermediate Using the K276R
Mutant of PLP-Dependent 3-Aminobenzoate Synthase PctV in the Biosynthe-
sis of Pactamycin
Authors: Akane Hirayama; Akimasa Miyanaga; Fumitaka Kudo; Tadashi Eguchi
This manuscript has been accepted after peer review and the authors have elected
to post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Digital
Object Identifier (DOI) given below. The VoR will be published online in Early View as
soon as possible and may be different to this Accepted Article as a result of editing.
Readers should obtain the VoR from the journal website shown below when it is pu-
blished to ensure accuracy of information. The authors are responsible for the content
of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201500426
Link to VoR: http://dx.doi.org/10.1002/cbic.201500426
A Journal of
www.chembiochem.org

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
Mechanism-based Trapping of the Quinonoid Intermediate Using
the K276R Mutant of PLP-Dependent 3-Aminobenzoate Synthase
PctV in the Biosynthesis of Pactamycin
Akane Hirayama,[a] Akimasa Miyanaga,[b] Fumitaka Kudo*[b] and Tadashi Eguchi*[a]
Abstract: Mutational analysis of the pyridoxal 5′-phosphate (PLP)-
dependent enzyme PctV has been carried out to elucidate the multi-
step reaction mechanism for the formation of 3-aminobenzoate (3-
ABA) from 3-dehydroshikimate (3-DSA). The results revealed that
the introduction of a K276R mutation led to the accumulation of a
quinonoid intermediate with an absorption maximum of 580 nm
following the reaction of pyridoxamine 5′-phosphate (PMP) with 3-
DSA. Furthermore, the chemical structure of this intermediate was
supported by X-ray crystallographic analysis of the complex formed
between the K276R mutant and the quinonoid intermediate. These
results clearly show that a quinonoid intermediate is involved in the
formation of 3-ABA, and also indicate that the Lys276 residue in the
active site of the PctV enzyme plays multiple roles including
acid/base catalysis during the dehydration reaction of the quinonoid
intermediate, as well as assisting in the PMP formation and
aminotransfer reaction.
followed by a ketimine intermediate, which can be hydrolyzed to
give pyridoxamine 5′-phosphate (PMP) and -ketoacid. PctV
uses L-glutamate to allow for the formation of PMP.[1a] The
amino group of the PMP generated in this way can then behave
as a nucleophile and attack the carbonyl carbon atom of 3-DSA
to form a ketimine intermediate. In a typical aminotransfer
reaction, this ketimine intermediate is converted to an aldimine
intermediate, which undergoes nucleophilic attack by the active
site lysine residue to give the initial enzyme-PLP aldimine. The
aminated product is then released from the enzyme. In the case
of the PctV reaction, additional dual dehydration steps after the
aminotransfer reaction would be required prior to the release of
the 3-ABA product (Scheme 1). Thus, specific catalytic reaction
mechanisms appear to be required for the dehydration stages of
the PctV-catalyzed reaction. Given that a better understanding
of the underlying mechanism of this dehydration process could
be used to expand the repertoire of PLP-dependent enzymatic
chemistry,[4] we recently became interested in the details of the
PctV reaction mechanism.
Introduction
O
NH2
H3C
N
NH
PctV is a pyridoxal 5′-phosphate (PLP)-dependent enzyme,
which catalyzes the formation of 3-aminobenzoic acid (1, 3-ABA)
from 3-dehydroshikimic acid (2, 3-DSA) during the biosynthesis
of antitumor antibiotic pactamycin (Scheme 1).[1] Given that
isotope-labeled 3-ABA has been shown to be efficiently
incorporated into pactamycin in feeding experiments, it seems
plausible that the PctV reaction initiates this particular secondary
metabolic pathway.[1a,2] PctV shows a moderate level of similarity
to glutamate-1-semialdehyde-2,1-aminomutase (GSAM, Figure
S1), which is involved in the formation of 5-aminolevulinic acid
from glutamate-1-semialdehyde during heme biosynthesis.[3]
Thus, PctV belongs to the aminotransferase family of enzymes,
which possess a critical lysine residue in their active site that
forms an internal aldimine complex with PLP (Scheme S1). The
enzyme-PLP aldimine complex formed in the active site of PctV
can react with a free amino acid to form an external aldimine
CH3
O
O
PctV
PLP
O
CH3HO
NH
O
O
H3C
HO
H2N
OH
CH3
OH
OH
CH3
HO
2
1
OH
O
L-Glu
a -KG 2 H2O
OH
pactamycin
Scheme 1. Formation of 3-aminobenzoate from 3-dehydroshikimate by PctV
during the biosynthesis of pactamycin.
Results
Model Structure and Mutational Analysis of PctV. A Swiss
model[5] structure of PctV based on GSAM (36% identity to PctV)
with the gabaculine-PLP complex[6] clearly showed that Lys276
is the key lysine residue in the active site (Figure S2). Several
other residues, including Tyr64, Lys159, Asn220, and Glu416
were located in close proximity to the gabaculine-PLP complex
in the original structure, which indicated that some of these
residues could be involved in the recognition of 3-DSA, as well
as assisting in the dehydration reactions. However, kinetic
analysis of the Y64F, K159Q, N220A, and E416A mutants
revealed that their kcat values (0.18–0.28 min−1) were similar to
that of the wild type enzyme (0.39 min−1) (Table 1). This result
therefore indicated that these amino acid residues are not
involved in the catalytic reactions, although Lys159 and Asn220
appear to be involved in the recognition of the substrate
because the KM values for the K159Q and N220A mutants were
[a]
[b]
A. Hirayama, Prof. Dr. T. Eguchi
Department of Chemistry and Materials Science
Tokyo Institute of Technology
2-12-1 O-okayama, Meguro-ku, Tokyo 152-8551 (Japan)
E-mail: eguchi@cms.titech.ac.jp
Dr. A. Miyanaga, Prof. Dr. F. Kudo
Department of Chemistry
Tokyo Institute of Technology
2-12-1 O-okayama, Meguro-ku, Tokyo 152-8551 (Japan)
E-mail: fkudo@chem.titech.ac.jp
Supporting information for this article is given via a link at the end of
the document.
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
7- and 62-fold higher than that of the wild type (2.1 µM),
respectively.
the observed accumulation of the quinonoid intermediate 4 in
the active site (Scheme 2).
H2N
HO
OH
HO
O
2-
OPO3
OH
H2N
OH
+
N
Table 1. Kinetic values of PctV and its mutants.
PMP
O
H
O
Wild
type
Y64F
K159Q
N220A
K276Q
E416A
PLP-PctV
PctV
K276R
H
N
KM [M]
2.1 ± 0.4
1.5 ± 0.1
15 ± 2
130 ± 30
N. D.
[a]
1.6
0.4
±
Arg276
H2N
NH
kcat [min-1
]
0.39
0.04
±
0.28
0.02
±
0.18
0.02
±
0.28
0.06
±
N. D.
[a]
0.20 ±
0.01
H
+
OH
OH
HO
HO
OH
OH
[a] N. D. means not detected. PctV (0.1 M) was reacted with L-glutamate
(200 mM), PLP (100 M), and 3-DSA at 28 °C. Kinetic values were
estimated by Michaelis-Menten plot.
OH
N
OH
N
O
O
H
NH2
N
N
H
HN
2-
H
Arg276
2-
N
OPO3
OPO3
H
PMP-3-DSA-imine
Quinonoid intermediate
3
4
As expected, the K276Q and K276R mutants did not afford
detectable amounts of 3-ABA from 3-DSA (data not shown),
which provided further confirmation that Lys276 is critically
important to the formation of a PLP-lysine aldimine complex and
the subsequent generation of PMP with L-glutamate. During the
aminotransfer reaction between PMP and an amino acceptor,
the lysine residue would act as a base to abstract a proton from
the ketimine intermediate of the PMP-amino acceptor complex
to give a quinonoid intermediate (Scheme S1). The resulting -
ammonium moiety of the lysine residue would then act as an
acid to supply a proton for the formation of an aldimine
intermediate, which would undergo nucleophilic attack by the
amino group of the lysine residue to give the initial internal PLP-
aldimine complex. Notably, the reaction of the K276Q mutant
with PMP and 3-DSA only resulted in the formation of PMP-3-
DSA 3, which was formed in the absence of enzyme (Figure S3).
Therefore, this result indicated that the glutamine residue in the
K276Q mutant does not act as base and thus the deprotonation
of 3 did not occur. In contrast, the reaction of the K276R mutant
with PMP and 3-DSA led to the accumulation of a reddish-purple
reaction intermediate, which gave a strong ultraviolet/visible
(UV/Vis) absorption at 580 nm (Figure 1). This result indicated
that the arginine residue in the K276R mutant was acting as a
base like as Lys276 in the wild type to abstract a proton from the
Scheme 2. Proposed mechanism for the reaction of PMP with 3-DSA in the
presence of the K276R mutant
PMP-3-DSA ketimine complex
3
to give
a
quinonoid
intermediate 4, as evidenced by the long wavelength of the
absorption (Scheme 2). The results of an ab initio calculation
using the Gaussian 09 program to determine the structure of the
reaction intermediate based on its UV/Vis absorption spectrum
supported the idea that a quinonoid intermediate was formed
prior to the dehydration step (Figure S4). Given that the acidity
of the guanidinium ion resulting from the protonation of the
arginine residue would be weaker than that of the lysine
ammonium ion, it is unlikely that the arginine residue would
behave as an acid catalyst and provide a proton for the
subsequent dehydration reaction. This would therefore lead to
Figure 1. K276R (40 M) reaction with PMP (40 M) and 3-DSA (400 M) at
room temperature. A) The UV/Vis spectra were measured at 30-min intervals
(0–3 hr and 9.5–12 hr). B) Time course for the absorbance at 580 nm.
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
Crystal Structure Analysis. To determine the structure of the
intermediate accumulated during the reaction of PMP with 3-
DSA in the presence of the K276R mutant, we solved the X-ray
crystal structure of the complex bound to the K276R mutant (2.4
Å resolution), as well as the wild type enzyme with PLP (Figure
2, Figure S5, Table S1). The overall structure of PctV is similar
to that of GSAM, as expected from the model structure (Figure
S5). The X-ray crystal structure of the wild type revealed that
PctV exists as a homodimer in its crystal form as well as in
solution (data not shown), and that the active site Lys276
residue is positioned at the dimer interface. The crystal structure
of the K276R mutant showed a significant electron density in the
active site corresponding to the structure of the PMP-3-DSA
complex (Figure 2B). The Asn220 residue was found to be
located in close proximity to the C-4 hydroxy group of 3-DSA
(2.6 Å) and the hydroxy group of PLP (2.8 Å), which indicated
that this residue is involved in the recognition of these two
substrates. The carboxylate moiety of the 3-DSA substrate
forms interactions with the Ser28 and Arg31 residues of the
enzyme. Furthermore, the six-membered ring structure derived
from 3-DSA contained two hydroxy groups, which indicated that
the accumulated intermediate was a ketimine or quinonoid
intermediate of PMP-3-DSA. However, the electron density
corresponding to the hydroxy groups was relatively weak, most
likely because of the partial dehydration of the intermediate
during X-ray irradiation. Taken together, these results suggested
that the K276R mutant trapped the quinonoid intermediate 4
because of the low acidity (pKa = 12.5 in aqueous solution) of
the resulting guanidinium ion, which would prevent the
occurrence of the subsequent dehydration reaction (Scheme 2).
Thus, the Lys276 residue in the wild type enzyme would act as a
base to allow for the formation of the quinonoid intermediate 4
with the concomitant formation of the -ammonium ion of the
lysine residue (pKa = 10.7 in aqueous solution), which would act
as an acid catalyst to facilitate the dehydration of the
intermediate to give the corresponding aldimine intermediate 6
(Scheme 3).
Figure 2. Crystal structure of K276R with PMP-3-DSA. (A) Overall structure,
(B) active site with the PMP-3-DSA. Monomeric subunits are colored in
green and cyan. An Fo-Fc electron density map contoured at 2.5  was
constructed prior to the incorporation of the PMP-3-DSA molecule.
Interactions with PMP-3-DSA are shown as broken lines.
Additional Mutational Analysis. The structure of the PMP-3-
DSA-K276R complex revealed that the Asn220, His411, and
Glu416 residues could interact with the C-5 hydroxy group of the
reaction intermediate through two water molecules and therefore
seemed to be involved in the second dehydration step (Figure 2).
However, the H411A, H411A/E416A, and N220A/H411A/E416A
mutants allowed for the efficient production of 3-ABA from 3-
DSA with PLP and L-glutamate (Figure S6), which indicated that
these residues are not important to the dehydration reaction.
Thus, it is still unclear whether the second dehydration occurs
enzymatically or non-enzymatically. The -amino group of the
Lys276 residue might attack the aldimine intermediate 6 to
promote the second dehydration reaction and the formation of
the ketimine intermediate 7 (Scheme 3). Last, deprotonation of
the C-6 position and aromatization, followed by the formation of
an internal PLP-lysine aldimine would give 3-ABA 1 as the final
product.
The reaction of the R31A mutant with L-glutamate, PLP
and 3-DSA did not afford 3-ABA, whilst the same reaction with
the S28A mutant produced 3-ABA, albeit at much slower rate
than the wild type enzyme (data not shown). It was difficult to
perform kinetic analysis experiments using the S28A mutant,
because the KM value for 3-DSA seems to be too high. In
contrast, both the S28A and R31A mutants reacted with PMP
and 3-DSA to produce 3-ABA with similar efficiency to the wild
type enzyme (Figure S7). However, the production rates of PMP
from PLP and L-glutamate were very slow for the S28A and
R31A mutants (Figure S8). These results therefore suggested
that the Ser28 and Arg31 residues are involved in the
recognition of the substrates, presumably via the carboxylate
moieties of L-glutamate and 3-DSA, rather than the dehydration
steps.
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
OH
HO
H+
OH
O
OH
Lys276
OH
H2N
OH
HO
HO
HO
N
H
HO
OH
OH
O
PctV
OH
O
OH
N
OH
N
O
H2N
HO
H
H
O
HO
Lys276
2-
N
H2N
OPO3
2-
H
OPO3
N
2-
H
OPO3
N
H
N
2-
OPO3
H
H+
OH
H
O
H+
OH
OH
OHHN
OH
N
H+
OH
OH
N
O
OH
N
H
H2N
Lys276
N
Lys276
O
H2N
H
N
O
N
Lys276
H
H
H2O
H2O
N
2-
PLP-aldimine
2-
H
OPO3
OPO3
2-
OPO3
Scheme 3. Proposed reaction mechanism of 3-ABA formation by PctV.
Stopped-flow Analysis with Wild Type of PctV. Notably,
stopped-flow analysis in the reaction of the wild type enzyme
with PMP and 3-DSA revealed that the same quinonoid
intermediate with an absorption band at 580 nm appeared to be
transiently generated (Figure 3). Therefore, it was concluded
that this quinonoid intermediate 4 is a genuine intermediate in
the PctV reaction (Scheme 3). The formation rates of the
quinonoid intermediate 4 (580 nm) by wild type of PctV and
K276R mutant were estimated to be 24 AU/min and 6.7 x 10-4
AU/min, respectively (Figure 1 and 3). This result indicates that
the K276R mutation significantly reduces the reaction rates
before the quinonoid formation, whilst the Arg residue of the
mutant could work as base like as Lys276 in the wild type to
generate the quinonoid intermediate 4. Since probable PMP-3-
DSA-imine intermediate 3 was not observed in the K276R
reaction with PMP and 3-DSA, the initial nucleophilic attack of
the amino group of PMP to the ketone of 3-DSA would be
affected by this mutation. Thus, -ammonium of the Lys residue
in the wild type might work as an acid catalyst to accelerate the
initial nucleophilic addition during the formation of the ketimine
intermediate 3. It is revealed that the ketimine intermediate 3 is
slowly generated in the K276Q reaction with PMP and 3-DSA
(Figure S3). Thus, in the K276R reaction with PMP and 3-DSA,
the ketimine intermediate 3 might be non-catalytically generated
and the Arg residue might work just as a base catalyst to form
the quinonoid intermediate 4.
The disappearance rate of the quinonoid intermediate 4 (580
nm) in the wild type was estimated to be ca. 18 min-1 by
assuming that the dehydration of the quinonoid intermediate is a
unimolecular reaction (Figure S9). This rate does not account for
the kcat value 0.39 min-1 in the steady state kinetics. Therefore,
Figure 3. Stopped-flow analysis of the wild type PctV (50 M) reaction
with PMP (50 M) and 3-DSA (200 M). A) The UV/Vis spectra were
measured at 0.5-sec intervals (0-5 sec) and 1-sec intervals (5-10 sec).
B) Time course for the absorbance at 580 nm.
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
the rate-limiting step seems to exist after the consumption of the
quinonoid intermediate. In parallel with the disappearance of the
absorption at 580 nm, the absorption at 330 nm and 415 nm
appeared to be increased to indicate that subsequent reaction
intermediates and/or products were formed before the release of
3-ABA (data not shown). However, it is hard to distinguish
several potential intermediates only from the obtained data,
because the ab initio calculation indicates that those absorptions
are overlapped (Figure S4). The aromatization reaction of PLP-
dihydro-ABA-aminal intermediate 7 that would show absorption
at 300 nm and 400 nm could proceed spontaneously, because
any amino acid residue does not interact with this region to
abstract a proton judging from the crystal structure of K276R-
quinonoid complex (Figure 2B). Thus, because only obscure
spectral change to indicate accumulation of reaction
intermediates was observed after the disappearance of the
quinonoid intermediate, we think that the product release would
be very slow and is the rate-limiting step in this reaction.
Crucial Role of Catalytic Lysine Residue in PLP-dependent
Enzyme. The catalytic importance of the active site lysine
residue has been highlighted in several other studies concerning
the mutational analysis of several other PLP-dependent
enzymes.[10] Once the lysine residue is mutated to another
amino acid in these enzymes, PMP is required to allow for the
analysis of their enzymatic reactions. Mutant enzymes
containing an arginine residue instead of a lysine usually
showed much lower levels of activity than the corresponding wild
type enzymes.[11] The results of these studies have therefore
allowed for the multiple functions of active site lysine residues to
be fully characterized.[12] However, there have been no report
providing clear evidence to show that the -amino group of the
lysine residue is critical to the formation of the quinonoid
intermediate or that the resulting -ammonium ion acts as an
acid catalyst to provide a proton for the conversion of the
quinonoid intermediate to an aldimine intermediate. The arginine
residue in the K276R complex with PMP-3-DSA was found to be
located in close proximity to the C-4 hydroxy group of the 3-DSA
substrate (3.7 Å, Figure 2B). Thus, in the reaction of the wild
type PctV enzyme, the protonation of the C-4 hydroxy group of
the quinonoid intermediate 4 appears to occur at the active site
lysine residue, followed by the dehydration (Scheme 3). Further,
it is suggested that the catalytic lysine would be involved in the
PMP-3-DSA-imine 3 formation as an acid catalyst, because the
formation rate of the quinonoid intermediate 4 between the wild
type and K276R was significantly different.
The formation of the quinonoid intermediate in the PLP-
dependent enzymatic reaction has been proposed in several
other studies.[13] From the structure of the PctV K276R mutant-
PMP-3-DSA complex, we could clearly observe the formation of
the native quinonoid intermediate structure during the formation
of 3-ABA. This result therefore confirmed that the pyridine ring of
PLP/PMP acts as a transient electron sink. The subsequent
acid-catalyzed flow of electrons from the electron-rich pyridine
ring is therefore the most likely mechanism for the dehydration
reaction (Scheme 3). In this sense, the acid-base properties of
lysine would be critical to controlling the formation of the
quinonoid intermediate and the subsequent protonation to assist
the dehydration. In the reaction of the K276R mutant with PMP
and 3-DSA, the quinonoid intermediate was accumulated as the
only product and any other reaction intermediate was not
detected. This result therefore demonstrated that the lysine
residue of the PctV enzyme is critical to the dehydration reaction
and the formation of 3-ABA.
Discussion
Mechanistic Comparison with ColD. ColD has been reported
to catalyze the mechanistically related PLP-dependent
dehydration reaction of GDP-4-keto-6-deoxymannnose to give
GDP-colitose in the biosynthesis of the O-antigen
lipopolysaccharide of a Gram-negative bacteria (Scheme S2).[7]
ColD has a histidine residue in its active site instead of lysine,
therefore the PMP cofactor is an important requirement for this
reaction.[8] In the ColD reaction, the histidine residue in the
active site abstracts a proton from the PMP-substrate ketimine
intermediate to assist in the dehydration step, with the
abstracted proton going on to act as an acid catalyst. The
switching of the active site histidine residue in ColD to lysine (i.e.,
H188K) in addition to the introduction of a S187N mutation
resulted in a change in the enzymatic function of the enzyme
from a dehydratase to an aminotransferase.[9] This change in the
function of the mutant ColD enzyme suggested that the lysine in
the active site was acting as a base, and that the resulting
protonated lysine was so far away from the dehydration site that
the enzyme could only facilitate the aminotransfer reaction.
These results therefore demonstrate that the positioning of the
active site amino acid residue (i.e., histidine or lysine) in the
ColD enzyme dictates the enzymatic reaction pathway. This
mechanism could also apply to the PctV-catalyzed reaction,
where the active site Lys276 residue adjacent to the C-4
hydroxy group of the 3-DSA substrate acts as an acid/base
catalyst to catalyze the dehydration of the PMP-3-DSA
intermediate. Thus, the location of the lysine residue is critical to
determining the reaction path. The K276H mutant of PctV
behaved as the K276Q mutant when PMP and 3-DSA were
reacted (data not shown). Thus, the His residue of the K276H
mutant does not work as a base catalyst to abstract a proton of
PMP-3-DSA-imine 3 to give quinonoid 4, presumably due to the
shorter length of the side chain than those of Arg and Lys.
Conclusions
We have successfully trapped a native form of the quinonoid
reaction intermediate formed during the PctV-catalyzed
conversion of 3-DSA to 3-ABA by introducing a K276R mutation
in the active site of the enzyme. Furthermore, we have provided
clear evidence that the Lys276 residue in the active site is
critical to the dehydration of the quinonoid intermediate, as well
as the formation of the initial PLP-lysine aldimine and PMP-3-
DSA- imine intermediates during the PctV-catalyzed formation of
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
Three mixtures were prepared for each concentration of 3-DSA. Reaction
mixtures were quenched by adding 50 L methanol and centrifuged
(15,000 rpm, 30 min). Supernatants of these samples were evaporated
and then dissolved in 10 L of MilliQ-water. These solutions were then
injected into an HPLC instrument described above. The elution
conditions were 0.1% trifluoroacetatic acid-7.5% CH3CN-H2O at a flow
rate 0.5 mL/min at room temperature. Obtained data were fitted to
Michaelis-Menten equation to determine their kinetic parameters.
3-ABA. It is envisaged that these results will be used to expand
the scope of PLP-dependent enzymatic chemistry.
Experimental Section
Materials and Instruments. All of the biological experiments including
DNA manipulation and enzyme treatments were performed according to
the standard protocols.[14] The restriction enzymes and ligase were
purchased from Takara (Shiga, Japan). DNA oligonucleotide primers
were synthesized by Fasmac (Kanagawa, Japan). KUBOTA 3700
(Kubota, Tokyo, Japan) and HIMAC CENTRIFUGE SCR20B (Hitachi,
Tokyo, Japan) centrifuges were used for centrifugation. UV/Vis spectral
analyses and absorbance measurements were recorded on a Shimadzu
UV-2450 spectrophotometer (Kyoto, Japan). Stopped flow analysis were
performed using FS-100 high-speed scan spectrophotometer (JASCO,
Tokyo, Japan), SFS-472 stopped flow system (JASCO, Tokyo, Japan),
and ED Heating Immersion Circulator (JULABO, Seelbach, Germany).
Calculation of UV/Vis Absorption. Gaussian 09 program package
(Gaussian, Connecticut, USA) was used for optimization of structures
and frequency calculation with the B3LYP method and the 6-31G+ basis
set. The solvation effect of water was considered by B3LYP/6-31G+.
Stopped Flow Analysis. Twenty five L of mixture A containing 50 mM
of Tris (pH 7.5), 10% glycerol, 100 M of PctV wild type, and 100 M of
PMP and 25 L of mixture B containing 50 mM of Tris (pH 7.5), 10%
glycerol, and 400 M of 3-DSA were mixed within 10 ms, and then
UV/Vis spectra were measured every 50 ms (total 15 s). Temperature of
reaction mixture was kept at 28 ± 0.5 °C.
Mutational Analysis of PctV. Mutagenesis of pctV was performed by
QuikChange® method. The plasmid pctV/pET28, which was obtained in
previous report[1a], was used as template for PCR. Sequences of used
primers were listed in Table S2. PCR conditions: 94 °C for 2 min 12 or 16
cycles of 98 °C for 10 s, 60 °C for 30 s, and 68 °C for 3.5 min; 0.2 L of
KOD -plus- Neo polymerase (TOYOBO, Osaka, Japan), 1.0 L of 10×
PCR Buffer for KOD -plus- Neo, 1.0 L of dNTPs (2 mM each), 0.6 L of
MgSO4 (25 mM), 0.3 L of primer-F (10 M), 0.3 L of primer-R (10 M),
0.5 L of DMSO, 10 ng of template, adjusted to 10 L with water. 5 L of
PCR reaction mixture was used for checking amplification and another 5
L was added 0.1 L of DpnI for digestion of template DNA. After
incubation at 37 °C, 2 L of reaction mixture was used for transformation
of E. coli DH5. Transformant was cultured on lysogeny broth (LB)
containing kanamycin (30 g/mL). Several colonies were cultured and
plasmid DNA was extracted. After confirming its sequence, E. coli
BL21(DE3) was transformed by the plasmid and cultured on LB
containing kanamycin (30 g/mL). H411A/E416A variant and
N220A/H411A/E416A variant were made using almost the same method
with pctV-H411A/pET28 and pctV-H411A-E416A/pET28 as template,
respectively. All recombinant proteins were expressed and purified
according to the previously described method.[1a]
Preparation of PctV-K276R Mutant for Crystallization. PctV-K276R
mutant was expressed and purified by the same method as wild type and
then concentrated by Amicon® ultra-4 (Millipore, Darmstadt, Germany).
PctV-K276R (150 M) was reacted with 200 M of PMP and 1.5 mM of
3-DSA in Tris (50 mM, pH 7.5) buffer containing 200 mM of L-glutamate
and 10% glycerol (28°C, 24 hr, 600 L). The reaction mixture (89.2 L)
was treated with 10 L of 10 × buffer containing 0.8 L of 0.2 unit/L of
thrombin (Merck, Darmstadt, Germany) for thrombin cleavage and the
mixture was incubated at 28 °C for 3.5 hr. Then streptavidin agarose was
added to remove thrombin. After removing of streptavidin agarose, the
solution was poured into column with His60 Ni superflow resin (Takara,
Shiga, Japan) and the unbounded fractions were used for gel filtration
(SuperdexTM 200 10/300GL, GE Healthcare UK, Buckinghamshire,
England; Tris (50 mM, pH 7.5) buffer containing 5% glycerol and 150 mM
of KCl). Fractions containing PctV-K276R were concentrated and then its
buffer was replaced with Tris (50 mM, pH 7.5) buffer containing 10%
glycerol.
Crystal Structural Analysis. Crystals of native PctV wild type were
grown from a 1:1 mixture of a protein solution (4.0 mg/mL in 10 mM Tris-
HCl (pH 7.5) and 10% glycerol) and a reservoir solution (0.2 M
magnesium chloride, 20% polyethylene glycol 4000, and 0.1 M Tris-HCl
(pH 8.5)) using the sitting-drop vapor-diffusion method at 26 °C. Crystals
of PctV K276R-PMP-3-DSA complex were grown from a 1:1 mixture of a
protein solution (7.5 mg/mL in 50 mM Tris-HCl (pH 7.5) and 10%
glycerol) and a reservoir solution (0.2 M magnesium chloride, 24%
polyethylene glycol 1000 and MES-NaOH (pH 5.8)) using the sitting-drop
vapor-diffusion method at 5 °C. Before X-ray data collection, the crystal
was transferred into the reservoir solution supplemented with 25%(v/v)
polyethylene glycol 400 or 10% (v/v) glycerol as a cryoprotectant and
flashfrozen in the liquid nitrogen stream. X-ray data were collected using
beamline AR-NW12A (Photon Factory, Tsukuba, Japan). All diffraction
data were indexed, integrated, and scaled using the program iMosflm.[15]
The initial phases were determined by the molecular replacement
method using the program Molrep[16] with the crystal structure of GSAM
(Protein Data Bank Code: 3FQ7). The program ARP/wARP[17] was used
for automatic initial protein model building and refinement. Coot[18] was
used for visual inspection and manual rebuilding of the model. Refmac[19]
was used for refinement. The figures were prepared using PyMOL (The
PyMOL Molecular Graphics System, DeLano Scientific LLC, Palo Alto,
CA.). The geometries of the final structures were evaluated using the
program Rampage.[20] The resulting coordinates and structure factors
To test enzymatic activity, PctV mutants were reacted with L-glutamate,
PLP, and 3-DSA at 28 °C. Reaction mixtures contained 1 M of PctV
mutant, 100 M of PLP, 200 mM of L-glutamate, and 200 M of 3-DSA
(total 50 L). Enzymatic reactions were quenched by adding 50 L
methanol. After centrifugation (15,000 rpm, 30 min), supernatants of
these samples were filtered through Millex LG filters (Millipore, Darmstadt,
Germany). The obtained solutions were then injected into an HPLC
instrument (Elite LaChrom L-2455 DAD Detector and L-2130 Pump,
Hitachi, Tokyo, Japan) equipped with a Phenomenex® Luna 3  PFP(2)
100 Å 150 × 4.6 mm column (Phenomenex, California, USA). The elution
conditions were 100 mM NaH2PO4-5% CH3CN-H2O, at a flow rate 0.5
mL/min at room temperature. Some PctV variants were also reacted with
PMP and 3-DSA at 28 °C. Reaction mixtures contained 10 M of PctV
variant, 100 M of PMP, and 200 M of 3-DSA (total 50 L). HPLC
analyses were performed as described above.
Kinetic Analyses of PctV Mutants. Tris buffer (10 mM, pH 7.5) was
used for kinetic analysis. 35 L of PctV variant containing PLP was
added to 15 L of mixture of L-glutamate and 3-DSA, and then incubated
at 28 °C for 30 min. Reaction mixtures contained 0.1 M of PctV variant,
100 M of PLP, 200 mM of L-glutamate, and various concentration of 3-
DSA (1.25, 2.5, 5.0, 6.25, 12.5, 25.0, 50.0, 75.0, 100, 150, and 250 M).
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
[1]
a) A. Hirayama, T. Eguchi, F. Kudo, Chembiochem 2013, 14,
1198-1203; b) K. H. Almabruk, W. Lu, Y. Li, M. Abugreen, J. X.
Kelly, T. Mahmud, Org. lett. 2013, 15, 1678-1681.
K. L. Rinehart, Jr., M. Potgieter, D. L. Delaware, H. Seto, J. Am.
Chem. Soc. 1981, 103, 2099-2101.
I. U. Heinemann, M. Jahn, D. Jahn, Arch. Biochem. Biophys. 2008,
474, 238-251.
M. D. Toney, Biochim. Biophys. Acta 2011, 1814, 1407-1418.
L. Bordoli, T. Schwede, Methods Mol. Biol. 2012, 857, 107-136.
G. L. Orriss, T. R. Patel, J. Sorensen, J. Stetefeld, FASEB J. 2010,
24, 404-414.
N. Beyer, J. Alam, T. M. Hallis, Z. Guo, H. W. Liu, J. Am. Chem.
Soc. 2003, 125, 5584-5585.
P. D. Cook, H. M. Holden, Biochemistry 2007, 46, 14215-14224.
P. D. Cook, R. L. Kubiak, D. P. Toomey, H. M. Holden,
Biochemistry 2009, 48, 5246-5253.
M. B. Bhatia, S. Futaki, H. Ueno, J. M. Manning, D. Ringe, T.
Yoshimura, K. Soda, J. Biol. Chem. 1993, 268, 6932-6938.
a) L. L. Ilag, D. Jahn, Biochemistry 1992, 31, 7143-7151; b) B.
Grimm, M. A. Smith, D. von Wettstein, Eur. J. Biochem. 1992, 206,
579-585.
have been deposited in the Protein Data Bank (PDB codes: 4ZM3 and
4ZM4).
[2]
[3]
UV/Vis Spectral Analyses. To investigate whether K276Q mutant
accumulates any intermediate, UV/Vis spectra analyses were performed.
3 L of 3-DSA was added to 147 L of mixture containing 10 mM Tris
(pH 7.5), 10% glycerol, PctV (wild type or K276Q) and PMP. Reaction
mixture contained 50 M of PctV, 50 M of PMP and 200 M of 3-DSA.
UV/Vis spectra were measured every 60 sec (total 1 hr) at room
temperature. UV/Vis spectra of mixture without enzyme were also
measured.
[4]
[5]
[6]
[7]
[8]
[9]
To investigate whether K276R mutant accumulates any intermediate, 6
L of 3-DSA was added to 144 L of mixture containing 10 mM Tris (pH
7.5), 10% glycerol, 200 mM L-glutamate, K276R variant, and PMP, and
then UV/Vis spectra were measured every 30 min (total 12 hr). Reaction
mixture contained 40 M of PctV, 40 M of PMP and 400 M of 3-DSA.
[10]
[11]
[12]
[13]
Z. Lu, S. Nagata, P. McPhie, E. W. Miles, J. Biol. Chem. 1993, 268,
8727-8734.
a) J. Lai, D. Niks, Y. Wang, T. Domratcheva, T. R. Barends, F.
Schwarz, R. A. Olsen, D. W. Elliott, M. Q. Fatmi, C. E. Chang, I.
Schlichting, M. F. Dunn, L. J. Mueller, J. Am. Chem. Soc. 2011,
133, 4-7; b) D. Milic, T. V. Demidkina, N. G. Faleev, R. S. Phillips,
D. Matkovic-Calogovic, A. A. Antson, J. Am. Chem. Soc. 2011,
133, 16468-16476; c) W. E. Karsten, P. F. Cook, Biochim. Biophys.
Acta 2009, 1790, 575-580; d) D. Milic, T. V. Demidkina, N. G.
Faleev, D. Matkovic-Calogovic, A. A. Antson, J. Biol. Chem. 2008,
283, 29206-29214; e) T. R. Barends, T. Domratcheva, V. Kulik, L.
Blumenstein, D. Niks, M. F. Dunn, I. Schlichting, Chembiochem
2008, 9, 1024-1028; f) N. C. Furumo, J. F. Kirsch, Arch. Biochem.
Biophys. 1995, 319, 49-54.
J. Sambrook, D. W. Russell, Molecular cloning. A laboratory
manual, 3 rd edn, Cold Spring Harbor Laboratory, Cold Spring
Harbor NY, 2001.
T. G. Battye, L. Kontogiannis, O. Johnson, H. R. Powell, A. G.
Leslie, Acta Crystallogr. D Biol. Crystallogr. 2011, 67, 271-281.
A. Vagin, A. Teplyakov, Acta Crystallogr. D Biol. Crystallogr. 2010,
66, 22-25.
R. J. Morris, A. Perrakis, V. S. Lamzin, Acta Crystallogr. D Biol.
Crystallogr. 2002, 58, 968-975.
P. Emsley, K. Cowtan, Acta Crystallogr. D Biol. Crystallogr. 2004,
60, 2126-2132.
G. N. Murshudov, A. A. Vagin, E. J. Dodson, Acta Crystallogr. D
Biol. Crystallogr. 1997, 53, 240-255.
S. C. Lovell, I. W. Davis, W. B. Arendall, 3rd, P. I. de Bakker, J. M.
Word, M. G. Prisant, J. S. Richardson, D. C. Richardson, Proteins
2003, 50, 437-450.
To investigate PMP forming ability of S28A and R31A mutants, 15 L of
L-glutamate was added to 135 L of the mixture containing 10 mM Tris
(pH 7.5), 10% glycerol, PLP, and PctV (wild type, S28A, or R31A), and
then UV/Vis spectra were measured every 60 sec (total 10 min).
Reaction mixture contained 7.5 M of PctV, 60 M of PLP and 1, 20, or
200 mM of L-glutamate.
Acknowledgements
[14]
This work was performed with the approval of the Photon
Factory Advisory Committee (Proposal Nos. 2012G508 and
2014G530) and was supported in part by Grants-in-Aid for
Scientific Research on Innovative Areas and a Grant-in-Aid for
the Japan Society for the Promotion of Science Fellows from the
Ministry of Education, Culture, Sports, Science and Technology
of Japan.
[15]
[16]
[17]
[18]
[19]
[20]
Keywords: PLP-dependent enzyme • quinonoid • pactamycin •
natural products • biosynthesis
For internal use, please do not delete. Submitted_Manuscript

ChemBioChem
10.1002/cbic.201500426
FULL PAPER
Table of Contents
FULL PAPER
Akane Hirayama, Akimasa Miyanaga,
Fumitaka Kudo,* and Tadashi Eguchi*
Page No. – Page No.
Mechanism-based Trapping of the
Quinonoid Intermediate Using the
K276R Mutant of PLP-Dependent 3-
Aminobenzoate Synthase PctV in the
Biosynthesis of Pactamycin
Purple quinonoid intermediate of pyridoxamine 5’-phosphate and 3-
dehydroshikimate was accumulated in the K276R mutant of 3-aminobenzoate
synthase PctV in the biosynthesis of pactamycin. X-ray protein structural analysis of
the complex revealed the quinonoid molecule trapped at the active site of the
mutant. Overall, the active site Lys found to have a multiple role including the
aminotransfer reaction and dehydration to give 3-aminobenzoate.
For internal use, please do not delete. Submitted_Manuscript