Identification of 2, 3-dihydrodipicolinate as the product of the dihydrodipicolinate synthase reaction from Escherichia coli

Article abstract of DOI:10.1016/j.abb.2018.06.011

Dihydrodipicolinate synthase (DHDPS) catalyzes the first step in the pathway for the biosynthesis of L-lysine in most bacteria and plants. The substrates for the enzyme are pyruvate and L-aspartate-β-semialdehyde (ASA). The product of the reaction was originally proposed to be 2,3-dihydrodipicolinate (DHDP), but has now generally been assumed to be (4S)-4-hydroxy-2,3,4,5-tetrahydro-(2S)-dipicolinate (HTPA). ASA is unstable at high pH and it is proposed that ASA reacts with itself. At high pH ASA also reacts with Tris buffer and both reactions are largely reversible at low pH. It is proposed that the basic un-protonated form of the amine of Tris or the α-amine of ASA reacts with the aldehyde functional group of ASA to generate an imine product. Proton NMR spectra of ASA done at different pH values shows new NMR peaks at high pH, but not at low pH, confirming the presence of reaction products for ASA at high pH. The enzymatic product of the DHDPS reaction was examined at low pH by proton NMR starting with either 3 h-pyruvate or 3 d-pyruvate and identical NMR spectra were obtained with four new NMR peaks observed at 1.5, 2.3, 3.9 and 4.1 ppm in both cases. The NMR results were most consistent with DHDP as the reaction product. The UV-spectral studies of the DHDPS reaction shows the formation of an initial product with a broad spectral peak at 254 nM. The DHDPS reaction product was further examined by reduction of the enzymatic reaction components with borohydride followed by GC-MS analysis of the mixture. Three peaks were found at 88, 119 and 169 m/z, consistent with pyruvate, homoserine (reduction product of ASA), and the reduction product of DHDP (1,2,3,6-tetrahydropyridine-2,6-dicarboxylate). There was no indication for a peak associated with the reduced form of HTPA.

Full text of DOI:10.1016/j.abb.2018.06.011

Accepted Manuscript
Identification of 2, 3-dihydrodipicolinate as the product of the dihydrodipicolinate
synthase reaction from Escherichia coli
William E. Karsten, Susan A. Nimmo, Jianguo Liu, Lilian Chooback
PII:
S0003-9861(18)30221-2
10.1016/j.abb.2018.06.011
YABBI 7757
DOI:
Reference:
To appear in: Archives of Biochemistry and Biophysics
Received Date: 20 March 2018
Revised Date: 27 May 2018
Accepted Date: 22 June 2018
Please cite this article as: W.E. Karsten, S.A. Nimmo, J. Liu, L. Chooback, Identification of 2, 3-
dihydrodipicolinate as the product of the dihydrodipicolinate synthase reaction from Escherichia coli,
Archives of Biochemistry and Biophysics (2018), doi: 10.1016/j.abb.2018.06.011.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Identification of 2, 3-Dihydrodipicolinate as the Product of the
Dihydrodipicolinate Synthase Reaction from Escherichia coli
William E. Karsten‡, Susan A Nimmo‡, Jianguo Liu*, and Lilian Chooback*
‡Department of Chemistry and Biochemistry, University of Oklahoma, 101 Stephenson Parkway,
Norman, Oklahoma 73019; *Department of Chemistry, University of Central Oklahoma, 100 N.
University Dr., Edmund, Oklahoma 73034
Research reported in this publication was supported by an Institutional Development Award
(IDeA) from the National Institute of General Medical Sciences of the National Institutes of
Health under grant number 8P20GM103447 and funds from the University of Central Oklahoma
office of Research and Grants to L.C.
*Corresponding Author
E-mail: lchooback@uco.edu. Phone (405) 974-5481. Fax (405) 974-3862

ACCEPTED MANUSCRIPT
ABBREVIATIONS
DL-ASA, DL-aspartate
β
-semialdehyde; Ches, 2-(N-cyclohexylamino) ethane-sulfonic acid;
DAP, meso-diaminopimelate; DHDP, dihydrodipicolinate; DHDPS, dihydrodipicolinate
synthase; DHDPR, dihydrodipicolinate reductase; Hepes, N-(2-hydroxyethyl)piperazine-N’-2-
ethane-sulfonic acid; HTPA, (4S)-4-hydroxy-2,3,4,5-tetrahydro-(2S)-dipicolinic acid; Mes, 2-(N-
morpholine) ethane-sulfonic acid; NADH, reduced nicotinamide adenine dinucleotide; Tris, Tris
(hydroxymethyl) aminomethane.
2

ACCEPTED MANUSCRIPT
ABSTRACT
Dihydrodipicolinate synthase (DHDPS) catalyzes the first step in the pathway for the
biosynthesis of L-lysine in most bacteria and plants. The substrates for the enzyme are pyruvate
and L-aspartate-β-semialdehyde (ASA). The product of the reaction was originally proposed to
be 2, 3-dihydrodipicolinate (DHDP), but has now generally been assumed to be (4S)-4-hydroxy-
2,3,4,5-tetrahydro-(2S)-dipicolinate (HTPA). ASA is unstable at high pH and it is proposed that
ASA reacts with itself. At high pH ASA also reacts with Tris buffer and both reactions are
largely reversible at low pH. It is proposed that the basic un-protonated form of the amine of
Tris or the α-amine of ASA reacts with the aldehyde functional group of ASA to generate an
imine product. Proton NMR spectra of ASA done at different pH values shows new NMR peaks
at high pH, but not at low pH, confirming the presence of reaction products for ASA at high pH.
The enzymatic product of the DHDPS reaction was examined at low pH by proton NMR starting
with either 3h-pyruvate or 3d-pyruvate and identical NMR spectra were obtained with four new
NMR peaks observed at 1.5, 2.3, 3.9 and 4.1 ppm in both cases. The NMR results were most
consistent with DHDP as the reaction product. The UV-spectral studies of the DHDPS reaction
shows the formation of an initial product with a broad spectral peak at 254 nM. The DHDPS
reaction product was further examined by reduction of the enzymatic reaction components with
borohydride followed by GC-MS analysis of the mixture. Three peaks were found at 88, 119
and 169 m/z, consistent with pyruvate, homoserine (reduction product of ASA), and the
reduction product of DHDP (1,2,3,6-tetrahydropyridine-2,6-dicarboxylate). There was no
indication for a peak associated with the reduced form of HTPA.
3

ACCEPTED MANUSCRIPT
KEYWORDS
lysine biosynthesis, dihydrodipicolinate synthase, NMR, GC-MS,
4

ACCEPTED MANUSCRIPT
INTRODUCTION
Dihydrodipicolinate synthase (DHDPS) (EC 4.2.1.52) is present in bacteria and the
chloroplasts of higher plants where it catalyzes the first reaction unique to the biosynthesis of
meso-diaminopimelate and L-lysine. Meso-diaminopimelate is a component of bacterial cell
walls and the absence of DHDPS has been shown to be lethal to bacteria [1, 2]. The enzyme is
not present in humans and has received attention recently as a potential target for the
development of new antibiotics or herbicides [3-5]. The substrates for the DHDPS reaction are
pyruvate and L-aspartate-β-semialdehyde. The D-isomer of aspartate-β-semialdehyde is reported
not to be a substrate or inhibitor of DHDPS [6]. Identity of the reaction product has been less
clear due to the lability of the product. The product of the DHDPS reaction was originally
proposed to be 2, 3-dihydrodipicolinate (DHDP) as shown in Scheme 1 [7, 8]. However, later
work by Blickling et al. [9], based upon proton and 13C NMR studies of the reaction, led to the
proposal that the immediate product of the enzymatic reaction is (4S)-4-hydroxy-2,3,4,5-
tetrahydro-(2S)-dipicolinic acid, HTPA (Scheme 1). Subsequent proton NMR results by others
led these workers to a similar conclusion that HTPA is the enzymatic reaction product [10, 11]
Reddy et al. [12] carried out an NMR study of the reaction of DHDPR, which catalyzes the next
step in the biosynthetic pathway after the DHDPS reaction, and the results were consistent with
DHDP as the substrate for DHDPR suggesting DHDP is the product of the DHDPS reaction.
More recently it has been proposed that HTPA is the product of the synthase reaction and
DHDPR uses HTPA as substrate and catalyzes the hydrolysis of HTPA to DHDP subsequent to
hydride transfer [10].
The kinetic mechanism for the enzyme is ping-pong at pH 8 and is supported by the
observation of a parallel initial velocity pattern and dead-end inhibition patterns for several
substrate analogs [13]. L-Lysine is an allosteric inhibitor of the enzyme that leads, at saturating
lysine concentration, to partial decrease in activity that is about 10% of the activity in the
absence of lysine. Crystal structures of the enzyme in the presence of pyruvate have shown
pyruvate covalently bound to lysine 161 [14]. The evidence suggests that pyruvate initially
forms a Schiff base with lysine 161 followed by subsequent loss of a proton from the β-methyl
group of the covalently bound pyruvate leading to formation of an enamine intermediate. The
loss of the proton comprises the irreversible step leading to the ping-pong kinetic mechanism at
5

ACCEPTED MANUSCRIPT
pH 8. Although the kinetic mechanism is ping-pong at high pH, it conforms to a sequential
kinetic mechanism at pH 5.7 [13]. After binding of ASA to enzyme the enamine is proposed to
attack the carbonyl carbon of ASA to initiate the second half reaction leading to the ultimate
formation of the enzymatic product. For the formation of HTPA, it has been proposed that after
attack of the enamine on ASA a transimination step occurs forming the final product. In
contrast, for formation of DHDP, it is proposed that prior to the transimination step a
dehydration step occurs forming the double bond seen in the product.
Our interest in the chemical mechanism of DHDPS prompted a desire to carry out pH
studies to investigate the acid-base chemical mechanism of the enzyme. However, it was
necessary to understand the factors affecting ASA stability in order to properly conduct the pH
studies and led to the current work. The results from the ASA stability studies led to a re-
examination of the identity of the product of the DHDPS reaction.
6

ACCEPTED MANUSCRIPT
MATERIALS AND METHODS
Chemicals and Reagents. Pyruvate, boric acid, NADH, DL-2-amino-4-pentenoic acid,
and rabbit muscle lactate dehydrogenase were purchased from SIGMA. The Tris base was from
RPI and potassium phosphate from Fisher. Hepes, Mes, and Ches buffers were from Research
Organics. The Ni-NTA was from Qiagen. Deuterium oxide 99% was purchased from
Cambridge Isotope Laboratories. DL-ASA was synthesized by ozonolysis of DL-2-amino-4-
pentenoic acid and purified on a DOWEX 50PW4-200 ion exchange column according to the
procedure of Black and Wright [15]. The ion exchange column removes formaldehyde which is
the second reaction product resulting from the ozonolysis step. Deuterated pyruvate was
synthesized using DHDPS. Protonated pyruvate was dissolved in 10 mL of deuterium oxide
(99%) containing 20 mM potassium phosphate buffer at pD 7.3. The final concentration of
pyruvate was 200 mM. DHDPS (80 µg) was added to the mixture and incubated at room
temperature for 72 hours, after which time the enzyme was removed by ultra-filtration. The
filtrate containing 3d-pyruvate was used without further purification. Protein concentrations
were determined by the method of Bradford using bovine serum albumin as protein standard
[16].
Enzymes. A PCR based method was used to clone the E.coli DHDPS gene (dapA) from
strain JM109 into the pET100/D-TOPO expression vector (Invitrogen) according to the
manufacturers recommended procedure. The cloned gene was transformed into E. coli bacterial
strain BL21-DE3. The expression of DHDPS to produce soluble protein was carried out in LB
media at room temperature in the presence of 0.5 mM IPTG. The soluble protein was purified
on a Ni-NTA column and eluted with imidazole using standard procedures. The resulting
enzyme was concentrated by ultra-filtration and dialyzed against two 1 liter volumes of enzyme
storage buffer (10 mM Hepes (pH 7.5), and 10% glycerol). The isolated enzyme is >95% pure
based upon SDS/PAGE. The gene (dapB) that encodes dihydrodipicolinate reductase was
cloned into a pET16b expression vector (Novagen) that incorporates an N-terminus 6-histidine
tag on the resulting enzyme. The gene sequence was used to design the forward and reverse
primers. E. coli strain JM109 chromosomal DNA and PCR were used to amplify the gene. The
PCR product was sequenced to confirm the identity of dapB. The dapB gene was cloned into the
expression vector pET16b using the BamH1 and Nde1 restriction enzyme sites. The histidine-
7

ACCEPTED MANUSCRIPT
tagged protein was expressed using IPTG and purified using Ni-NTA agarose. The resulting
enzyme was concentrated and dialyzed as described for DHDPS. The enzyme is estimated to be
at least 95% pure based upon SDS/PAGE analysis.
Enzyme Assay. All data were collected on either a Beckman DU 640 spectrophotometer
or a Hewlett Packard 8453 diode array spectrophotometer. The assays were done at 25º C. The
temperature was maintained with a Neslab RTE7 circulating water bath connected to the
Beckman spectrophotometer or a Hewlett Packard Peltier temperature controller connected to the
8453 diode array spectrophotometer. The assays were carried out in 1 cm path-length quartz
cuvettes and a 1 mL volume. A typical assay contained 150 mM buffer at the desired pH, 0.2
mM NADH, 10 µg DHDPR, 0.5 µg DHDPS, and varied concentrations of pyruvate and ASA.
The assay is a coupled enzyme assay whereby the product of the DHDPS reaction is reduced by
DHDPR using NADH and the reaction is followed by the decrease in absorbance at 340 nm
resulting from the conversion of NADH (ε
340 = 6220 M-1 cm-1) to NAD.
Reaction of ASA with Tris Buffer. The reaction of Tris buffer with ASA was studied in
the following manner. Initial rates were measured in a standard coupled enzyme assay at pH 8.1
containing 100 mM Hepes, and 0, 20, 40, or 80 mM Tris at 25 ˚C. All assays also contained 1
mM pyruvate, 0.2 mM L-ASA, 0.2 mM NADH, 10 µg DHDPR and 0.5 µg DHDPS. The
reactions were followed for 10 minutes. Another assay was carried out by pre-incubating in the
cuvette 80 mM Tris with 0.2 mM L-ASA for 5 minutes and then adding the remaining assay
components to initiate the reaction. Pre-incubation studies with Tris and DL-ASA were also
done. A pre-incubation mixture was made containing 100 mM Hepes, 10 mM DL-ASA, 80 mM
Tris in a total volume of 200 µL at pH 8. Immediately after adding DL-ASA, a 10 µL aliquot of
the pre-incubation mixture was transferred to an end-point assay to determine the L-ASA
concentration. After five-minutes of incubation at 22 ºC, a second 10 µL aliquot was transferred
to an L-ASA end-point assay. At the same five minute mark two additional 10 µL aliquots were
transferred to two separate 1 mL assay mixes containing buffers only. These two remaining
assay mixes were allowed to incubate at 22 ºC for either 20 minutes or 80 minutes at which time
the remaining assay components were added to initiate the reaction and determine the amount of
L-ASA present. The end-point assays contained 100 mM Hepes, 50 mM Mes, 2 mM pyruvate,
8

ACCEPTED MANUSCRIPT
0.2 mM NADH, 5 µg DHDPS, 10 µg DHDPR, and DL-ASA at pH 6.8. The pH of assays and
pre-incubation mixtures was routinely determined after completion of the reactions.
ASA Stability. The stability of ASA at different pH values was determined as follows.
A 10 mM solution of DL-ASA was prepared at different pH values using 150 mm Hepes or Ches
buffer. After a 10 minute incubation time, a 10 µL aliquot of the DL-ASA mixture was
transferred to a DHDPS end-point reaction assay. The concentration of L-ASA in the assay was
about 0.05 mM. The pH values for the pre-incubations were 6.8, 8.1, and 8.8. In another
experiment, DL-ASA (10 mM) was pre-incubated 10 minutes at pH 8.8, a 10 µL aliquot of the
pre-incubation mix was transferred to an end-point assay to determine the amount of L-ASA
present. A second 10 µL aliquot was removed at 10 minutes and incubated one hour at pH 6.8
prior to adding the remaining assay components to start the reaction. A control assay with no
pre-incubation was also carried out.
Proton NMR Studies. The proton NMR samples of ASA contained in 0.7 mL 10 mM
DL-ASA and 150 mM potassium phosphate at either pH 6.5, 8.0 or 8.8. Prior to acquiring the
NMR spectra, the samples were allowed to incubate for 10 minutes at room temperature after
addition of DL-ASA. The pH of the NMR samples were determined after collecting the NMR
spectra. The reported pH values are based upon the pH meter reading, but since the samples
were in D2O, the pD values are 0.4 units higher due to the isotope effect on the pH electrode
[17]. The samples for monitoring the proton NMR of the DHDPS reaction contained in 0.7 mL,
10 mM DL-ASA, 7 mM 3h-pyruvate or 3d-pyruvate, and 150 mM potassium phosphate, pD 7.5.
DL-ASA was added last just prior to collecting an initial control spectrum. Immediately after
acquisition of the control spectrum, 80 µg of DHDPS was added to the sample and NMR spectra
were collected at 2 minute intervals for 8 minutes. The pD of the sample was determined after
completion of the experiment.
The NMR data were collected using VNMRJ2.2 C on a Varian VNMRS-500 MHz NMR
spectrometer equipped with a 5 mm triple resonance probe. The proton data were collected
using a standard one pulse experiment with an acquisition time of 2 seconds, a relaxation delay
of 1 second, a 45o pulse width and 8 scans. The 2D experiments were collected with the gCOSY
pulse sequence as supplied by Agilent. The data were collected with 1 scan and 128 increments.
9

ACCEPTED MANUSCRIPT
Ultraviolet-Spectra of the DHDPS Reaction Product. In order to determine the
ultraviolet-spectra of the reaction product an enzyme assay was set-up using a 1 cm path length
tandem cuvette. One side of the tandem cuvette contained 30 mM potassium phosphate buffer,
pH 7.4, and 5 µg DHDPS in a final volume of 0.9 mL. The other side of the cuvette contained
30 mM potassium phosphate, pH 7.4, 0.4 mM pyruvate, and 0.4 mM L-ASA in a final volume of
0.9 mL. The cuvette was inserted into the HP-diode array spectrophotometer and the instrument
was blanked on the cuvette. To initiate the reaction, the cuvette was removed from the
spectrophotometer, mixed, and reinserted followed immediately by data acquisition. The first
spectrum was acquired ten seconds after mixing followed by the acquisition of eight additional
spectra at one minute intervals to determine the UV-spectra of the initially formed enzymatic
product. Additional spectra were then acquired at ten minute intervals for one hour. A spectrum
of dipicolinate was determined in a cuvette containing 25 mM potassium phosphate, pH 7.4, and
0.05 mM dipicolinate and a final volume of one mL.
Gas Chromatography-Mass Spectrometry. For the experiment to determine the mass
of the product an assay was set-up containing 30 mM ammonium bicarbonate, pH 7.4, 7 mM
pyruvate, 10 mM DL-ASA (5 mM L-ASA), and 10.4 µM DHDPS subunit concentration in a
total volume of 10 mL. The reaction was allowed to proceed for 8 minutes at which time sodium
borohydride was added to a final concentration of 10 mM. The enzyme was removed by ultra-
filtration using an Amicon centrifugal concentrator. The pH of the flow through was adjusted to
7.0 with HCl, frozen at -20 ºC, and lyophilized to dryness using a Labconco Freezone Model
77400 lyophilizer.
For the GC-MS experiment about 0.1 g of the dried crude sample was placed inside a test
tube. Two mL of methylene chloride was added to the test tube and mixed thoroughly. The
dissolved sample was then collected and filtered into a GC vial through a 0.45 µM Millipore
filter. The sample was immediately injected into the GC-MS system at 250º C in the split less
mode. The GC-MS system consisted of an Agilent 7890A Gas Chromatography and an Agilent
5975 Mass Selective Detector (Agilent Technologies, Wilmington, DE, USA). Chromatographic
separation was achieved on a high resolution GC HP-5ms capillary column (30 mm length ×
0.25 mm i. d. × 0.25 µm film thickness) from Agilent J&W (Agilent Technologies, Wilmington,
DE, USA). The GC oven temperature was programmed to increase from 60º C to 220º C at a
10

ACCEPTED MANUSCRIPT
rate of 10º C min-1 with initial and final hold times of 1 min, respectively. Helium was used as a
carrier gas at a flow rate of 0.75 mL min-1.
The mass spectrometer source temperature was set to 230º C. Mass spectrometric
detection was accomplished with the scanning mode first to scan for all components with the
mass range m/z 40 – 300 in the sample, and then Selective Ion Monitoring (SIM) mode was used
to detect the specific ions with m/z 88, 119, and 169 in the sample. The ionization energy was
70ev. The data from GC/MS were analyzed using Agilent MSD Chem. Station Data analysis
software.
11

ACCEPTED MANUSCRIPT
RESULTS AND DISCUSSION
DHDPS is a potential target for antibiotic or herbicide development. A thorough
understanding of the chemical mechanism of the enzyme would be an invaluable aid in this
endeavor. The most fundamental aspect for a study to elucidate the chemical mechanism of the
enzyme is to understand what reaction the enzyme catalyzes. The substrates of the DHDPS
reaction are pyruvate and L-ASA, but the identity of the reaction product has been less clear. In
early work on the enzyme it was proposed the product of the reaction is DHDP [7, 8]. Based
upon the more recent NMR studies of the reaction by Blickling et al. [9] and others [10, 11] it
has now generally been assumed that HTPA is the immediate reaction product of DHDPS [6, 7,
11, 12, 20, 21, 24, 25]. Blickling et al. [9] proposed that the immediate product of the reaction is
HTPA that subsequently converts non-enzymatically to DHDP. This proposal leads to several
apparent inconsistencies. Reddy et al [12] carried out an NMR study of the reaction of DHDPR,
that catalyzes the next step in the biosynthetic pathway after the DHDPS reaction, and the results
were consistent with DHDP as the substrate for DHDPR. The results indicated hydride transfer
from NAD(P)H is to C4 and proton addition to C5 of DHDP. Since the substrate for DHDPR
appears to be DHDP the simplest explanation is that DHDP is the product of the DHDPS
reaction. When DHDPR is used as a coupling enzyme for assaying DHDPS no lag is observed
in the assay. This observation suggests that, if HTPA is the product of the synthase reaction,
conversion of HTPA to DHDP is fast, at least on the time scale of the initial velocity assay.
Slow conversion of HTPA to DHDP could be expected to lead to a lag in the coupled reaction as
DHDP slowly builds-up in the assay. Blickling et al. proposed rapid non-enzymatic conversion
of HTPA to DHDP. However, in their NMR studies directed at identifying the reaction product
the same fast conversion of HTPA to DHDP would be expected and consequently in the NMR
one would expect to observe DHDP or at least a mixture of HTPA and DHDP. Blickling et al.
reported detection of HTPA only and is a major problem for the idea of fast conversion of HTPA
to DHDP. Another potential difficulty with the Blickling et al. study is the reported use of 200
mM Tris buffer in their NMR experiments carried out at pH 9. As shown in the current work
Tris reacts with ASA (see below) and would affect the NMR results. More recently it has been
proposed that HTPA is the product of the synthase reaction and that DHDPR uses HTPA as
substrate and catalyzes the hydrolysis of HTPA to DHDP subsequent to hydride transfer [10]. In
12

ACCEPTED MANUSCRIPT
consideration of the inconsistencies associated with the previous results, we chose to investigate
the identity of the product of the DHDPS reaction.
Reaction of DL-ASA with Tris. The effect of Tris buffer on the DHDPS reaction initial
rates at pH 8 are shown in Figure 1A. The control reaction in the absence of Tris gave an initial
rate of 7.1 µM/min. The other assays that were done in the presence of increasing Tris
concentrations and initiated by the addition of DL-ASA, showed similar initial rates as indicated
by the similar slopes of the lines near time zero. Although the initial rates were similar, distinct
and increasing curvature of the initial rates were observed for those determined in the presence
of increasing Tris concentrations. At 80 mM Tris the rate over the last two minutes decreased to
0.83 µM/min compared to the control rate in the absence of Tris of 4.9 µM/min over the same
time period. Lastly an initial rate was determined by pre-incubating DL-ASA with 80 mM Tris
in the assay for 5 minutes before initiating the reaction by addition of the remaining assay
components and is the uppermost initial rate in the figure. The initial rate for the assay is
0.93 µM/min and is similar to the final rate for the assay done in the presence of 80 mM Tris and
initiated by addition of L-ASA. Pre-incubation of the other assay components with 80 mM Tris
at pH 8 did not display a decrease in the initial rate (data not shown). Enzymatic rates
determined at pH 6.8 in the presence of 80 mM Tris were within error identical to rates
determined in the absence of Tris and no evidence of increased curvature with Tris present (not
shown).
The effect of pre-incubating DL-ASA with Tris at pH 8 followed by removal of an
aliquot to an end-point assay at pH 6.8 to determine the L-ASA present is shown in Figure 1B.
If an aliquot is removed at time zero, the end-point assay indicates about 35 µmoles of L-ASA is
present in the end-point assay. After a 5 minute incubation of DL-ASA with Tris, the amount of
L-ASA remaining was about 11 µmoles or 31 % of the time zero control. Additional aliquots
were removed from the pre-incubation mixture at five minutes and allowed to incubate with
buffer alone in the end-point assay at pH 6.8 for 20 or 80 minutes. After incubating for the
allotted time at pH 6.8 the remaining assay components were added to initiate the reaction. The
resulting end-point assays indicate the L-ASA can be recovered upon incubation at pH 6.8 with
about 85% of the L-ASA recovered after an 80 minute incubation.
13

ACCEPTED MANUSCRIPT
Tris buffer reacts with DL-ASA at pH 8 and the reaction is dependent on Tris
concentration. The pKa for Tris is about 8, and thus under the experimental conditions about
50% of the Tris present is in the basic form. At low pH DL-ASA shows little reactivity with Tris
and suggests it is the basic form of Tris that reacts with DL-ASA. The most likely possibility is
attack of the un-protonated primary amine of Tris on the aldehyde carbon of DL-ASA followed
by loss of water to give a Schiff base between Tris and DL-ASA. Imines such as these are
generally readily reversible in water and this is what is observed with the DL-ASA-Tris reaction
at pH 6.8. It is likely other accessible basic primary amines would lead to similar results. The
results indicate care must be exercised in the choice of assay buffer, enzyme storage buffer, and
other assay components that may contain primary amines. Reaction of an amine with L-ASA
could possibly result in artifacts in enzyme assays if not taken into consideration, especially
since the enzyme is most commonly assayed at the relatively high pH of 8.
Stability of DL-ASA. In order to investigate the stability of DL-ASA at different pH
values, a set of pre-incubation experiments were carried out at different pH values followed by a
removal of an aliquot to an L-ASA end-point assay at pH 6.8. The results are shown in Figure
2A. The control end-point assay indicated the presence of about 48 µmoles of L-ASA. After a
10-minute incubation of DL-ASA at the indicated pH values the amount of L-ASA decreased to
24, and 11 µmoles at pH 8.1, and 8.8, respectively. Since the rates in the end-point assays were
characterized by an initial burst followed by a slower final rate, attributed to slow hydrolysis of
the imine adduct to regenerate L-ASA, the amount of L-ASA present was determined by
extrapolation of the linear rate back to time zero to calculate the amount of L-ASA present
initially. At the highest pH tested only about 20% of the L-ASA present in the control was still
present. The slower linear rate after the initial burst suggests the reaction that reduces the
available L-ASA is reversible. To test this possibility a pre-incubation experiment was done by
incubating DL-ASA with buffer at pH 8.8 and removing aliquots to an L-ASA end-point assay to
determine the L-ASA remaining (Figure 2B). The control assay done at time zero indicated 45
µmoles of L-ASA present. After a ten minute incubation period at pH 8.8 and based upon the
size of the initial burst, the amount of L-ASA present in the end-point assay was decreased to
about 13 µmoles, indicating more than 70% of the L-ASA was not present compared to the
control. A second aliquot of ASA was removed at ten minutes and allowed to incubate for two
14

ACCEPTED MANUSCRIPT
hours with buffer alone at pH 6.8. The remaining assay components were added to initiate the
end-point assay. The amount of L-ASA was determined to be 39 µmoles after the two hour
incubation. The recovery of most of the L-ASA indicates the reaction can be reversed in the pre-
incubation mix at low pH.
The reaction of DL-ASA with itself shows similar characteristics as the reaction of DL-
ASA with Tris base. The reaction occurs at high pH and is reversible at low pH. A similar
explanation is likely for this reaction as for the reaction of ASA with Tris whereby at high pH the
α-amino group of DL-ASA could become deprotonated and attack the aldehyde of another
molecule of ASA as shown in Scheme 2. After loss of water there is the possibility of ring
closure as a result of nucleophilic attack of the deprotonated amine on the aldehyde of the second
attached ASA leading to an eight membered ring. The free amino group could also potentially
attack a third ASA molecule, although formation of the eight membered ring is favored
considering the intramolecular nature of this reaction. Intramolecular attack of the α-amine on
the β-carbonyl of ASA might be possible, but is unlikely given the bond strain associated with
the resulting 4-membered ring. There are a number of possible products for this reaction. There
are also a number of possible combinations of products since the ASA used in these studies is a
mixture of D-ASA and L-ASA. Consequently, there is the possibility for different isomeric
compositions of the reaction products. The reaction of ASA with itself is another potential
complicating factor that should be taken into consideration in studying DHDPS.
Proton NMR Studies of ASA and the DHDPS Reaction. To further examine the DL-
ASA self-reaction, proton NMR spectra were acquired at different pH values and are shown in
Figure 3. At the lowest pH value the major proton peaks are consistent with ASA and similar to
previous proton NMR spectra of the molecule [10]. The minor peaks in the spectrum are
associated with DL-2-amino-4-pentenoic acid, the un-reacted precursor of DL-ASA used in the
ozonolysis reaction to generate DL-ASA. The multiplets at about 2.0 and 2.5 ppm represent the
aliphatic protons on the β-carbon of DL-ASA. The peak at 3.7 ppm is assigned to the α-proton
and is shifted downfield due to the amino group on the α-carbon. The other major peak at 5.1
ppm is the proton associated with the diol form of the aldehyde. Another small peak is also
observed at 9.5 ppm (not shown) that represents the aldehyde proton. As the pH is raised new
peaks appear that are not present at low pH. The most prominent of these are a doublet of
15

ACCEPTED MANUSCRIPT
doublets at about 2.85 ppm and new multiplets at about 3.2, 3.4 and 3.6 ppm. The new peaks are
more pronounced as the pH is increased. The most interesting of the new peaks is the doublet of
doublets observed at 2.85 ppm. Previous proton NMR studies intended to identify the product of
the DHDPS reaction have assigned a peak with a similar chemical shift to the proposed product
HTPA [9-11]. These studies were all reportedly done at pH 8 or above, conditions where the
ASA self-reaction readily occurs and suggests that in the previous work the 2.85 ppm peak could
have resulted from the ASA non-enzymatic reaction rather than arising from the enzymatic
reaction.
To investigate this possibility, proton NMR spectra of the DHDPS reaction were acquired
at a lower pH and high enzyme concentrations to minimize the non-enzymatic ASA reaction.
The results are shown in Figure 4. Two reactions were done using either 3h-pyruvate or 3d-
pyruvate as substrate. As can be seen in Figure 4 the reaction containing 3d-pyruvate does not
display the proton peaks associated with 3h-pyruvate. For comparison an initial spectrum was
acquired prior to adding enzyme. Regardless of whether the enzymatic reaction used protonated
or deuterated pyruvate the number and chemical shifts of the new peaks that appear after adding
enzyme are the same. There are four new peaks including a doublet of doublets at 1.45 ppm, and
fairly broad peaks at 2.31, 3.9 and 4.12 ppm. The NMR spectra were collected at two minute
intervals (2 – 8 minutes) and all spectra were identical except for the size of the NMR peaks.
There is no evidence in either case for the appearance of a new peak at about 2.9 ppm or other
peaks associated with the non-enzymatic reaction of ASA. The gCOSY for the reaction started
with 3d-pyruvate is shown in Figure 5. The gCOSY results indicate coupling between the
protons located at 4.17 ppm and 3.9 ppm with the peak located at 2.31 ppm. The 2.31 ppm
proton is coupled to the proton at 1.5 ppm. The gCOSY for the 3h-pyruvate reaction (not
shown) displays an identical coupling pattern as the 3d-pyruvate reaction.
The NMR studies of DL-ASA are consistent with the pre-incubation and kinetic studies
of DL-ASA stability at different pH values. The appearance of new NMR peaks at high pH
confirms the generation of reaction products resulting from the reaction of DL-ASA with itself at
high pH. The most important of the DL-ASA product peaks is the doublet of doublets at 2.85
ppm since a similar doublet of doublets with a similar chemical shift has been assigned
previously to the DHDPS reaction product in experiments reportedly done at pH 8 or above. The
16

ACCEPTED MANUSCRIPT
2.85 ppm peak has been assigned to one of the protons on C5 of the assumed product HTPA. It
appears likely that in the previous studies the 2.85 ppm peak arises from the ASA non-enzymatic
reaction rather than being associated with the product. Consequently, the product of the DHDPS
reaction was studied by proton NMR at a pH and conditions that preliminary results suggested
would not lead to significant generation of the ASA self-reaction product.
In the NMR studies of the DHDPS reaction, the reaction was done using either 3h-
pyruvate or 3d-pyruvate. As shown in Figure 3 and 4, if the DHDPS reaction is done using 3h-
pyruvate, assuming HTPA is the product, this should result in protons being present at C5 of
HTPA since this carbon arises from pyruvate. The NMR spectrum should then have new
resonances at relatively high field for the protons at C3 and the protons at C5 of HTPA. One
could expect four different sets of peaks for these protons at similar up-field positions in the
spectra. On the other hand, if the reaction uses 3d-pyruvate, the two protons at C5 would be
replaced with deuterons that would be unobservable and thus the spectrum would show only the
C3 protons. The two spectra, started with either 3h-pyruvate or 3d-pyruvate, should be different.
However, the two spectra are identical and display only two sets of high field peaks in both
cases. This observation is inconsistent with HTPA as the reaction product.
However, the number and pattern of new peaks associated with the DHDPS reaction
started with either 3h-pyruvate or 3d-pyruvate is consistent with DHDP as the reaction product.
If DHDP is the reaction product and the reaction uses 3h-pyruvate as substrate one proton would
be predicted at C5 of the product. The doubly bonded C4 and C5 protons could have similar
chemical shifts and would be located relatively downfield. The proton at C1 would also be
shifted downfield and thus, along with the C4 and C5 protons, are likely to be the two peaks
located at about 4 ppm. The two protons on C3 would correspond to the NMR peaks at 2.31 and
1.5 ppm. If the reaction uses 3d-pyruvate, the proton at C5 would be replaced with a deuteron
and NMR silent, however, the proton on C4 would still be observable downfield. The remaining
protons will still be observed and consequently the two product spectra starting with either 3h-
pyruvate or 3d-pyruvate will be identical with regards to the number and position of the peaks.
The coupling of both peaks observed near 4 ppm with the 2.31 ppm peak and the coupling of the
2.31 peak with the 1.5 ppm peak is consistent with DHDP. The same coupling pattern would be
expected regardless of whether the reaction uses 3h-pyruvate or 3d-pyruvate and that is the
17

ACCEPTED MANUSCRIPT
observed result. The proton NMR results are most consistent with DHDP as the product of the
DHDPS reaction, but with one ambiguity, the chemical shift of the C4 and C5 protons on the
carbon-carbon double bond appear to be located at about 4 ppm and is further up-field than is
typically observed for protons attached to an internal carbon-carbon double bond. Given the
resulting uncertainty to the conclusions associated with the chemical shift of the C4 and C5
protons additional methods were chosen to investigate the identity of the enzymatic product.
UV-Spectra of the DHDPS Reaction Product. The product of the DHDPS reaction
was also examined by UV-spectroscopy. The DHDPS reaction was examined over time and is
shown in Figure 6. A tandem cuvette was prepared containing enzyme on one side and the
substrates pyruvate and L-ASA on the other. The cuvette was blanked in the diode array
spectrophotometer and then mixed to initiate the reaction. The first spectrum was acquired ten
second after mixing and eight more spectra were acquired at one minute intervals. The spectral
data below 200 nM is unreliable due to the high absorbance in this region resulting from
absorbance by the cuvette, substrates, and buffer. However there is a clear significant increase in
absorbance over time at about 200 nM as the reaction proceeds. In addition, a second broad
absorbance peak is observed with a λmax at about 254 nM. After the eight minute reaction
period, the mixture was allowed to incubate for an additional one hour and a final spectrum
acquired at this time. The one hour spectrum is different from those at the earlier time points and
shows increases in absorbance at about 230 and 270 nM. Subtracting the spectrum at eight
minutes from the one acquired at one hour yields the spectrum shown in the inset of Figure 6A.
Two peaks are observed with λmax values at about 225 and 271 nM. For comparison the
spectrum of dipicolinate is shown in Figure 6B. Dipicolinate displays significant absorbance at
about 200 nM with a longer wavelength shoulder. A second absorbance peak is seen at longer
wavelength with a λmax at ~271 nM and two shoulders associated with this peak at about 265 and
279 nM.
An investigation of the DHDPS reaction reveals at earlier time points spectra with
significant absorbance around 200 nM and a second broad spectral peak with a λmax at ~254 nM.
The proposed product HTPA has two carboxyl groups and a single double bond, both of which
would likely absorb below 200 nM. In contrast, DHDP has two conjugated double bonds within
the ring and this feature typically absorbs at longer wavelength due to the delocalization in the
18

ACCEPTED MANUSCRIPT
extended π system. The broad 254 nM peak would be consistent with the conjugated double
bonds. The presence of the conjugated double bonds in the more rigid planar cyclic system of
DHDP would improve the π system overlap and favor the lower energy absorption.
For comparison, the UV-spectrum of dipicolinate is shown in Figure 6B. Dipicolinate
displays a significant absorbance at ~200 nM with a longer wavelength shoulder and a second
peak that absorbs at 271 nM. The longer wavelength peak has two shoulders at 265 and 279 nM.
The more conjugated double bond system in dipicolinate would be expected to show a lower
energy peak than the 254 nM peak seen for DHDP. The difference spectrum shown in the inset
of Figure 6A has a lower energy absorption peak that is very similar to the lower energy peak
displayed by dipicolinate. It is unlikely the product of the DHDPS reaction converts to
dipicolinate, but does suggest conversion of the enzymatic product to a more conjugated system.
There seems to be two likely ways in which DHDP could react. Hydrolysis of the imine linkage
between N1 and C6 would open the ring to generate a carbon-oxygen double bond at C6 and a
free amino group at C2 of the straight chain product. A second possibility could be
deprotonation at C3 of DHDP. A protonated imine nitrogen of DHDP could serve as an electron
sink which would stabilize the anion formed if deprotonation occurs at C3. The resulting anion
would lead to a more conjugated π system that would be expected to absorb at a longer
wavelength than the parent compound. The longer wavelength peak shown in the inset of Figure
6A could represent the proposed anion. Hydrolysis and deprotonation could both occur during
the long term incubation of the product and the resulting spectrum would be the combination of
the two. The important point for the current work is the direct observation of the breakdown of
the DHDPS product over time.
Mass of the DHDPS Product by GC-MS. The DHDPS reaction was examined by GC-
MS in order to determine the mass of the product. A reaction mixture was made up containing 7
mM pyruvate, 10 mM DL-ASA (5 mM L-ASA), and 1.7 µM DHDPS in 10 mL. Since the
product of the DHDPS reaction is labile, sodium borohydride was added after an eight minute
reaction time to reduce the carbon nitrogen double bond in the product. The borohydride could
also be expected to reduce the carbon-oxygen double bond present in the substrates ASA and
pyruvate. For the substrates, pyruvate would be converted to lactate and ASA to homoserine.
The potential product DHDP would be converted to 1,2,3,6-tetrahydropyridine-2,6-dicarboxylate
19

ACCEPTED MANUSCRIPT
and HTPA to 4-hydroxypiperidine-2,6-dicarboxylate. The GC-MS results are shown in Figure 7.
As shown in Figure 7A a GC peak is observed at a retention time of 10.7 minutes. The masses
of the molecules associated with the GC peak are indicated in Figure 7B. Three peaks are
observed with m/z values of 88, 119 and 169.
Since the UV-spectral studies indicates the product of the DHDPS reaction is unstable a
method was needed to stabilize the product for the GC-MS analysis. Borohydride reduction of
the imine linkage in the product was chosen to stabilize the product. Borohydride would reduce
the imine in the product to an amine and also reduce the carbonyl functional groups in the
substrates pyruvate and ASA. The enzymatic reaction for GC-MS analysis was allowed to
proceed long enough to generate sufficient product, but not so long as to generate significant
breakdown products. After borohydride reduction the expected products are lactate, (m/z 89.07),
homoserine (m/z 119.12), and the reduced form of either DHDPS (m/z 169.14) or HTPA (m/z
187.5) as shown in Scheme 5. The GC-MS results yield three compounds with m/z values of 88,
119, and 169. These are consistent with pyruvate, homoserine, and the reduced form of DHDP.
There is no indication of an m/z peak at ~187 that would represent the reduced form of HTPA.
The GC-MS results are consistent with the NMR and UV-spectral results, all of which support
the conclusion that DHDP is the product of the DHDPS reaction.
Chemical Mechanism of DHDPS. The first half-reaction of the DHDPS reaction is
shown in Scheme 3. Based upon crystal structures of the enzyme [14], it appears that Y133 is
the only active site residue that is correctly positioned to donate a proton to the carbonyl oxygen
of pyruvate in the formation of the Schiff base between pyruvate and K161. Y133 would then be
in the correct protonation state to accept a proton from the methyl group of pyruvate to generate
the enamine intermediate. In the second half-reaction (Scheme 4) Y133 most likely would play a
similar role by donating a proton to the aldehyde oxygen in the enamine attack on L-ASA to
generate the intermediate shown in (IIA) or (IIB). This is where the two alternative mechanisms
diverge. In the proposed formation of HTPA, the α-amino group of L-ASA would need to be in
position to attack the Schiff base between the intermediate and K161 and is shown that way in
Scheme 4. A crystal structure of the enzyme with succinic-semi-aldehyde (SSA) covalently
bound to the enzyme:pyruvate complex is likely a reasonable model for the intermediate [26].
The structure shows that the α-carboxyl group of SSA interacts with R138 and N248 and is
20

ACCEPTED MANUSCRIPT
bound in an extended conformation that would place a modeled in α-amino group, as in L-ASA,
at least 4.5 Å away from and poorly oriented for attack on the Schiff base. L-ASA likely makes
a similar interaction with R138 and, even if bound somewhat differently than SSA, would still
not place the α-amino group in position to attack the Schiff base in an extended configuration.
The question becomes how to get the α-amine in position to attack the Schiff base in what
appears to be a stable complex. There is no apparent driving force to accomplish the movement
necessary to generate HTPA. On the other hand, dehydration and generation of a trans-double
bond as shown in IIB would provide the conformational change needed in the intermediate to
bring the α-amine in position to attack the Schiff base. Since one end of the intermediate is
tethered to the enzyme by the Schiff base, the planar double bond should shorten the
intermediate sufficiently to disrupt the interaction between the α-carboxyl and R138 allowing for
the α−amine to swing into position and initiate the transimination reaction to generate DHDP.
Conclusions. The preponderance of evidence supports the conclusion that DHDP is the
direct product of the DHDPS catalyzed reaction. Previous NMR studies of the DHDPR reaction
provides evidence that DHDP is the substrate for the reductase and suggests the product of the
synthase reaction is DHDP. The kinetic behavior of the coupled enzyme assay using DHDPR
supports the conclusion that DHDP is the product of the synthase reaction. The recognition that
one or more of the NMR peaks previously assigned to the product of the DHDPS reaction,
presumed to be HTPA, actually arises from a non-enzymatic reaction. The proton NMR studies
in the current work using either protonated or deuterated pyruvate are most consistent with
DHDP as the DHDPS reaction product. The UV-spectral studies of the DHDPS reaction support
a conjugated product for the reaction consistent with DHDP and the GC-MS results provide
further evidence for DHDP as the DHDPS reaction product. A knowledge of the reaction
catalyzed by an enzyme is fundamental in order to properly interpret kinetic and mechanistic
studies carried out with the enzyme. In addition, the design of inhibitors for an enzyme such as
DHDPS that is a potential target for antibiotic development would be greatly enhanced by the
identification of intermediates along the reaction pathway, something only possible with the
knowledge of the overall reaction catalyzed by the enzyme.
21

ACCEPTED MANUSCRIPT
ACKNOWLEDGMENTS
The authors wish to thank Dr. Paul Sims for use of his instruments and lab space
to carry out these studies and Dr. Paul F. Cook for a critical reading of the manuscript.
22

ACCEPTED MANUSCRIPT
REFERENCES
[1] D. Meadow, D. S. Hoare, E. Work, Interrelationship between lysine and α, ε-
diaminopimelic acid and their derivatives and analogues in mutants of Escherichia coli,
Biochem. J. 66 (1957) 270-282.
[2] L. E. Rhuland, Role of α, ε-diaminopimelic acid in the cellular integrity of Escherichia
coli, J. Bacteriol. 73 (1957) 778-783.
[3] C. V. Coulter, J. A.Gerrard, J. A. E. Kraunsoe, A. J. Pratt, Escherichia coli
dihydrodipicolinate synthase and dihydrodipicolinate reductase: kinetic and inhibition studies
of two putative herbicide targets, Pestic. Sci. 55 (1999) 887-895.
[4] R. J. Cox, A. Sutherland, J. C. Vederas, Bacterial diaminopimelate metabolism as a
target for antibiotic design, Bioorg. Med. Chem. 8 (2000) 8843–871.
[5] C. A. Hutton, T. J. Southwood, J. J. Turner, Inhibitors of lysine biosynthesis
as antibacterial agents, Mini ReV. Med. Chem. 3 (2003) 115-127.
[6] R. C. J. Dobson, J. A. Gerrard, and F. G. Pearce, Dihydrodipicolinate synthase is
not inhibited by its substrate S-aspartate-β-semialdehyde, Biochem. J. 377 (2004) 757-762.
[7] Y. Yugari, C. Gilvarg, The condensation step in diaminopimelate synthesis, J. Biol.
Chem. 240 (1965) 4710-4716.
[8] J. G. Shedlarski, C. Gilvarg, The pyruvate-aspartic semialdehyde condensing enzyme of
Escherichia coli, J. Biol. Chem. 245 (1970) 1362-1373.
[9] S. Blickling, C. Renner, B. Laber, H.-D. Pohlenz, T. A. Holak, R. Huber, Reaction
mechanism of Escherichia coli dihydrodipicolinate synthase investigated by X-ray
crystallography and NMR spectroscopy, Biochemistry 36 (1997) 24-33.
23

ACCEPTED MANUSCRIPT
[10] S. R. A. Devenish, J. W. Blunt, J. A. Gerrard, NMR studies uncover alternate substrates
for dihydrodipicolinate synthase and suggest that dihydrodipicolinate reductase is also a
dehydratase, J. Med. Chem. 53 (2010) 4808-4812.
[11] T. P. Soares da Costa, A. C. Muscroft-Taylor, R. C. J. Dobson, S. R. A. Devenish, G.
B. Jameson, and J. A. Gerrard, How essential is the ‘essential’ active-site lysine in
dihydrodipicolinate synthase? Biochimie 9 (2010) 837-845.
[12] S. G. Reddy, J. C. Sacchettini, and J. S. Blanchard, Expression, purification, and
characterization of Escherichia coli dihydrodipicolinate reductase, Biochemistry 34 (1995)
3492-3501.
[13] W. E. Karsten, Dihydrodipicolinate synthase from Escherichia coli: pH Dependent
changes in the kinetic mechanism and kinetic mechanism of allosteric inhibition by L-
lysine, Biochemistry 36 (1997) 1730-1739
[14] S. R. A. Devenish, J. A. Gerrard, G. B. Jameson, R. C. J. Dobson, The high-resolution
structure of dihydrodipicolinate synthase from Escherichia coli bound to its first substrate,
pyruvate, Acta Cryst F64 (2008) 1092-1095.
[15] S. Black, N. G. Wright, Aspartic β-semialdehyde dehydrogenase and aspartic β-
semialdehyde, J. Biol. Chem. 213 (1955) 39-50.
[16] M. M. Bradford, A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding, Anal. Biochem. 72
(1976) 248-254.
[17] K. B. Schowen, R. L. Schowen, Solvent isotope effects on enzyme systems, Methods
Enzymol. 87 (1982) 551-606.
24

ACCEPTED MANUSCRIPT
[18] R. C. J. Dobson, K. Valegard, J. A. Gerrard, The crystal structure of three site-directed
mutants of Escherichia coli dihydrodipicolinte synthase: Further evidence for a catalytic
triad, J. Mol. Biol. 338 (2004) 329-339.
[19] R. C. J. Dobson, M. D. W. Griffin, S. R. A. Devenish, F. G. Pearce, C. A. Hutton, J. A.
Gerrard, G. B. Jameson, Conserved main-chain peptide distortions: A proposed role for
Ile203 in catalysis by dihydrodipicolinate synthase, Protein Science 17 (2008) 2080-2090.
[20] R. C. J. Dobson, M. A. Perugini, G. B. Jameson, J. A. Gerrard, Specificity versus
catalytic potency: The role of threonine 44 in Escherichia coli dihydrodipicolinate synthase
mediated catalysis, Biochimie 91 (2009) 1036-1044.
[21] A. C. Muscroft-Taylor, T. P. Soares da Costa, J. A. Gerrard, New insights into the
mechanism of dihydrodipicolinate synthase using isothermal titration calorimetry,
Biochimie 92 (2010) 254-262.
[22] L. Shames, F. C. Wedler, Homoserine kinase of Escherichia coli. Kinetic mechanism
and inhibition by L-aspartate semialdehyde, Arch. Biochem. Biophys. 235 (1984) 359-370.
[23] C. P. Phenix, D. R. J. Palmer, Isothermal titration microcalorimetry reveals the
cooperative and noncompetitive nature of inhibition of Sinorhizobium meliloti L5-30
dihydrodipicolinate synthase by (s)-lysine, Biochemistry 47, (2008) 7779-7781.
[24] C. J. T. Conly, Y. V. Skowpen, S. Li, D. R. J. Palmer, D. A. R. Sanders, Tyrosine 110
plays a critical role in regulating the allosteric inhibition of Campylobacter jejuni
dihydrodipicolinate synthase by lysine, Biochemistry 53 (2014) 7396-7406.
[25] P. Shrivastava, V. Navratna, Y. Silla R. P. Dewangan, A. Pramanik, S. Chaudhary, G.
Rayasam, A. Kumar, B. Gopal, S. Ramachandran, Inhibition of Mycobacterium tuberculosis
dihydrodipicolinate synthase by alpha-ketopimelic acid and its other structural analogues,
Sci. Rep. 6 (2016) 30827; doi:10.1038/srep30827.
25

ACCEPTED MANUSCRIPT
[26] B. A. Boughton, R. C. J. Dobson, C. A. Hutton, The crystal structure of
dihydrodipicolinate synthase from Escherichia coli with bound pyruvate and succinic
semialdehyde: Unambiguous resolution of the stereochemistry of the condensation product,
Proteins 80 (2012) 2117-2122.
26

ACCEPTED MANUSCRIPT
Scheme 1. Proposed Reaction Products of DHDPS
27

ACCEPTED MANUSCRIPT
Scheme 2. Proposed ASA Reaction Products
28

ACCEPTED MANUSCRIPT
Scheme 3. First Half-Reaction for DHDPS
29